The Dunnage zone of central Newfoundland records the closing of an oceanic tract — the Iapetus Ocean. Within this tract, the Red Indian line separates arc terranes developed close to Laurentia (Notre Dame subzone) from those associated with the peri-Gondwanan terrane Ganderia (Exploits subzone). Sandstones and conglomerates of the Badger Group were deposited on the Exploits subzone, southeast of the Red Indian line, from Late Ordovician (Katian) to early Silurian (Llandovery) time. Two samples were collected from the base and top of the Badger Group for detrital zircon U/Pb dating. The lower sample yielded detrital zircon populations with a large statistical peak at ca. 1.0 Ga, and other features characteristic of derivation from Laurentian sources. Paleozoic zircons with ages from Late Cambrian to Late Ordovician were probably derived from the Notre Dame arc on the margin of Laurentia. The upper sample yielded a comparable age distribution, but with a much smaller proportion of Mesoproterozoic relative to Paleozoic detrital zircon. These results date the earliest arrival of Laurentia-derived detritus on the peri-Gondwanan Exploits subzone crust in the early Katian (ca. 455 Ma). However, the absence, in both samples, of Neoproterozoic grains suggests that there was still a seaway separating the Gander margin from arc terranes accreted to Laurentia until at least the middle Llandovery (ca. 436 Ma).
The rocks of central Newfoundland record the closing of an oceanic tract — the Iapetus Ocean. The closing of Iapetus involved a large number of generally short-lived accretionary and collisional events, evidence for which has been preserved in the Dunnage Zone of the Newfoundland Appalachians, a complex mosaic of supra-subduction oceanic crust, volcanic arcs, and microcontinents derived from both Laurentia and Gondwana. Typical mid-ocean ridge crust is rarely preserved and is assumed to have been largely subducted. Collisions between arc and microcontinental elements were commonly oblique and rarely led to pervasive high-grade metamorphism. Hence, syntectonic sedimentary successions are relatively well preserved; their provenance has the potential to constrain terrane collisions, especially where sediments derived from one terrane are transported and deposited on another.
The Red Indian line is a major tectonic boundary in the Newfoundland Appalachians (Fig. 1), separating arc terranes developed close to the margin of Laurentia (Notre Dame subzone) from those formed or associated with the margin of Ganderia (Exploits subzone). In this paper, we examine the detrital zircon provenance of Late Ordovician to early Silurian sandstones of the Badger Group, collected from outcrops immediately southeast of the Red Indian line, in the Exploits subzone. We show that the provenance of these sands is consistent with derivation from the Notre Dame subzone, to the northwest, demonstrating that elements of the Exploits subzone were in close proximity to the Laurentian margin by Late Ordovician (Katian) time, ∼450 Ma.
Exploits subzone arc basement
The Exploits subzone (Fig. 1) is divided into tectonic slices by numerous thrust faults (O’Brien 2003); many of these faults bound slices that show different tectonic and stratigraphic histories. Nonetheless, Zagorevski et al. (2007a, 2010) were able to identify common elements between many of the slices, including two distinct arc successions locally separated by a disconformity. The older Penobscot arc is of Cambrian to Tremadocian (∼515–485 Ma) age, whereas the younger Victoria arc is mainly Dapingian to Sandbian (472–455 Ma). The two arcs are separated by the Penobscot orogeny, which resulted in the emplacement of Penobscot back-arc basin ophiolites onto the margin of Ganderia (Colman-Sadd et al. 1992a; Zagorevski et al. 2010).
The oldest rocks in the Exploits subzone, located in an eastern slice, are mainly felsic and mafic volcanic rocks with minor siliciclastic sedimentary rocks of the Sandy Brook Group (Rogers et al. 2006), which are intruded by mafic and intermediate plutonic rocks of the Crippleback Intrusive Suite (ca. 564 Ma). These rocks bear a close resemblance in their age and isotopic characteristics with basement of the microcontinent Ganderia, which lies to the southeast and south (e.g., Dunning and O’Brien 1989; Valverde-Vaquero et al. 2006a, 2006b). This and the presence of occasional Ediacaran xenocrystic zircons in the Cambrian to Tremadocian arc rocks farther west suggest that, where older basement is present in the subzone, it is of peri-Gondwanan affinity.
The Neoproterozoic rocks are overlain with interpreted unconformity in the Grand Falls map area by basaltic and minor felsic volcanic rocks assigned to the Noel Paul’s Brook Group of the Victoria Lake Group, representing the younger Victoria arc succession. Representatives of the Penobscot and Victoria arcs occur in most of the tectonic units that lie to the northwest, as far as the Red Indian line, although paleomagnetic evidence (e.g., Todaro et al. 1996) suggests that some of these arc rocks may have been erupted in widely separated locations. However, in the area of New World Island, a mélange unit, the Dunnage mélange, occurs in an equivalent stratigraphic position to the arc rocks.
Late Ordovician black shale
Most units of the Victoria arc and the Dunnage mélange are overlain by a laterally persistent unit of black shale, generally assigned to the Shoal Arm, Dark Hole, Rogers Cove, and Lawrence Harbour formations in Notre Dame Bay, which have yielded Sandbian to early Katian graptolites (Fig. 2). This blanket of black shale has been interpreted to represent the base of a foreland basin developed on the Victoria arc during the Late Ordovician, while it was progressively underthrusted beneath the Red Indian Lake arc (upper plate) of the Notre Dame subzone on the leading edge of Laurentia (Zagorevski et al. 2007b, 2008). This arc–arc collision coincides with a general break in arc activity on both sides of the Red Indian line, although localized Sandbian to Katian volcanic rocks are present (van Staal et al. 2007; Zagorevski et al. 2008). The presence of this black shale interval is one of the major distinguishing features of the Exploits subzone; an unconformity is generally present at this level in the Notre Dame subzone to the northwest, interpreted as a result of progressive uplift owing to underthrusting by the arc rocks of the Exploits subzone (Zagorevski et al. 2008) and overlain by mainly Silurian terrestrial deposits (van Staal et al. 2009).
Katian black shale in the southwestern part of the Expoits subzone is commonly truncated tectonically at the structural top and is commonly associated with mélange that has black shale matrix, suggesting that the shale formed a regional detachment horizon during thrusting (Zagorevski et al. 2007b). However, to the northeast, a thick succession of siliciclastic sedimentary rocks, assigned to the Badger Group by Williams et al. (1995) overlies the black shales. The incoming of these coarser clastic sediments is also diachronous, as already noted by Dean (1977), Kusky and Kidd (1996), Williams (1991), Colman-Sadd et al. (1992b), and Williams et al. (1995). Basal parts of the Badger Group (included in the Point Leamington and Samson Formations) are Katian sand-rich turbidite successions with abundant folds and other deformation structures, which have been variably interpreted as soft-sediment (Pickering 1987; Blewett 1991) or tectonic in origin (Elliott and Williams 1988). Higher units include conglomerate (Goldson Formation). Despite the difficulty of interpreting the deformed successions, the turbidites include strata on both sides of the Ordovician–Silurian boundary, extending up into the late Llandovery (O’Brien 2003).
There are no indications of major angular unconformities between the Badger Group and the underlying Lawrence Harbour shale, on which it seems to rest with a conformable and gradational, though diachronous, contact. However, locally on New World Island, Badger Group rocks assigned to the Goldson Formation have been reported to unconformably overlie pre-Sandbian volcanic rocks of the Summerford Group (Reusch 1987; Williams 1995; O’Brien 2003). Internally, there are indications of tectonic activity during deposition of the Badger Group, in the form of olistostromal deposits interpreted as polymictic debris flows adjacent to seafloor breaching faults (e.g., Arnott 1983; Karlstrom et al. 1983, Van der Pluijm 1986), and folds interpreted as soft-sediment structures by some authors (e.g., Arnott 1983; Pickering 1987) but as D1 tectonic folds by others (e.g., Elliott and Williams 1988). Regardless, all previous authors agree that the first major cleavage-forming tectonic deformation of the Badger Group postdated deposition of the Goldson Formation. There are indications, in parts of the Exploits subzone, that initial tectonic deformation of the Badger Group occurred in late Llandovery time, before deposition of redbeds of the mainly Wenlock Botwood Group, which overlies the Badger Group and older rocks of the Wild Bight Group and Dunnage Mélange with angular unconformity at many locations (van der Pluijm 1990; Rogers and van Staal 2005). However, elsewhere, the Botwood Group has been inferred to overlie the Badger Group conformably and transitionally (Williams 1991). Despite this variability, the Badger Group is reasonably interpreted as recording a 20 million year interval of syntectonic sedimentation during convergence of the Notre Dame subzone with elements of the Exploits subzone and the Gander Zone (van Staal et al. 1998, 2009).
There has been one previous attempt to date detrital zircons from the Badger Group. Pollock et al. (2007) dated a small number of grains (<15) from each of two samples in the Goldson Formation. They recorded a small number of Mesoproterozoic grains and a predominance of Late Cambrian to Ordovician grains in each case; although the small sample sizes precluded conclusive inferences on provenance. Pollock et al. (2007) favoured a source in the Notre Dame subzone on the basis of regional stratigraphy, which is supported by the presence of large tonalite cobbles in the Point Leamington Formation that have an obvious source in the Notre Dame arc (van Staal et al. 2007, 2009, fig. 9a).
Reef limestone clasts are prominent as transported cobbles and large olistoliths in both Late Ordovician and Silurian parts of the Badger Group (B.H. O’Brien, personal communication, 2011). In situ reefs of this age are not known in Newfoundland but are known elsewhere from basins overlying both Laurentian and peri-Gondwanan terranes in Quebec and New Brunswick (e.g., Wilson et al. 2004; Malo 2004).
Detrital zircon geochronology
The Badger Group overlies black pyritic mudstone, presumably recording a tectonically quiet interval, and represents an influx of clastic sediment into an area previously dominated by intra-oceanic and peri-Gondwanan arcs (though also including earlier intervals of black shale). Because this succession has often been compared with that in the Southern Uplands of Scotland (e.g., van der Pluijm 1986; Williams 1991), where a similar succession has been interpreted as the result mainly of southward progradation of Laurentia-derived sand (e.g., Waldron et al. 2008 and references therein), we chose two stratigraphically separated but paleontologically well-dated locations near the base and top of the Badger Group to sample for detrital zircon geochronology (Figs. 1, 2).
Sample VL99-21 is from the basal sandstone of the Point Leamington Formation of the Badger Group (Fig. 2). The sample was taken from the lowest sandstone exposed at Little Red Indian Falls in the Exploits River, 15 km south of the town of Badger (coordinates 556528E, 5413134N; all coordinates in the text are Zone 21, Universal Transverse Mercator, North American Datum 27). The sampled sandstone immediately overlies black shale of the Lawrence Harbour Formation, which has an early Katian (equivalent to late Caradocian) depositional age (Dicranograptus clingani Zone; Williams 1991). Graptolites from this zone are also recorded by Williams from locations higher in this section, constraining the age of the sample as within that of the D. clingani Zone, estimated as 455.8–453 Ma in the time scale of Ogg et al. (2008).
Sample VL99-23 (Fig. 2) was collected from coarse to very coarse, poorly sorted sandstone of the Goldson Formation in its type area on the east side of Toogood Arm (coordinates 680628E, 5277481N). The sample was collected at locality 13 of van der Pluijm et al. (1987), originally described by McKerrow and Cocks (1978) near Toogood Arm on New World Island. The rocks at this location contain Stricklandia, assigned by McKerrow and Cocks to the Fronian stage of the Llandovery Series. In more modern time scales, the Fronian corresponds approximately to latest Aeronian (possibly earliest Telychian) stage, corresponding to a depositional age of ∼437–435 Ma (e.g., Ogg et al. 2008).
Heavy mineral concentrates were prepared from the sandstone samples by standard crushing, grinding, Wilfley table, and heavy liquid techniques. Mineral separates were sorted by magnetic susceptibility using a FrantzTM isodynamic separator and hand-picked using a binocular microscope. Zircon grains from the sandstones were cast in a 2.5 cm diameter epoxy mount (GSC mount #188) along with fragments of the GSC laboratory standard zircon (z6266, with 206Pb/238U age = 559 Ma) and were imaged in back-scattered electron (BSE) and catholuminescence (CL) modes utilizing a scanning electron microscrope (SEM) (Fig. 3). U–Pb sensitive high-resolution ion microprobe (SHRIMP) analytical procedures were carried out at the Geological Survey of Canada (GSC) following Stern (1997), with standards and U–Pb calibration methods following Stern and Amelin (2003). Further details of the analytical techniques are presented in Appendix A. U–Pb SHRIMP analytical data are presented in Table 1 and are displayed in cumulative probability plots (Fig. 4).
Abundant detrital zircons were retrieved from the Point Leamington Formation sandstone (VL99-21). Zircon morphologies range from well-rounded grains to delicate elongate crystals (Fig. 3a). Seventy-three zircons were analyzed from this sample; analyses range from Archean to Ordovician in age (Table 1). The largest proportion of analyses are between 0.9 and 1.2 Ga, comprising over 60% of the grains analyzed (Fig. 4). A statistical age peak occurs at ∼1.0 Ga, with Grenville-aged zircons analyzed clustering between 0.9 and 1.2 Ga. A significant number of grains (14%) fall in the 1.3–1.6 Ga age range. There are a small number of Paleoproterozoic and Neoarchean detrital grains analyzed (n = 3). This spectrum of Mesoproterozoic and scarce Paleoproterozoic ages, and particularly the asymmetric peak at ca.1.0 Ga, closely resembles numerous age distributions recorded from the Neoproterozoic and early Paleozoic eastern margin of Laurentia (e.g., Cawood et al. 2007; Waldron et al. 2008) (Fig. 5). Detrital analyses from the sample also include two grains with ages of ca. 615 and ca. 640 Ma, which probably reflect input from igneous rocks associated with Iapetan rifting on the Laurentian margin. About 16% of the zircons analyzed are Late Cambrian to Ordovician in age, ranging from ca. 492 to ages coincident within error with the depositional age at ca. 450 Ma. These detrital zircons are probably derived from the adjacent Notre Dame arc, western parts of which were undergoing rapid exhumation during the Middle to Late Ordovician (van Staal et al. 2007). An alternative is derivation from the underlying Victoria arc of the Exploits subzone, but this is unlikely because of the absence of grains with ages characteristic of the immediately underlying mainly Cambrian Penobscot arc, which directly underlies the Badger basin in some areas (e.g., Rogers et al. 2005; van Staal et al. 2005; Zagorevski et al. 2007a, 2008, 2010). In addition, derivation from peri-Gondwanan terranes, such as Avalonia and Ganderia, with Neoproterozoic basement is unlikely because there is a complete absence of ages around 680 Ma and between 600 and 530 Ma that are characteristic of these terranes.
Detrital zircons retrieved from the coarse sandstone of the Goldson Formation (VL99-23) include a range of morphologies, from euhedral elongate to stubby crystals and rounded grains (Fig. 3b). Analyses range from Archean to Ordovician in age (n = 62, Table 1). The detrital zircon population in the Goldson Formation sandstone is dominantly Middle Cambrian to Late Ordovician in age (Fig. 4). Like those from the Point Leamington Formation sandstone (VL99-21), these detrital grains are interpreted to be mainly derived from the Notre Dame arc, but the larger range of Cambrian ages raises the possibility that Exploits arc – back-arc volcanic rocks or units farther to the southeast might have contributed to the detritus. Nonetheless, the Notre Dame subzone also contains potential sources for these grains in the Baie Verte Oceanic tract, and we prefer this explanation because of the lack of Neoproterozoic to Early Cambrian (600 to 530 Ma) grains characteristic of Ganderian and Avalonian crust. In comparison with VL99-21, Mesoproterozoic detrital grains make up a much smaller component of the total population (∼30%); however, the relative proportions of ages are similar. The largest statistical age peak is close to 1.0 Ga, and a distinct smaller peak is present at ∼1.4 Ga. There is a gap in the distribution from 2.0 to 2.5 Ga and a small number of Neoarchean grains. The Precambrian part of this distribution resembles that of VL99-21, and the same arguments for a predominantly Laurentian source apply; the combination of a strong, asymmetric “Grenville” peak at 1.0 Ga, a scarcity of Paleoarchean grains (especially in the range 2.0 to 2.5 Ga), and a small contribution from Archean crust is characteristic of Laurentia-sourced sands throughout the Appalachian–Caledonide region (Fig. 5). We, therefore, envisage that the majority of these grains were transported eastward to the marine Badger basin by rivers sourced in the Notre Dame arc and uplands of the deformed Dashwoods and Humber zones to the west.
Both samples clearly show evidence for derivation from the margin of composite Laurentia, where, at the time of deposition, the Taconian orogen was already present and undergoing major erosion. The alternative explanation for the origin of the Late Cambrian to Ordovician detritus, that it was derived by erosion of the Victoria arc located stratigraphically below the Badger Group, is unlikely on several counts. Firstly, wherever the Victoria arc is seen, it is overlain by black shales of the Lawrence Harbour Formation and equivalents. Although these do locally rest with disconformity on the Victoria and Penobscot arcs (e.g., Rogers et al. 2005; van Staal et al. 2005; Zagorevski et al. 2007a, 2008, 2010), the deep-water black shales everywhere separate the arc rocks from the Badger Group, suggesting that the arcs were not exposed during deposition of the latter. Evidence instead suggests that, during the Late Ordovician, the Victoria arc was being underthrusted beneath the leading edge of composite Laurentia (van Staal et al. 1998; Zagorevski et al. 2007a, 2008). Secondly, if the Victoria arc were exhumed to the point where it could provide detritus to the Badger Group, it is likely that the underlying Penobscot arc and Ganderian basement would also be exposed.
Many authors have made comparisons between the Exploits subzone of the Newfoundland Appalachians and the Southern Uplands terrane of Scotland. In general, the Badger Group sandstones resemble in their provenance and stratigraphic position those described in the northern and central belts of the Southern Uplands of Scotland (McKerrow et al. 1977; Leggett et al. 1979; Colman-Sadd et al. 1992b; Floyd 2001; Waldron et al. 2008). However, the abundance of Late Cambrian and Ordovician detrital zircon in both Badger Group samples is a significant difference, and it indicates that the predominant source of sediment for the Badger Group lay in the adjacent Notre Dame subzone, rather than the Laurentian margin rocks of the Humber Zone to the west, in which igneous rocks of this age are scarce. Although we have only analyzed two samples, our results from the Goldson Formation are supported by those of Pollock et al. (2007), which also showed a predominance of Paleozoic zircon, suggesting that the prominence of the Notre Dame arc as a source increased over time during deposition of the Badger Group, relative to the contribution of Laurentian basement. We speculate that the abundance of ancient continentally derived Laurentian detritus in the UK Southern Uplands may reflect its position offshore of the mouth of a major orogen-draining river (similar to the modern Indus or Ganges–Brahmaputra), which filled the convergent zone with turbidites. Each of the tectonic slices in the Southern Uplands records rapid deposition of turbidites closely followed by accretion. In contrast, the closure of the Iapetus Ocean along the Red Indian line was marked by more moderate amounts of continent-derived clastic sediment and an abundance of arc fragments on both sides of the closing oceanic domain, leading to accretion of relatively more abundant igneous sources. The Badger Group as a whole appears to record deposition over a much longer time interval (∼20 million years) and contains much greater stratigraphic variation (O’Brien 2003) than any one of the Southern Uplands accretionary tracts.
The Badger Group thus represents the earliest definite record of deposition of Laurentia-derived detritus on crust of the peri-Gondwanan domain during closure of the Iapetus Ocean. To the east, in the equivalent turbidite successions of the Southern Uplands terrane of the British Isles, exposed basement is rare, as most tectonic slices have undergone décollement along the black shales. Rare basal volcanic rocks have been mainly interpreted as within-plate or ocean-island basalts (e.g., Phillips et al. 1995). However, possible arc-related basalt occurs in very small areas intercalated in lower Katian black slate at Gabsnout burn (Barnes et al. 1995; Phillips et al. 1995) and in contact with Sandbian (Emo and Smith 1978; Morris 1987) black shale at Tattinlieve and Slieve Aughty inliers in Ireland (Winchester and van Staal 1995), suggesting possible arc involvement. The region of extended and fragmented arcs represented by the Exploits subzone in the Newfoundland portion of the orogen appears less developed in the British Caledonides. There, the first clear record of deposition of Laurentia-derived siliciclastic sediments over peri-Gondwanan crust does not occur until the Wenlock, when turbidites of the Windermere Supergroup invaded the Lake District of Northern England (e.g., King 1994).
In New Brunswick, the Popelogan arc is the equivalent to the Victoria arc; it was also overlain by Sandbian black shale, which is in turn overstepped in the late Katian by the Grog Brook Formation and other rocks of the Matapedia overstep sequence (Fyffe and Fricker 1987; Wilson et al. 2004), which are Badger group correlatives, and which contain reef limestones comparable to those found as clasts in the Badger Group. The collision of the Popelogan arc is, thus, virtually coeval with that of the Victoria arc in Newfoundland (van Staal 1994). The top of the black shale cover also appears to become younger (up to D. clingani Zone) to the southeast (van Staal et al. 2003, fig. 4) and is at least locally overlain by Sansom-like greywackes (e.g., Tomogonops Formation). These rocks were incorporated together with Late Ordovician – early Silurian (447–435 Ma) blueschists and other high-pressure rocks in a southeast-facing subduction complex related to northwest-directed subduction of the Gander margin (van Staal 1994; van Staal et al. 2008).
In what type of basin was the Badger Group deposited? Early workers suggested that the area immediately south and east of the Red Indian line represented trench deposits (Pickering 1987; Kusky and Kidd 1996). However, Williams et al. (1995) and O’Brien (2003) highlighted the variability of timing, thickness, and facies of clastic sedimentary rocks in the Badger belt, suggesting a level of complexity inconsistent with a simple southeast-prograding accretionary prism. Subsequently, the basin has also been interpreted as a successor basin (e.g., Zagorevski et al. 2008) that subsided post-tectonically in Katian–Llandovery time following collision of the Victoria arc with composite Laurentia, but syntectonically with respect to subduction of the Gander margin. However, the lack of evidence for penetrative deformation prior to Badger Group deposition, combined with the lack of peri-Gondwanan input from Exploits subzone arc basement, suggest that this was not a typical successor basin, although we agree with these and other authors (Williams et al. 1995) that it was a syntectonic marine basin.
Regional relationships indicate that the Badger Group was diachronously deposited over the Victoria arc. Deposition overlaps with northwest-directed subduction of lithosphere of the oceanic Exploits back-arc basin and adjacent Gander margin to the southeast (van Staal 1994; van Staal et al. 1998, 2009; Valverde-Vaquero et al. 2006b), which generated a short-lived Late Ordovician – early Silurian arc built upon composite Laurentia to the northwest (Whalen et al. 2006; van Staal et al. 2007). Such an oceanic remnant of Iapetus also explains the absence of Ganderian or Avalonian basement sources in the Badger Group. If the Iapetus Ocean had entirely closed at the time of deposition, grains with ages from 650 to 530 Ma would be expected in the Badger Group. Combined, the available evidence suggest that the Badger Group formed southeast of this arc and probably represents either forearc basin or foredeep deposits, or a combination of the two. We suggest that the locus of accretion stepped outboard (i.e., southeast) of the basin following deformation prior to deposition of the overlying Botwood Group in the Wenlock, consistent with closure of the final marine remnant of the Iapetan realm along the Dog Bay line later in the Silurian (Williams et al. 1993).
U–Pb SHRIMP data from detrital zircons were examined from two samples of syntectonic sediments of the Badger Group along and east of the Red Indian line, a fundamental tectonic boundary between the Notre Dame and Exploits subzones in Newfoundland. At the base, Katian dark volcaniclastic sandstones sampled near the falls in the Exploits River mark the beginning of syntectonic sedimentation. At the top of the Badger Group, coarse, gritty Llandoverian sandstone of the Goldson Formation, sampled on New World Island, record the final stages in the filling of the Badger Group basin. The detrital zircon population is dominantly Late Cambrian to Ordovician in age, mainly derived from sources in the Notre Dame arc that formed on the Laurentian margin of Iapetus. Derivation from Exploits arc – back-arc volcanic rocks is less likely because of the lack of ages corresponding to the Penobscot and older arcs, abundant in the basement of the Exploits subzone. The input of the Laurentian basement significantly decreases upwards in the section and is probably due to collision-induced uplift of the Notre Dame arc that progressively diminished the contribution of Laurentian basement, exposed further to the west. The absence of many zircons in the 680–530 Ma age range is significant and suggests there was still a seaway separating the Gander margin and Avalonia from the arc terranes accreted to Laurentia (van Staal 1994).
JWFW acknowledges the support of Natural Sciences and Engineering Research Council Discovery Grant A8508. We are grateful to Alex Zagorevski for material for Fig. 1. We acknowledge the helpful comments of reviewers Doug Reusch and Brien O’Brien, and of Associate Editor Brendan Murphy.
- Received April 5, 2011.
- Accepted April 5, 2011.
- Published on the NRC Research Press Web site at http://cjes.nrc.ca on December 19, 2011.
- Published by NRC Research Press
Appendix A: U–Pb analytical techniques
U–Pb SHRIMP (sensitive high-resolution ion microprobe) analytical procedures were carried out at the Geological Survey of Canada (GSC) following Stern (1997), with standards and U–Pb calibration methods following Stern and Amelin (2003). Briefly, zircon grains from the sandstone were cast in 2.5 cm diameter epoxy mounts (GSC mount #188) along with fragments of the GSC laboratory standard zircon (z6266, with 206Pb/238U age = 559 Ma). The mid-sections of the zircons were exposed using 9, 6, and 1 µm diamond compound, and the internal features of the zircons were characterized in back-scattered electron (BSE) and catholuminescence (CL) modes, utilizing a scanning electron microscope (SEM). Mount surfaces were evaporatively coated with 10 nm of high purity Au. Analyses were conducted using an 16O– primary beam, projected onto the zircons at 10 kV. The sputtered area used for analysis was ca. 25 µm in diameter. The count rates at 10 masses, including background, were sequentially measured over six scans with a single electron multiplier and a pulse counting system with deadtime of 23 ns. Off-line data processing was accomplished using customized in-house software. The 1σ external errors of 206Pb/238U ratios reported in the data table incorporate a ±1.1% error in calibrating the standard zircon (see Stern and Amelin 2003). No fractionation correction was applied to the Pb-isotope data; common Pb correction utilized the Pb composition of the surface blank (Stern 1997). Isoplot v. 3.00 (Ludwig 2003) was used to generate the cumulative probability plots.