- UW Department of Geology and Geophysics
Proterozoic metamorphic and igneous rocks in the northern Colorado Front Range display evidence for significant but localized 1.4-Ga deformation. Initial metamorphism and deformation occurred during crustal assembly ∼1.7 Ga, and the area was subsequently affected by widespread “anorogenic” granitic plutonism at ∼1.4 Ga. Although there is little evidence for penetrative 1.4-Ga deformation in northern Colorado, data from the northeast-striking Moose Mountain shear zone indicate localized 1.4-Ga contractional strain. The shear zone deforms both pre-1.70-Ga supracrustal rocks and the ∼1.4-Ga St. Vrain granite. Kinematic indicators within the deformed granite show south-side-up reverse motion. A continuum from magmatic to solid-state mylonitic fabrics indicates that deformation occurred during the emplacement and cooling of the granite at ∼1.4 Ga. Thrust-sense deformation, coupled with synchronous northeast-southwest extension recorded in dike swarms, is consistent with a model for regional northwest-directed contractional deformation during emplacement of the 1.4-Ga plutonic suite. Paleoproterozoic supracrustal rocks within the shear zone show evidence for the 1.4-Ga shearing as well as for an older phase of deformation apparently dominated by sinistral motion. Supracrustal rocks on opposite sides of the shear zone record different structural histories, and detrital zircon samples differ dramatically in age populations across the shear zone. These data lead us to suggest that the shear zone separates two unrelated packages of rocks that were juxtaposed against one another during ∼1.7-Ga assembly of the region. Thus, original assembly of this region around 1.7 Ga probably involved previously undocumented transcurrent movements that juxtaposed “terranes” with differing Paleoproterozoic histories and resulted in large-scale zones of crustal weakness that localized subsequent deformation.
A suite of ∼1.4-Ga granitic to anorthositic intrusions extends across North America from southern California to Labrador (e.g., Anderson, 1983). These plutons perforate crust that was added to the craton 300–400 m.y. earlier, but the tectonic setting at the time of this widespread magmatic event is poorly understood. Explanations for the 1.4-Ga magmatic event must account for two seemingly opposing sets of observations: (1) the intracontinental setting, mantle-derived geochemical signatures, and apparent absence of penetrative deformational fabrics are consistent with an “anorogenic” origin, in which supercontinent breakup and/or regional extension caused mantle upwelling and anatectic melting of underplated mafic material (e.g., Anderson, 1983; Hoffman, 1989; Windley, 1993; Frost and Frost, 1997; Frost et al., 1999); and (2) detailed structural studies in and around several 1.4-Ga plutons in Colorado, Arizona, New Mexico, Nevada, and Ontario indicate that northwest-directed contractional deformation was associated with the emplacement and cooling of the plutons (Vernon, 1987; Graubard and Mattinson, 1990; Gonzales et al., 1994; Nyman et al., 1994; Duebendorfer and Christensen, 1995; Nyman and Karlstrom, 1997; Fueten and Redmond, 1997). Nyman et al. (1994) proposed that this contractional deformation was a regional event recording an inboard tectonic response to subduction or transpressional stresses along the southeastern margin of Laurentia. Any viable model for the 1.4-Ga thermal event must account for both the geochemical data and the structural setting of the plutons.
In the western United States, many of the 1.4-Ga granitoid bodies are spatially associated with northeast-striking shear zones. Development of major shear zones at 1.4 Ga would imply an orogenic setting, particularly if there also is evidence of regional fabric development at 1.4 Ga. In contrast, some or all of these shear zones may represent major structures that were active during initial ∼1.8–1.7-Ga amalgamation of the Proterozoic continent (Bickford, 1988; Condie, 1992; Bowring and Karlstrom, 1990; Van Schmus et al., 1993) and were subsequently reactivated during 1.4-Ga magmatism. For example, Graubard and Mattinson (1990), Aleinikoff, Reed, and DeWitt (1993), and Nyman et al. (1994) documented emplacement of the 1442-Ma Mt. Evans batholith in central Colorado during reversesense reactivation of the preexisting Idaho Springs–Ralston shear zone. Emplacement of the 1.4-Ga granites along preexisting structures could have resulted in shear zone reactivation at relatively low stresses and need not imply regional orogeny. Detailed analysis of 1.4-Ga shear zones and rocks affected by 1.4-Ga heating is thus necessary to assess whether 1.4-Ga deformation was penetrative, localized along new shear zones, or localized along preexisting zones of weakness. Each of these alternatives has different implications for the setting of 1.4-Ga magmatism. Documentation of earlier motion histories on shear zones associated with the 1.4-Ga granites also has implications for the models of Paleoproterozoic continental accretion.
We focus in this paper on the Moose Mountain shear zone and adjacent country rocks in the northern Colorado Front Range (Figs. 1 and 2). The Moose Mountain shear zone is located about 10 km south of Big Thompson Canyon, between the Buckhorn Creek and Skin Gulch shear zones to the north and the Idaho Springs–Ralston shear zone to the south. The shear zone deforms rocks of the ∼1.4-Ga Longs Peak–St. Vrain batholith and juxtaposes high-temperature supracrustal rocks to the south against low- to medium-temperature rocks to the north. The presence of multiple shear fabrics in pre-St. Vrain rocks indicates that the Moose Mountain shear zone also had a significant history prior to emplacement of the granite. We report on: (1) indications that the Moose Mountain shear zone was a major tectonic boundary during initial assembly of the region prior to 1.70 Ga; (2) reactivation of the shear zone during ∼1.4-Ga granite emplacement; (3) the lack of regional fabrics associated with widespread 1.4-Ga reheating of the area; and (4) strain axes during ∼1.4-Ga magmatism. Our observations are consistent with models for regional northwest–southeast contraction in the Mesoproterozoic, but only if most of the contractional strain was concentrated along preexisting zones of crustal weakness. The fabric anisotropy developed by earlier movement on the proto-Moose Mountain shear zone probably facilitated migration of the 1.4-Ga granitic magma to middle and upper crustal levels, and heat from the magma likely facilitated reactivation of the shear zone (e.g., Davidson et al., 1992; D'Lemos et al., 1997; Hanmer, 1997).
Proterozoic Rock Types
Rocks presently exposed within the Mesozoic and Cenozoic uplifts of the northern Front Range belong to the Proterozoic Colorado province (Tweto, 1987; Bickford, 1988; Reed et al., 1993), which is directly juxtaposed against the Archean Wyoming province along the Cheyenne belt suture in southern Wyoming (Karlstrom and Houston, 1984; Fig. 1). Paleoproterozoic schists and gneisses within the Front Range were affected by multiple phases of deformation, metamorphic recrystallization, and plutonism (e.g., Braddock and Cole, 1979) that are presumed to be related to assembly of the Colorado province and collision with the Wyoming craton between 1.75 and 1.70 Ga (e.g., Reed et al., 1987, 1993; Houston et al., 1989; Aleinikoff, Reed, and Wooden, 1993b; Chamberlain et al., 1993; Chamberlain, 1998). The details of this assembly and collision are poorly documented in Colorado, however, particularly in comparison with data from Wyoming (e.g., Houston et al., 1989; Premo and Van Schmus, 1989; Chamberlain et al., 1993; Scoates and Chamberlain, 1997) and Arizona and New Mexico (Karlstrom and Bowring, 1988; Bowring and Karlstrom, 1990).
The Moose Mountain shear zone (Figs. 1 and 2) is one of a number of Proterozoic shear zones that are subparallel to the Cheyenne belt. North of the shear zone, a sequence of quartz-rich to semi-pelitic metasedimentary rocks with scattered layers of amphibolite and calcsilicate rock is exposed in and adjacent to Big Thompson Canyon; graded bedding is well-preserved in the metasedimentary rocks at all metamorphic grades. Based on the geochemical signature of these metasedimentary rocks, Condie and Martell (1983) argued that the sequence represents mature sediments deposited in a forearc setting. Reed et al. (1987) noted that the Big Thompson rocks are located between two ∼1.8–1.7-Ga magmatic arc systems, and suggested that these rocks were most probably deposited in a back-arc setting. Detrital zircon ages place a maximum age of 1758 ± 26 Ma on deposition of the Big Thompson sequence (this study, see below). Metasedimentary sequences south of the Moose Mountain shear zone contain a higher overall percentage of intercalated metavolcanic material and lack the well-preserved primary sedimentary structures that characterize the Big Thompson Canyon sequence (e.g., Gable and Sims, 1969; Gable, 1996).
The entire northern Front Range was intruded by calcalkaline plutons of the Routt plutonic suite (Tweto, 1987) 25–50 m.y. after deposition of the sediments. In the area of this study, the Routt suite includes granodioritic and quartz monzonitic phases of the Boulder Creek batholith and related plutons, which have been dated by sensitive high-resolution ion microprobe (SHRIMP) analysis of zircons at 1717 ± 5 Ma (Premo and Fanning, 2000). These plutons are variably foliated, and generally they are interpreted to be syntectonic with respect to closing of a back-arc basin between the Cheyenne belt and the southern part of the Colorado province (Reed et al., 1987; Premo and Fanning, 1997). Within an area bounded to the north by the Buckhorn Creek shear zone and to the south by the Moose Mountain shear zone, the supracrustal succession also was intruded by sills and small plutons of trondhjemite at 1726 ± 15 Ma (U-Pb zircon age; Barovich, 1986; Fig. 2). Trondhjemitic intrusions are unknown from other parts of the Front Range.
Emplacement of the 1.7-Ga Routt suite was followed by a hiatus in igneous activity until the Mesoproterozoic. During the interval from 1.5 to 1.3 Ga, voluminous granitic plutons perforated the preexisting Proterozoic crust along a belt extending from California to eastern Canada (e.g., Bickford and Anderson, 1993). Within the region of this study, reported crystallization ages of the Mesoproterozoic granites are in the range of 1.42–1.44 Ga and include, from north to south, the Sherman Granite (1433 ± 1.5–1437.8 ± 3.2 Ma, U-Pb zircon; Frost et al., 1999), Log Cabin batholith (1390 ± 30 Ma, Rb-Sr; Peterman et al., 1968), Longs Peak–St. Vrain batholith (1423 ± 30 Ma, Rb-Sr; recalculated from value of 1450 ± 30 Ma given by Peterman et al., 1968), Silver Plume batholith (1422 ± 3 Ma, U-Pb zircon; Graubard adn Mattinson, 1990), and Mt. Evans pluton (1442 ± 2 Ma, U-Pb zircon; Aleinikoff, Reed, and DeWitt, 1993). All of these granitic bodies contain alkali-feldspar phenocrysts up to several centimeters long that locally define a trachytic flow foliation. These are the granites that Anderson (1983) and Anderson and Cullers (1999) characterized as “anorogenic.”
We focus in particular on the Longs Peak–St. Vrain granite (hereafter referred to as the St. Vrain granite), which is spatially associated with the Moose Mountain shear zone along its northeastern margin (Fig. 2). The St. Vrain granite is a porphyritic two-mica granite with a prominent flow foliation (Cole, 1977). Geochemical and Sm-Nd isotopic studies indicate that the source material for this granite probably was peraluminous metasedimentary crust high in potassium and enriched in incompatible elements (DePaolo, 1981; Anderson and Morrison, 1992). Anderson and Thomas (1985) argued that the St. Vrain magma was hot and dry, and it migrated from a melt site at 25–37 km depth to emplacement conditions of 9–11 km and 740–760° C.
Mesoproterozoic igneous activity also included emplacement of a swarm of basalt, andesite, and dacite dikes that crop out from southern Wyoming south to the Moose Mountain shear zone (Fig. 2). Individual dikes are typically a few meters wide and up to several kilometers long; they generally trend N10° W to N20° W and have subvertical dips (Nutalaya, 1966; Punongbayan, 1972; Tweto, 1987). The dikes are not controlled noticeably by preexisting structure in the country rocks, and their orientation probably was governed by the regional stress field at the time of emplacement (Cole, 1977). These dikes crosscut the main granite of the Sherman batholith, but they are themselves largely cut by Silver Plume-type granites of the Log Cabin batholith (Peterman et al., 1968). Within the Rustic quadrangle, however, mafic dikes appear to have intruded while the Log Cabin granite was still molten (Shaver et al., 1988). These cross-cutting relationships suggest dike emplacement between 1433 ± 1.5 and 1390 ± 30 Ma (Rb-Sr age of Log Cabin granites) or 1422 ± 3 Ma (U-Pb zircon age of Silver Plume granite; see above).
Rocks of the Big Thompson Canyon area exhibit three stages of folding with development of associated axial surface foliations and/or crenulation cleavages. Graded bedding, which is preserved locally from low grades up to the migmatite zone, allows identification of large-scale F1 folds that are east- to northeast-trending, isoclinal, and overturned to the north (Gawarecki, 1963; Braddock and Cole, 1979). A strong axial surface cleavage (S1) generally is subparallel to bedding (S0). F2 folds are northeast-trending, tight, upright to shallowly northeast-plunging folds with an axial surface crenulation cleavage (S2) that is subparallel to S1 at most localities (Gawarecki, 1963; Nutalaya, 1966; Punongbayan, 1972). F3 folds are northwest-trending, shallowly to steeply northwest-plunging with axial surface cleavage or crenulation cleavage (S3) that is locally well-developed in micaceous horizons (Gawarecki, 1963; Nutalaya, 1966; Punongbayan, 1972). At some localities, S3 is the dominant fabric and has obliterated traces of the earlier fabrics except within porphyroblast inclusion trails. The timing of development of the S3 cleavage is interpreted to be syn- to post-intrusion of the 1726 ± 15-Ma trondhjemite of Big Thompson Canyon (Barovich, 1986). The F1 and F2 folding events are restricted to a narrow time interval between the 1758 ± 26 Ma deposition of the Big Thompson sediments (age of the youngest detrital zircon, see below) and F3 folding around 1726 ± 15 Ma. There is no evidence for further deformation in the area until movement on the Moose Mountain shear zone, which is described in detail below.
The deformational history of foliated metasedimentary rocks immediately south of the Moose Mountain shear zone is not as well-known, owing in large part to the presence of large bodies of 1.4- and 1.7-Ga granitic rocks. Approximately 30 km south of the Moose Mountain shear zone, Tweto and Sims (1963) and Gable and Sims (1969) recognized two phases of regional folding that were pre-to syn-emplacement of plutonic rocks of the Routt suite. Neither of these episodes of folding corresponds in orientation or style to the structures preserved in Big Thompson Canyon. Mapping by Punongbayan et al. (1989) indicates a sharp discordance in foliation orientations across the Moose Mountain shear zone. Mapping also shows intersecting northeast-striking S2 and northwest-striking S3 fabrics north of the shear zone juxtaposed against rocks with only a single foliation that is folded around broad, steeply east-plunging axes just south of the shear zone. The Moose Mountain shear zone thus appears to separate regions that had different Paleoproterozoic structural histories.
Metamorphic isograds mapped in the supracrustal rocks north of the Moose Mountain shear zone define an apparent east-plunging synformal structure that is discordant to the regional fold pattern of the rocks (Fig. 2; Hutchinson and Braddock, 1987). The metamorphic grade ranges from biotite zone adjacent to the unconformity with Paleozoic sediments at the range front to second-sillimanite zone migmatites to the north, west, and south. The isograd pattern was interpreted originally as representing peak metamorphism associated in time with deformation and intrusion of trondhjemite ∼1.72 Ga (Barovich, 1986; Hutchinson and Braddock, 1987). Early interpretations (e.g., Gawarecki, 1963; Nutalaya, 1966) suggested that metamorphism of the supracrustal sequence defined a low-pressure, high-temperature facies series with biotite, garnet, staurolite, cordierite, andalusite, and sillimanite occurring as porphyroblasts at different grades. However, studies by Hodgins (1997), Selverstone et al. (1997), and Shaw et al. (1999) document a complex polymetamorphic history involving early metamorphism at 7–11 kbar, followed by synkinematic decompression to ∼4 kbar by 1.72 Ga.
The initial prograde metamorphic assemblages were replaced locally by chlorite + white mica ± cordierite pseudomorphs during post-deformational fluid-rock interaction; subsequent growth of euhedral garnet and staurolite crystals in some of these pseudomorphs provides evidence for a younger heating event. Partially reset 40Ar/39Ar age spectra from hornblende and completely reset spectra from muscovite and biotite indicate that this reheating event occurred around 1.4 Ga (Shaw et al., 1999). Modeling of the 40Ar/39Ar data is most consistent with reheating to a maximum temperature of 525–600° C for a short period (< 20 m.y.) followed by relatively rapid cooling (Shaw et al., 1999). Although other authors (e.g., Hodges et al., 1994) have suggested that the 1.4-Ga heating event was followed by very slow cooling, identical muscovite and biotite 40Ar/39Ar ages in the area of this study are more consistent with rapid cooling (Shaw et al., 1999).
The metamorphic mineral development in this region is thus interpreted to result from a combination of 1.7- and 1.4-Ga metamorphism. Neither the isograd pattern (Hutchinson and Braddock, 1987) nor the locations at which ∼1.4-Ga resetting of Ar age spectra occurred (Shaw et al., 1999) are spatially related to exposed 1.4-Ga plutons. It is therefore probable that the entire region experienced a widespread but short-lived mid-crustal heating event ∼1.4 Ga. Metamorphic recrystallization, however, occurred only at localities at which mineral nucleation was facilitated by earlier retrogression of the rocks.
Metamorphism in the region south of the Moose Mountain shear zone was described by Gable and Sims (1969) before many dates were available for the Colorado province. However, based on the detailed fabric and mineral associations described by Gable and Sims (1969), a Paleoproterozoic heating ± decompression path from garnet + staurolite to sillimanite + K-feldspar ± cordierite ± spinel zones can be reconstructed. Some samples show further decompression into the andalusite field, but it is unclear whether the andalusite is 1.7 or 1.4 Ga in age.
REGIONAL MESOPROTEROZOIC FABRICS
There is little evidence for ∼1.4-Ga penetrative deformation outside of the northeast-striking shear zones. Minerals such as staurolite and garnet, which grew during ∼1.7-Ga regional metamorphism (Hutchinson and Braddock, 1987), typically were partially to completely pseudomorphed by finegrained aggregates of white mica + chlorite during or prior to ∼1.4-Ga crustal reheating (Selverstone et al., 1997; Shaw et al., 1999). Conversion of the relatively strong original minerals into sheet silicates should have promoted overall weakening of the supracrustal succession and facilitated strain localization during orogenesis (e.g., Passchier and Trouw, 1996). Thus, any regional deformation during or following the 1.4-Ga heating should be readily apparent in rocks containing abundant pseudomorphs. However, our examination of more than 250 thin sections from throughout the region shown in Figure 2 revealed only two samples in which post-hydration deformation is apparent. In the vast majority of cases, mica and chlorite in the pseudomorphs randomly overprint the 1726 ± 15-Ma S3 foliation; the two exceptions are closely associated with shear zones (Moose Mountain and Buckhorn Creek) and show development of a new foliation in the pseudomorphs that is similar in orientation to that of the shear zones. Thus, there is little evidence for pervasive deformation of the region ∼1.4 Ga; strain of this age appears to be localized in shear zones.
MOOSE MOUNTAIN SHEAR ZONE
The Moose Mountain shear zone is northeast-striking, 1–2 km wide, and exposed for > 15 km southwest of the unconformity with Paleozoic sediments at the range front (Figs. 2 and 3). The shear zone is characterized by an anastomosing mylonitic foliation with an average present-day orientation of N55° E, 65° SE. At its northeastern end, the shear zone deforms Paleoproterozoic amphibolite and schist and locally contains a zone of tectonic mélange (Fig. 3). To the southwest, the Paleoproterozoic supracrustal succession is absent, and the shear zone is confined to the 1.4-Ga St. Vrain granite. The mylonitic fabric feathers out to the southwest within the granite, and the shear zone as a whole becomes difficult to trace southwest of ∼105° 30'W (J. Cole, personal communication, 2000). The northern margin of the Moose Mountain shear zone is marked by a zone of brittle deformation related to the Laramide-age Skinner Gulch and Rattlesnake Park faults (Fig. 3). The southern boundary of the zone is gradational, with mylonite and ultramylonite horizons becoming progressively less abundant in that direction.
Two distinct lineations are present within the shear zone: (1) a pervasive stretching lineation that plunges steeply to the southwest; and (2) a shallowly east-plunging mineral/stretching lineation that is locally present. The steep lineation occurs in all rock types affected by the shear zone, whereas the shallow lineation has been observed only within the supracrustal succession present at the northeastern end of the exposed shear zone (Fig. 4). We thus infer that the shallow lineation is the older of the two and that it developed prior to emplacement of the St. Vrain granite. This lineation is referred to below as L1. The steeply plunging lineation is clearly the same age as, or younger than, emplacement of the St. Vrain granite and is referred to as L2. Neither lineation is present in rocks outside the limits of the shear zone.
Rock Types Affected
St. Vrain Batholith
The Moose Mountain shear zone is most easily recognized within the 1.4-Ga St. Vrain granite. This megacrystic granite has a prominent flow foliation that is overprinted by mylonitic to ultramylonitic fabrics within the shear zone (Figs. 5A and 5B). Significant grain-size reduction is evident in the field along with asymmetric clasts and S-C fabrics. Pseudotachylite and microbreccia are present locally. The L2 stretching lineation is well-developed in the granite and has an average plunge of 62° toward S13° W. If the ∼20° E dip of the adjacent Paleozoic unconformity is back-corrected to horizontal, the L2 lineation plunge is essentially downdip (corrected orientation of shear zone: N63° E, 60° SE; corrected L2 lineation: 59° toward S24° E; Fig. 4).
Amphibolite, Quartzite, and Schist
Mafic amphibolite in the northeasternmost exposures of the Moose Mountain shear zone is prominently foliated and contains the assemblage hornblende + plagioclase ± garnet ± biotite ± ilmenite ± sphene ± quartz. Small amounts of semipelitic schist (quartz + biotite + muscovite + plagioclase) occur within the mafic amphibolite near its northern edge. Lenses of felsic amphibolite (plagioclase + hornblende + epidote + quartz) become common toward the south. In addition, lenses of massive quartzite (quartz + hematite ± garnet ± grunerite ± stilpnomelane ± epidote ± hornblende) tens of meters long and a few meters thick are interlayered with the amphibolite (and occur as clasts in the mélange, see below; Fig. 3). Based on their mineralogy and absence of detrital zircons in samples that we have separated, we interpret the quartzites to be metamorphosed ferruginous cherts. At least one lens of variably altered and metamorphosed ultramafic rock is also present in the vicinity of the metachert lenses (Fig. 3). Assemblages within the ultramafic horizon include anthophyllite + magnetite ± hornblende ± cummingtonite, clinopyroxene + talc ± hornblende, and hornblende + actinolite + chlorite + Cr-rich magnetite. Whole-rock analyses of these rocks yield < 45% SiO2 and > 20% MgO, confirming their ultramafic nature (Selverstone, unpublished data). We suspect that the zone characterized by lenses of metachert and ultramafic rock is part of the adjacent tectonic mélange, but sparse outcrop precludes detailed mapping in this area. Neither metachert nor ultramafic rock had been reported in the area prior to this study.
The L2 stretching lineation is generally well-developed throughout the supracrustal succession, and the shallowly east-plunging L1 lineation also is present locally. In the field, L1 is most obvious as an amphibole mineral lineation (anthophyllite and hornblende) within the ultramafic horizon (Fig. 3), but it also is present locally as a stretching lineation throughout the supracrustal succession. On average, L1 plunges ≤40° toward N55° E (≤23° toward N60° E when corrected for dip on the Paleozoic unconformity; Fig. 4).
The mélange zone within the Moose Mountain shear zone (Figs. 3 and 6), which ranges from 3 to 500 m in width, has not been described previously. The host rock is an extremely fine-grained calcite marble that locally contains porphyroclasts and/or -blasts of garnet + diopside + microcline + quartz ± epidote ± hornblende sphene. Deformed clasts and lenses within the marble include anthophyllite-magnetite schist, clinopyroxenite, calcic gneiss, epidote amphibolite, garnet amphibolite, metachert, and pegmatite. The clasts range from millimeters to tens of meters in length, and they are themselves mylonitized, isoclinally folded, and/or boudinaged (Figs. 6A and 6B).
The marble host rock has a prominent mylonitic foliation that is parallel to the shear zone fabric. Both the L1 and L2 lineations are variably developed on foliation surfaces (Fig. 6C), with L2 being more common along lithologic contacts. Foliation within small clasts typically is discordant with respect to the marble host, but large blocks (> 10 m long) of amphibolite/ultramafic schist, metachert, and calcic gneiss are foliated parallel to the marble host. These larger blocks preferentially preserve the shallow L1 stretching/mineral lineation, but most also show composite fabrics.
Shear Sense and Inferred Temperature History
St. Vrain Granite
Feldspar imbrication, asymmetric porphyroclast tails, and S-C fabrics evident in outcrops viewed parallel to the L2 lineation and perpendicular to the foliation consistently yield a southeast-side-up reverse shear sense, which was confirmed through thin section examination of numerous mylonitic samples (e.g., Fig. 7).
The least-deformed granite within the Moose Mountain shear zone shows field evidence for apparent magmatic tiling of feldspars (e.g., Blumenfeld, 1983; Blumenfeld and Bouchez, 1988; Tikoff and Teyssier, 1994; McCaffrey, 1994); such tiling is rare outside the boundaries of the shear zone. The generally undeformed nature of these feldspars is consistent with a magmatic origin for the tiling. Analogue experiments suggest that tiling features will be best-developed at relatively low shear strains (γ <6); at higher shear strains, K-feldspar phenocrysts will tend to be oriented subparallel to the shear plane (Arbaret et al., 1996). In outcrops that show tiling, K-feldspar phenocrysts are dominantly imbricated in a pattern that indicates southeast-side-up shearing during emplacement of the granitic magma. Based on the conditions of emplacement estimated by Anderson and Thomas (1985) for the St. Vrain granite, the feldspar tiling probably developed at temperatures around 750° C. In most places within the Moose Mountain shear zone, solid-state deformational fabrics overprinted the magmatic fabrics to varying degrees, but relic imbrication is still evident (Fig. 7B).
Solid-state fabrics preserved in feldspars from more highly strained parts of the St. Vrain granite range from plastic to brittle in character, and all are associated with asymmetric features showing southeast-side-up shear sense. Plastic deformational features comprise myrmekitic intergrowths, development of flame perthite, core-and-mantle structure, subgrain formation, and undulatory extinction.
Myrmekitic intergrowths of plagioclase and quartz are developed on the S-surfaces of K-feldspar porphyroclasts in a few well-developed S-C mylonites in the St. Vrain granite. As pointed out by Simpson and Wintsch (1989), such localized development of myrmekite can be used as a shear sense indicator, as myrmekite preferentially develops at sites of high stress concentration. The localized occurrence of myrmekites in St. Vrain granite samples is consistent with the southeast-side-up shearing deduced from S-C relationships in the same samples. Myrmekite is inferred to develop at temperatures ≥500° C in deformed granitic rocks (Simpson and Wintsch, 1989).
Mylonitic to ultramylonitic rocks within the St. Vrain granite show well-developed core-and-mantle structure in K-feldspar (Fig. 7C) and, to a lesser extent, in plagioclase. This texture indicates that dislocation climb was active during deformation, suggesting ultramylonite formation at temperatures in excess of 400–500° C (e.g., Hirth and Tullis, 1992; Passchier and Trouw, 1996). Sigma and delta porphyroclasts of feldspar with core-and-mantle structure are all consistent with a southeast-side-up shear sense during recrystallization; muscovite fish that occur in close association with dynamically recrystallized feldspars show the same kinematics.
Lower temperature plastic microstructures also are present in K-feldspar and plagioclase and comprise local development of flame perthite parallel to the inferred shortening direction (Fig. 7D; Pryer and Robin, 1996), scattered subgrain development, and undulose extinction. These features represent deformation assisted by dislocation glide and by diffusion of alkalis within the feldspars. Brittle deformation of feldspars also is common. Many of these feldspars show antithetic “bookshelf” faults. Muscovite in samples with brittlely deformed feldspars is kinked and fractured. Temperature estimates for all these features are in the 300–400° C range (Pryer, 1993; Pryer and Robin, 1996; Passchier and Trouw, 1996).
Quartz in many of the granitic samples is prominently ribboned (e.g., Figs. 7A and 7B) with a strong lattice-preferred orientation and local development of core-and-mantle structure. These fabrics are characteristic of deformation in Regimes 1 and 2 of Hirth and Tullis (1992) at temperatures in the 300–350° C range. Shear sense determined from the lattice-preferred orientation in the ribboned quartz is again southeast-side-up.
Many samples show a complete succession of fabrics from apparently magmatic tiling of feldspars through ribboning of quartz and brittle faulting of feldspars. As indicated above, however, there is no change in kinematics from the high-temperature fabrics to lower temperature fabrics. These observations are interpreted to imply either: (1) continuous deformation along the Moose Mountain shear zone during emplacement and cooling of the St. Vrain granite; or (2) an increase in strain rate and/or fluid pressure to produce “low-temperature” fabrics while the rocks were still at elevated temperature. We consider the former explanation to be the more likely, but we cannot rule out some contribution from elevated fluid/magma pressure or strain rate.
he supracrustal rocks preserved in northeastern exposures of the Moose Mountain shear zone display both the L1 and L2 lineations, and many of the samples contain composite fabrics evident in thin section. In quartzite (metachert) horizons, the L2 lineation is dominant as a prominent stretching lineation in outcrop. However, most samples cut parallel to this lineation either yield no shear sense or show a low-temperature reorientation of quartz grains (marked by extreme undulatory extinction) into southeast-side-up reverse-sense shear bands. In contrast, sections cut parallel to the L1 lineation show a well-developed, sinistral S-C fabric (Fig. 8A). Shear bands present in interlayered hornblende amphibolites and quartzites also yield a sinistral shear sense (Fig. 8B), whereas L2-parallel sections show ambiguous shear sense.
In anthophyllite schists, L1 is defined in the field by a pronounced alignment of amphibole and Mgchlorite grains, and rare ductile shear bands yield sinistral shear sense. The anthophyllite lineation is disrupted locally by microbreccia lenses that are parallel to the L2 direction, but no shear sense is apparent in these structures. Metachert samples that are cut in the plane containing both lineations show a composite fabric. They exhibit strong lattice- and shape-preferred orientation of quartz grains parallel to L1 and partial recrystallization into parallelism with L2 (Fig. 8C). Alignment of anthophyllite + Mg-chlorite grains in amphibolite and the development of a quartz lattice- and shape-preferred orientation in metachert suggest that the L1 lineation developed at temperatures >500° C (Spear and Rumble, 1986; Hirth and Tullis, 1992). The brittle disruption of the amphibole lineation and undulatory quartz in L2-parallel shear horizons suggest that L2-related deformation occurred at considerably lower temperatures than L1-related deformation.
Although determination of shear sense is not possible in every sample cut parallel to L1, all samples that do yield shear sense record motion that is sinistral with respect to present-day coordinates. It is, of course, difficult to determine whether the shear planes have maintained their orientation since the Proterozoic. The L1 lineation and associated kinematics could have been produced originally by dominantly reverse dip-slip motion followed by subsequent counterclockwise rotation of the shear planes by > 40°. However, no evidence exists for such rotation. Normal dip-slip motion followed by rotation is also improbable, as it would have dropped high-grade metamorphic rocks down against lower grade rocks. In the absence of convincing evidence to the contrary, we believe that sinistral transcurrent history recorded by Paleo-proterozoic rocks within the Moose Mountain shear zone does reflect a Proterozoic strike-slip history.
Age Constraints and Net Strain
The microstructural evidence indicating deformation within the Moose Mountain shear zone during emplacement and cooling of the St. Vrain granite requires that southeast-side-up reverse motion occurred around 1423 ± 30 Ma. The disappearance of mylonitic fabrics in the St. Vrain batholith to the southwest of the study area implies either that magmatism there outlasted deformation or that the 1.4-Ga deformation decreased in intensity to the southwest. The 1317 ± 50 Ma Iron Dike shows no offset where it crosses the extrapolated trace of the Moose Mountain shear zone (J. Cole, personal communication, 2000; Braddock and Cole, 1990) and places an absolute limit on the youngest age of movement in the zone.
Hodgins (1997) used the percentages of deformed and undeformed St. Vrain granite present in a north–south transect across the shear zone, coupled with shear strains (γ, tangent of the angular shear Ψ) calculated from S-C angles, to estimate the amount of 1.4-Ga offset on the Moose Mountain shear zone. Within the 2 km-wide shear zone, Hodgins (1997) observed 76% undeformed granite (γ = 0), 14% protomylonite (γ = 0.2), 9% mylonite (γ = 0.7), and ∼1% ultramylonite (γ = 7). Using these values, a total southeast-up reverse offset of ≤5 km was calculated. This calculated value clearly has a large associated uncertainty, but at least it gives an approximation of the net strain associated with development of the 1.4-Ga L2 lineation.
The presence of pre-St. Vrain granite deformational fabrics within supracrustal rocks at the northeast end of the Moose Mountain shear zone indicates that an earlier deformational event, probably dominated by sinistral motion, occurred prior to 1423 Ma. In the absence of age dates on minerals crystallized in the L1 lineation, it is difficult to determine directly whether this earlier phase of movement was Mesoproterozoic or Paleoproterozoic in age. The shallowly plunging L1 lineations and asymmetric fabrics within the mélange suggest a movement history dominated by sinistral strike-slip motion. The strong lattice-preferred orientation of quartz and amphiboles associated with this lineation further suggests that it developed during relatively high-temperature metamorphism. Because the 1.4-Ga metamorphism in the area appears to have been only a short-lived thermal pulse (Shaw et al., 1999), we believe that the early L1-associated fabrics most probably developed during the ∼1.7-Ga metamorphic history of the region.
The age of transcurrent motion along the proto-Moose Mountain shear zone can be further refined by consideration of structural and petrologic features of the blocks on either side of the shear zone (Figs. 2 and 10). The Big Thompson region to the north of the Moose Mountain shear zone preserves evidence for three stages of folding, the youngest of which has been dated as synchronous with emplacement of trondhjemite at 1726 ± 15 Ma (Barovich, 1986). Similar folds are absent in rocks to the south. Trondhjemite is present only on the north side of the shear zone, but bodies of the 1714 ± 5-Ma Boulder Creek batholith (Premo and Fanning, 2000) occur on both sides of the shear zone. These features suggest that the blocks on opposite sides of the Moose Mountain shear zone had separate histories at 1726 ± 15 Ma, but the blocks shared a common history by 1714 ± 5 Ma. Transcurrent movement probably was thus confined to the interval between these two times. If this interpretation is correct, it implies that disparate crustal fragments were juxtaposed across a proto-Moose Mountain shear zone around 1.71 Ga.
In an attempt to constrain the magnitude of total offset on the Moose Mountain shear zone, we collected samples of quartzitic metasedimentary rocks from opposite sides of the Moose Mountain shear zone for detrital zircon age determinations. Identical age populations in the zircon suites would imply that metasedimentary units can be correlated across the shear zone (consistent with the modest reverse offset calculated above from the St. Vrain granite). Different zircon populations, in contrast, would support relatively large offset, including pre-St. Vrain movement, across the shear zone. Approximately 50 zircon grains from each sample were analyzed using the sensitive high-resolution ion microprobe (SHRIMP) at the Australian National University (following methods outlined in Compston et al., 1984 and Williams and Claesson, 1987). Age frequency data for each of these samples are shown in Figure 9 as relative concordia plots, histograms, and probability curves.
Sample CLl-96 was collected along the range front ∼500 m north of the Moose Mountain shear zone (Fig. 2). This sample shows a wide range of detrital zircon ages, from 1758 ± 26 to ∼3450 Ma, with a pronounced peak at about 1800 Ma (Fig. 9B; Table 1). This is the first sample from the northern Colorado Front Range to yield significant evidence of Archean detritus (e.g., Premo and Van Schmus, 1989; Ball and Farmer, 1991; Aleinikoff, Reed, and Wooden, 1993). In contrast, sample BT151-LY from 1 km south of the Moose Mountain shear zone (Fig. 2) shows a pronounced peak at about 1825 Ma (Fig. 9D; compare with slightly younger main peak in Fig. 9B) and lacks evidence for zircons older than 2650 Ma (Fig. 9D; Table 1). One subhedral, prismatic grain from this sample (analysis BT-17.1, Table 1) has a nearly concordant age of about 1.14 Ga, significantly younger than the known age of deposition (the metasedimentary host rock was intruded by the 1714 ± 5-Ma Boulder Creek granodiorite). This grain may have formed in the quartzitic metasediment during a 1.1-Ga heating event or it may be a contaminant. Samples CL1-96 and BT151-LY were collected from outcrops previously mapped as the same unit (Paleoproterozoic quartzofeldspathic schist, Xqs, of Braddock et al., 1988a, 1988b). The difference in provenance age patterns recorded in these samples is statistically significant, however, and provides evidence that rocks on opposite sides of the Moose Mountain shear zone cannot be directly correlated, despite apparent similarities in the field.
∼ 1.7-Ga TECTONISM
The differences both in structural histories and detrital zircon age populations from samples on opposite sides of the Moose Mountain shear zone raise the possibility that Paleoproterozoic offset juxtaposed two crustal blocks with very different early histories. The source of the Archean component in sample CL1-96 (Fig. 9; Table 1) probably is reworked material from the Wyoming province (e.g., Mueller et al., 1998), but sediments originally deposited to the south of the present-day Moose Mountain shear zone apparently were not derived from this same source region. It is possible that the different age ranges of the zircon suites on opposite sides of the shear zone simply reflect different drainage systems within a single coherent crustal terrane. Likewise, the structural differences across the zone could simply reflect variations in local strain fields. However, L1-related sinistral shear indicators and the presence of an exotic mélange zone in the Moose Mountain shear zone are more consistent with a tectonic explanation for the data. Specifically, the explanation would involve transcurrent juxtaposition of unrelated blocks against one another.
Assuming plausible slip rates for a major strikeslip boundary, sinistral displacement along the proto-Moose Mountain shear zone could have been a few hundred kilometers during the period from 1726 ± 15 to 1714 ± 5 Ma. Although there is no known evidence to support such large translations, amalgamation of far-traveled blocks may have been a key part of Paleoproterozoic tectonism in the region.
Contractional deformation along the Cheyenne belt in southern Wyoming occurred between 1.78 and 1.74 Ga (Chamberlain, 1998). This age span has been viewed traditionally as the age at which the Colorado province was accreted onto the Archean Wyoming province. Data from vicinity of the Moose Mountain shear zone, however, show that additional fragments were added, probably via transcurrent motion, as recently as 1.71 Ga. These data suggest that Proterozoic lithosphere was accreted piecemeal over a significant interval (Fig. 10A) and that the Colorado province should not be regarded as a single crustal block. Considerable additional work is necessary to determine whether other shear zones within the Colorado province have similar early histories to that of the Moose Mountain shear zone (as speculated upon by Tweto and Sims, 1963).
Gravity data summarized by Smithson and Boyd (1998) in the northern Front Range show a decrease of ∼30–50 mGal in vicinity of the Moose Mountain shear zone, corresponding to a modeled increase in crustal thickness from ∼43 km north of the shear zone to ∼48 km south of the zone. These data, combined with the apparent differences in source and structural history of rocks on opposite sides of the Moose Mountain shear zone, support the interpretation that the shear zone is a major crustal structure that penetrates the entire crust.
Evidence for 1.4-Ga deformation in the Colorado Front Range comes from northeast-striking shear zones, such as the Moose Mountain zone, mafic and intermediate dike swarms, and from upright folds in the St. Vrain granite and nearby schist in Rocky Mountain National Park (F4; Cole, 1977). The Moose Mountain shear zone records southeast-side-up reverse motion associated with emplacement of the 1423 ± 30-Ma St. Vrain granite. The northeast-trending F4 folds deform the flow foliation of St. Vrain granite and are inferred to have formed during granitic emplacement (Cole, 1977). The dike swarms have an average orientation of N15° W, 90° (Nutalaya, 1966; Braddock et al., 1970b; Punongbayan, 1972), and their age of emplacement is constrained by mutually crosscutting relationships with the 1433 ± 1.5-Ma Sherman Granite (Frost et al., 1999) and the 1390 ± 30 Ma Log Cabin batholith (Peterman et al., 1968; possibly 1422 ± 3 Ma if Peterman et al.'s correlation with main phase of Silver Plume Granite is correct).
The combined data from shear zone-, dike-, and fold-orientations in the northern Colorado Front Range indicate a 1.4-Ga strain field characterized by northwest–southeast shortening and northeast–southwest extension (Fig. 10B). This strain field is consistent with the model by Nyman et al. (1994) for a regional contractional orogeny at ∼1.4 Ga. Nyman et al. (1994) argued that contractional deformation occurred throughout the southwestern United States in response to far-field transpressional stresses from the southern Laurentian. plate boundary (Fig. 10B). Alternatively, these data also are consistent with relatively minor jostling along preexisting zones of weakness in the absence of a major contractional orogenic event. The apparent absence of 1.4-Ga deformational fabrics in supracrustal rocks between the discrete shear zones is evidence that deformation was not penetrative at 1.4 Ga.
DISCUSSION AND CONCLUSIONS
Mesoproterozoic contractional deformation is locally evident throughout a great expanse of Proterozoic North American lithosphere, suggesting an orogen on the order of 1200 km wide. However, the deformation associated with the 1.4-Ga event was largely localized into discrete shear zones; pervasive evidence for 1.4-Ga orogeny is therefore lacking. In the case of the Moose Mountain shear zone, strain was accommodated by reactivation of an older structure that separates blocks with significantly different pre-1.71-Ga histories.
Many studies have shown that within-plate oceanic and continental crust largely respond to horizontal compressive stresses by reactivation of older structures such as high-angle reverse faults and/or strike-slip faults (e.g., Karner et al., 1993; Ziegler et al., 1995; Allen and Vincent, 1997; Murphy et al., 1999). Such reactivation can affect preexisting structures over enormous distances. Within the oceanic crust of the central Indian Ocean, reverse faults occur every 5–20 km for distance of ∼1500 km (Karner et al., 1993), apparently reflecting compressive plate interactions along the Himalayan zone; these faults developed by reactivation of normal faults initially formed at a spreading center. Within the Alpine foreland, Ziegler et al. (1995) described Tertiary reverse reactivation of older structures up to hundreds of kilometers away from the main zone of Alpine orogenic activity. In both examples, there is little discernible deformation of rocks between individual reactivated structures, an observation that is similar to the pattern of Mesoproterozoic deformation in the southwestern United States.
It is thus tempting to argue that the data from the Moose Mountain shear zone and surrounding rocks directly support the model of Nyman et al. (1994) and reflect far-field compressive stresses transmitted from an active orogen along the southern margin of Laurentia. However, a similar pattern of widespread compressive reactivation of preexisting structures (basin inversion) is also evident across much of central Africa today, where it probably reflects a response to the nonorogenic divergent stress regime in the West and Central African rift systems (Ziegler et al., 1995).
Examination of the igneous geochemical record in conjunction with the structural record suggests a complex scenario for Mesoproterozoic tectonism. Studies of numerous ∼1.4-Ga shear zones show northwest–southeast contraction, probably largely reflected in reactivation of preexisting structures (e.g., Graubard and Mattinson, 1990; Aleinikoff et al., 1993a; Nyman et al., 1994; this study). That is consistent with transmission of far-field compressive stresses from the southern margin of Laurentia. The granitic magmatism of the same age, however, can be explained most easily in a setting involving large-scale mantle upwelling, either in a plume or extensional setting (e.g., Hoffman, 1989; Windley, 1993; Frost et al., 1999). Such settings also are capable of transmitting horizontal compressive stresses over large distances in the crust, as pointed out above. Thermal “priming” of the crust (e.g., Hand and Sandiford, 1999) by magmatic underplating and partial melting in such a setting may sufficiently weaken preexisting penetrative shear zones that reactivation may occur at relatively low compressive stresses. Magmas also are likely to use subvertical fabrics and sharp lithologic contacts as conduits, leading to a feedback effect between melt localization into older shear zones and shear zone localization in plutons (e.g., D'Lemos et al., 1997). The setting for 1.4-Ga magmatism and deformation in the Colorado Front Range thus remains ambiguous, but clearly it must involve some component of mantle upwelling at depth in addition to shortening at middle- to upper-crustal levels.
Despite the prominent mylonitic fabrics present in the St. Vrain granite, much of the history of the Moose Mountain shear zone must be older. The relict strike-slip indicators in Paleoproterozoic rocks within the zone, coupled with differences in sediment source, structural history, and igneous history in rocks on opposite sides of the shear zone, point to its role as a transcurrent structure during Paleoproterozoic accretion. In addition to its early and Mesoproterozoic movement histories, the Moose Mountain shear zone also was utilized as the locus of a brittle Laramide fault in late Cretaceous–early Tertiary times (Punongbayan, 1972), despite its relatively unfavorable orientation with respect to dominant Laramide structures in the region. As pointed out by Karlstrom and Humphreys (1998), much of the tectonic history of western North America has involved reactivation of structures initially established during assembly of the continent in Precambrian times. Faults that initially were steeply dipping with strike-slip displacements are likely to represent profound, through-going crustal discontinuities that could be reused readily during subsequent orogenic cycles.
Financial support for this work was provided by NSF grants EAR-9596218 and EAR-9804712 to JS and by a GSA Research Grant to MH. We thank Laura Pletsch-Rivera and Ben Gutzler for assistance in the field. Discussions with Lang Farmer, Karl Karlstrom, Colin Shaw, Jack Reed, and Aaron Cavosie contributed to our thinking about the Proterozoic in Colorado. Journal reviews by Kevin Chamberlain and Ernie Duebendorfer and editorial assistance from Art Snoke resulted in significant clarification of the manuscript and are gratefully acknowledged. We also thank Jack Reed and Jim Cole for helpful comments on the manuscript.
- Received January 13, 2000.
- Revision received August 21, 2000.
- Accepted September 12, 2000.