Detrital zircon data from the upper parts of the Proterozoic Hess Canyon Group of southern Arizona reveal abundant 1600–1488 Ma detrital zircons, which represent ages essentially unknown from southern Laurentia. This basinal succession concordantly overlies a >2-km-thick-section of 1657 ± 3 Ma rhyolite of the Redmond Formation. The rhyolite is intercalated with and hence contemporaneous with the lower parts of the overlying White Ledges Formation, a 300-m-thick orthoquartzite unit at the base of the Hess Canyon Group. These quartzites contain a unimodal detrital zircon age probability distribution with peak ages of 1778, 1767, and 1726 Ma, supporting regional correlation with other ca. 1.65 Ga quartzite exposures in southwestern Laurentia. However, the ∼900-m-thick argillaceous Yankee Joe and minimum 600-m-thick quartzite-rich Blackjack Formations contain younger detrital zircons, with peak ages ranging from 1666 to 1494 Ma and a maximum depositional age of 1488 ± 9 Ma. Prominent age peaks at 1582–1515 Ma and 1499–1488 Ma represent detritus that is exotic and not derived from known southern Laurentian sources. The Blackjack Formation is cut by the 1436 ± 2 Ma Ruin Granite, indicating that deposition, deformation, and intrusion occurred between 1488 and 1436 Ma. This basin likely developed before or in the early stages of the 1.45–1.35 Ga intracontinental tectonism in southwestern Laurentia. Our findings necessitate the presence of an ∼170 m.y. disconformity within the Hess Canyon Group and document a previously unrecognized episode of Mesoproterozoic basin sedimentation (>1.5 km of section) between 1488 and 1436 Ma in southern Laurentia. This new record helps to fill the 1.60–1.45 Ga magmatic gap in southern Laurentia and supports hypotheses for a long-lived Proterozoic tectonic margin along southern Laurentia from 1.8 to 1.0 Ga. The 1.6–1.5 Ga detrital zircon ages offer important new constraints for ca. 1.5 Ga Nuna reconstructions and for the paleogeography of contemporaneous basins such as the Belt Basin in western Laurentia.

The Precambrian assembly of the core of Laurentia involved assembly of Archean cratons during 1.8 Ga continent-continent collisions (e.g., Trans-Hudson orogen) that formed the supercontinent Nuna (Zhao et al., 2002; Whitmeyer and Karlstrom, 2007; Reddy and Evans, 2009). Beginning at about the same time, Proterozoic orogenic belts were successively added “southward” to Laurentia (present coordinates) between 1.8 Ga and 1.0 Ga along a convergent margin that may have extended from Baltica to Australia (Karlstrom et al., 2001; Evans and Mitchell, 2011). By analogy to, for example, the modern Cordilleran convergent system, which extends spatially from Alaska to Chile and temporally from 400 Ma to present, the long-lived orogen hypothesis for Proterozoic Laurentia needs continued testing via evidence for near-continuous tectonism at one place or another along the Laurentian margin as well as the nature of other craton blocks that were located both along strike and outboard during the postulated 800 m.y. evolution of the orogenic system.

Proterozoic sedimentary basins, many containing quartzite and rhyolite successions, have emerged as key tectonic marker units for distinguishing the timing and character of events that resulted in the growth and stabilization of continental lithosphere in the southwestern United States (Hoffman, 1988; Karlstrom and Bowring, 1988, 1991; Jessup et al., 2006; Barth et al., 2009; Jones et al., 2009). Studies of similar successions across multiple cratons have provided new insights into patterns of sedimentary provenance, dynamics of basin evolution within long-lived orogenic systems, crustal recycling during continental evolution, and regional crustal architecture (Giles et al., 2004; Griffin et al., 2006; Barth et al., 2009; Jones et al., 2009; Shufeldt et al., 2010; Howard et al., 2011). Proterozoic metasedimentary successions also represent key constraints in reconstructions of Rodinia and pre-Rodinian supercontinents (e.g., Karlstrom et al., 2001; Sears and Price, 2003; Zhao et al., 2004; Betts et al., 2008; Li et al., 2008; Stewart et al., 2010).

In this paper, we document for the first time a sedimentary basin in Arizona that evolved between 1488 and 1436 Ma, a time of basin development and tectonism that has been previously unrecognized in southwestern Laurentia. Hence, the new data provide insight into what was happening along the long-lived margin during a proposed 1.60–1.45 Ga “tectonic gap” in southern Laurentia (e.g., Karlstrom et al., 2004). Deformation of this basin also occurred before 1436 Ma, compatible with the possibility of a near continuum of deformation between the previously described 1650–1600 Ma Mazatzal orogeny and 1450–1350 Ma tectonism associated with intracratonic granitic magmatism (Karlstrom and Bowring, 1988; Nyman et al., 1994; Karlstrom et al., 2004; Jones et al., 2010). Because of the scarcity of 1600–1500 Ma detritus globally (Condie et al., 2009), we also explore the implications of zircons of this age as constraints for correlations with other contemporaneous successions in Laurentia and Australia, and for reconstructions of the Nuna supercontinent (Hoffman, 1988).

Proterozoic basement rocks are exposed throughout the Transition zone of central Arizona (Fig. 1), a 500-km-long by 90-km-wide physiographic province bounded by flat-lying Paleozoic cover to the north and by highly dissected rocks of the extensional Basin and Range Province to the south (Fig. 1 inset). Outcrops of Proterozoic basement preserve thick successions of metasedimentary and metavolcanic rocks and crosscutting granitoids that record a prolonged history of sedimentation, magmatism, deformation, and metamorphism between ca. 1840 Ma and 1630 Ma (Karlstrom and Bowring, 1988, 1993; Conway and Silver, 1989). Two Paleoproterozoic orogenic events resulted in the assembly and accretion of dominantly juvenile crust in a southward progression. The ca. 1.70 Ga Yavapai orogeny involved assembly and accretion of pre–1.70 Ga crust of the Yavapai Province in northwestern Arizona. The ca. 1.65 Ga Mazatzal orogeny represents the addition of younger than 1.70 Ga terranes to the southeast (Karlstrom and Bowring, 1988). The intervening time involved development of syntectonic basins and deposition of quartzite-rhyolite successions, formed during episodes of slab rollback along the active margin (Jones et al., 2009). Paleoproterozoic rocks were perforated by Mesoproterozoic granitoid plutons over a prolonged interval from 1.45 to 1.35 Ga. Thus, the emerging model is one of a tectonic semicontinuum during progressive crustal assembly with a tectonic gap between 1.6 and 1.45 Ga (Karlstrom et al., 2004; Duebendorfer et al., 2006).

The Tonto Basin Supergroup is a >8-km-thick package of sediment and volcanic material deposited ca. 1720–1700 Ma on top of ca. 1760–1730 Ma ophiolitic and arc plutonic rocks (Conway and Silver, 1989; Cox et al., 2002; Dann, 1997). Deformation of this succession produced northwest-verging folds and fold-and-thrust belt geometries at greenschist facies during the Mazatzal orogeny (Doe and Karlstrom, 1991). The >2-km-thick Hess Canyon Group (Fig. 2), the main focus of this paper, is a younger succession of quartzite, argillite, arkosic quartzite, and slate exposed in the southern Tonto Basin (Fig. 1). In an early study, Trevena (1979) correlated the Hess Canyon Group with the Mazatzal Group in the northern Tonto Basin. Later, Conway and Silver (1989) proposed that the Hess Canyon Group may be correlative with either the Mazatzal Group or the Houdon Formation, which underlies the Mazatzal Group, or the Hess Canyon Group is not correlative with either. Karlstrom and Bowring (1993) noted that the Hess Canyon Group rests on, and is intercalated with, the ca. 1660 Ma Redmond Formation, a >2-km-thick succession of rhyolite and interlayered volcanic sediment (Figs. 2 and 3). The White Ledges, Yankee Joe, and Blackjack Formations are interpreted as nearshore fluvial and tidal deposits (Cuffney, 1977; Trevena, 1979) that reach a maximum thickness of ∼3 km (Fig. 2), and they have also been inferred to be ca. 1.65 Ga in age (Karlstrom and Bowring, 1993). The upper Blackjack Formation is truncated by the crosscutting ca. 1436 ± 2 Ma Ruin Granite (Figs. 3 and 4; Isachsen et al., 1999).

Bedding orientations in Hess Canyon Group exposures define a syncline with an axial plane striking northeast, a north limb dipping 40° to 60° southeast (Fig. 4), and the south limb dipping ∼20° northwest. Deformation styles range from brittle to ductile depending on lithology and proximity to the Ruin Granite aureole. Quartzite in the White Ledges Formation is cut by minor, low-displacement, top-to-the-north, high-angle reverse faults, whereas more argillaceous sections are pervasively folded and variably foliated. The shale-rich Yankee Joe Formation contains abundant meter-scale tight to isoclinal folds, complex internal thrusting, and penetrative slaty cleavage. Fold asymmetries indicate northwest vergence, with the exception of rare southeast-vergent folds, which are interpreted to be associated with back thrusts. Internal folds and fabrics are parallel to larger (1–3 km) asymmetric folds with steeply east-southeast–plunging axes that indicate northwest-southeast–directed shortening with an oblique sinistral component. Davis et al. (1981) estimated the minimum thrust translation within the Yankee Joe Formation to be 500 m. Relative to a present unit thickness of ∼800 m (Fig. 4), this amount of thrusting is compatible with ∼50% shortening, as is often observed in foreland thrust belts (Boyer and Elliott, 1982). Deformational features in the Hess Canyon Group are cut by the unfoliated 1436 ± 2 Ma Ruin Granite (Isachsen et al., 1999). Bedding locally steepens against the intrusive contact, possibly indicating thermal softening and deformation associated with emplacement. A 1000-m-thick contact aureole is crosscut by granite-filled veins (Fig. 5) and locally contains andalusite and staurolite, suggesting contact metamorphic temperatures of ∼500 °C and a likely emplacement depth of 5–10 km (Cuffney, 1977).

The Mesoproterozoic Apache Group was deposited unconformably on dipping beds of the Hess Canyon Group (Figs. 2 and 4). Shale of the Pioneer Formation was deposited over local monadnocks of Hess Canyon Group and Ruin Granite, with relief up to 80 m causing local pinch-outs and onlaps (Fig. 2; Cuffney, 1977). Tuff in the lowest part of the Apache Group has been dated at 1328 ± 5 Ma (Stewart et al., 2001), and the succession is cut by dikes and sills of ca. 1100 Ma diabase (Silver, 1978). Thus, Hess Canyon Group sediments were buried to 5–10 km depth, deformed into broad folds and top-to-the-NW thrusts between 1488 and 1436 Ma, and then unroofed to the surface between 1436 and 1328 Ma.

We collected eight samples of the Proterozoic Hess Canyon Group and one sample of the underlying Redmond Formation to better constrain the age of deposition and determine the provenance of the succession. Our primary aim was to test correlations and compare provenance of the Hess Canyon Group with other similar ca. 1.70 and 1.65 Ga quartzite-rhyolite successions in the region. Location coordinates for each sample are provided in the GSA Data Repository (Table DR11), and the approximate sample locations are shown in Figures 3 and 4 in the context of the stratigraphic sections of Cuffney (1977). For seven of the samples, processing and preparation followed the methodology of Jones and Connelly (2006). U-Pb analyses of these samples were done at the Geological Survey of Denmark and Greenland (GEUS) using laser ablation–magnetic sector field–inductively coupled plasma–mass spectrometry (LA-SF-ICP-MS) techniques described by Frei and Gerdes (2009). Two samples, BJ-2-2011 and 20110306-1 (Redmond rhyolite), were processed and analyzed using laser ablation–inductively coupled plasma–mass spectrometry (LA-ICP-MS) techniques at the Arizona LaserChron Center of the University of Arizona via methodologies described by Gehrels et al. (2008). Complete analytical data are available in the GSA Data Repository (Table DR1 [see footnote 1]) together with age histograms and U-Pb concordia diagrams. Detrital zircon data are summarized in Table 1.

The sample of rhyolite from the uppermost Redmond Formation (Fig. 3) yielded a simple population of euhedral to subhedral zircon. Cathodoluminescence imaging of the grains revealed consistent internal oscillatory zonation with little to no evidence of xenocrystic cores. Thirty-nine grains analyzed yielded ages that overlap within uncertainty and produce a weighted mean 207Pb/206Pb age of 1657 ± 3 Ma (mean square of weighted deviates [MSWD] = 0.23; Fig. 3). We interpret this age to represent crystallization of the rhyolite and, because the rhyolite was collected from the uppermost Redmond Formation just below where it interfingers with the lower White Ledges Formation, the onset of Hess Canyon Group deposition.

Eight samples were collected throughout the Hess Canyon Group for detrital zircon geochronology. Figure 2 shows normalized age probability distribution curves for ≤10% discordant detrital zircon for these samples together with the generalized stratigraphy and available paleocurrent data from Cuffney (1977). Peak ages represent the weighted average of clusters of three or more grains with overlapping ages. All eight samples predominantly contained subrounded to rounded, equant zircons with up to 10% elongate or prismatic grains. Euhedral grains were rare. The range of 207Pb/206Pb ages for all samples is 3688 to 1427 Ma.

Age probability distributions indicate that all samples are dominated by Paleoproterozoic and Mesoproterozoic grains, with age peaks ranging from 1972 to 1494 Ma (Fig. 2). Early Paleoproterozoic and Archean zircons define relatively small peaks with ages of 2724–2463 Ma in four of the samples (Fig. 2) and make up less than 20% of the population in any single sample. Peak ages >2.4 Ga match well with the ages of dominant peaks in the ca. 1.75 Ga Vishnu Schist exposed in the Grand Canyon to the north (Fig. 1; Shufeldt et al., 2010), suggesting that Archean zircon might have been recycled from older sedimentary successions in the region. Alternatively, these grains could have been derived from older cratonic sources such as the Wyoming Province, Gawler craton, North China craton, and Antarctica, or from more proximal 2.7–2.4 Ga sources not presently exposed (Shufeldt et al., 2010, and references therein).

Circa 2000–1800 Ma detrital zircons define subsidiary age peaks in all but one of the samples (Fig. 2). Zircons with these ages are also present in, and could have been derived from, Grand Canyon rocks (Shufeldt et al., 2010). The dominant age peaks in all samples are Late Paleoproterozoic to Early Mesoproterozoic, with an age range of 1778 to 1567 Ma (Fig. 2). We interpret 1778–1709 Ma age peaks to represent sources in the Yavapai Province of southwestern Laurentia, with the 1726–1709 Ma peaks representing abundant ca. 1760–1700 Ma igneous rocks exposed in Arizona (Karlstrom and Bowring, 1993). With the exception of the ca. 1840 Ma Elves Chasm Gneiss (Hawkins et al., 1996), plutons older than 1760 Ma are generally not found in Arizona. Thus, the older 1778–1773 Ma peaks in the upper White Ledges Formation could be attributed to more distal sources, for example, the Needle Mountains in southwestern Colorado (Gonzales and Van Schmus, 2007; Jones et al., 2009). Detrital zircon age peaks younger than 1700 Ma are only found in the Yankee Joe and Blackjack Formations, and 1667–1610 Ma detrital zircon populations are interpreted to represent local sources in the Mazatzal Province of southeastern Arizona (Karlstrom and Bowring, 1993). These sources may also include 1657 ± 3 Ma rhyolite of the underlying Redmond Formation (Fig. 3) that is intercalated with the basal section of the White Ledges Formation.

The most surprising finding was abundant Mesoproterozoic detrital zircons in the Yankee Joe and Blackjack Formations that define age peaks of 1495 Ma and 1582–1488 Ma, respectively (Fig. 2). Because of the significant overlap in the probability distribution of different age populations, particularly on the younger end of the spectra, we used the “unmix ages” tool in Isoplot to determine the ages of the youngest populations in these samples. This tool uses Gaussian deconvolution to model data sets containing multiple age components (Ludwig, 2004). We used the Age Pick macro of Gehrels (2009) together with visual inspection of the age probability spectra to estimate the number of age components in each sample. The youngest detrital zircon population in the upper Yankee Joe Formation produced the most prominent age probability peak among the upper four samples (Fig. 2), and the weighted average age of this population was 1495 ± 10 Ma. The youngest probability peaks in the three Blackjack Formation samples are less obvious because they define humps or shoulders on the side of more dominant peaks. The Mesoproterozoic zircon ages in these samples represent a nearly continuous spectrum with an abundance of ages as young as ca. 1450 Ma (Table DR1 [see footnote 1]). The deconvolution approach yielded weighted average ages for the youngest zircon populations in the three Blackjack Formation samples of 1515 ± 17 Ma, 1488 ± 9 Ma, and 1499 ± 10 Ma, listed in order from oldest to youngest (Fig. 2). Peak ages of 1495 ± 10 Ma and 1488 ± 9 Ma from the Yankee Joe and Blackjack Formation, respectively, provide robust new estimates of the maximum depositional age of these two units, and we interpret 1488 ± 9 Ma to represent the best maximum age constraint on deposition of the entire upper part of the Hess Canyon Group.

In terms of source regions for the 1.6–1.5 Ga zircons, the >1.5 Ga grains found in these two units are somewhat older than, and hence were probably not derived from, the 1.48–1.44 Ga Eastern Granite-Rhyolite Province of southern Laurentia or extrusive equivalents of contemporaneous granites exposed in the southwestern United States (Van Schmus et al., 1993). Goodge and Vervoort (2006) and Condie et al. (2009) noted abundant 1475–1450 Ma and 1370 Ma zircon crystallization ages from A-type granites throughout the south-central United States, and both peaks are prominent in the post–1.35 Ga detrital zircon record (Gehrels et al., 2011; Timmons et al., 2005). Deposition of the Hess Canyon Group either predates or is in the early phases of this magmatic event, and the detrital zircon ages of 1582–1488 Ma are not known from any igneous rocks of southern Laurentia. Thus, certainly the detritus, and possibly basin deposition itself, fall within or at the edges of the well-documented magmatic gap ca. 1600–1500 Ma found in the southwest (Karlstrom et al., 2004).

Age and Evolution of the Hess Canyon Group

The depositional age of the White Ledges Formation and, thus, the lower Hess Canyon Group is constrained by interfingering relationships with the 1657 ± 3 Ma Redmond Formation. Detrital zircon age spectra for the White Ledges samples are similar to other ca. 1.70 Ga and 1.65 Ga quartzite successions exposed in the surrounding region (Jones et al., 2009), though the dominant age peaks are somewhat older than those from contemporaneous rocks exposed in the Mazatzal Province of southern New Mexico (Luther et al., 2005; Luther, 2006; Amato et al., 2008). However, they are consistent with the age of rocks exposed throughout the surrounding region and may indicate derivation from local Yavapai-aged sources together with significant recycling of older metasedimentary successions in the region ca. 1657 Ma. The compositional purity of the White Ledges Formation is somewhat surprising given its close association with voluminous rhyolite and associated volcanic rocks, but this is a regional, not local, conundrum, as similar associations are well documented throughout southwestern Laurentia (e.g., Bauer and Williams, 1994; Williams, 1991; Jones et al., 2009, 2011).

The Yankee Joe and Blackjack Formations were previously interpreted to be conformable with the White Ledges Formation (e.g., Cuffney, 1977), and deformation of the entire succession has been attributed to the Late Paleoproterozoic (1.65–1.60 Ga) Mazatzal orogeny (Cuffney, 1977). However, detrital zircon data indicate a pronounced shift in provenance from the White Ledges to the Yankee Joe Formation and a Mesoproterozoic depositional age for the upper Yankee Joe and entire Blackjack Formation (Fig. 2). Deposition of the lower Yankee Joe Formation was influenced by derivation of zircons from local Mazatzal-aged sources, possibly including the underlying Redmond Formation. Age probability peaks between 1639 and 1610 Ma in the upper Yankee Joe and Blackjack Formations may also indicate derivation from local Mazatzal-aged sources. However, the age peaks of <1600 Ma represent sources exotic to southwestern Laurentia that are discussed in more detail in the following.

The shift in provenance corresponds with a change in lithology from quartz litharenite and orthoquartzite of the White Ledges Formation to shale, siltstone, and arkosic sandstone of the Yankee Joe Formation (Fig. 2). There is no obvious outcrop evidence for an angular unconformity within the section, and differences in the style of deformation throughout the section could be attributed to rheological contrasts, different deformation events, or both. The age of the uppermost White Ledges Formation is not well established, and the detrital zircon data require that only the uppermost Yankee Joe Formation be Mesoproterozoic in age. Thus, one or more disconformities may be present within the section as it seems unlikely that the succession represents a continuum of deposition from 1657 to 1488 Ma. Stratigraphic data show a change in composition to finer-grained and more feldspathic rocks at the contact between the White Ledges and Yankee Joe Formations that likely indicates a significant change in depositional environment. We speculate that this is also where the depositional age likely changed, and there is a significant disconformity here between 1660 and 1600 Ma rocks below this contact to 1488– 1435 Ma sedimentary rocks above, a hiatus of ∼170 m.y.

We propose a model in which a basin formed ca. 1660 Ma and accumulated several kilometers of volcanic rocks and quartzite until ca. 1630 or 1600 Ma on the basis of constraints for contemporaneous deposits in New Mexico (Luther et al., 2005; Luther, 2006; Amato et al., 2008). A hiatus followed until ca. 1.5 Ga, when deposition began again with the accumulation of several kilometers of sediment that was initially locally derived based on the unimodal 1667 Ma peak in the lower Yankee Joe Formation. Regional topography created during prolonged accretionary orogenesis ca. 1.8–1.6 Ga in southwestern Laurentia would have been substantially eroded during the 1.6–1.5 Ga period of tectonic quiescence, leading to the development of larger continental river systems or epicontinental pediments (i.e., Sears, 2007) and much broader dispersal of sediment throughout Laurentia to the continental margins. The influx of locally derived sediment in the lower Yankee Joe Formation may reflect the initial response to basin formation or reactivation. Sediment sources changed up section to distal, exotic continental sources with an abundance of 1.6–1.48 Ga rocks. For example, non-Laurentian sources of 1.6–1.48 Ga zircons were potentially sustained by relatively continuous uplift and unroofing in the source region ca. 1.60–1.48 Ga. Hence, these zircons may reflect a distant “orogenic” unroofing and deformational interval not recorded within Laurentia. Such 1.6–1.48 Ga deformational events are recognized on a few cratons worldwide.

In terms of deformational timing recorded within the Yankee Joe Basin, the entire succession including the White Ledges Formation was folded and thrust-faulted in response to NW-SE contraction and buried to granite emplacement depths prior to emplacement of the Ruin Granite ca. 1436 Ma. Based on known Laurentian deformational ages, it seems most likely that the layer-parallel shearing and folding of the succession represent basin inversion and thrusting that was likely syntectonic with the early stages of 1.45–1.35 Ga regional tectonism related to granite emplacement in the surrounding region. As magmatism, deformation, and metamorphism continued in southwestern Laurentia until ca. 1350 Ma in some areas (Jones et al., 2010), erosion and exhumation of more than 5 km of crust in the southern Tonto Basin tilted and exposed the Ruin Granite and metasedimentary succession prior to deposition of the Apache Group.

Speculations on the Sources of ca. 1.6–1.48 Ga Detrital Zircon

The ca. 1488–1436 Ma age range of late Mesoproterozoic basin formation in Arizona may have overlapped with ca. 1470–1454 Ma deposition of the lower Belt Supergroup in western Laurentia (Fig. 6; Sears et al., 1998; Evans et al., 2000), a succession that also contains abundant ca. 1600–1500 Ma detrital zircons of probably exotic (i.e., non-Laurentian) provenance (Fig. 6; Ross and Villeneuve, 2003; Link et al., 2007). The presence of a nearly complete spectrum of predepositional ages (1600–1490 Ma) as detrital components in both basins suggests that the source areas were magmatically active over a considerable duration of time before sedimentation began (Ross and Villeneuve, 2003).

As one possible Laurentian source region, Van Schmus et al. (1993) identified a region in the midcontinent of southern Laurentia in which ca. 1.48–1.35 Ga rocks of the Granite-Rhyolite Province have ca. 1.5 Ga Nd model ages. This region is separated from other parts of the Granite-Rhyolite Province with older Nd model ages by an isotopic boundary informally called the “Missouri Line” (ML in Fig. 6; Van Schmus et al., 1993). Whitmeyer and Karlstrom (2007) showed this line as the approximate northern boundary of a 1.5 Ga accretionary province, raising the possibility that rocks from Yankee Joe/Blackjack Basin could have been a relatively local Laurentian source for ca. 1.5 Ga detrital zircons in Arizona. However, to date, no actual zircons of this age have been reported from subcrop from the Granite-Rhyolite Province (e.g., Reed et al., 1993; Barnes et al., 1999; Bickford et al., 2000, 2010; Barker and Reed, 2010). Additionally, rocks of the Granite-Rhyolite Province rocks have not been substantially unroofed, as shown by the extensive preservation of ca. 1.48–1.35 Ga rhyolite and sedimentary rocks in the shallow subsurface.

Other possible 1.6–1.5 Ga sources in western Laurentia are limited to ca. 1580 Ma orthogneiss in the Priest River Complex in the northwestern United States (PRC in Fig. 6; Doughty et al., 1998) and ca. 1600–1590 Ma volcanic rocks and diabase in northwestern Canada (WB and WCD in Fig. 6; Thorkelson et al., 2001; Hamilton and Buchan, 2010). Even if these possible sources did contribute 1.60–1.58 Ga zircon to any of the 1.49–1.44 Ga basins in western Laurentia, this still would not explain abundant <1.58 Ga detrital zircons in both the Hess Canyon Group and lower Belt Supergroup (Fig. 6; Ross and Villeneuve, 2003; Link et al., 2007; this study). Furthermore, available paleocurrent data for both successions indicate source regions to the west or south that suggest outboard sources relative to the rifted western margin of Laurentia (Trevena, 1979; Ross and Villeneuve, 2003). Thus, we interpret the 1.6–1.48 Ga detrital zircons to have been derived from non-Laurentian sources that were relatively proximal to western Laurentia at ca. 1.5 Ga.

Australia contains the most abundant and complete record of magmatism and tectonism between 1600 and 1490 Ma, represented most prominently by the 1600–1560 Ma Hiltaba Event (Betts et al., 2002) and the 1590–1490 Ma Isan orogeny (Betts and Giles, 2006). Following the Isan orogeny, regions of the Mount Isa Inlier and Gawler craton underwent widespread cooling and exhumation from 1470 Ma to 1370 Ma (Betts and Giles, 2006). Multiple sedimentary basins were developed at this time across Australia that include the Cariewerloo basin (Gawler), the South Nicholson basin (Mount Isa), and the Edmund basin (Western Australia), and 1.60–1.49 Ga ages are widely represented in the detrital zircon record of these basins (Griffin et al., 2006). Ross and Villeneuve (2003) suggested Australia as the most likely source for non-Laurentian detrital zircon in the lower Belt Supergroup on the basis of detrital zircon ages and paleocurrent data indicating predominant sediment influx from a western craton (present-day coordinates). Some reconstructions of pre-Rodinian supercontinents show Australia west of Laurentia (e.g., Zhao et al., 2002; Giles et al., 2004) either in a northerly (SWEAT-like [southwest U.S.–East Antarctic]) position or a more southerly (AUSWUS [Australia–southwest U.S.]) position, supporting the possibility of Australia as a sediment source in western Laurentia ca. 1490–1430 Ma (Fig. 6).

Global age compilations (Fig. 6; Condie et al., 2009) identify South America as the only other continent with a significant 1.60–1.50 Ga component––ca. 1565–1535 Ma tonalite, granodiorite, and granite of the Cachoeirinha plutonic suite of the Amazonia craton (Geraldes et al., 2001). Paleomagnetic data indicate that the Amazonia craton might have been adjacent to the Llano region of southern Laurentia by ca. 1.25 Ga during the initial phase of Rodinia assembly (Tohver et al., 2002), thus permitting Amazonia as an outboard source of detritus in southwestern Laurentia ca. 1.49 Ga (Fig. 6). However, the ages of exposed rocks in Amazonia do not match all of the 1.60–1.50 Ga age peaks observed in the Hess Canyon Group, and the position of Amazonia relative to other conjugate cratons is not well understood. Goodge et al. (2010) recently identified a ca. 1.58 Ga granitic clast from Quaternary tills of the central Transantarctic Mountains, indicating the presence of Proterozoic crust beneath East Antarctica, and, thus, Antarctica is another possible source for 1.60–1.50 Ga detrital zircon (TAM in Fig. 6).

Paleogeographic Implications

In light of our new evidence for 1600–1488 Ma detrital zircons in Arizona and age similarities linking sources in Australia with both the Hess Canyon Group and lower Belt Supergroup, we prefer a position for Australia more directly adjacent to western Laurentia as shown in Figure 6 and speculate that Australia was a possible source for non-Laurentian zircons in both the upper Hess Canyon Group and Belt Basin. The relative scarcity of 1.6–1.5 Ga zircon sources worldwide (Fig. 6 compilation; Condie et al., 2009) makes detrital zircon within this age range a particularly useful constraint in ca. 1.50–1.45 Ga reconstructions. This is especially true along the western Laurentian margin, where 1.6–1.5 Ga detrital zircons are now recognized in multiple locations (Belt Basin and Yankee Joe Basin), and where the positions of formerly adjacent continents throughout the Proterozoic are widely debated (see Evans and Mitchell, 2011; Dalziel, 2010). In our favored reconstruction shown in Figure 6, almost all Australian provinces with documented 1.60–1.50 Ga activity are proximal to western Laurentia and could have provided sediment to basins in both Arizona and the Belt Basin to the north. Widespread magmatism and tectonism in Australia ca. 1.60–1.49 Ga would have created fertile sources and topography to sustain sediment supply even as orogenesis waned.

For Paleoproterozoic times, correlation of volcanic breccias between Australia and northwestern Laurentia has been proposed as a connection between the two cratons as early as ca. 1.60 Ga, but with Australia in a more northerly (SWEAT) position at this time (Fig. 7A; Thorkelson et al., 2001). However, a sharp change in provenance of NE Australian at 1.65 Ga from nonjuvenile to juvenile (1650 Ma) sources is recorded in Nd studies, raising the possibility of a more southerly (AUSWUS) connection at 1650–1600 Ma with a sediment dispersal from the juvenile Mazatzal Province to NW Australian basins (Barovich and Hand, 2008).

Existing paleomagnetic data are permissive of associations between Australia and Laurentia until as late as ca. 1.20 Ga (Payne et al., 2009) in either a SWEAT configuration (Dalziel, 2010; Goodge et al., 2010), closer to the AUSWUS connection (Evans and Mitchell, 2011), or with migration of Australia-Antarctica relative to the Laurentian margin from 1.65 to 1.2 Ga. It is worth noting that existing paleomagnetic constraints are permissive of a relatively wide range of craton configurations between ca. 1.6 and 1.0 Ga, such that viable Nuna reconstructions need to evolve into viable Rodinia configurations over the time period from 1.6 to 1.0 Ga. Our preferred configuration for the 1.5–1.3 Ga time frame (Fig. 6) does not negate key geologic correlations highlighted by Goodge et al. (2010) in their modified SWEAT reconstruction (Fig. 7B), but it provides a testable alternative with Australia-Antarctica farther south along the “western” Laurentian margin. Thus, several source regions containing abundant 1.6–1.5 Ga zircons in Australia-Antarctica would be opposite two major basins containing abundant 1.6–1.5 Ga zircons within Laurentia. This is testable by continued U-Pb and Hf characterization of zircons from both source regions and depositional basins, and additional studies of the sedimentary distributive systems.

This paper stems from field work by the first author and complementary field work by Jones and Karlstrom. Laboratory support was provided by Frei and Thrane at the Geological Survey of Denmark and Greenland, and Gehrels at the Arizona LaserChron Center of the University of Arizona. Funding for analytical work was provided by the College of Science and Mathematics at University of Arkansas at Little Rock. The authors would like to thank Bob and Debbie Buecher for their assistance in the field. Karlstrom acknowledges support from National Science Foundation grant EAR-0607808. Bruce Trudgill is thanked for constructive comments on earlier versions of the manuscript. Formal reviews by Pat Bickford, Geoff Fraser, and an anonymous reviewer together with editorial and review comments by John Goodge helped to clarify and improve the manuscript.

1GSA Data Repository Item 2012050, Age probability and U-Pb Concordia diagrams, is available at www.geosociety.org/pubs/ft2012.htm, or on request from [email protected], Documents Secretary, GSA, P.O. Box 9140, Boulder, CO 80301-9140, USA.