Detrital zircon U-Pb geochronology and Hf isotope geochemistry of the Roberts Mountains allochthon: New insights into the early Paleozoic tectonics of western North America

The Roberts Mountains allochthon (RMA) consists of internally deformed Cambrian through Devonian rocks, and structurally overlies coeval passive margin strata in northeastern and north-central Nevada (Schuchert, 1923; Kay, 1951; Roberts et al., 1958; Madrid, 1987; Burchfiel et al., 1992) (Figs. 1 and 2). Roberts Mountains allochthon rocks include chert, argillite, arenite, quartzite, limestone, and mafic volcanic rocks. The RMA is often thought to have been deposited in an ocean basin outboard of coeval passive margin strata in western Laurentia and to have been tectonically emplaced onto this margin during the Late Devonian to Early Mississippian Antler orogeny (e.g., Roberts et al., 1958; Burchfiel and Davis, 1972; Madrid, 1987). Various workers have suggested wildly disparate sources for the RMA strata. Some workers (e.g., Roberts et al., 1958; Burchfiel and Davis, 1972; Poole et al., 1992) suggested that the RMA strata originated in western Laurentia (Fig. 1) and deposited in an ocean basin to the west. Speed and Sleep (1982) hypothesized that the RMA strata are the accretionary prism of a far-traveled arc. Gehrels et al. (2000a) proposed that the RMA originated in the Peace River Arch region of western Canada. Wright and Wyld (2006) suggested that the RMA was deposited as far afield as Avalonia or Gondwana and subsequently was tectonically transported to western Laurentia along its southern margin (Fig. 3). Colpron and Nelson (2009) proposed that RMA strata could have originated in the northern Baltica–southern Caledonides region and been tectonically transported along the northwest margin of Laurentia (Fig. 3). Determining the provenance of the RMA units will unravel this puzzle and provide new insight into early Paleozoic tectonics in the western Cordillera.

The gaps in understanding about the RMA strata—their provenance, sedimentary transport to depositional basin, and possible subsequent tectonic transport—can be addressed using detrital zircon analyses. We analyzed detrital zircons to obtain both uranium-lead ages and hafnium isotope ratios. U-Pb ages are important for identifying and then characterizing the provenance of sedimentary strata, and for comparison between sedimentary units (Gehrels et al., 2000b; Fedo et al., 2003; Gehrels, 2012, 2014). Hafnium isotope compositions are used to determine the geochemical character of the magma in which the zircons crystallized. When combined with U-Pb ages, Hf isotope composition provides a powerful complement for interpreting sedimentary provenance (Bahlburg et al., 2011; Gehrels and Pecha, 2014).

In this study, we determined the U-Pb ages and Hf isotope compositions of detrital zircons in six samples of RMA strata in north-central Nevada. We use these data to interpret provenance, sedimentary transport to depositional basins, possible subsequent tectonic transport, and relationships between RMA units. Our study builds on an earlier analysis of RMA samples that determined U-Pb ages using isotope-dilution–thermal ionization mass spectrometry (ID-TIMS) (Gehrels et al., 2000a, 2000b). Using detrital zircons from the same samples, the original data set was enlarged and enhanced. We analyzed a significantly larger number of grains per sample, changed and updated grain selection methods, and added Hf isotope composition analyses. We used laser-ablation–inductively coupled plasma mass spectrometry (LA-ICPMS) for all analyses. We report here 1151 new U-Pb ages and 228 new Hf isotope analyses.

Detrital zircon analyses allow us to resolve the original sources of these units. We show that some RMA units in some cases share an origin, while others units do not.

The North American craton contains several Proterozoic and Archean age provinces, thus providing geologically distinguishable crustal provinces that are source terranes for the upper Proterozoic and lower Paleozoic continental margin sedimentary section (e.g., Gehrels et al., 2011, and references cited therein) (Fig. 1). The Yavapai-Mazatzal Province (1.8–1.6 Ga) extends across central North America (Fig.1). It is bounded on the north and northwest by the Trans-Hudson orogenic terrane (2.0–1.8 Ga) and Archean rocks (>2.5 Ga) of the Wyoming and Superior Provinces (Fig. 1). It is bounded on the south and east by the terranes of the Grenville orogen (1.2–1.0 Ga) and on the west by the Mojavia terrane (>2.5 Ga with 1.6–1.7 Ga granitoids) (Fig. 1).

Detrital zircon sources for the passive margin section changed in the upper Proterozoic–Lower Cambrian (Linde et al., 2014, and references cited therein). The 1.2–1.0 Ga Grenville orogen of southern and eastern North America (Fig. 1) was a significant sediment source for western Laurentia throughout the Neoproterozoic (Rainbird et al., 1997, 2012), including the upper Proterozoic passive margin section from the northwest United States to Sonora, Mexico (e.g., Lawton et al., 2010; Gehrels and Pecha, 2014; Yonkee et al., 2014; Linde et al., 2014). In contrast, the 1.8–1.6 Ga Yavapai-Mazatzal and 1.48–1.34 Ga mid-continent granite-rhyolite provinces within the North America craton (Fig. 1) were the more predominant sediment sources for strata higher in the passive margin section (e.g., Lawton et al., 2010; Gehrels and Pecha, 2014; Yonkee et al., 2014; Linde et al., 2014).

The RMA is often interpreted as a package of oceanic sediments emplaced structurally eastward onto the western Laurentian craton during the Late Devonian–Early Mississippian Antler orogeny (Roberts et al., 1958; Poole et al., 1992). Roberts Mountains allochthon strata are exposed in north-central Nevada between the Golconda thrust on the west and the Roberts Mountains thrust on the east; some units crop out west of the Golconda thrust in tectonic windows (Fig. 2). Rocks of the allochthon structurally overlie coeval rocks of the western Laurentian passive margin (e.g., Schuchert, 1923; Kay, 1951; Roberts et al., 1958; Madrid, 1987) (Fig. 3). Roberts Mountains allochthon strata are highly deformed, and include imbricated older-over-younger thrust sheets (Evans and Theodore, 1978; Oldow, 1984; Noble and Finney, 1999). The metamorphic grade of the strata is generally greenschist facies or lower (Gehrels et al., 2000a). The RMA was emplaced along the Roberts Mountains thrust during the Late Devonian to Early Mississippian Antler orogeny (Roberts et al., 1958). The Antler foreland basin, west of the Laurentian craton and east of the Antler orogen, was filled between Devonian and Early Mississippian time by sediments shed from the uplifting Antler highlands (Poole, 1974; Trexler et al., 2003) (Fig. 2).

The plate tectonic setting of the Antler orogeny has been variously interpreted as continent-continent collision, continent-arc collision, backarc thrusting, and polarity reversal of a subduction zone (e.g., Nilsen and Stewart, 1980; Speed and Sleep, 1982; Dickinson et al., 1983). The RMA is often interpreted as an accretionary prism formed due to plate convergence at the continental margin (Speed and Sleep, 1982; Oldow, 1984; Dickinson, 2000).

Evidence for Antler-age tectonism has been reported along the western Laurentian margin, in Alaska and Canada (e.g., Nilsen and Stewart, 1980; Gehrels and Smith, 1987; Dusel-Bacon et al., 2006; Nelson et al., 2006; Paradis et al., 2006; Piercey et al., 2006; Colpron et al., 2007). Middle to Late Devonian continental arc magmatism occurred in the Alaska Range and central Yukon (Piercey et al., 2006) (Fig. 1). Upper Devonian–Early Mississippian felsic igneous and metaigneous rocks record bimodal volcanism in east-central Alaska and the Yukon (Dusel-Bacon et al., 2006) (Fig. 1). In south-central British Columbia (Fig. 1), a Late Devonian continental arc and backarc developed (Paradis et al., 2006).

Colpron et al. (2007) and Colpron and Nelson (2009) have proposed a direct link between the Antler orogeny and coeval tectonism of western Laurentia. They propose that a “Northwest Passage” opened in mid-Paleozoic time between Laurentia and Siberia, and a Scotia-style arc developed along the northern Laurentian margin in the Early Devonian (Fig. 3). The Alexander terrane, and other fragments such as the eastern Klamath and northern Sierran terranes, were transported from a Baltica origin to northwestern Laurentia through the Northwest Passage via the westward migration of the arc’s subduction zone (Fig. 3). By Middle Devonian time, a sinistral transform fault developed at the southern end of this passage and extended southward along western Laurentia. This system transported these terranes and fragments south along the margin. Colpron and Nelson (2009) note progressively younger deformation southward along the Laurentian margin, from Alaska and the Yukon to Nevada, and suggest that this records the southward propagation of the transpressional system. They propose that this fault system could have provided the weakness along which Devonian subduction initiated.

The RMA strata sampled (Figs. 2 and 4; Table 1) are arenite beds within units that are predominantly chert and argillite with some limestone and mafic volcanic rocks. Most contacts between and within units are structural, and the stratigraphic bases and tops of units are not known. The strata of the RMA are described briefly below, as evidence of their depositional environments.

The Snow Canyon Formation and the McAfee Quartzite in the Independence Mountains (Figs. 2 and 4; Table 1) are the equivalent of the upper Vinini and Valmy formations, respectively (Holm-Denoma et al., 2011). Both units are Middle Ordovician based on graptolite fauna (Churkin and Kay, 1967). The Snow Canyon Formation is predominantly chert with arenite, shale, and siltstone layers, and basaltic lavas with interbedded limestone (Churkin and Kay, 1967). The McAfee Quartzite is predominantly massive cliff-forming quartzite with intervals of shale and siltstone and bedded chert (Churkin and Kay, 1967). The arenite intervals in these formations are interpreted as turbidites (Miller and Larue, 1983).

The Vinini Formation (Figs. 2 and 4; Table 1) was first mapped in the Roberts Mountains by Merriam and Anderson (1942) along Vinini Creek. Merriam and Anderson (1942) recognized two informal units (upper and lower) based on lithology and graptolite fauna and described the extreme structural disruption of these rocks. In later work, Noble and Finney (1999) used precise radiolarian biostratigraphy to demonstrate a high degree of structural imbrication both within the Vinini Formation and within Devonian cherts. In the Toquima Range, near Petes Summit (Fig. 2), the Vinini Formation is divided into two informal units (upper and lower), which are mapped in depositional contact, and the extreme structural complexity and repetition of thrust slices is also mapped (McKee, 1976). We observed the depositional contact at Petes Summit, where the quartz arenite of the upper Vinini rests on shale of the lower Vinini. The lower Vinini Formation is predominantly quartz arenite, with siltstone, shale, chert, and limestone (Finney et al., 1993). The lower Vinini Formation is Upper Lower to Lower Middle Ordovician in age, based on graptolite and conodont fauna (Finney et al., 1993). The arenite intervals in the lower Vinini Formation are interpreted as turbidites (Finney et al., 1993). The upper Vinini Formation is predominantly shale and bedded chert, with some siltstone and arenite (Finney et al., 1993). The unit is Middle Middle to Upper Ordovician, based on graptolites and conodonts (Finney et al., 1993). Graptolites and conodonts of the lower Vinini Formation are similar to those found in coeval Laurentian shelf carbonate deposits (Finney and Ethington, 1992; Finney, 1998). At Petes Summit, we observed low-angle cross lamination and hummocky cross stratification in the arenite of the upper Vinini Formation. We therefore interpret the upper Vinini as having been deposited in a high-energy environment at a depth above storm wave base on the continental shelf and probably at less than 100 m depth.

The Elder Sandstone (Figs. 2 and 4; Table 1) is predominantly fine-grained sandstone and siltstone, with some cherty shale and quartzite (Gilluly and Gates, 1965). Fossils are sparse in the unit; the age is Lower Silurian based on graptolites (Gilluly and Gates, 1965). The Elder Sandstone is interpreted as a turbidite deposit (Madrid, 1987).

The Slaven Chert (Figs. 2 and 4; Table 1) is predominantly black, bedded chert with shale beds and some limey sandstone and siltstone (Gilluly and Gates, 1965). The unit is Middle Devonian based on a variety of fossils (Gilluly and Gates, 1965). The arenite intervals in the Slaven Chert are interpreted as turbidites (Madrid, 1987).

Zircon grains from six arenite samples were analyzed for U-Pb ages and Hf isotope ratios (Figs. 2 and 4; Table 1). A small number of zircon grains from these samples were previously analyzed for U-Pb ages by Gehrels et al. (2000a), using ID-TIMS (Fig. 5). Zircon grains were separated and analyzed at the University of Arizona LaserChron facility using standard techniques described by Gehrels and Pecha (2014) to yield a best age distribution reflective of the true distribution of detrital zircon ages in each sample. Approximately 200 randomly selected grains were analyzed in each sample for U-Pb ages. Approximately 50 of these grains were subsequently analyzed for Hf isotopes. Hf analyses were conducted on top of the pits left after U-Pb analysis, to ensure that Hf isotope data were collected from the same domain as the U-Pb age. Analyses were conducted by LA-ICPMS using the Photon Machines Anlyte G2 excimer laser connected to the Nu Plasma high-resolution inductively coupled plasma-mass spectrometer, using methods identical to those described by Gehrels and Pecha (2014).

Analytical results are displayed graphically on normalized probability plots (Figs. 5 and 6), which allow visual comparison between zircon populations. U-Pb geochronology results are displayed in Figure 5, which contains both data from the original ID-TIMS analyses of these samples (Gehrels et al., 2000a) and the LA-ICPMS analyses of the current study, and in Figure 6, which displays the U-Pb results and Hf isotope analyses of the current study on the same chart. The essential U-Pb isotope information and ages are reported in Supplemental Table 1¹.

We compared detrital zircon age distributions both visually and statistically. Our initial appraisal was visual comparison of the probability plots. We also compared age distributions using the Kolmogorov-Smirnov (K-S) statistic (Guynn and Gehrels, 2006) (Table 2). The K-S statistic calculates whether a statistically significant difference exists between two distributions. P<0.05 indicates >95% probability that two U-Pb distributions are not the same. The K-S statistic is sensitive to proportions of ages present, and a low-P value may indicate that the proportions of age peaks are different, even though the ages are similar (Gehrels, 2012).

Hafnium isotope data are shown on Hf-evolution diagrams (Fig. 6) that display epsilon Hf_(t) (εHf_(t)) values at the time of crystallization. The essential Hf isotope information is reported in Supplemental Table 2².

Although the ID-TIMS data (Gehrels et al., 2000a) are similar to the new LA-ICPMS data (Fig. 5), there are variations in the proportions of age groups. The two studies used different grain-selection procedures. For the ID-TIMS study, zircon crystals were selected from color and morphology groups, without regard to the number of grains in each group. For our LA-ICPMS study, we attempted to select grains at random from the entire population of grains. This procedure resulted in a more representative age distribution because the grains are chosen randomly. The results and interpretations that follow are all based upon LA-ICPMS ages from our current study.

U-Pb geochronology and Hf isotope analyses reveal that the RMA strata are in two distinct groups. Five of the six samples (the Snow Canyon Formation, the McAfee Quartzite, the upper Vinini Formation, the Elder Sandstone, and the Slaven Chert) yield similar U-Pb age spectra, while the remaining sample, the lower Vinini Formation, yields significantly different U-Pb age spectra (Figs. 5 and 6). The Hf data from the five samples with similar U-Pb ages are similar, while the lower Vinini Formation sample, because of its different age spectra, yields significantly different Hf ratios (Fig. 6).

To interpret provenance, we compared the data from our study to known U-Pb ages and Hf isotope data from Laurentian basement provinces and other sedimentary units.

The detrital zircon age spectra of the Snow Canyon Formation, the McAfee Quartzite, the upper Vinini Formation, the Elder Sandstone, and the Slaven Chert are consistent with provenance in the Peace River Arch (PRA) region of western Canada (Fig. 7). The 1820–1960 Ma grains are similar in age to magmatic arcs in the PRA region, including the Fort Simpson, the Rimbey, the Ksituan, and the Great Bear arcs (Hoffman, 1989; Ross, 1991; Villeneuve et al., 1993) (Fig. 7). The 2060–2120 Ma grains are similar in age to accreted terranes in the PRA region, including the Buffalo Head and Chincaga terranes (Hoffman, 1989; Ross, 1991; Villeneuve et al., 1993) (Fig. 7). The 2650–2750 Ma grains are similar in age to Archean terranes in the PRA region, including the Nova and Hearne terranes (Hoffman, 1989; Ross, 1991; Villeneuve et al., 1993) (Fig. 7).

The Hf isotope data are consistent with provenance in the PRA region. The 1820–1960 Ma grains have a wide range of values, from juvenile and moderately juvenile through evolved (εHf_(t) +10 to –15), similar to those of other units interpreted to originate in the PRA region (Gehrels and Pecha, 2014). The 2060–2120 Ma grains are more narrowly grouped, with moderately juvenile to evolved values of εHf_(t) +3 to –6, compatible with other units originating in the PRA region (Gehrels and Pecha, 2014). The 2560–2750 Ma grains have juvenile, moderately juvenile, and evolved values of εHf_(t) +6 to –15, also compatible with PRA origin (Gehrels and Pecha, 2014). The ages of basement terranes that comprise the PRA region (Fig. 7) are all represented in the age spectra of the RMA samples (exclusive of the lower Vinini Formation).

The detrital zircon U-Pb ages and Hf isotope data from these RMA strata are similar to selected passive margin strata and RMA strata analyzed in other studies (Fig. 8). The RMA strata sampled in this study (exclusive of the lower Vinini) have U-Pb age spectra similar to those of the Ordovician Valmy Formation of the RMA (Gehrels and Pecha, 2014), as well as the Eureka Quartzite and the Mount Wilson Formation (Gehrels and Pecha, 2014), and the Kinnikinic Quartzite (Barr, 2009), Ordovician units of the western Laurentian passive margin (Figs. 8 and 9). The K-S analyses of the RMA and the Ordovician passive margin units discussed above do not contradict our interpretation that the RMA strata have a common provenance with the Ordovician passive margin sandstones (Table 2). These RMA strata also show similar Hf isotope ratios to the Valmy Formation (Gehrels and Pecha, 2014) and to the Eureka Quartzite and the Mount Wilson Formation (Gehrels and Pecha, 2014) (Fig. 8).

The Peace River Arch region of western Canada is the source for the RMA units in this study, exclusive of the lower Vinini Formation, and for the Ordovician passive margin sandstones. The Peace River Arch region was an uplifted region from late Neoproterozoic through Middle Devonian time (Cant, 1988; Cant and O’Connell, 1988; Cecile et al., 1997). Igneous bodies in the PRA region have ages similar to the U-Pb ages of zircons in the RMA rocks sampled (Figs. 7 and 8). The U-Pb age spectra of the RMA rocks sampled are not consistent with derivation from the central Laurentian craton; the Yavapai-Mazatzal terranes are 1.6–1.8 Ga and cannot serve as a source of the 1.8–2.0 Ga grains in the samples.

The U-P age spectra of the lower Vinini Formation are consistent with provenance in north-central Laurentia. The 490–500 Ma grains are similar in age to plutonic suites in roof pendants and inliers within the Challis volcanic-plutonic complex and the Idaho batholith (Lund et al., 2010). The 1110–1120 Ma grains are consistent with the Grenville orogen; the 1420 Ma grains are consistent with the central Laurentian anorogenic granites; the 1660–1800 Ma grains are consistent with the Yavapai-Mazatzal terranes; and the 2470–2750 Ma grains are consistent with the Archean craton (Bickford et al., 1986; Hoffman, 1989; Ross, 1991; Anderson and Morrison, 1992; Bickford and Anderson, 1993; Van Schmus et al., 1993) (Fig. 1). River systems traversing the north-central craton from east to west transported sediments from these crystalline bedrock sources—or from sediments recycled from them—and subsequently deposited them off the western Laurentian margin as the lower Vinini Formation.

The Hf isotope data of the lower Vinini grains are also consistent with origin in north-central Laurentia. The 490–500 Ma grains have mostly moderately juvenile to evolved values (εHf_(t) +3 to –5), with two grains highly evolved (εHf_(t) –20 to –25). The moderately juvenile to evolved grains are compatible with the plutonic suites in Idaho; however, the highly evolved grains are unlike any analyzed in these suites (Todt and Link, 2013). The 1110–1120 Ma grains have moderately juvenile values (εHf_(t) +4 to +6), similar to those of the Grenville orogen (Mueller et al., 2008; Bickford et al., 2010). The 1420 Ma grains have juvenile to moderately juvenile values (εHf_(t) +7 to +3), compatible with the anorogenic granitoids of the mid-Laurentian craton (Goodge and Vervoort, 2006). The 1660–1800 Ma grains have juvenile to moderately juvenile values (εHf_(t) +10–0), similar to the Yavapai-Mazatzal terranes (Bickford et al., 2008). The 2470–2750 Ma grains have moderately juvenile to evolved values (εHf_(t) +6 to –6), compatible with those in northern Greenland and Arctic Canada (Rohr et al., 2008, 2010) (Fig. 1).

The Early Cambrian uplift of the Transcontinental Arch altered the drainage patterns in western Laurentia; this change is recorded in the changing detrital zircon age patterns between upper Neoproterozoic and Lower Cambrian passive margin strata (Linde et al., 2014, and references cited therein) (Fig. 10). The uplift of the arch blocked the transport of Grenville-age grains and created, on the west flank of the arch itself, a new highland and source of sand, consisting of Yavapai-Mazatzal basement rocks and sedimentary rocks recycled from this basement. In many older passive margin strata that predate the uplift of the arch, Grenville-age grains predominate (Fig. 10). These grains were transported by continent-spanning rivers that drained the central craton and Grenville orogenic terrane to the western Laurentian margin through the late Neoproterozoic (Rainbird et al., 1997, 2012). In many younger passive margin strata, deposited after the uplift of the arch, Yavapai-Mazatzal–age grains dominate (Fig. 10). Rivers originating in the central craton were blocked from flowing to the west by the uplifted arch, which blocked the transport of many Grenville-age grains (Amato and Mack, 2012; Gehrels and Pecha, 2014; Linde et al., 2014; Yonkee et al., 2014).

The detrital zircon U-Pb ages and Hf isotope data of the lower Vinini Formation resemble those of the younger, post–Transcontinental Arch uplift, passive margin strata, such as the Geersten Quartzite of Utah and the Osgood Mountains Quartzite of Nevada (Fig. 11). These are the only post–arch uplift passive margin data sets for which we have both U-Pb ages and Hf isotope data. The lower Vinini Formation U-Pb age spectra and Hf isotope ratios are similar to those of the younger passive margin strata. The provenance of the lower Vinini Formation is central Laurentian, shed from the western flanks of the Transcontinental Arch and the regions to the west of the arch, after the uplift of the arch (Fig 1).

Sedimentological analyses provide a further constraint and suggest that the Ordovician passive margin sandstones are not the source of the RMA strata, but rather that these strata have a common source. Finney and Perry (1991) proposed that the Eureka Quartzite (an extensive Ordovician passive margin unit) was the source of the sandstones in the younger sections of the Vinini and Valmy formations of the RMA. However, the grains of the Ordovician passive margin sandstones are more texturally mature than those of the RMA strata, whose grains are coarser, larger, and more poorly sorted (Ketner, 1966). The more mature shelf sands such as the Eureka Quartzite and Mount Wilson Formation could not be the source of the more immature RMA sandstones. The RMA and passive margin sandstones have similar U-Pb age spectra and Hf isotope ratios (Fig. 7) and share a common source in the PRA region.

The Ordovician passive margin sands and the RMA strata sampled have different depositional histories (Fig. 12). The Mount Wilson Formation was deposited in a nearshore to shelf environment immediately outboard of the Peace River Arch (Kent, 1994). Other Ordovician passive margin sandstones, now preserved as the Eureka Quartzite and the Kinnikinic Quartzite, were shed from the Peace River Arch, and subsequently moved southward along the western Laurentian margin via longshore transport to the depositional basin (Ketner, 1968) (Fig. 12B). The evidence for this transport is that grain size decreases and sorting improves in Ordovician arenites from near the PRA source (the Mount Wilson Formation) southward through Idaho (the Kinnikinic Quartzite) and into Nevada and California (the Eureka Quartzite) (Ketner, 1968). The texturally immature arenites of the RMA did not undergo the extensive reworking of this longshore transport. Sediments of the RMA strata, other than the lower Vinini Formation, were deposited as turbidites (Miller and Larue, 1983; Madrid, 1987; Finney et al., 1993) offshore of the Peace River Arch (Figs. 12A–12C).

To reach their current geographic location, the RMA strata were tectonically transported south along the western Laurentian margin, in Latest Devonian time (Fig. 12E). This is consistent with a sinistral transpressional fault system, as proposed by Colpron and Nelson (2009) (Figs. 12D and 12E). Subsequent shortening moved the RMA up onto the craton in the Antler orogeny of Latest Devonian–Earliest Mississippian time (Figs. 12E and 12F).

These U-Pb geochronology and Hf isotope analyses of RMA strata give new insight into their provenance. We confirmed previous work that had indicated different detrital zircon U-Pb ages among strata of the RMA, implying different sources for these units. New data indicate that provenance of the lower Vinini Formation is north-central Laurentia, shed from the western flanks of the uplifted Transcontinental Arch. Other RMA strata sampled, Ordovician–Devonian, are similar to Ordovician passive margin sandstones that crop out widely through western North America. These units share a common source in the Peace River Arch region.

Combining sedimentology with detrital zircon data reveals the relationship between the RMA strata and passive margin shelf sands, making it possible to distinguish between sedimentary transport and tectonic transport of the RMA strata. The Ordovician passive margin sands were deposited in the PRA region, and some sand was carried south and reworked by longshore transport. In contrast, the Ordovician–Devonian RMA strata (exclusive of the lower Vinini) were deposited in a shelf, slope, or basin environment offshore of the PRA region; the arenite intervals in these units were deposited as shelf sands or turbidites. These RMA strata, along with the lower Vinini Formation, were tectonically transported in Late Devonian time southward along the margin on a sinistral transpressional fault system. The entire RMA package was subsequently emplaced eastward onto the craton during the Late Devonian–Early Mississippian Antler orogeny.

We would first like to acknowledge the essential role that Bill Dickinson played in all of our RMA studies. Bill was an incomparable mentor and model for all of us and is already sorely missed. Requiem aeternam dona ei, Domine, et requiescat in pace.

Additionally, we thank the researchers and staff at the Arizona LaserChron center, especially Mark Pecha and Nicole Giesler and volunteer laboratory assistants Dan Sturmer of Shell Oil Company and University of Nevada, Reno, graduate students Connor Newman and Kyle Basler-Reeder for critical help in Tucson. This research was funded under Arizona LaserChron National Science Foundation (NSF) grant EAR-1338583 to Gehrels. The original collecting effort for these samples was supported by NSF grants EAR-9116000 and EAR-9416933. Linde also thanks the following organizations for generous scholarship support of her graduate work: Raytheon Corporation Student Veterans Scholarship, the Rocky Mountain Association of Geologists Veteran Memorial Scholarship, the Nevada Petroleum and Geothermal Society, the Graduate Student Association of the University of Nevada, Reno, and the Cordilleran Section of the Geological Society of America.