- UW Department of Geology and Geophysics
A major geologic boundary has been proposed in the Southern Rocky Mountains separating Proterozoic crustal provinces with different ages and tectonic histories. These provinces probably correlate with the Yavapai (1.8–1.7 Ga) and Mazatzal (1.7–1.6 Ga) provinces of Arizona. Geologic, geochemical, geochronologic, and xenolith data suggest that the boundary lies within a ∼300 km-wide zone that trends northeastward through southern Colorado and northern New Mexico. This zone also seems to have focused later tectonic and thermal effects. However, no major shear zone that might represent a discrete tectonic suture has been identified in the area, and there is no agreement on precisely where the boundary is or what tectonic significance it may have.
We present a review of evidence supporting extrapolation of the Yavapai-Mazatzal boundary through the Southern Rocky Mountains. Limitations in the precision, quantity, and interpretation of available data probably contribute to disagreement over the location of the boundary. However, the disparity in boundaries defined by different data sets may partly reflect a complex or gradational transition between crustal domains. We propose a speculative model for the boundary based on a preliminary structural analysis. Tectonic fabrics appear to be consistent with the initial juxtaposition of arc terranes of the Yavapai and Mazatzal provinces on a low-angle thrust system with later modification and steepening of the boundary during continued crustal shortening. This model explains the diffuse isotopic boundary as a manifestation of a vertically heterogeneous crustal column that might promote isotopic mixing. The cryptic structural expression of the suture may result from a layer-parallel style of suturing and complex post-accretionary tectonic overprinting.
- province boundary
- continental accretion
- island arcs
- isotopic provinces
- Southern Rocky Mountains
- shear zones
Plate tectonic theory postulates that new continental lithosphere is formed by differentiation of basaltic material in magmatic arcs and assembly of the buoyant arc terranes at convergent margins (Hamilton, 1969, 1981, 1988). This process should produce continents comprising discrete terranes separated by tectonic boundaries (sutures) representing the closure of oceans or back-arc basins. Post-accretionary shuffling of tectonic blocks may also produce structural boundaries or reactivate primary accretionary structures (Jones et al., 1983; Hamilton, 1988). A complete model of continental accretion and its effects on lithospheric architecture requires an understanding of the boundaries between terranes and the processes that knit separate arc terranes together into a coherent continent.
In the southwestern United States, a > 1000 km wide zone of juvenile crust was accreted to the rifted southern margin of the Wyoming craton between about 1.8 and 1.6 Ga (Karlstrom and Bowring, 1988, 1993). Rocks of this orogen are exposed in Laramide uplifts along a 700 km-long transect in the Southern Rocky Mountains. This is an important natural laboratory for the study of accretionary boundaries and processes because a variety of possible boundaries is exposed and because mid-crustal levels that are inaccessible in younger accretionary orogens are now at the surface.
This paper reviews evidence supporting a proposed arc/arc-accretionary boundary in central Colorado. This boundary is one of several potential eastward extensions of an orogenic and isotopic boundary first recognized in Arizona that separates the ca. 1.8–1.7 Ga Yavapai province from the ca. 1.7–1.6-Ga Mazatzal province (Silver, 1965, 1968; Karlstrom and Bowring, 1988, 1993). The province boundary has been recognized by previous workers on the basis of geochronologic, isotopic, petrologic, and structural discontinuities and/or gradients (e.g., Condie, 1982, 1986; Bennett and DePaolo, 1987; Wooden and DeWitt, 1991; Karlstrom and Bowring, 1993; Karlstrom and Daniel, 1993). However, in the Rocky Mountains, no discrete structural suture has been identified, and different data sets identify different locations for the boundary (Fig. 1). In an effort to stimulate and focus future work on the boundary, we have tried to integrate published isotopic, geochronologic, and xenolithic results with compiled metamorphic and structural data to produce a speculative model for the juxtaposition of two or more arc terranes along initially low-angle structures. This model is an attempt to reconcile conflicting proposed locations for the boundary, to explain the lack of an obvious structural suture, and to provide a testable cross section for further geological and geophysical studies.
REGIONAL TECTONIC SETTING
The best documented example of a fundamental accretionary boundary exposed along the Rocky Mountain transect is the suture between Archean and Proterozoic crust, the Cheyenne belt of southern Wyoming (CB in Fig. 1). Steeply-dipping shear zones of the Cheyenne belt mark the juxtaposition of the rifted Archean basement (> 2.5 Ga) of the Wyoming craton and its Paleoproterozoic passive margin cover (> 2.0 Ga) with a succession of 1.8–1.7-Ga arc terranes (Hills and Houston, 1979; Condie, 1982; Karlstrom and Houston, 1984; Duebendorfer and Houston, 1986, 1987; Hoffman, 1989; Houston et al., 1989; Van Schmus et al., 1993a). Sharp Nd and Pb isotopic discontinues between the two provinces coincide with the tectonic suture (Zartman, 1974; Bennett and DePaolo, 1987; Ball and Farmer, 1991).
Within the Proterozoic provinces of Colorado and New Mexico, it has been more difficult to identify terranes and sutures. By analogy with modern arcs, individual terranes should rarely exceed several hundred kilometers in width suggesting that the southwestern Proterozoic provinces should be composed of a number of distinct terranes separated by tectonic suture zones. Numerous shear zones in the Southern Rocky Mountains divide the Proterozoic crust into tectonic blocks (Fig. 1), but these zones record a long history of 1.8–1.4-Ga tectonism and it has been difficult to distinguish the more fundamental paleo-subduction boundaries from later accommodation structures. Indeed, many early accretionary structures may have been reactivated or overprinted during progressive shortening after accretion, especially at the middle crustal levels now exposed (Karlstrom and Williams, 1998).
Silver (1965) first proposed a northeast-trending geochronologic boundary between an older 1.8–1.7-Ga block to the north and a younger 1.7–1.6-Ga block to the south based on work in central Arizona. The boundary was imprecisely located, but extended from Arizona through the Four Corners region and was postulated to continue through central Colorado (Fig. 1). Variations of this boundary have appeared in regional compilations ever since (e.g., Silver et al., 1977; Van Schmus and Bickford, 1981; Nelson and DePaolo, 1985; Bennett and DePaolo, 1987; Van Schmus et al., 1987, 1993b; Karlstrom and Bowring, 1993). At least six other versions of this boundary have been proposed and, taken together, the proposed boundaries define a 300 km-wide transition zone between the Yavapai and Mazatzal provinces (Fig. 1). Both the northern and southern limits of the province correspond roughly to a geophysical lineament. The northern limit follows the Holbrook lineament in western Arizona and the southern limit coincides with the Jemez lineament in northern New Mexico (Fig. 1; Karlstrom and Conway, 1986). The coincidence of these geophysical features with the proposed geochemical and geochronologic boundaries suggests that at least some of these boundaries may have important deep-crustal significance.
In evaluating tectonic boundaries within orogenic belts it is important to apply a multi-disciplinary approach to clearly define provinces and boundaries. Different techniques may “see” different aspects or levels of a complex transition between tectonic blocks. The following sections review geochronologic, isotopic, xenolithic, and structural evidence for the proposed Yavapai-Mazatzal boundary where it intersects the Southern Rocky Mountains. We also present compiled metamorphic and structural data from general geologic studies in the area of the proposed boundary, focusing on the northern part of the transition in central and southern Colorado (see Williams et al., this issue, for overview of northern New Mexico). No geophysical studies have specifically targeted the proposed Yavapai-Mazatzal boundaries in the Rocky Mountains, but seismic lines are planned as part of the upcoming Continental Dynamics Rocky Mountain transect study (see Introduction to this issue).
GEOCHRONOLOGIC AND ISOTOPIC STUDIES
Geochronologic and isotopic studies provide the primary line of evidence for defining the Yavapai-Mazatzal boundary in the Southwest (Silver, 1965; Condie, 1981, 1982; Bennett and DePaolo, 1987). Proposed isotopic boundaries in the Southern Rockies are broadly consistent at a regional scale but differ significantly in detail (Fig. 1; DePaolo, 1981; Condie, 1982, 1987; Nelson and DePaolo, 1985; Bennett and DePaolo, 1987; Reed et al., 1987; Wooden and DeWitt, 1991; Aleinikoff et al., 1993).
Following the pioneering work of Silver (1965, 1968), Condie (1981, 1982) identified two Proterozoic crustal provinces that stretch from Arizona to the Rocky Mountain region. The northern province is characterized by supracrustal volcanic rocks with U-Pb zircon and Rb-Sr isochron ages of 1.80 to 1.72 Ga and the southern province by ages of 1.72 to 1.65 Ga (Condie, 1982). Condie placed the boundary between these provinces in northern New Mexico south of the Sante Fe Range (C1 in Fig. 1). In a later modification, Condie (1987) proposed that a northern arc province of 1.80–1.76-Ga crust extended as far south as the Manzano and Pedernal mountains of New Mexico. It was overlapped by a 1.70–1.68-Ga continental arc or back-arc volcanic province that reached from the Grenville front to central Colorado near Salida (C2 in Fig. 1) creating a composite transition zone in southern Colorado and northern New Mexico. Condie also identified smaller arc and back-arc terranes inset into the larger provinces in New Mexico (1.72 and 1.65 Ga) and the Salida-Gunnison area of central Colorado (1.74–1.73 Ga).
Reed et al. (1987) subdivided the Yavapai (Colorado) province into three subprovinces on the basis of lithologic contrasts (Tweto, 1987) and generally southward-decreasing U-Pb ages (Fig. 1). The three subprovinces are: (1) the Green Mountain arc characterized by syntectonic pluton ages greater than about 1.75 Ga; (2) an intervening composite back-arc basin with deformed plutons older than ca. 1.70 Ga [except for the ca. 1.67-Ga Kroenke pluton near the southern edge of the subprovince, see Premo and Fanning (1997) for revised age of Boulder Creek quartz monzonite]; and (3) the Salida-Gunnison magmatic-arc complex in central and southern Colorado with pluton and metavolcanic ages from 1.76–1.60 Ga (Reed et al., 1987).
Wooden and DeWitt (1991) proposed a common Pb boundary in Arizona (W&D in Fig. 1) that roughly coincides with Condie's (1982) geochronologic boundary. Rocks (mainly plutonic) from southeast Arizona (Mazatzal province) have lower 208Pb/204Pb for a given value of 206Pb/204Pb than rocks from central Arizona (Yavapai province). The data implies a lower time-integrated Th/U for the source of Yavapai rocks (∼2) than for the Mazatzal source (∼4). There is no corresponding difference in 207Pb/204Pb vs. 206Pb/204Pb (Wooden and DeWitt, 1991). The data of Stacey and Hedlund (1983) and Wooden and Aleinikoff (Wooden and Aleinikoff, 1987) suggest that the differences in common lead persist into New Mexico and Colorado with the Pb province boundary passing through northern New Mexico (Fig. 1; Wooden et al., 1988; Wooden and DeWitt, 1991).
Aleinikoff et al. (1993) interpreted differences in Pb isotopic ratios between rocks of southern Colorado and northern New Mexico and rocks of northern Colorado as evidence for a Pb transition zone in central Colorado supporting Reed's boundary between a composite back-arc domain (northern Colorado) and the Salida-Gunnison magmatic arc (Fig. 2). Inferred Th/U of the source calculated from 208Pb/204Pb and 206Pb/204Pb ratios are scattered from 1 to 9 in the northern back-arc area and more tightly clustered between 1 and 3 in the Salida-Gunnison area and northern New Mexico. However, the mean values for the two populations are similar. The tighter clustering of inferred Th/U in the southern samples might partly reflect the fact that about 50 percent of the southern samples came from a relatively small area in the Taos Range (Aleinikoff et al., 1993, figs. 2 and 6, table 1). The 207Pb/204Pb vs. 206Pb/204Pb are similar in the northern and southern sample populations. Any real differences in the common lead signatures of the back arc domain in northern Colorado and the Salida-Gunnison area could reflect different source rock lithologies (Aleinikoff et al., 1993) or variable uranium loss due to different thermal and metamorphic conditions (cf. Wooden et al., 1988; Aleinikoff et al., 1993).
It is important to note that the Pb transition zone proposed by Aleinikoff et al. (1993) does not correspond to the boundary drawn by Wooden and DeWitt (1991). The latter boundary separates a domain with calculated Th/U values of ∼2 in the north (Yavapai) from a domain with values of ∼4 in the south (Mazatzal), whereas the former marks a perceived transition between variable Th/U (north) and more typical Yavapai province ratios in the Salida-Gunnison area and northern New Mexico. If a statistically significant difference between the common Pb signatures of rocks in southern Colorado and those in northern Colorado does, in fact, exist, the Pb-transition zone of Aleinikoff et al. (1993) may mark an internal lithotectonic boundary within the Yavapai province as proposed by Reed et al. (1987).
Nd Model Ages
Nd model ages (TDM) are interpreted to date the time of crustal differentiation from a depleted mantle source (DePaolo, 1981). Bennett and DePaolo (1987) proposed two Proterozoic “crustal formation” provinces on the basis of Nd data (Fig. 1). Nd provinces are broadly consistent, at a regional scale, with crystallization age provinces (Condie, 1981, 1982, 1986) and Pb isotopic provinces (Nelson and DePaolo, 1985; Bennett and DePaolo, 1987; Wooden et al., 1988). A northern province (Nd province 2) with model ages (TDM) of 2.0–1.8 Ga corresponds roughly to the 1.8–1.7-Ga supracrustal age province (Yavapai) and a southern province (Nd province 3) with model ages of 1.8 Ga, corresponds to the 1.7–1.6-Ga age province (Mazatzal). However, in the Southern Rockies the location of the Nd boundary is only approximately known because it is delimited by only three sample localities, separated by about 150 km (Fig. 2; DePaolo, 1981; Bennett and DePaolo, 1987). The limited number of samples in this area makes it difficult to assess whether the Nd-boundary is gradational or discontinuous and how well it coincides with other isotopically defined boundaries in the area.
The interpretation of Nd model ages as “crust-formation ages” has been questioned by Arndt and Goldstein (1987). Nonetheless, the regionally consistent pattern of Nd ages, rarity of documented mantle additions to the crust, and the general lack of significantly older potential source material in the Southwest are cited as justifications for using Nd isotopes to define crustal provinces in this region (e.g., Farmer and DePaolo, 1983; Nelson and DePaolo, 1985; Bennett and DePaolo, 1987). The Nd crust-formation provinces of Bennett and DePaolo (1987) were constructed using Nd model ages of crustally derived granitoids ranging in age from Paleoproterozoic to early Tertiary. The consistency of results from this diverse suite of rocks suggests that they may be derived from long-lived, stable, lower-crustal sources with distinct Sm-Nd systematics (Livaccari and Perry, 1993).
Synthesis of Isotopic and Geochronologic Studies
Despite the discordance of proposed isotopic boundaries based on different data sets, the balance of isotopic evidence from geochronology, common Pb, and Nd model ages seems to support the conclusion that a regionally significant crustal boundary passes through southern Colorado or northern New Mexico. The three most probable explanations for disparity in the boundaries determined by different isotopic systems and methods are: (1) experimental and statistical uncertainties limit the precision of isotopic boundaries to worse than about ± 100 km; (2) the grouping of samples into populations with arbitrary age or compositional limits may not be consistent between studies employing different geochemical tools; or (3) different isotopic systems and methods may be sensitive to different aspects of a complex, heterogeneous transition between isotopic provinces. The first two possibilities explain the range of proposed boundaries as an artificat of sampling, experimental, and statistical methods. If this is the case, with continued detailed work, the profusion of boundaries should converge on a single discrete province boundary representing the interface between crustal blocks with distinct geochronologic and isotopic properties. The third possibility is a suggestion that the range of proposed isotopic boundaries may, in fact, reflect real complexities of the boundary zone. More detailed isotopic work is needed to better understand the nature of the isotopic changes in the boundary zone and refine the positions of the various proposed boundaries. However, we suggest that a complex or gradational boundary zone should be considered as a viable hypothesis to explain discrepancies in proposed isotopic province boundaries.
Consistent differences in xenolith populations from different parts of the Navajo Volcanic Field (NVF in Fig. 1) in the Four Corners region have recently been interpreted as evidence of a lower-crustal petrologic boundary coinciding with the northern limit of the Yavapai-Mazatzal boundary zone (Selverstone et al., 1997b, in press). Xenoliths from diatremes in the northwest part of the volcanic field show evidence of hydration and deformation and include a wide variety of rock types including metasedimentary gneisses and eclogites. Some xenoliths from the northwestern diatremes preserve evidence of counterclockwise P–T paths (Tmax before Pmax) reaching maximum temperatures of ∼850°C and maximum pressures of ∼10 kbar. The southeastern diatremes contain only undeformed granulite xenoliths with no evidence of hydrous alteration or penetrative deformation. A few xenoliths from the southeast preserve fragmentary evidence for clockwise P–T evolution. Selverstone et al. (in press) interpret the differences between the xenolith populations as evidence of a transition between lower-crustal blocks with significantly different Proterozoic tectonic histories (Xe in Fig. 1). Moreover, they postulate that the observed range of rock types, hydration, deformation, and P–T history of the northwestern xenoliths suggest a north-dipping paleo-subduction zone in the Four Conrners area. A north-dipping Proterozoic subduction zone is consistent with xenolith and other evidence, but this interpretation is not unique given the inherently fragmentary nature of the xenolith record, the lack of reliable age constraints, and the unknown geometry of deep-crustal structures.
METAMORPHISM AND STRUCTURE
Karlstrom and Bowring (1988) defined the Yavapai and Mazatzal orogenic provinces based on contrasting deformational and metamorphic histories of fault-bounded tectonic blocks exposed in Arizona. The Yavapai orogenic province comprises 1.8–1.7-Ga juvenile crust deformed before about 1.7 Ga. The Mazatzal province is characterized by 1.7–1.6-Ga crust deformed ca. 1.66–1.60 Ga (Karlstrom and Bowring, 1988, 1993). These orogenic provinces correspond roughly to Silver's (1965, 1968) Yavapai and Mazatzal age provinces. The boundary between the provinces is marked by steep shear zones separating tectonic blocks, but does not necessarily coincide exactly with isotopic province boundaries (Karlstrom and Bowring, 1993; Karlstrom and Danniel, 1993). Instead, the Mazatzal deformation front extends beyond the 1.7–1.6-Ga supracrustal rocks into 1.8–1.7-Ga rocks more characteristic of the Yavapai province indicating that deformation was transmitted into older rocks of the foreland during continental assembly (Karlstrom and Danniel, 1993).
The Southern Rocky Mountains lie along the trend of the Proterozoic accretionary orogens from the Precambrian exposures of Arizona and may be correlative (Fig. 1; Silver, 1965; Karlstrom and Bowring, 1993; Karlstrom and Daniel, 1993). However, the boundary zone in the Rockies appears much more diffuse and lacks the steep foliation and sub-vertical shear zones typical of Precambrian exposures in Arizona. In the Southern Rockies, steep shear zones are concentrated in the northeast-trending Colorado mineral belt (CMB in Fig. 1; Tweto and Sims, 1963; Warner, 1978), which lies north of the inferred isotopic and geochronologic boundaries. The relationship between the isotopic boundaries of central and southern Colorado and the Colorado mineral belt is uncertain. The following sections and accompanying figures present compilations of metamorphic and structural data from the region of the boundary. These data are essential to unraveling the thermal and kinematic history of accretion and modification that may be responsible for the cryptic nature of the boundary zone.
A generalized compilation of metamorphic grade in the area of the boundary zone shows the inferred “peak” metamorphic conditions (maximum temperature) reported by a number of authors (Fig. 2). In central Colorado, no-one has yet investigated possible composite effects of superimposed metamorphic events, as has recently been documented to the north (Selverstone et al., 1997a; Shaw et al., in press) and south (Pedrick et al., 1998). The limited available data seem to reflect a predominance of amphibolite-grade conditions, typical of much of Colorado and New Mexico (e.g., Tweto, 1987; Grambling, 1988; Reed et al., 1993), although anomalous granulite-grade rocks occur in the Wet Mountains (Fig. 2; Brock and Singewald, 1968; Tweto, 1987) and northern Taos Range (Grambling et al., 1989; Pedrick, 1995; Pedrick et al., 1998).
Steep gradients in the conditions of peak metamorphism from greenschist and amphibolite to granulite grade in these areas are consistent with similar observed field gradients and inferred pluton-enhanced metamorphism elsewhere in the Southwest (Williams and Karlstrom, 1996). There is no obvious pattern to help define a tectonic boundary. However, the occurrence of granulites within an otherwise medium-grade metamorphic domain is consistent with differential uplift and tectonic juxtaposition of different crustal levels on large-scale structures (Reed et al., 1987). In the Wet Mountains, the Ilse fault zone (IF in Fig. 2), a subvertical, north-trending structure with Precambrian ancestry separates granulites on the west from amphibolite grade rocks on the east (Singewald, 1966; Tweto, 1987). It is possible that the Ilse fault played a role in juxtaposing granulite grade and amphibolite grade rocks, but the tectonic significance of the Ilse fault remains unclear. The paucity of quantitative thermobarometric data, systematic analyses of metamorphic overprinting, and the relation of metamorphism to deformation is a serious impediment to interpreting the metamorphic history of the proposed boundary zone. Further work in this area is needed.
The Mazatzal-age deformation front identified in Arizona may be extrapolated into the Rocky Mountain region (Fig. 1; Karlstrom and Daniel, 1993). A line representing the Mazatzal-age (ca. 1700–1650 Ma) orogenic front can be drawn between 1690–1650-Ma plutons that suffered significant synemplacement deformation and those that are relatively underformed (Fig. 3). South of the proposed deformation front, we observe penetratively deformed plutons and pluton margins. The Garrell Peak pluton (1663 ± 4 Ma) is strongly, but variably deformed and the 1672 ± 5 Ma Trout Creek pluton north of Salida is highly deformed only at its southern margin (Fig. 3; Boardman, 1976; Bickford et al., 1989; Shaw and Karlstrom, 1997). In the Needle Mountains, the ca. 1690-Ma Tenmile Creek and Bakers Bridge plutons crosscut early fabrics, but were deformed by post-1690-Ma tectonism that also affected the unconformably overlying Uncompahgre group (Gibson, 1987, 1990; Gibson and Simpson, 1988; Harris, 1990). This event may coincide with the 1.67–1.65-Ga ductile thrusting and imbrication of the Hondo group of New Mexico (Williams et al., this issue) and syncontractional emplacement of 1.65-Ga plutons in southern New Mexico (Silver, 1965; Karlstrom and Bowring, 1988, 1993; Bauer and Williams, 1994). Tectonism of this age has been correlated to the Mazatzal orogeny of Arizona (e.g., Karlstrom and Bowring, 1993; Karlstrom and Daniel, 1993). In contrast 1.65-Ga deformation does not seem to have strongly affected the area north of the proposed deformation front. For example, the 1676 ± 5 Ma Cochetopa granite (Fig. 3) is virtually undeformed (Bickford et al., 1989; Shaw and Karlstrom, 1997), and the primary phase of deformation farther north in Colorado probably occurred before 1.70 Ga (e.g., Barovich, 1986; Premo and Fanning, 1997). The northern limit of Mazatzal-age deformation is still poorly defined in the Southern Rockies, but this orogenic front may represent an important structural discontinuity.
Our preliminary analysis of foliation trends in southern Colorado and northern New Mexico suggest the regional development of an initially gently-dipping, west- to northwest-striking tectonic foliation (tentatively designated S1) usually parallel to primary compositional layering (S0). These early fabrics are variably overprinted by a steeper tectonic fabric (S2) in much of the area south of the Mazatzal front (Fig. 3; Shaw and Karlstrom, 1997). In less deformed areas S1 is well preserved. For example, foliations near Salida and in the northern Wet Mountains dip shallowly north (Fig. 3). Steeply dipping northeast- and northwest-striking fabrics are developed in some areas (e.g., south and northeast of Gunnison, Fig. 3), but seem to be the result of folding of S1 (Afifi, 1981). Early top-to-the-south shear-sense has been identified in the Taos, Cimarron (Pedrick et al., 1998), and Rincon ranges (Read et al., this issue) of New Mexico, but we have not yet identified sufficient kinematic indicators to confirm this vergence in central Colorado. In other areas steep northeast-striking S2 fabrics predominate. Steeply dipping, northeast-striking S2 foliations and related upright folds are developed in the central Wet Mountains, the northern Sangre de Cristo Mountains, Gunnison area, and Needles Mountains. The F2 folds and S2 foliation are interpreted to reflect a dominantly pure-shear northwest-southeast shortening of the region (D2) subsequent to S1 development. This pattern of shallow fabrics overprinted by steeper ones is similar to the pattern developed near the Yavapai-Mazatzal boundary in central Arizona (Karlstrom, 1989).
The superposition of these two dominant regional fabrics, as well as other locally important fabrics has created intricate composite structural patterns. A comparison of two areas in central Colorado illustrates the inferred regional deformation sequence and the southeast to northwest decreasing D2 deformation gradient (Fig. 3). In the central Wet Mountains (near the granulite exposures) complex interference patterns indicate at least three generations of superimposed folds (Fig. 3; Brock and Singewald, 1968; Lanzirotti, 1988). F2 is the dominant generation of fold with shallow to moderately northeast-plunging axes and steep northeast-striking axial planes consistent with northwest-southeast shortening. A steep S2 axial-planar cleavage is variably developed. Both northwest-trending F1 fold axes and axial planes are tightly folded by F2 folds. The closed figures formed by the traces of form-surfaces apparent in the map patterns of some folds are consistent with a hybrid type 2 (mushroom)–type 3 (convergent-divergent) interference pattern (Ramsay, 1967; Ramsay and Huber, 1987). This indicates that the axial planes of F1 folds were probably at a high angle to the subvertical flow vector for F2 folding (Ramsay and Huber, 1987) consistent with an initial low-angle axial plane for F1 folds and a 90° change in the direction of maximum compression between D1 and D2 (Lanzirotti, 1988). Originally shallowly northwest plunging F1 fold axes were rotated about northeast-trending axes into steep northwest and southeast plunging orientations (Fig. 3). The northeast fabrics of the central Wet Mountains are abruptly truncated to the east by the Ilse fault.
Near the proposed Mazatzal deformation front immediately north of Salida shallowly dipping, east-west-striking metavolcanic and associated rocks are relatively undeformed and preserve primary volcaniclastic and intrusive textures (Fig. 3; Boardman, 1976, 1980, 1986). However, as the margin of the 1672 ± 5 Ma Trout Creek pluton is approached the shallow fabrics in these rocks (S1) are progressively transposed into a sub-vertical, northeast-trending fabric parallel to magmatic and solid state fabrics within the pluton (S2). The deformation is clearly enhanced by proximity to the pluton, and we interpret the age of deformation to be about 1670 Ma.
We tentatively correlate the F2 folds and S2 foliation in the central Wet Mountains with the transposition of shallow (S0 and S1) fabrics into steep, northeast-trending fabrics north of Salida on the basis of trend and structural style. Both of these structures seem to be manifestations of the regionally extensive D2 deformation. If this correlation holds up it implies that D2 deformation occurred ca. 1670 Ma, coeval with the Mazatzal orogeny in New Mexico and Arizona, and also that a deformation gradient (or discontinuity) exists between the penetrative deformation of the central Wet Mountains and the limited, pluton-enhanced deformation north of Salida.
Our regional synthesis of structural data supports the following tentative kinematic model. (1) Initial thrust-sense deformation took place on a low-angle, layer parallel S1 foliation. If any high-strain shear zones developed during this phase they might be difficult to identify because they cut compositional and tectonic layering at a low angle and may have been pre-peak metamorphism. (2) S1 foliation was variably overprinted by a steep, northeast-striking S2 fabric and folded by F2 upright folds during a pure-shear phase of deformation, perhaps related to large-scale northwest-southeast crustal shortening after ca. 1690 Ma. The D2 deformation was largely limited to the area south of the Salida-Gunnison area (Mazatzal Front (MF) in Fig. 3).
YOUNGER PLUTONISM AND TECTONISM
In addition to the isotopic, xenolith, and structural data there are indications that the proposed boundary zone in southern Colorado has been an important tectonic and magmatic boundary at various times since the accretion of the continent (Karlstrom and Humphreys, 1998). For example, ca. 1.4 Ga, when regional “anorogenic” magmatism was dominantly granitic at exposed crustal levels, mafic plutons reached the middle crust in only two places–the Laramie anorthosite of the Cheyenne belt area and the Electra Lake gabbro of the Needle Mountains, within the proposed Yavapai–Mazatzal boundary zone, (Fig. 2). We speculate that older zones of weakness, related to the accretionary boundaries, facilitated emplacement of mafic rocks at middle crustal levels at these localities. Likewise at about 1.1 Ga (Grenville age), most of the magmatic activity was concentrated to the south in Texas and Arizona (Smith et al., 1997), yet the voluminous Pikes Peak pluton was emplaced at the northern margin of the proposed boundary zone (Figs. 2 and 3). Again in the Tertiary, calderas of the San Juan volcanic field penetrated the crust along this line (Fig. 2). Although the location and nature of Proterozoic structures remains uncertain, we speculate that at different times segments of the boundary acted both as zones of enhanced fertility for magmas and as conduits for magma emplacement, potentially because of early accretionary structures. A major center of anomalously low velocity mantle underlying the boundary zone (Fig. 2) may also be influenced by lithospheric architecture inherited from the assembly of the continent (Lerner-Lam et al., 1998).
Present physiography and mantle structure also suggest that accretionary structures may have a persistent influence. A transition from the closely spaced uplifts of the Colorado Rockies to the narrow and widely spaced ranges of New Mexico broadly coincides with the Yavapai-Mazatzal boundary zone. This transition echoes the change in style between the broad basins of the Wyoming Rockies and the denser Colorado Rockies that occurs across the Cheyenne belt (see Pazzaglia and Kelley, 1998, Fig. 1). Recent work has also identified important gradients in denudation history across the zone (Pazzaglia and Kelley, 1998), and the highest peaks in the Southern Rocky Mountains occur at the intersection of the Rio Grande rift and the Colorado Mineral belt immediately north of the Yavapai-Mazatzal boundary zone (Reed, 1996). All of these observations suggest that Proterozoic accretionary structures have had a persistent influence on later tectonism, although the mechanisms remain unclear.
SPECULATIVE MODEL FOR THE BOUNDARY
Despite the obvious need for more data, the evidence for a province boundary in southern Colorado or northern New Mexico seems moderately strong at the regional scale. Assuming that a boundary does exist, any model must either dismiss the disparity of proposed geochronologic, isotopic, and structural boundaries as an artifact of the limited data set and inherent limits of precision of the methods used, or explain it in terms of crustal geometry and geologic processes. We think it is most constructive to propose a testable working hypothesis that is consistent with the available data, expecting that the model will be modified or rejected as new information is discovered.
We suggest that the best explanation for differences in proposed province boundaries is a gradational or complex transition zone. This transition could be a laterally heterogeneous zone composed of a tectonic mosaic of small terranes permitting isotopic mixing at a scale below the resolution of isotopic methods, or a gradational change in crustal properties produced by a continuous process without distinct terranes or sutures. The lack of major shear zones associated with the isotopic boundaries argues against the mosaic model, which predicts abundant sutures. Furthermore, most plate tectonic models as well as geophysical and geologic observations suggest that large scale (> 1000 km) continental growth usually proceeds by the discontinuous accretion of discrete tectonostratigraphic terranes rather than by continuous processes (e.g., Hamilton, 1981, 1988; Jones et al., 1983; Hoffman, 1989).
Another possible model for a gradational transition is a vertically heterogeneous crustal section with crustal blocks juxtaposed on low-angle sutures and modified by later tectonism. As a working hypothesis, we tentatively propose that, in the Southern Rockies, terranes of the Mazatzal and Yavapai provinces were juxtaposed along a low-angle boundary that evolved from a north dipping subduction zone (Fig. 4). During collision, overthrusting of Mazatzal terranes may have been facilitated by midcrustal detachments in thermally weakened arc lithosphere (Fig. 4C). In this model the transition zone defined by the range of isotopic boundaries (Fig. 1) broadly coincides with a region of overlap between the Yavapai and Mazatzal provinces which constitute the hanging wall and footwall, respectively, of a crustal-scale thrust system (Fig. 4E). In our model, the surface expression of the suture lies in northern New Mexico coincident with Condie's (1982) supracrustal boundary. The postulated fossil subduction zone underlying the northern limit of the boundary zone in the Four Corners area might represent oceanic crust preserved at the intersection of the suture with the moho (Fig. 1; Fig. 4E; Selverstone et al., 1997b, in press). Moreover, we suggest the possibility that a mantle suture separating Yavapai and Mazatzal mantle lithosphere may continue dipping NW under Colorado (cf. Calvert et al., 1995). This mantle suture, and perhaps a similar southeast-dipping structure related to the Cheyenne belt subduction zone, may have influenced the development of Proterozoic shear zones and Laramide mineralization in the Colorado mineral belt (CMB, Fig. 1; Fig. 4E).
This model explains the disparity in proposed isotopic boundaries as a manifestation of vertical crustal heterogeneity (Fig. 4E). Different isotopic and geochronologic techniques sample different portions of the crustal column. The Nd boundary (Bennett and DePaolo, 1987) defined using plutonic rocks derived largely from the lower crust lies significantly farther north than the province boundary defined by the ages of supracrustal rocks exposed at the surface (Condie, 1981, 1982). Common Pb systematics are strongly influenced by the integrated crustal column through which the host magma has passed Hence, either Yavapai, Mazatzal, or intermediate Pb signatures might be expected in the boundary zone.
The lack of an obvious structural suture at the surface may be the result of initial near-layer-parallel thrusting creating a cryptic boundary between crustal blocks. Subsequent folding and/or imbrication of the boundary in response to continued crustal shortening might have locally steepened the boundary, compositional layering, and early tectonic foliations. The integrated effects of early low-angle thrusting and later pure-shear shortening may have produced a geometrically complex tectonic interfingering of Yavapai and Mazatzal blocks in the mid-crust. Anomalously high-grade metamorphic rocks, as observed in the Wet Mountains and Taos Range, could be parts of the higher-grade hanging wall preserved as klippen above a mid-crustal detachment surface (Fig. 4E). Depending on the initial geometry of ramps and flats in the subduction zone, and the nature of post-accretion deformation in the middle crust, different rheological layers may have acted as detachments, and the suture zone may step down through the lithosphere. If so, reactivation of different segments of the suture in the mantle, lower crust, or middle crust at different times could have focused magmatism or deformation in different parts of the transition zone (Fig. 4E).
Although our model is highly speculative it has the following advantages. First, it provides a coherent explanation for the discrepancy in province boundaries inferred from different methods. Secondly, it is consistent with structural observations. Thirdly, it proposes a framework for understanding Proterozoic accretionary tectonics that is consistent with modern, low-angle convergent margin structures (Hamilton, 1981, 1988) and seismic images of other accretionary boundaries (e.g., Calvert et al., 1995). Finally, it presents a testable model cross-section for future geophysical and geologic investigations.
IMPLICATIONS FOR STYLE OF SUTURING
A comparison of the cryptic Yavapai-Mazatzal boundary in the Southern Rocky Mountains and the more obvious Archean-Proterozoic suture at the Cheyenne belt reveals two fundamentally different structural styles of suturing. It may also be that the nature of the Yavapai-Mazatzal boundary varies significantly along trend, with fabrics and bounding structures chanigng from steep (in Arizona) to shallow (in Colorado). These observations demonstrate that tectonic sutures within continental crust are both complex and diverse. In addition, the differences between the Yavapai-Mazatzal and Cheyenne belt accretionary boundaries suggest that the lithospheric properties of the colliding blocks may partly control the structural style of accretionary orogenes. During collision, the rifted cratonic footwall of the Cheyenne belt may have acted as a tectonic backstop dictating the geometry of thrusting and promoting the development of a discrete suture zone characterized by steep fabrics. In contrast, for the Yavapai-Mazatzal boundary, the lack of a rigid cratonic footwall and high heat flow related to arc magmatism may have permitted a low-angle suture to develop. Overthrusting of one arc terrane onto another and penetrative pure shear shortening of the boundary could have been facilitated by a thermally weakened middle crust.
If our preliminary model of the Rocky Mountain segment of the Yavapai-Mazatzal boundary as a folded, initially low-angle suture that is detached in the mid-crust is correct, it would imply that the strength of the mid-crust is an important control on tectonic style and that sutures in arc-arc collision zones may be wide zones of tectonic and chemical mixing of adjacent terranes. This would have important implications for modeling the geodynamic processes of accretion, for understanding suspect terranes, and for evaluating the effects that accretionary structures may have on latter tectonism and plutonism.
Research for this paper was supported by NSF Continental Dynamics Program grant number EAR-9614787 (Karlstrom) and a Kelly-Silver graduate research fellowship from the Caswell Silver Foundation (Shaw). Helpful reviews by Christine Siddoway and John Geissman greatly improved the manuscript.
- Received February 9, 1998.
- Revision received March 13, 1998.
- Accepted April 22, 1998.