Abstract

The Laramide orogen of the U.S. Cordillera formed in the latest Cretaceous, and deformation lasted into the earliest Oligocene. Along and proximal to the eastern and northern margins of the Colorado Plateau, deformation associated with this event mainly took place along reactivated structures. Related tectonic models invoke some role for the plateau either as a stress guide transmitting compression to the foreland or as a freely rotating microplate. Models dominated by northward displacements of the Colorado Plateau also require covariance between timing and magnitude of dextral strike-slip deformation in the eastern domain and thrust deformation in the northern domain. Here we show that fault-zone materials that are exposed in a major, large-magnitude-displacement strike-slip fault zone east of the plateau contain a well-defined magnetization of late Paleozoic age, suggesting that the fault zone has not been strongly modified since the late Paleozoic. Given that these fault-zone materials include indurated metagranitic crush breccias that must have been at or near the surface at the onset of Carboniferous sedimentation, it is likely that the observed large-magnitude displacements are the result of a poorly understood Precambrian tectonic event. Large-magnitude dextral-slip estimates along this and similar structures may be incorrectly assigned to younger tectonic events. In this context, Laramide strain estimates north of the plateau should not be linked with these older displacements and may instead have resulted from a complex combination of Laramide plateau rotation and general east-directed shortening associated with the formation of the Sevier fold-and-thrust belt salient.

INTRODUCTION

Tectonism along the eastern margin of the Colorado Plateau of the southwest United States has long been thought to be associated with plateau rotation (Hamilton, 1981) and reactivation of bounding structures that first formed during the Neoproterozoic (Tweto and Sims, 1963). Understanding the displacement history of these structures requires integrating fault kinematics with assessments of the timing and magnitude of reactivation during the Precambrian and three importantPhanerozoicevents:thePennsylvanian to Permian Ancestral Rocky Mountain orogeny (Kluth and Coney, 1981; Budnik, 1986); the latest Cretaceous to earliest Oligocene Laramide orogeny (Dickinson and Snyder, 1978; Hamilton, 1988; Constenius, 1996; Erslev, 1999, 2005; Wawrzyniec et al., 2002); and Neogene extension (Woodward, 1977; Chapin and Cather, 1994; Wawrzyniec et al., 2002). Central to the evaluation of deformation during these time periods is the relative importance and timing of strike-slip faulting within the Cordilleran foreland. The Picuris-Pecos fault system, exposed in north-central New Mexico (Fig. 1) has the largest demonstrable strike-slip offset of any structure in the Cordilleran foreland, with a cumulative dextral offset of 37 km based on piercing points in Precambrian rocks (Miller et al., 1963; Karlstrom and Daniel, 1993, 1994; Woodward, 1994; Fankhauser and Erslev, 2004), and thus it plays a very crucial role in this evaluation. Field relations indicate repeated reactivation, but no field-based observations have been reported that can convincingly demonstrate that the major strike-slip event is younger than Precambrian in age. To refine estimates of the age of the large-magnitude displacement on the Picuris-Pecos fault system, we have obtained paleomagnetic data from several rock types exposed within the shear zone that clearly formed in response to displacement along the fault system. If these deformed rocks have a well-defined characteristic remanence (ChRM) that can be ascribed to a particular time period, then the age of this material, and thus the major faulting event, can be no younger than the age of the ChRM.

The accurate determination of the age of displacement on the Picuris-Pecos system bears on the tectonics of the Colorado Plateau and adjacent regions, in particular in the context of a long-standing debate concerning the net northward translation of the Colorado Plateau. Based on similar data sets (e.g., Cenozoic basin geometry, isopach trends of Mesozoic strata, magnetic anomaly patterns), estimates of the net dextral offset during the Laramide orogeny between the Colorado Plateau and cratonic North America range from ∼20–41 km (Woodward et al., 1997; Yin and Ingersoll, 1997; Woodward, 2000; Wawrzyniec et al., 2002) to 55–170 km (Chapin and Cather, 1981, 1983; Karlstrom and Daniel, 1993; Cather, 1999). More recently, a cut-and-paste correlation of magnetic anomaly patterns in northern New Mexico, which are assumed to reflect basement magnetic character, has been used to construct an estimate of 50–90 km of combined Laramide and Ancestral Rocky Mountain dextral offset east of the plateau (Cather et al., 2006). This estimate, as well as others that call for larger magnitudes of offset, also invokes direct balancing of shortening along Laramide structures north of the plateau with lateral-displacement estimates from the transpressional eastern margin of the plateau (Cather et al., 2006), and thus they call on large, (tens of kilometers) Laramide-age, dextral slip on north-south–trending foreland shear zones (Karlstrom and Daniel, 1993; Cather, 1999; Cather et al., 2006).

GEOLOGIC SETTING

The Colorado Plateau represents a major physiographic province within the western Cordillera that can be characterized as a region of relatively modest deformation when compared to the amount of deformation that has taken place at its margins. Previous hypotheses have suggested (1) that foreland deformation immediately east and north of the plateau may have been associated with plateau displacement described by a modest vertical-axis rotation defined by either a proximal (Hamilton, 1981) or a more distal Euler pole (Hamilton, 1988; Molina-Garza et al., 1998; Cather, 1999; Wawrzyniec et al., 2002) located east of the Colorado Plateau near a latitude of 34°N and longitude of 105°W and 100°W, or (2) that plateau rotation occurred as a consequence of a northward-increasing displacement field (Erslev, 2005). Specifically, Euler pole rotation models require that the cumulative magnitude of shortening north of the Colorado Plateau be balanced with more oblique shortening and strike-slip offset observed within the eastern margin of the plateau. In all cases, foreland deformation is placed in the context of an integrated strain field extending from Wyoming to central New Mexico.

The Picuris-Pecos fault system is a major zone of dextral, strike-slip offset within the eastern margin of the plateau that shares a common N-S trend with several basement structures in this part of the Cordilleran foreland (Miller et al., 1992). In the southern Sangre de Cristo Mountains (Fig. 1), Deer Creek Canyon exposes a complex zone of deformation that merges with the Picuris-Pecos fault to the north and consists of well-exposed, indurated, metagranitic crush breccias and shattered rocks up to 100 m wide juxtaposed against folded but much less penetratively deformed late Paleozoic strata across high-angle, gouge-filled faults and lower-angle, conformable contacts. Locally, the breccias are cut by small carbonate veins (Fig. 2A) and by subvertical, locally brecciated, oxidized, hematite-bearing, tabular carbonate bodies up to 25 m long that cut obliquely through the shattered granitic rocks of the Dear Creek fault (Fig. 2B). These carbonate bodies have been explained as rotated clasts derived from previously overlying limestone beds or as carbonate fissure-fill deposits (Fankhauser and Erslev, 2004). If these tabular, carbonate features are fissure-fill deposits, these deposits may mark the earliest phase of preserved Paleozoic deposition, which is associated with a regional nonconformity between eroded Precambrian rocks and overlying Carboniferous limestones. In addition to fault-related rocks within the zone of cataclasis, the red bed–dominated Pennsylvanian-Permian strata of the Sangre de Cristo Formation and underlying limestone-dominated Pennsylvanian Madera Formation are folded and tilted into a syncline and provide additional sampling opportunities that may further demonstrate the antiquity of deformation of these strata.

METHODS

Oriented samples for paleomagnetic analysis were collected from all types of fault-zone rocks, as well as from late Paleozoic strata (carbonates and red beds) with a range of bedding orientations, in order to assess the possibility of localized vertical axis rotation of rocks adjacent to the fault zone that might have accompanied strike-slip deformation. Specimens from independent samples were prepared as cubes cut from oriented blocks or cylinders cut from cores drilled in the field. Each site consisted of at least seven specimens from seven cores or 30 specimens derived from four to eight blocks. After several low-temperature, liquid nitrogen demagnetization (Dunlop and Argyle, 1991) steps, specimens were subjected to progressive thermal demagnetization up to ∼680 °C, because alternating field demagnetization failed to randomize more than a few percent of the natural remanent magnetization (NRM). Demagnetization results were interpreted using conventional means; directions of specimen magnetizations were determined using principal component analysis (Kirschvink, 1980). Site means and associated statistics were determined using spherical statistics (Fischer, 1953; Onstott, 1980).

To address the issue of strain balancing between deformation north and east of the plateau, we generated a model of plateau motion based on the present-day outline of the Colorado Plateau (Fig. 3). The trace was picked in ArcGIS©2006 and converted to point values in units of longitude and latitude and ported into Stereonet v.6.3.2 (Allmendinger, 2004). The points were then rotated about various Euler pole axes of rotation to remove the effects of Cenozoic foreland deformation and to determine late Mesozoic paleopositions for the plateau. The rotated data were then imported back into ArcGIS©2006, which provided a basis for comparative analysis to determine the finite particle-motion trajectories produced by plateau rotation. Such a reconstruction assumes that there is only one Euler pole for Laramide-age plateau rotation and that deformation within the plateau is insignificant when compared to the deformed margins. In this context, we consider the trajectories to represent possible modeled finite shortening directions associated with Laramide orogeny and assume that they are reasonable approximations of the associated modeled incremental shortening between the plateau and the North American craton.

PALEOMAGNETIC RESULTS

Interpretable paleomagnetic data were obtained from 199 samples derived from 17 sites in Paleozoic sedimentary rocks, 7 sites in tabular, nonvein, depositional carbonates hosted by brecciated Precambrian metagranite, and 2 sites from the brecciated metagranite 01(Table 1). Fault-zone materials showed a range of responses to progressive demagnetization (Fig. 4). The tabular carbonate material yielded the most internally consistent paleomagnetic results, and ChRM values were well-grouped at the site level and typically showed southeast declination and shallow negative inclination that was readily isolated in thermal demagnetization above ∼600 °C. Thermal demagnetization of three-component isothermal remanent magnetization (IRM) (Lowrie, 1990) verified the principal magnetic phase in these rocks as hematite. IRM acquired in fields of 0.03 T and 0.3 T is always less than ∼30% of that acquired by the 3.0 T component. The lower-field IRM components showed little if any inflection over the ranges of laboratory unblocking temperatures that would be consistent with the presence of lower-coercivity magnetic phases (Fig. 5). In many carbonate rocks, the remanence is typically carried by magnetite (McCabe and Elmore, 1989); the dominance by hematite as the principal remanence carrier in these rocks suggests that an oxidizing, iron-bearing fluid was responsible for secondary magnetization acquisition in these rocks (Elmore et al., 1999).

Two sites from the granitic breccia, all sites in the tabular carbonates, and most sites in sedimentary rocks provided interpretable data. In situ results from all sites yielding interpretable data can be summarized by a grand mean with declination of D = 138.3°, inclination of I = 3.8°, α95 = 12.4°, k = 8.3, and N = 19 of 25 total sites (Fig. 6A). For sites from rocks with discernible bedding (Pennsylvanian-Permian strata and some sandstone inclusions within the breccia), a tilt-corrected mean, using the present strike axis, of D = 140.1°, I = −5.5°, α95= 11.0°, k = 20.1, and N = 11 of 14 total sites (Fig. 6A) was obtained. Data from sites in seven spatially distinct carbonate fissure fills in granitic breccia yielded an in situ mean direction of D = 130.9°, I = −18.7°, α95 = 7.5°, k = 74.9, and N = 7 of 7 sites accepted (Fig. 6A). A plausible structural correction for the result from the carbonate fis-sure fills involved restoring the fissures to vertical, and this yielded a mean of D = 124.6°, I = − 19.5°, α95 = 7.2°, k = 82.2, and N = 7, which is statistically indistinguishable from the in situ result. With the exception of one site in granite breccia, all sample/site ChRM directions showed reverse polarity and, after tilt correction, southeast declination and shallow, negative inclination. The site-mean directions provided virtual geomagnetic poles that are similar to paleomagnetic poles that define the late Paleozoic (Pennsylvanian to Permian) apparent polar wander path of North America (Fig. 6B).

KINEMATIC MODEL RESULTS

Results from Euler pole rotation calculations can be summarized in the context of two end-member cases for plateau rotation. A synthesis of paleomagnetic data collected from pre-Cenozoic strata on the Colorado Plateau and comparison with comparable age data for the North American craton and eastern seaboard suggest a maximum of 5–8° clockwise rotation of the plateau about any Euler pole east of 105°W and near 34°N (Molina-Garza et al., 1998). However, the farther to the east the pole is located, the greater the magnitude of dextral shear that is required along the eastern margin of the plateau (Molina-Garza et al., 1998). This concept is demonstrated by comparing two Euler poles: one located proximal to the plateau at 105°W, 34°N, and one located ∼600 km east of the plateau at 100°W, 34°N (Fig. 3). A clockwise, 5–8° rotation about the proximal pole results in 17–27 km of dextral shear east of the plateau (within north-central New Mexico) and 64–114 km of concomitant NE-directed shortening north of the plateau. A similar rotation about the more distal Euler pole results in 48–77 km of dextral shear east of the plateau and 87–140 km of NE-directed shortening to the north.

DISCUSSION

Map relations involving Mesoproterozoic and younger rocks in the southern Sangre de Cristo Range imply that slip (and reactivation) along the Picuris-Pecos fault system is poorly constrained to have taken place between ca. 1.4 Ga and ca. 35 Ma (Karlstrom and Daniel, 1993, 1994; Woodward, 1994). The fault zone is thought to have been reactivated during every major tectonic event that affected this part of the Cordillera (Erslev et al., 2004; Cather et al., 2005, 2006). We interpret the ChRM values from tabular carbonates to be of late Paleozoic age (ca. Pennsylvanian to Early Permian), acquired during the Permian-Carboniferous reverse superchron (Opdyke and Channel, 1996; Opdyke et al., 2000), which spanned the time period between ca. 318–316 and 262 Ma. This age assignment for the ChRM is critical to our assessment of the age of the youngest possible phase of substantial fault offset, because, on the basis of geometric relations with host granitic breccias, formation of the tabular carbonates took place when most deformation within the fault zone had ceased. Although carbonate within these tabular bodies is locally brecciated, these bodies have a subvertical orientation that trends to the NW, obliquely across the N-S trace of the breccia zone (Fankhauser and Erslev, 2004). Notably, the carbonate bodies are cut by minor faults with small dextral separations similar to gouge-filled, planar fault surfaces that crosscut the early breccia fabric.

The ChRM directions from late Paleozoic Madera and Sangre de Cristo Formation strata are also inferred to be of late Paleozoic age (Fig. 6) and demonstrate that the area has not experienced appreciable, local-scale, vertical-axis rotation in response to faulting. Also, given the overall agreement between the in situ and tilt-corrected results, it is possible that some of the tilt of the Phanerozoic strata pre-dated magnetization acquisition in these materials. Unfortunately, the exposures of late Paleozoic strata along the Deer Creek locality do not lend themselves to a proper fold test to further evaluate this possibility. Although these data do not directly bear on the age of the youngest phase of appreciable slip on the fault system, we note that these rocks do not contain any additional well-defined and internally consistent magnetizations that can confidently be ascribed to a younger age, and thus be associated with a younger phase of faulting. We also note that magnetizations of similar, late Paleozoic age have been identified in several types of Precambrian crystalline rocks in the general southern Rocky Mountain area (Geissman and Harlan, 2002) and have been interpreted as having a chemical origin indicative of low-temperature fluid migration in response to deformation during the Ancestral Rocky Mountain orogeny.

Our understanding of the western U.S. Cordilleran foreland is largely based on studies of processes related to regionally NE-directed shortening during the Laramide orogeny, which affected areas north and east of the Colorado Plateau, and subsequent regional extension, including the formation of the Rio Grande rift, which approximately defines the eastern margin of the plateau. Most regional models consider the plateau as a relatively undeformed feature that, since the mid-Cretaceous, has played some role in transferring strain over 1000 km inboard from the southwestern margin of North America, where, during the latest Cretaceous to early Cenozoic, parts of either the Kula or the Farallon plates were being subducted to the northeast relative to North America (Engebretson et al., 1984; Stock and Molnar, 1988), although the location of their plate boundary remains subject to debate (Bunge and Grand, 2000). In this context, our interpreted age of the large-magnitude offset recorded along the Picuris-Pecos fault and the paleomagnetic results from the Deer Creek locality have direct bearing on models that address latest Cretaceous to early Cenozoic evolution of this part of the western Cordillera.

The presence of exclusively late Paleozoic magnetizations in the tabular carbonates and the overall agreement between these results and those from adjacent late Paleozoic strata (Fig. 1; 01Table 1) support the hypothesis that no significant deformation event since the late Paleozoic has remagnetized or greatly modified the magnetic character of the fault-zone materials. We do know that the fault system was reactivated during the Laramide orogeny, resulting in vertical separation to form an inverted monocline with Triassic strata (Fankhauser and Erslev, 2004) exposed south of Deer Creek. However, this deformation does not require a large lateral motion or large fault reactivation beyond that required to form the monoclinal flexure. Although an exact estimate of structural relief cannot be demonstrated, such flexure can result from small offsets (tens to hundreds of meters), and deformation may not be associated with a chemically reactive fluid front that would have affected the rocks found along this segment of the Picuris-Pecos fault zone. Such an interpretation is supported by other studies that indicate that large fault zones tend to act as fluid conduits during and after deformation (e.g., Knipe, 1993; Blumstein et al., 2005). The Moine thrust of NE Scotland is one example where paleomagnetic analysis of fault-related rocks has demonstrated that a large, regional-scale structure has on multiple occasions served to transmit fluids during regional tectonic or local igneous activity (Blumstein et al., 2005). Given the regional scale of the Picuris-Pecos fault system, it seems untenable that any young deformation event that produced large strains would have left no magnetic signature in the fault-zone material of the Deer Creek fault. We interpret these data to suggest that the youngest phase of large-magnitude lateral slip along the Picuris-Pecos fault system took place during or before the Ancestral Rocky Mountain orogeny (Pennsylvanian to Permian; Miller et al., 1992b). Moreover, because the estimated cumulative dextral slip along this fault system is based on piercing points of rocks of Mesoproterozoic age and the observation that the unusually well-indurated basement fault-zone materials must have been very near the surface during the Pennsylvanian-Permian Ancestral Rocky Mountain orogeny, it is entirely possible that the age of most of the dextral slip is Precambrian. This conclusion can be further tested by fault-scaling relationships and regional kinematic framework based on Euler pole geometries that describe the interaction between the plateau and the Cordilleran foreland.

Scaling relationships between displacement and fault length (Cowie and Scholz, 1992; Schultz and Fossen, 2002) require that a fault with ∼37 km of observed displacement during a single phase of tectonism must have an overall length of some 200–400 km. Such relationships principally apply to structures with known end points of zero displacement and likely apply in this case. The known surface trace of the Picuris-Pecos fault system is ∼100 km along a N-S strike across northern New Mexico. We envision two end-member possibilities to explain the apparent discrepancy between fault trace length and the observed large dextral offset. First, the offset observed within Precambrian basement rocks reported by Karlstrom and Daniel (1993) was associated with a Precambrian event, and the related shear zone is partially obscured by Phanerozoic cover rocks. In this case, Laramide lateral offset was insignificant and associated with a small, reactivated fraction of a much larger shear zone. Alternatively, the observed lateral offset is younger and was transferred to the visible trace of the Picuris-Pecos fault by a series of fault interactions along the entire eastern margin of the Colorado Plateau (e.g., Cather et al., 2006). In the case of the latter hypothesis, all major dextral faults should demonstrate clear transfer geometries and an overall absence of slip gradients unless the faults interacted strictly as lateral fault stepovers, where major fault zones spatially overlapped but did not merge. If the faults interacted as stepovers, then each individual strand should show slip gradients from zero offset at the tips to maximum offsets near the middle of the trace, and scaling relationships should apply. If the faults interacted as releasing or shortening bends serving as slip terminations or relays to other structures, then the fault-bend evolution should include a significant amount of structural relief (either extended or shortened depending on geometry), and/or distributed pull-apart basin deposits (e.g., Wakabayashi et al., 2004).

In the context of the Picuris-Pecos fault, the southern end could connect with the SW-NE–trending Tijeras fault zone to form a right-stepping or bend geometry. However, although this region shows a complex pattern of deformation and some volcanic activity of appropriate age, there is no field-based evidence to indicate major NW-directed extension associated with large-scale structural relief (e.g., core-complex formation). In the case of lateral fault stepovers, all of the principal zones of offset should show demonstrable slip gradients, and thus the scaling relationships should then apply. In other words, if the lateral offset on the Picuris-Pecos fault system is younger than the Ancestral Rocky Mountain orogeny, then the scaling relationships must apply, and we are left with a discrepancy between the offset estimate and the known length of the structure. We do note that the Picuris-Pecos fault may have been truncated by Neogene rifting to the north, but if the scaling relationships apply, then the system must extend out of the rift in southern Colorado. To date, no structure of comparable dextral offset has been reported in Colorado. Thus, we must conclude that the large dextral offsets within the basement rocks must pre-date the Laramide orogeny (if not also the Ancestral Rocky Mountain orogeny). All of this evidence is consistent with 40Ar/39Ar thermochronology data derived from the southern Sangre de Cristo Range that demonstrate that crystalline basement rocks were uplifted to within a few kilometers of the surface between 1000 and 800 Ma (Sanders et al., 2006).

Finally, we must address the concept of an integrated strain field within the Cordilleran foreland. Since hypotheses considering the possibility of Laramide-age plateau rotation were first postulated (e.g., Chapin and Cather, 1981; Hamilton, 1981, 1988), it has been largely assumed that shortening estimates north of the plateau must balance with dextral translation east of the plateau. Cather et al. (2006, p. 318) claimed that the requirement for strain balancing is the “most robust” argument supporting an interpretation of large, Laramide dextral offsets east of the plateau. Regardless of the validity of such an argument, kinematic, plate-scale modeling of plateau motion demonstrates that the magnitude of dextral shear east of the plateau required to balance with shortening north of the plateau is largely a function of the position of the Euler pole and the amount of allowable plateau rotation (Fig. 3). In this context, we can examine the range of estimates to evaluate the relative importance of plateau rotation and how it may or may not relate to an integrated strain field in the Laramide foreland.

Present estimates of dextral shear east of the plateau range from 20 to 35 km (Woodward et al., 1997; Wawrzyniec et al., 2002) to some ∼55–95 km (Cather, 1999; Cather et al., 2006), where the latter value incorporates an estimated 37 km of proposed Cenozoic slip on the Pecos-Picuris fault system. Both groups of estimates necessarily include any lateral offset related to the formation of the Rio Grande rift, which is largely a Neogene tectonic feature that has accommodated a variable amount but on average ∼9 km of E-W–directed extension along its entire length and may be associated with an estimated 7 km of rift-related, northward translation of the plateau (Wawrzyniec et al., 2002). This estimate of northward translation is largely based on the inversion of fault kinematic data to calculate plateau motion, but the faults used may only represent part of the extensional history of the rift. Conversely, Cather et al. (2006) prefer a sinistral component of Colorado Plateau/craton interaction during rifting (Chapin and Cather, 1994; Cather et al., 2006), but such a sense of slip has never been verified to be pervasive or dominant along the entire length of the rift. Recent and ongoing studies of our own seem to support long-lived, E-W– (Erslev, 2001) or W-NW– to E-SE–directed extension south of central Colorado and more SW-directed extension north of central Colorado. In other words, the major rift basins are, as previously argued by Aldrich et al. (1986) and Lewis and Baldridge (1994), associated with at least a minor component of dextral shear along the N-S–trending Rio Grande rift.

Given the current uncertainty regarding the magnitude and style of strike-slip deformation along the rift in Cenozoic time, we prefer to discount the relative importance of any Neogene-age lateral slip between the plateau and the North American craton. Assuming a considerably older, late Paleozoic to Precambrian age of dextral offset along the Pecos-Picuris fault zone, then the net prerift, late Cretaceous to Cenozoic offset along the eastern margin of the plateau is limited to between 20 and 41 km and would appear to be distributed across several structures. If we assume a proximal Euler pole with a maximum of ∼4° of clockwise rotation (Molina Garza et al., 1998; Cather, 1999), then the predicted balancing strains north of the plateau would be 10–15 km of dextral shear east of the plateau (within north-central New Mexico) and 30–60 km of concomitant NE-directed shortening north of the plateau. A similar rotation about the more distal Euler pole results in 25–40 km of dextral shear east of the plateau and 45–70 km of NE-directed shortening to the north. Using the same reasoning applied to the recent estimate of 55–95 km of northward translation of the Colorado Plateau during the Laramide orogeny (Cather et al., 2006), the distal pole would generate 179 km of NE-directed shortening north of the plateau associated with an ∼8° clockwise rotation, and the proximal pole would generate at least 400 km of NE-directed shortening associated with an ∼30° clockwise rotation. Rotations of this magnitude are testable by paleomagnetic techniques, and the only rotations approaching these magnitudes are found within fault-bounded blocks within the eastern margin of the plateau (Wawrzyniec et al., 2002), and such large rotations of the plateau remain untenable. Also, careful construction of balanced cross sections using appropriate seismic lines and other subsurface data (Stone, 1993) has demonstrated that the maximum shortening strains on thrust faults found north of the plateau, which are generally thought to be Laramide foreland structures, have no more than ∼70 km of NW-directed shortening. Therefore, it would seem that if the strain fields north and east of the plateau are integrated about a single Euler pole, then the amount of Laramide dextral shear between the plateau and the craton must be less than ∼14–38 km, regardless of which Euler pole is used to model the rotation of the plateau. Such a conclusion assumes of course that all of these structures have a history of NW-directed shortening. If the Everging, Sevier salient has driven any of the strain north of the plateau, it would result in more easterly vergence on foreland structures, and the derived estimate based on our analysis of Stone (1993) would not necessarily apply. More importantly, a single Euler pole for plateau rotation cannot be used to describe the trends and magnitudes of strain throughout the foreland.

CONCLUSIONS

Structures that were reactivated in association with motion of the Colorado Plateau relative to the foreland since the mid-Cretaceous have played a clear role in accommodating large strains within the Cordilleran foreland. Therefore, it is critical that the timing of documented displacements be well understood before processes related to such deformation are fully comprehended. The presence of a well-defined and well-grouped magnetization of late Paleozoic (Pennsylvanian to Permian) age within fault-zone materials of the Picuris-Pecos fault is consistent with relatively small, if not negligible, post–Ancestral Rocky Mountain orogeny displacement along this fault. The nonconformity at the base of the Carboniferous section and the well-exposed metagranitic crush breccias along Deer Creek imply that the fault has an important Precambrian history. Taken together, it is likely that most of the observed strike-slip motion along this structure is Precambrian in age, and the last significant chemically reactive front that affected these rocks took place during the Ancestral Rocky Mountain orogeny. This conclusion strongly favors hypotheses calling on limited Laramide dextral slip east of the plateau generated by the oblique convergence of the plateau on the craton, not on slip along discrete large-magnitude strike-slip faults like the Picuris-Pecos fault. The inadequacy of the Euler pole models to predict distal Laramide displacements suggests that the Colorado Plateau was caught up in a heterogeneously shortening orogen and was not the primary cause of foreland shortening. If this hypothesis is acceptable, then we must reconsider the nature of the integrated strain field during Laramide tectonism. An alternative hypothesis, for example, is that the Sevier fold-and-thrust belt salient in part directly drives shortening in the foreland of Wyoming north of the Colorado Plateau, and thus this strain field is the product of larger Cordilleran deformation. Meanwhile, the eastern margin of the plateau may more strongly reflect deformation directly coupled to ENE plateau motion. Such a case provides an explanation for reports of E-SE–directed shortening for the Hogback thrust north of the Uintas (Bradley and Bruhn, 1988) and NW-SE–directed shortening reported for some uplifts in the northern plateau (Bump and Davis, 2003). It is also compatible with most hypotheses that describe how regional foreland deformation occurred (e.g., convergence linked to multilevel crustal detachment [Erslev, 2005], gravitational spreading of the Sevier salient [Livaccari, 1991], subduction of an oceanic plateau during low-angle subduction [Livaccari et al., 1981; Saleeby, 2003], etc.). Most, if not all, of these hypotheses should be testable, but it will require a concerted effort to more adequately determine the magnitude and timing of slip events along reactivated structures of the Cordilleran foreland. These efforts should improve our understanding of Cordilleran foreland tectonics and address some of the many complexities related to tectonic terrains dominated by multiply reactivated shear zones.

We thank an anonymous reviewer and Ronald Bruhn for thoughtful reviews of this manuscript. We also thank the University of New Mexico Paleomagnetism Laboratory, and the Department of Earth and Planetary Sciences for financial and material support in completing this research. The lead author also thanks Ben Swanson and Amy Ellwein for invaluable assistance developing the Euler pole modeling approach.