Abstract

Studies of pre-Grenvillian (1.4–1.3 Ga) plutons offer insight into the dynamics of arc amalgamation and backarc rifting prior to continental collision during the Ottawan orogeny. The Central Metasedimentary Belt boundary thrust zone (CMBbtz) is a northeast-southwest–trending thrust zone consisting of metaplutonic thrust sheets enveloped in gneissic tectonites and calcitic-dolomitic marble. Tonalitic CMBbtz thrust sheets (Dysart and Redstone), located in the southern Ontario Grenville Province (Canada) are made up of upper amphibolite facies, foliation-concordant metatonalite (+ amphibole ± biotite ± accessory zircon and titanite) and amphibolite (± biotite ± clinopyroxene). These thrust sheets are thought to have formed and amalgamated onto the Laurentian margin prior to Ottawan orogeny. Major and trace element analyses show that the metatonalite rocks have calc-alkaline affinity and amphibolite rocks have both calc-alkaline and tholeiitic affinities, suggesting an arc environment. Laser ablation–inductively coupled plasma–mass spectrometry (LA-ICP-MS) geochronology of zircon from the two thrust sheets yield igneous ages of ca. 1350–1300 Ma for diorite and granodiorite, ca. 1150–1100 Ma ages from Dysart tonalite interpreted to record metamorphic zircon growth, and a ca. 1086 Ma Ottawan metamorphic age from a Dysart amphibolite. The ca. 1150–1100 Ma metamorphic event has not been previously documented within these thrust sheets of the CMBbtz, but correlates well with thermal events in the allochthonous Parry Sound domain to the west, as do ca. 1350 Ma igneous ages of tonalite in both areas. These data support the hypothesis that the CMBbtz and Parry Sound domain may have been initially linked. Widespread ca. 1350 Ma crust along with distinct 1460–1400 Ma depleted mantle model ages (TDM) are also consistent with a shared genesis with the Dysart–Mount Holly suite in New York and Vermont, and support the correlation between the CMBbtz thrust sheets and the Adirondack Highlands–Mount Holly belt as a rifted arc. Alternatively, the CMBbtz thrust sheets and the Adirondack Highlands–Mount Holly belt may represent contemporaneous arc development at different parts of the convergent margin; however, we support the correlation between the CMBbtz thrust sheets and the Adirondack-Highlands–Mount Holly belt.

INTRODUCTION

Studies of ancient continental collisions offer insight into the evolution and architecture of modern-day mountain belts such as the Himalaya or Andes. In particular, the now-exhumed cores of ancient orogens facilitate the investigation of lower crustal processes, as well as the preorogenic histories of mountain belts. In southern Ontario, Canada, the Grenville Province consists of multiple terranes (Fig. 1) representing different arc environments that have been amalgamated at the Laurentian margin and metamorphosed prior to and during the Grenvillian orogeny (1090–980 Ma; as defined by Rivers, 2008; Easton, 1992; Carr et al., 2000; Rivers and Corrigan, 2000). Detailed investigations of these terranes provide clues into the early tectonic history of the Grenville Province and shed light on the lower crustal levels of modern-day convergent plate margins.

This paper contributes new geochemical and geochronological data from tonalitic thrust sheets of the Central Metasedimentary Belt boundary thrust zone (CMBbtz) in Ontario (Fig. 1). Here we use the thrust sheet terminology of Hanmer (1988) to describe the tectonic contact between the metatonalitic rocks discussed here and surrounding igneous and metasedimentary tectonites. These tonalitic rocks play an important role in tectonic models of the Grenville Province, and are thought to represent remnants of an early continental arc rifted by later extension (McEachern and van Breemen, 1993). In most models, this rifted arc is called the Dysart–Mount Holly arc (e.g., Hanmer et al., 2000; Rivers and Corrigan, 2000; McLelland et al., 2010), and the CMBbtz tonalitic rocks represent the western part of the rifted arc, whereas tonalitic rocks in the Adirondack Highlands (New York) and the Green Mountains (Vermont) are the eastern part. However, there are few published constraints on the geochronology and geochemistry of tonalite-dominated thrust sheets in the CMBbtz. The geochronology of these rocks is limited to two isotope dilution–thermal ionization mass spectrometry determinations of zircon performed in the 1980s: one poorly constrained from the Redstone thrust sheet (1344 +93/–32 Ma; van Breemen and Hanmer, 1986), and one unpublished date from the Dysart sheet. The unpublished date is described as 1370–1350 Ma by Lumbers et al. (1990) and by McNutt and Dickin (2012) as 1337 Ma. The geochemical data used to correlate these rocks with tonalite in New York and Vermont are limited to a major element average and select trace elements presented by Lumbers et al. (1990) and Ratcliffe et al. (1991).

This study presents new geochronologic and geochemical data on the tonalite-dominated thrust sheets of the CMBbtz. These data provide new insights into the development and tectonic evolution of the oldest element of the CMBbtz and into the potential links between currently distal Grenvillian terranes.

GEOLOGIC SETTING

Regional Framework

The southern Grenville Province consists of a series of distinct terranes that were juxtaposed during closure of an ocean basin leading up to, and during, the ultimate collision between Laurentia and another continent (thought to be Amazonia; e.g., Hoffman, 1991; Tohver et al., 2006; Jamieson et al., 2007; McLelland et al., 2010, and references therein). From northwest to southeast in southern Ontario, pre-Grenvillian terranes (older than ca. 1090 Ma as defined by Rivers, 2008, and references therein) are found within the Central Gneiss Belt (CGB), Central Metasedimentary Belt (CMB), and the Adirondack Highlands–Mount Holly belt (Fig. 1). Here we use the Grenville terminology following the original CGB and CMB boundaries originally proposed by Wynne-Edwards (1972) and the Adirondack Highlands–Mount Holly belt as proposed by McLelland et al. (2010). The CGB consists of dominantly Paleoproterozoic and Mesoproterozoic rocks thought to have formed within or close to Andean-style arcs on the Laurentian continental margin (Carr et al., 2000, and references therein). Farther east and structurally higher, the CMB, which includes the composite arc belt of Carr et al. (2000), consists of early (1370–1350 Ma) and late (1280–1230 Ma) plutonic rocks indicative of Andean-style magmatism followed by backarc development (e.g., Lumbers et al., 1990; Carr et al., 2000; Hanmer et al., 2000). Closure of the backarc basin was followed by thrusting along the CMBbtz. On the southeastern margin of the Grenville Province, the Adirondack Highlands–Mount Holly belt includes the Mount Holly suite in Vermont and may represent a rifted continuation of the CMB (McLelland et al., 2010; Fig. 1). These terranes were formed and subsequently juxtaposed in a complex series of collisional events, which include events related to arc formation and amalgamation, i.e., the Elzevirian (ca. 1245–1225 Ma) and Shawinigan (ca. 1190–1140 Ma) orogenies, and two Grenvillian events related to later continental collisional orogenic phases: the Ottawan (ca. 1090–1020 Ma) and Rigolet (ca. 1000–980 Ma; Rivers, 1997, 2008; McLelland et al., 2010).

CGB

The older than 1400 Ma pre-Grenvillian terranes of southern Ontario consist of the regions within the CGB, between the Grenville front to the northwest and the CMBbtz to the southeast (Fig. 1; e.g., Wynne-Edwards, 1972; Carr et al., 2000). The CGB consists predominantly of ca. 1900–1450 Ma orthogneiss representative of the Laurentian craton margin that formed as a result of magmatism and metamorphism associated with multiple Andean-style arcs (Carr et al., 2000, and references therein). The terranes within the CGB are characterized by upper amphibolite to granulite facies migmatitic orthogneiss (Culshaw et al., 1997; Carr et al., 2000, and references therein) and contain voluminous 1450–1420 Ma plutons with later ca. 1240 Ma Sudbury metadiabase, eclogite remnants, and ca. 1170–1150 Ma coronitic olivine metagabbro (Culshaw et al., 1997; Ketchum and Davidson, 2000). The allochthon boundary thrust, the boundary between Laurentia and the Laurentian margin, is defined as the thrust separating ca. 1240 Ma Sudbury metadiabase on the footwall and eclogite remnants and ca. 1170–1150 Ma coronitic olivine metagabbro on the hanging wall of this thrust zone (Ketchum and Davidson, 2000). Grenvillian metamorphism within the CGB is dominated by Ottawan (1090–1020 Ma) metamorphic fabrics found within terranes southeast of the allochthon boundary thrust (with the exception of the Parry Sound domain), whereas those terranes northwest of the allochthon boundary thrust preserve both Ottawan and Rigolet (1005–980 Ma) metamorphic signatures (e.g., Carr et al., 2000; Ketchum and Davidson, 2000; Rivers, 2008, and references therein). In contrast, the allochthonous Parry Sound domain of the CGB (Fig. 1) contains mostly 1400–1330 Ma rocks along with younger plutonic rocks, and records ca. 1160 Ma and 1120 Ma metamorphic events (van Breemen et al., 1986; Tuccillo et al., 1992; Wodicka et al., 2000), leading some to correlate the Parry Sound domain with the CMBbtz (Wodicka et al., 1996; Culshaw et al., 1997).

CMB and Its Link to the Adirondack–Mount Holly Belt

The CMB represents a series of allochthonous arcs that were subsequently emplaced onto Laurentia during the Shawinigan orogeny and later metamorphosed during the continent-collisional Grenvillian orogeny (e.g., Culshaw et al., 1997; Timmermann et al., 1997; Carr et al., 2000), and provides an important link between the pre-Grenvillian Laurentian margin and its offshore terranes. These juvenile arc terranes are characterized by abundant calcitic–dolomitic marble and can be divided, from west to east, into the CMBbtz (Hanmer and McEachern, 1992) and the Elzevir terrane, which includes the discrete Harvey-Cardiff, Belmont, Grimsthorpe, Mazinaw, and Sharbot Lake domains (Easton, 1992; Carr et al., 2000; Fig. 1). The CMB is characterized by metasedimentary, metavolcanic, and metaplutonic rocks, the oldest being the ca. 1370–1300 Ma tonalitic rocks in the western CMB (CMBbtz) that are the focus of this study, followed by 1280–1220 Ma granitic to gabbroic plutons and 1090–1060 Ma late plutons across the central and eastern CMB (e.g., Lumbers et al., 1990; Easton, 1992; Carr et al., 2000). For purposes of readability, the prefix “meta” has been removed from the rock names herein. Farther east, as originally proposed by Wynne-Edwards (1972), the Frontenac terrane and Adirondack Lowlands are included within the CMB based on their supracrustal packages and the preservation of 1180–1160 Ma Shawinigan metamorphism and fabrics (following the location of these terranes within the CMB by McLelland et al., 2010). Tonalitic and trondhjemitic 1400–1300 Ma plutons of the Adirondack Highlands and Mount Holly suite have been proposed to be geochemically correlative with other terranes along the pre-Grenvillian Laurentian margin, including the CMBbtz (Ratcliffe et al., 1991; McLelland et al., 2010), and suggest the possibility of a more spatially related genetic origin.

The amalgamation of the various CMB terranes is recorded through pre-Grenvillian orogenic events, including the Elzevirian (ca. 1245–1225 Ma) orogeny involving the terranes and domains containing ca. 1270 Ma plutons (e.g., Corfu and Easton, 1995; Easton and Kamo, 2011) and indicative of initial closure of the continental margin ocean basin, and the Shawinigan (ca. 1190–1140 Ma) contractional orogeny, which has been interpreted as when the CMB and Adirondack Highlands–Mount Holly belt were accreted to the Laurentian margin (e.g., Hanmer et al., 2000; Rivers, 2008; McLelland et al., 2010). Although the orogenic episodes have been well preserved within the CMB, the ultimate origin of arc terranes of the CMB is disputed, and at least two end-member models exist for the formation of the CMB. Carr et al. (2000) suggested that parts of the CMB and Adirondack Highlands–Mount Holly belt (separated by them into the Central Arc Belt and Frontenac-Adirondack Belt) formed outboard of the western Laurentian margin and were later amalgamated onto the Laurentian margin. In contrast, Hanmer et al. (2000) and McLelland et al. (2010) suggested that the CMB and Adirondack Highlands–Mount Holly belt represent elements of an Andean Laurentian margin with backarc basins prior to subsequent reamalgamation with the Laurentian margin during the Shawinigan orogeny. Furthermore, the model proposed by McLelland et al. (2010) involves the rifting of the CMB and Adirondack Highlands–Mount Holly belt from the Laurentian margin prior to the Andean-style Elzevirian orogeny and the Shawinigan contractional orogeny.

CMBbtz and Its Link to the Parry Sound Domain

Following arc amalgamation and juxtaposition of the CMB with the Laurentian margin, the CGB and parts of the CMB were further deformed during Grenvillian continental collisional orogenesis (e.g., Culshaw et al., 1997; Carr et al., 2000; McLelland et al., 2010, and references therein). The extensively studied Grenvillian orogeny is characterized by two orogenic phases: the Ottawan (1090–1020 Ma; pertinent to this study) phase associated with a hot, thick, long-duration orogen, and the Rigolet (1000–980 Ma) phase associated with a rapidly cooled, lower temperature orogen showing overprinting fabrics in regions closer to the Grenville front tectonic zone (Rivers, 2008, and references therein). In parts of the CMB the Ottawan orogenic phase overprints earlier fabrics formed during the Elzevirian and Shawinigan orogenies (Rivers, 2008), and is characterized by granulite facies metamorphism correlated with thrusting of far-traveled allochthonous terranes, including the Parry Sound domain onto the Laurentian margin and into the CGB (Wodicka et al., 1996; Culshaw et al., 1997). Geochronology from the Parry Sound domain indicates the presence of ca. 1430 Ma or younger quartzite, 1400–1330 Ma tonalitic to granodioritic orthogneiss, and anorthosite emplaced both ca. 1350 Ma and ca. 1160 Ma, all of which were overprinted by metamorphic events at 1160 and 1120 Ma (van Breemen et al., 1986; Wodicka et al., 1996, 2000; Culshaw et al., 1997). U-Pb titanite cooling ages ca. 1080 Ma have also been recorded in the Parry Sound domain (e.g., Wodicka et al., 2000, and references therein). The nature and ages of plutonism and sedimentation suggest a genetic link between the Parry Sound domain and the CMBbtz (e.g., van Breemen et al., 1986; Wodicka et al., 1996).

Because the CMBbtz marks the tectonic boundary between dominantly 1900–1450 Ma gneissic rocks of the CGB to the west and the dominantly 1350–1250 Ma CMB to the east, this boundary thrust zone is central to our understanding of the construction and evolution of the Laurentian margin at the time. In detail, the CMBbtz consists of a series of coherent plutonic thrust sheets enveloped in carbonate-rich tectonites; here we follow Hanmer and McEachern (1992), who defined the Redstone, Dysart, and Glamorgan as individual thrust sheets within the 30-km-wide CMBbtz. The northwesternmost thrust sheet, the Redstone, is characterized by coarse-grained, biotite-hornblende tonalitic orthogneiss (1344 +93/–32 Ma; van Breemen and Hanmer, 1986) with rare, relict strongly pleochroic metamorphic orthopyroxene (Hanmer, 1988). The Redstone thrust sheet is the only thrust sheet within the CMBbtz surrounded by orthogneiss tectonites (Fig. 2; Hanmer, 1988), leading some (e.g., Lumbers et al., 1990) to place the Redstone thrust sheet within the easternmost domains of the CGB corresponding to the Laurentian margin. Structurally higher and farther east, the Dysart thrust sheet consists of tonalitic orthogneiss with foliation-parallel amphibolite sheets (Hanmer, 1988) and is surrounded by marble tectonic mélange. Lumbers et al. (1990) categorized the Dysart thrust sheet as an early (ca. 1370–1350 Ma) low-Al2O3-type trondhjemite suite with hornblende as its major mafic component and with some metamorphic clinopyroxene, formed at the expense of hornblende. The Dysart trondhjemite rocks are juxtaposed with the Glamorgan trondhjemite rocks of the adjacent Glamorgan thrust sheet, which are younger (ca. 1280–1270 Ma) and geochemically different (high-Al2O3 type; Lumbers et al., 1990).

RESULTS

Field Relationships and Petrography

Field relationships within the Redstone and Dysart thrust sheets were examined along road cuts northwest of Haliburton, Ontario (Fig. 2). The Redstone thrust sheet exhibits variably strained zones of foliation-concordant tonalite and amphibolite, which are all intruded by late granitic dikes with large K-feldspar. Tonalite (+ amphibole ± biotite ± relict clinopyroxene ± accessory zircon and titanite) and amphibolite (±biotite ± relict clinopyroxene) show foliation defined by the mineral alignment of biotite and amphibole. Locally, these samples display relict clinopyroxene breaking down to amphibole, and annealed fabrics suggestive of a prolonged high-temperature environment following transitional granulite to upper amphibolite facies metamorphism.

The Dysart thrust sheet outcrops are predominantly composed of medium-grained biotite-amphibole tonalitic orthogneiss intercalated with foliation-concordant coarse-grained granite and amphibolite layers to 1 m thick (Fig. 3A). Tonalite (+ amphibole ± biotite ± accessory opaques, zircon, and titanite) displays foliation defined by the mineral alignment of biotite and amphibole, whereas foliation in the amphibolite (± clinopyroxene) is defined by the mineral alignment of amphibole and shows locally developed amphibole rims on relict clinopyroxene. Hanmer (1988) interpreted these continuous and unbranching amphibolite layers to have been mafic sills that intruded into the Redstone and Dysart tonalites. Locally, weakly foliated and intricate contact relationships characterized by coarse amphibole crystals between amphibolite and mafic-poor granodiorite (Fig. 3B) may represent original intrusive mingling relationships or felsic intrusion and metamorphism of the adjacent amphibolite. Regardless, this delicately preserved contact relationship indicates that intrusion of at least some of the felsic plutonic rocks postdate earlier foliation-forming deformation events, which is supported by geochronologic data discussed herein. Late granitic dikes with large K-feldspar (e.g., sample KA11) and large (to 30 cm) biotite laths cover as much as 20% of outcrops and crosscut foliation in the tonalitic and amphibolitic orthogneiss.

Geochemistry

We selected 20 rock samples from 7 well-characterized localities (Fig. 2) for major element analysis. In addition, a subset of 11 samples from the Dysart and Redstone thrust sheets were selected for trace and rare earth element (REE) analysis. Hand samples were collected from low-strain rocks and as far from lithologic boundaries as possible to reduce the possibility of samples being tectonic mixtures between adjacent rock types. Major element and trace element data were collected via X-ray fluorescence and inductively coupled plasma–mass spectrometry (ICP-MS), respectively, by SGS Mineral Services (Toronto, Ontario; see http://www.geochem.sgs.com for details). Whole-rock geochemistry and trace element tables are provided in the Supplemental File1 (Tables DR1 and DR2).

The Redstone tonalite and amphibolite display a linear calc-alkaline trend on the AFM (alkali, Fe, Mg oxides) diagram with some amphibolite samples overlapping into the tholeiitic field (Fig. 4A). The Redstone tonalite suite, consisting of 61%–67% SiO2 and having 3.6–5.5 Na2O/K2O ratios, is tonalite as plotted on a normative Ab-An-Or (albite-anorthoclase-orthoclase) diagram, and quartz diorite and diorite as plotted on a QAP (quartz, alkali feldsar, plagioclase feldspar) diagram (Figs. 4B, 4C). The Redstone amphibolite is slightly more variable than the tonalite and has 46%–47% SiO2 and Na2O/K2O ratios of 1.5, 5.8, and 6.8 (samples KA21, KA28, and KA19, respectively). The Redstone tonalite suite plots predominantly as a medium-K calc-alkaline series (Fig. 4D). The Redstone diorite (KA25) exhibits a negatively sloping chondrite-normalized REE profile and plots in the volcanic arc granite field on the Pearce et al. (1984) granitic discrimination diagram, which utilizes fluid-immobile trace elements (Figs. 5A and 6A). The Redstone amphibolite (KA28) has a relatively flat chondrite-normalized REE profile and plots in the mid-oceanic ridge basalt (MORB) field on the Pearce and Cann (1973) basaltic discrimination diagram, which utilizes fluid-immobile trace elements (Figs. 5B and 6B). Due to the small number of samples, Figures 5 and 6 may not represent the full geochemical variability of the Redstone thrust sheet.

In comparison to the Redstone thrust sheet, the Dysart thrust sheet displays more variable geochemistry with tonalite suite analyses forming a sublinear trend spread throughout a greater portion of the calc-alkaline field on an AFM diagram (Fig. 4A). Dysart tonalite suite samples plot in the tonalitic, trondhjemitic, and granodioritic fields on an Ab-An-Or diagram (Fig. 4B) and consist of tonalite, granodiorite, and one diorite as classified by a QAP diagram (Fig. 4C). These samples contain 66%–72% SiO2 and have 1.3–5.1, 11.7, and 12.4 Na2O/K2O ratios, whereas the Dysart amphibolite has 44%–50% SiO2 and 3.2–4.4 and 6.3 Na2O/K2O ratios. Large ion lithophile element (LILE) concentrations are also variable within the Dysart tonalite samples (Fig. 5C); however, high field strength element (HFSE) concentrations are more uniform and all samples plot consistently in the syncollisional and volcanic arc granitic field based on the Pearce et al. (1984) discrimination diagram (Fig. 6A). Chondrite-normalized REE plots display both profiles with steep heavy rare earth element (HREE) slopes lacking a Eu anomaly and flat HREE slopes with subtly negative Eu anomalies. In contrast to Dysart tonalite, the coarse-grained granodiorite (KA5) does not match the trends within the rest of the suite, as it is Fe and Mg poor, depleted in HFSEs, and has a strong positive Eu anomaly (Figs. 5A, 5C). These characteristics do not fit well within the Dysart suite and suggest that this sample may have been derived from a different magma source that postdates intrusion of Dysart tonalite (see following).

There are two distinct types of Dysart amphibolite: those that have primitive compositions on an AFM diagram and those with more evolved tholeiitic compositions (Fig. 4A). Even in our limited data set, these amphibolite types display an array of trace element signatures. Tholeiitic sample KA2 displays a flat, enriched REE trend and plots in the MORB field on the basaltic discrimination diagram (Figs. 5B and 6B). The other tholeiitic sample, KA31, also displays a flat, enriched REE trend relative to chondrite, but plots in the calc-alkaline field on this diagram (Figs. 5B and 6B). Sample KA38 plots near the calc-alkaline–tholeiitic division on an AFM diagram, displays a flat, relatively less enriched REE trend, and plots in the MORB field on the basaltic discrimination diagram (Figs. 4A, 5B, and 6B). Sample KA9 displays a steep negative REE pattern that is relatively enriched in light (L) REEs and depleted in HREEs, TiO2, and Y with respect to the other Dysart amphibolite samples and plots in the calc-alkaline field on the AFM and basaltic discrimination diagrams (Figs. 4A, 5B, and 6B).

Zircon Geochronology

Zircon from five samples from the Dysart thrust sheet and one from the Redstone thrust sheet were analyzed for U-Pb isotopes to place constraints on the timing of plutonism and metamorphism within the CMBbtz; ∼40 zircon grains from magnetic and nonmagnetic splits of each sample were picked and mounted in epoxy, and imaged using cathodoluminescence (CL) to observe chemical zoning and determine core-rim morphologies within individual zircon grains. U-Th-Pb isotopes were analyzed at the University of California, Santa Barbara, using laser ablation (LA) ICP-MS with 15 and 20 µm spots following the methods of Kylander-Clark et al. (2013). Raw unknown analyses were corrected according to analyses of external zircon standard 91500 (accepted age of 1065 Ma; Wiedenbeck et al., 1995) measured after every fifth unknown; 64 analyses of secondary zircon standard GJ1 (accepted age of 608.5 Ma; Jackson et al., 2004), measured throughout all analytical sessions and using the same correction curve, yielded a 207Pb/206Pb weighted average age of 604.7 ± 6.7 Ma (95% confidence). Reported ages on unknowns are 207Pb/206Pb ages determined by applying the TuffZirc algorithm of Ludwig (2003) to concordant analyses. Interpretations of zircon ages as representative of either igneous or metamorphic crystallization are discussed individually for each sample in the following, and are based on isotopic ratios as plotted on concordia diagrams, core verses rim morphology as observed in CL, and U/Th ratios, such that upper intercept analyses from grain cores are considered to represent igneous crystallization, whereas lower intercept analyses from grain rims and/or with high U/Th ratios characteristic of metamorphic zircon (e.g., Rubatto, 2002) are considered to represent zircon growth in the presence of a metamorphic fluid. The U-Th-Pb isotope values for all unknown analyses are listed in Table DR3 in the Supplemental File (see footnote 1).

Nonmagnetic zircon crystals separated from Redstone diorite KA25 are to 300 µm in diameter, clear, prismatic, and euhedral. Small opaque inclusions within the Redstone zircon are present and may be the cause for the magnetic zircon fraction. Well-preserved oscillatory zoning is seen in all zircon cores, whereas dark rims <20 µm thick on <15% of the zircon grains are poorly developed (Fig. DR1 in the Supplemental File [see footnote 1]). Isotopic analyses from the Redstone zircon cores and mantles yield a group of 20 concordant upper intercept analyses with low U/Th ratios (Fig. 7A) and give an age of 1327 +14/–16 Ma (96% confidence) interpreted as the igneous crystallization age of the Dysart diorite. Seven nearly concordant younger analyses range from 1267 to 1238 Ma, although these younger analyses are not chemically different from the older population with respect to U and Th concentrations, and do not display any apparent spatial relationship with respect to the core-rim morphology observed in CL. It is not clear how these younger analyses may relate to postintrusive thermal events.

Dysart granodiorite sample KA4 contains large, 175–300-µm-diameter, pink and amber euhedral zircon. Imaged using CL, the zircon grain cores have well-preserved oscillatory zoning that is cut by thin (<25 µm thick) U-rich dark rims that are developed in several imaged grains (Fig. DR2 in the Supplemental File [see footnote 1]). Isotopic analyses from zircon cores in sample KA4 yield a group of 19 concordant analyses (Fig. 7B) with low U/Th ratios (<3) that define an age of 1343 +15/–14 Ma (94% confidence) interpreted as the timing of igneous intrusion. Five nearly concordant younger analyses with relatively high U/Th ratios of 7–160 from grain rims yield ages from 1200 to 1100 Ma and are suggestive of metamorphic zircon growth or recrystallization during one or more younger thermal events.

Dysart tonalite KA30 and diorite KA37 has smaller, 175- and 125-µm-diameter, respectively, amber, subhedral zircon crystals. The magnetic fraction of both samples is more fractured, includes more inclusions, exhibits relict oscillatory zoning, and has thin rims, to 20 µm thick (Figs. DR3 and DR4 in the Supplemental File [see footnote 1]). Zircons from the nonmagnetic fraction display homogeneously dark rims to 30 µm thick with mottled cores that display relict oscillatory zoning in CL. U-Pb analyses from cores and rims of both magnetic and nonmagnetic fractions reveal high U concentrations consistent with significant metamictization and Pb loss from grain cores. However, 4 nearly concordant analyses from KA30 grain cores display U/Th < 4 and yield 207Pb/206Pb ages of 1330–1370 Ma, and KA37 yields 6 concordant analyses with U/Th < 4 from grain cores that define a 207Pb/206Pb age of 1313 +21/–26 Ma (97% confidence; Figs. 7C, 7D), and is interpreted as the timing of original igneous intrusion. Analyses associated with grain mantles and rims from both samples give younger ages: KA30 rims yield a group of 13 concordant lower intercept analyses that define an age of 1131 +15/–16 Ma (98% confidence), whereas KA37 rims show a cluster of 5 nearly concordant analyses ranging from 1150 to 1100 Ma (Figs. 7C, 7D). Higher U/Th ratios typically >5 in younger concordant analyses from both of these samples are consistent with zircon growth or recrystallization in the presence of a metamorphic fluid, and are interpreted to date a younger metamorphic event with the Dysart thrust sheet.

Zircon crystals and fragments recovered from the weakly foliated, Fe-Mg–poor Dysart granodiorite KA5 are to 300 µm in diameter, pink, euhedral, and partially fragmented by the mineral separation process. In CL, zircon fragments display oscillatory zoning in grain cores characterized by broad homogeneously dark rims (Fig. DR5 in the Supplemental File [see footnote 1]). Isotopic analyses from KA5 zircon cores and rims yield a group of 27 concordant (Fig. 7E) analyses giving an age of 1112 +14/–13 Ma (95% confidence). Despite this young age, oscillatory zoning observed through grain cores in combination with similar ages for core and rim analyses and the lack of older zircon in this sample imply that this date is the igneous age of this sample, and supports outcrop relationships (Fig. 3B) indicating late intrusion of this weakly foliated granodiorite.

Zircon separated from Dysart tholeiitic amphibolite KA31 is pink to amber and <125 µm in diameter. CL images display well-developed sector zoning in thick U-rich rims that overgrow rarely preserved rounded ∼30 µm U-poor cores with straight or oscillatory zoning (Fig. DR6 in the Supplemental File [see footnote 1]). Analyses from grain cores yield older ages to 1313 Ma suggestive of the original igneous component, whereas ages from 27 concordant grain rims (Fig. 7F) with low U/Th ratios (85% of analyses <7) yield an age of 1086 +13/–11 Ma (95% confidence) associated with the timing of metamorphic zircon overgrowths.

DISCUSSION

Evolution of the CMBbtz Thrust Sheets

The Redstone and Dysart thrust sheet tonalitic rocks display calc-alkaline trends and give igneous ages of 1327 +14/–16 Ma (KA25, Redstone diorite), 1343 +15/–14 Ma (KA4, Dysart granodiorite), and 1313 +21/–26 Ma (KA37, Dysart diorite; Figs. 4A and 7) that are within range of each other and are similar to previous age estimates for CMBbtz tonalite (1344 +93/–32 Ma; van Breemen and Hanmer, 1986; 1370–1350 Ma; Lumbers et al., 1990). The Redstone and Dysart thrust sheet granodiorite-diorite samples plot within the syncollisional and volcanic arc granite field on granitic discrimination diagrams, and have steep negative REE profiles (Figs. 5A and 6A). This suggests an arc setting for the formation of the Redstone and Dysart tonalitic rocks. Redstone diorite sample KA25 and Dysart granodiorite and tonalite samples KA4 and KA10 are depleted in HREEs; this suggests that garnet was stable in the residue during the generation of tonalite melts. In contrast, Dysart tonalite and diorite samples KA30 and 37 have flat HREE profiles, suggesting that garnet was not present during tonalite magma formation. These data indicate the presence of a variety of geochemically distinct magmatic source regions and initially variable magma compositions that are suggestive of an evolving arc setting. It is interesting that there is a marked lack of an inherited zircon component with ages older than 1350 Ma in all of the geochronology samples, which would be expected if the Redstone and Dysart sheets were built on the margin of Laurentia and interacted with older rocks of the CGB. This may indicate that the Redstone and Dysart arc was built on juvenile crust that was subsequently accreted to the continental margin.

The primitive amphibolite from the Dysart thrust sheet (KA9) displays a steeply negative REE profile similar to the HREE-depleted tonalitic rocks and plots within calc-alkaline basalt fields on discrimination diagrams, suggesting that at least some of the Redstone and Dysart amphibolite rocks are of arc affinity. Although the Redstone and Dysart thrust sheets display calc-alkaline trends for tonalitic and amphibolitic samples suggesting arc affinity, there is also evidence for the presence of different suites of tholeiitic amphibolite, variable in trace elements, in both thrust sheets. Although the age of most mafic bodies within the Dysart is uncertain due to sometimes ambiguous crosscutting relationships, the tholeiitic amphibolite rocks suggest MORB affinity, based on flat chondrite-normalized REE patterns and discrimination diagrams, whereas relict zircon cores in sample KA31 suggest a possible ca. 1300 Ma igneous age (Figs. 5B, 6B, and 7F). Some samples have ambiguous trace element compositions. For example, both samples KA2 and KA31 have flat, enriched REE profiles and have 3.04% and 3.51% TiO2 and 187 and 408 ppm Zr, respectively, perhaps indicating a cumulate component in these samples (Figs. 5B, 5D; Tables DR1 and DR2 in the Supplemental File [see footnote 1]). Also, amphibolite samples KA28 (Redstone) and KA38 (Dysart) both plot on the calc-alkaline–tholeiitic division on an AFM diagram, exhibit flat chondrite-normalized REE profiles, and plot in the MORB field on basaltic discrimination diagrams, which may be suggestive of a more primitive tholeiitic composition (Figs. 4A, 5B, and 6B). Therefore, variability exists within the Redstone and Dysart suites and in this study we observe three populations of amphibolite found within the arc: REE-enriched tholeiitic rocks, primitive tholeiites with flat REE profiles, and calc-alkaline rocks with steep REE profiles (Figs. 5 and 6).

Following igneous emplacement of arc rocks in the CMBbtz, the Redstone and Dysart thrust sheets underwent at least two subsequent metamorphic and/or intrusive events. The timing of accretion of the CMB to the CGB is controversial, as there is evidence for two deformation events in the CMBbtz ca. 1190 and 1080–1060 Ma (McEachern and van Breemen, 1993). The 1080–1060 Ma event is interpreted to be the most pervasive deformation, and CGB rocks structurally underlying the CMBbtz only record 1080–1075 Ma metamorphism, which was interpreted by Timmermann et al. (1997) as resulting from accretion at or shortly before ca. 1080 Ma (Bussy et al., 1995; Timmermann et al., 1997). Metamorphic U-Pb 1150–1100 Ma ages determined here from the Dysart tonalite zircon rims (best represented by high U/Th analyses that define an age of 1131 +15/–16 Ma in sample KA30) were unexpected for the CMBbtz and are between the 1190 and 1080–1060 Ma ages observed by others. The ages of these metamorphic rims are within error of igneous zircon with an age of 1112 +14/–13 Ma from REE-depleted Dysart granodiorite KA5. This latter age provides a minimum age for the intrusive contact relationships observed between the granodiorite and adjacent amphibolite (Fig. 3B), and suggests that 1150–1100 Ma metamorphism in the CMBbtz may be related to intrusive events during that period. This heretofore unrecognized intrusive and metamorphic event may represent an isolated event limited to exposures within the Dysart thrust sheet. However, ca. 1120 Ma metamorphic ages have also been reported from the Parry Sound allochthon (Wodicka et al., 2000, and references therein), an area that has been linked to the CMBbtz in tectonic reconstructions (Wodicka et al., 1996; Culshaw et al., 1997). Culshaw et al. (1997) and Wodicka et al. (2000) proposed that the ca. 1120 event may have been the initial encounter of the Parry Sound and CMBbtz with Laurentia, which may be supported by the new ages reported here. The older (1190 Ma) metamorphic ages, observed by McEachern and van Breemen (1993), were not observed in the rocks analyzed in this study.

Metamorphic zircon rims from Dysart amphibolite sample KA31 (1086 +13/–11 Ma), along with previous data (ca. 1080–1060; McEachern and van Breemen, 1993), record the final stages of CMBbtz collisional orogenesis. Metamorphic ages of ca. 1080 Ma have been observed in amphibolite and orthogneiss throughout much of the CGB, and this is interpreted as the age of thrusting of the CMB onto the CGB during the Ottawan orogenic phase (Culshaw et al., 1997; Timmermann et al., 1997; Rivers, 2008).

Regional Correlations of the CMBbtz

Our new data are consistent with the correlation of CMBbtz tonalite thrust sheets with the 1400–1300 Ma calc-alkaline trondhjemite and tonalite of the Adirondack Highlands–Mount Holly belt (Lumbers et al., 1990; Ratcliffe et al., 1991; McLelland et al., 2010). Geochemically, the Dysart and Redstone tonalitic rocks are similar with respect to major and trace elements, with similar high Na2O (3.82%–6.32%) and low K2O (0.37%–2.93%), and only the Al2O3 of the Dysart thrust sheets plotting outside of the fields of Adirondack and Green Mountain tonalitic rocks (13.7%–19.6%, probably reflecting differences in late crystallization history). The correlation between Redstone tonalite and the Mount Holly Complex in Vermont is especially good, but should be considered with caution due to the small sample size. They are coincident on AFM diagrams (Fig. 4A) and have similarly high-Al2O3, low-Yb, and LREE-enriched rocks. Published Nd model mantle extraction ages show the contrast between the CGB terranes (ca. 1500 Ma depleted mantle, TDM, model ages; Muskoka domain), the CMB terranes (ca. 1280 Ma TDM model age), and the strongly correlated TDM model ages (ca. 1460–1400 Ma; ɛNd 1.35 Ga ≈ 4) between the Redstone, Dysart, and Adirondack suites (Daly and McLelland, 1991; Moretton and Dickin, 2013). All of these previous and new data support the tectonic model of the ca. 1350–1300 Ma Dysart–Mount Holly arc, later rifted and now separated by the CMB (e.g., McLelland et al., 2010). Alternatively, the CMBbtz thrust sheets may have formed in a similar environment, resulting in similar geochemical signatures, and during the same interval (ca. 1350–1300 Ma) as the Adirondack Highlands–Mount Holly belt, but were formed at different parts of the convergent margin. Although this is a possibility, we favor a temporal, geochemical, and spatial correlation between the CMBbtz and the Adirondack Highlands–Mount Holly belt.

Based on geochronology and geochemistry in the allochthonous Parry Sound domain, a correlation has been proposed with the CMBbtz (Wodicka et al., 1996). Our new data support this correlation. The Parry Sound domain contains ca. 1314 Ma tonalitic gneiss and 1400–1330 Ma granitic to tonalitic gneiss (Wodicka et al., 1996); these ages overlap the ca. 1343 and 1313 Ma ages obtained in this study for dioritic to granodioritic gneiss within the Redstone and Dysart thrust sheets of the CMBbtz. Parry Sound domain zircon grains recorded granulite and amphibolite facies metamorphic events ca. 1160 and 1120 Ma (van Breemen et al., 1986; Tuccillo et al., 1992; Wodicka et al., 2000). The Parry Sound 1120 Ma age, which is broadly equivalent to the magmatic age of the late Dysart granodiorite and recorded in metamorphic rims on tonalite zircon, is an unusual metamorphic age for the southwestern CMBbtz (McEachern and van Breemen, 1993). The Ottawan metamorphic overprint ca. 1080 Ma observed throughout much of the CGB (Carr et al., 2000, and references therein), in mafic rocks structurally beneath the Parry Sound domain (Bussy et al., 1995; Timmermann et al., 1997), and in U-Pb titanite cooling ages within the Parry sound domain (Tuccillo et al., 1992; Culshaw et al., 1997; Wodicka et al., 2000) correlates with the 1086 Ma metamorphic age found within Dysart amphibolite. In addition to geochronological data, new major and trace element data for mafic metaigneous rocks from the Parry Sound domain also appear to correlate well with the amphibolite rocks from the Redstone and Dysart thrust sheets (Marsh et al., 2012). These similar igneous and metamorphic histories and geochemistry support a link between the CMBbtz and the Parry Sound domain and suggest an early pre-Ottawan deformation (ca. 1120 Ma) possibly associated with thrust emplacement of the Parry Sound domain above the CGB (e.g., Culshaw et al., 1997) followed by subsequent early Grenvillian deformation ca. 1080 Ma (Dysart amphibolite KA31).

CONCLUSIONS

New geochemistry and zircon geochronology data for the Dysart and Redstone tonalitic thrust sheets of the Central Metasedimentary Belt boundary thrust zone in southern Ontario indicate the following.

  1. Tonalitic-dioritic plutonism ca. 1343–1313 Ma formed the major rock types within the Dysart and Redstone thrust sheets.

  2. The Dysart and Redstone tonalite-diorite and the Redstone amphibolite are calc-alkaline with geochemical affinity to an arc environment, but the lack of inherited zircons older than 1350 Ma suggests development on juvenile crust.

  3. The Dysart amphibolite have both calc-alkaline and tholeiitic affinities.

  4. Late 1112 Ma granodioritic intrusions are found within the Dysart thrust sheet, and ca. 1130 Ma metamorphic ages in Dysart tonalite correlate well with Parry Sound thermal events, and may date the initial collision of the Parry Sound and CMBbtz with Laurentia (Culshaw et al., 1997; Wodicka et al., 2000).

  5. The Dysart amphibolite records a ca. 1086 Ma Ottawan metamorphic age.

The calc-alkaline geochemical trends and geochemical volcanic arc affinity suggest an arc origin for the thrust sheets within the CMBbtz. Also, widespread ca. 1350 Ma crust in association with volcanism along with distinct 1460–1400 Ma TDM model ages are consistent with a shared genesis with Dysart–Mount Holly suite plutons. These interpretations support the correlation between the Dysart-Redstone suite and the Adirondack Highlands–Mount Holly belt as a rifted arc (e.g., McLelland et al., 2010). An alternative explanation may be that these similar geochemical trends and similar ages are results of spatially unrelated suites within belts that have similar formation environments at the same time. However, considering the presented data and discussion, we support the former hypothesis as presented in McLelland et al. (2010, and references therein).

The allochthonous Parry Sound domain of the CGB, which contains 1400–1330 Ma orthogneiss with trace element trends and a metamorphic overprint ca. 1120 Ma similar to our results from the Redstone and Dysart thrust sheets presented here, is thought to be closely related to the CMBbtz (van Breemen et al., 1986; Wodicka et al., 1996; Culshaw et al., 1997). Our new data suggest that the Parry Sound domain and the 1350 Ma thrust sheets of the CMBbtz may have been initially linked and both underwent early pre-Ottawan deformation.

We acknowledge the years of work James McLelland has dedicated to studying the region, work that was in part the impetus for this study. Our research in the Grenville Province of Ontario was made possible by the support of the Keck Geology Consortium and National Science Foundation Research Experiences for Undergraduates (NSF-REU) grant 1005122. We thank Steve Dunn and Michelle Markley, the two codirectors of the summer 2011 Keck Consortium project, and project participants Bo Montanye, Naomi Barshi, Callie Sendek, Calvin Mako, Neva Fowler-Gerace, Jacquelyne Nesbit, and Edward Marshall for assistance in the field and the laboratory. We also thank David Linsley, Di Keller, and Sarah Lemon at Colgate for their laboratory support, and Graham Baird, Natasha Wodicka, and two anonymous reviewers for their detailed reviews, which greatly improved this manuscript.

1Supplemental File. PDF file of three tables and six figures. Table DR1: Whole rock geochemistry data for samples from the Dysart and Redstone thrust sheets. Table DR2: Trace element geochemistry data. Table DR3: Zircon U-Pb isotope data from individual spot analyses. Figure DR1: CL images with LA-ICP-MS spots for Dysart granodiorite KA25. Figure DR2: CL images with LA-ICP-MS spots for Dysart granodiorite KA4. Figure DR3: CL images with LA-ICP-MS spots for Redstone diorite KA30. Figure DR4: CL images with LA-ICP-MS spots for Dysart tonalite KA37. Figure DR5: CL images with LA-ICP-MS spots for Dysart amphibolite KA5. Figure DR6: CL images with LA-ICP-MS spots for Dysart diorite KA31. If you are viewing the PDF of this paper or reading it offline, please visit http://dx.doi.org/10.1130/GES00868.S1 or the full-text article on www.gsapubs.org to view the Supplemental File.