Geochronology of igneous rocks in the Sierra Norte de Córdoba (Argentina): Implications for the Pampean evolution at the western Gondwana margin

The tectonic history of the Eastern Sierras Pampeanas metamorphic complex in western Argentina (Fig. 1) has been interpreted in two fundamentally different ways. The first set of models interprets the complex as a continental terrane. This crustal fragment was referred to as the allochthonous Pampean terrane (Ramos, 1988) or “Pampia” (e.g., Ramos, 1999; Ramos et al., 2010), which collided with the western margin of Gondwana in Early Cambrian time (e.g., Rapela et al., 1998a, 2001; Baldo et al., 1999) along a collision zone exposed in the Sierras de Córdoba in the easternmost part of the Eastern Sierras Pampeanas (Fig. 1). The interpretation of the Pampean terrane as a fragment of the autochthonous Río de la Plata craton that rifted from the margin and subsequently collided again (Rapela et al., 1998a, 2001; von Gosen et al., 2002) has been excluded by the models of Rapela et al. (2007) and Casquet et al. (2012). Pampia has more recently been interpreted as a cratonic fragment of Rodinia that collided with the Río de la Plata craton in the east (Ramos et al., 2010, with references therein). As an alternative, it has been proposed that the continental terrane, named MARA (Casquet et al., 2012), consists of basement of Mesoproterozoic or “Grenville” age and represents the Western Sierras Pampeanas (Rapela et al., 2007; Casquet et al., 2012). In this interpretation, the terrane collided with the Eastern Sierras Pampeanas during the Pampean orogeny (Casquet et al., 2012; Iannizzotto et al., 2013). Similar models have been proposed by Siegesmund et al. (2010) and Escayola et al. (2011).

A second set of models suggests that contractional deformation, metamorphism, and magmatism recorded in the Eastern Sierras Pampeanas metamorphic complex resulted from ridge subduction and subsequent ridge collision (Simpson et al., 2003; Gromet et al., 2005; Schwartz et al., 2008). Models that do not invoke collision of a continental fragment (Simpson et al., 2003; Whitmeyer and Simpson, 2004; Schwartz et al., 2008) are supported by the lack of clear basement rocks of the Río de la Plata craton exposed in the Sierras de Córdoba. Paleoproterozoic zircon ages from drill cores in the central and southeastern Córdoba Province, east of the Sierras de Córdoba, provide the first definitive evidence for the western extent of the Río de la Plata cratonic basement (Rapela et al., 2005, 2007). East- or northeast-directed subduction led to widespread arc-related magmatism and formation of an accretionary prism in the western portion of the complex (e.g., Northrup et al., 1998; Rapela et al., 1998a), all of which were accreted to the western Gondwana margin during contractional Pampean deformation (e.g., Piñán-Llamas and Simpson, 2006; Steenken et al., 2011).

Schwartz and Gromet (2004) and Rapela et al. (2007, 2011) proposed that the Río de la Plata craton was displaced along dextral shear zones during or after the Pampean orogeny. In this scenario, the craton did not represent the western Gondwana margin in the Córdoba area before Pampean deformation. Verdecchia et al. (2011) recently inferred that the Río de la Plata craton reached its present relative position during the mid- to Late Cambrian, after the main Pampean tectonothermal event, whereas Iannizzotto et al. (2013) proposed a strike-slip transfer of the craton during dextral mylonitization in the Early Cambrian.

A central factor in these tectonic models is the inferred role of a thick clastic and mostly turbiditic sequence in northwest Argentina known as the Puncoviscana Formation (sensu lato). Deposition of the sequence comprises the late Neoproterozoic–Early Cambrian interval (e.g., Aceñolaza and Miller, 1982; Aceñolaza, 2003; Aceñolaza and Aceñolaza, 2005; Adams et al., 2008, 2011; Escayola et al., 2011; Toselli et al., 2012) before the Early Cambrian Pampean orogeny and subsequent unconformable clastic deposition of the Mesón Group in the early Late Cambrian (Adams et al., 2011, with discussion therein). The Puncoviscana Formation succession grades into higher-grade metamorphic rocks to the south (e.g., Willner et al., 1985; Willner and Miller, 1986; Aceñolaza et al., 1988; Willner, 1990). Metaclastic sequences in the southern and southeastern parts of the Eastern Sierras Pampeanas (e.g., Sierras de San Luis, Sierras de Córdoba; Fig. 1) have been interpreted as time equivalents of this formation (e.g., Prozzi and Ramos, 1988; Prozzi, 1990; Baldo et al., 1996; Rapela et al., 1998a, 2007; von Gosen and Prozzi, 1998; Söllner et al., 2000; von Gosen et al., 2002; Schwartz and Gromet, 2004).

Deposition of the Puncoviscana Formation sediments in a large approximately N-S–trending basin was related either to the passive margin of western Gondwana (e.g., Willner et al., 1987; Aceñolaza et al., 1988; Rapela et al., 1998a; Piñán-Llamas and Simpson, 2006; Adams et al., 2008, 2011) or a foreland basin in front of the developing Pampean orogen in the east (e.g., Kraemer et al., 1995; Keppie and Bahlburg, 1999; Zimmermann, 2005). Escayola et al. (2011) proposed that Puncoviscana Formation sediments older than 530 Ma were deposited in the arc-trench gap of the west-facing Pampean arc and/or the associated trench, whereas sediments younger than 530 Ma were deposited in a syncollision foreland basin. In the model of Rapela et al. (2007), the Neoproterozoic sediments represent a forearc sequence along the proto-Pacific Gondwana margin, affected by dextral strike-slip displacement at its eastern border, whereas deposition of the Puncoviscana Formation in an aulacogenic basin was proposed by Toselli et al. (2012).

In the northeastern continuation of the Sierras de Córdoba, the Sierra Norte de Córdoba is underlain by widely distributed granitoid intrusions (Fig. 1) commonly referred to as the Sierra Norte–Ambargasta batholith. The intrusions are thought to represent a calc-alkaline arc (e.g., Lira et al., 1997) emplaced from ca. 555 to 525 Ma, with later peraluminous magmatism between ca. 525 and 515 Ma (Miró et al., 2005; Schwartz et al., 2008). The magmatism was related to east-directed subduction beneath the western Gondwana margin, perhaps the Río de la Plata craton (cf. Schwartz et al., 2008; von Gosen and Prozzi, 2010), and possibly combined with the approach of the Pampean terrane in the west (e.g., von Gosen and Prozzi, 2010). Subduction is interpreted as dextral oblique (von Gosen and Prozzi, 2010), leading to final oblique collision (e.g., Rapela et al., 2007; Casquet et al., 2012) with dextral transpressive kinematics (e.g., von Gosen and Prozzi, 2010; Iannizzotto et al., 2013).

The Cambrian magmatic rocks are intruded by Ordovician granitoids in the Sierra Norte and Sierras de Córdoba (e.g., Baldo et al., 1998; Rapela et al., 1998a; Bonalumi and Baldo, 2002), which are related to the Famatinian magmatic arc in the western part of the Eastern Sierras Pampeanas complex (Fig. 1). The Famatinian continental arc (Pankhurst et al., 1998, 2000; Dahlquist et al., 2008, 2013; Ducea et al., 2010) is interpreted as the product of the Upper Cambrian–Middle Ordovician subduction beneath the Eastern Sierras Pampeanas, and it was demonstrably affected by compressive deformation and metamorphism during subsequent collision of the Cuyania (Precordillera) terrane (e.g., Pankhurst et al., 1998; Rapela et al., 1998c; von Gosen and Prozzi, 1998, 2005c; von Gosen et al., 2002; Sato et al., 2003; Thomas and Astini, 2003; Astini and Dávila, 2004; González et al., 2004; Ramos, 2004; Steenken et al., 2006).

Herein, we provide U-Pb zircon data from magmatic rocks with different structural relationships to determine the timing of sedimentation and different stages of compressive deformation(s) and igneous activity in the Sierra Norte de Córdoba, representing the easternmost segment of the Eastern Sierras Pampeanas close to or at the (unexposed) Río de la Plata craton as western pre-Andean Gondwana margin. The relationships define a continuum of partitioned contractional and dextral deformation that affected clastic sedimentary sequences deposited on this margin. The new data allow regional correlation of stratigraphic and deformational histories and greatly strengthen earlier conclusions for the importance of dextral transpression during Pampean orogenesis (von Gosen and Prozzi, 2010; Iannizzotto et al., 2013).

Intrusive rocks exposed in the Sierra Norte de Córdoba and their northern continuation in the southernmost part of the Santiago del Estero Province enclose thin N-S– to NE-SW–striking screens of low-grade metamorphic rocks (Fig. 2) that represent relics of a former widely distributed clastic sequence. The metaclastic rocks have been extensively studied (Lucero, 1958, 1969; Methol, 1958; Quartino et al., 1978; Lucero Michaut, 1979, 1981; Lucero Michaut and Daziano, 1985, 1988; Llambías et al., 2003; Martino et al., 2004; von Gosen and Prozzi, 2005b, 2009, 2010; von Gosen et al., 2009) and either assigned a Proterozoic or Proterozoic–early Paleozoic age (e.g., Lucero, 1958, 1969; Lucero Michaut, 1979, 1981) or interpreted as time equivalents of the Neoproterozoic–Early Cambrian Puncoviscana Formation of northwest Argentina (e.g., Llambías et al., 2003; von Gosen and Prozzi, 2005b, 2009, 2010; von Gosen et al., 2009) with a southeastern extent in the Sierras de Córdoba (see previous section).

The metasedimentary rocks were assigned to the La Lidia–Simbol Huasi Formation by Lucero Michaut (1979) based on comparison with the La Lidia Formation of Beder (1922). More recent studies have shown that the sequence can be separated into two different units associated with acid meta-igneous rocks (e.g., von Gosen and Prozzi, 2005a, 2005b, 2005d, 2009, 2010). The Lower Unit, exposed in the La Lidia, Talayaco, and Agua del Río areas (Fig. 2), consists of thick metaconglomerate layers associated with metasandstone, quartzite, and quartz schist interlayered with metarhyolite to metarhyodacite volcanic units. The Upper Unit includes quartzite and metasandstone, with minor metasiltstone to mudstone, and some intercalated metarhyolite to metadacite layers. Metaconglomerate in the Upper Unit only occurs at Rodeito and west and northwest of Simbol Huasi (Fig. 2). Minor local metaconglomerate units were also found in the El Tío and Cerro Resbaladero sections.

The clastic members of the Lower Unit have been interpreted as a fluvial to alluvial-fan section deposited in proximity to synsedimentary normal faults (von Gosen and Prozzi, 2005b, 2005d, 2009). Deposition of the Upper Unit presumably took place under continuing subsidence. The entire succession is related to a passive-margin setting, most likely on the western margin of the Río de la Plata craton (cf. von Gosen and Prozzi, 2005b, 2005d, 2009, 2010; Schwartz et al., 2008; von Gosen et al., 2009). A metarhyolitic ignimbrite at the base of the section in the La Lidia area gave a conventional thermal ionization mass spectrometry (TIMS) U-Pb zircon age of 584 +22/–14 Ma (Llambías et al., 2003) that was considered to be the age of crystallization by the authors. However, the ignimbrite could have included different zircon populations that might have led to a mixed age. U-Pb secondary ionization mass spectrometry (SIMS) dating of detrital zircons from a quartzite gneiss in the southernmost part of the Sierra Norte de Córdoba suggests a 560 Ma maximum age for deposition of its protolith (Iannizzotto et al., 2013). Both dates point to a maximum Neoproterozoic depositional age for the clastic units.

The clastic succession was affected by at least one compressive deformation event (D₁) that led to folding and cleavage or foliation formation. D₁ deformation is interpreted to record strain partitioning with a dextral component related to distinct shear zones (von Gosen and Prozzi, 2010). The clastic units and magmatic rocks that intrude them were locally deformed during subsequent dextral shearing and mylonitization (D₂ event: von Gosen and Prozzi, 2010; see also Miró et al., 1999; Martino, 2003; Iannizzotto et al., 2011, 2013). Metamorphism accompanying deformation ranged from subgreenschist- to lower-greenschist-facies conditions. The metasedimentary rocks have also been locally overprinted by contact metamorphism of variable grade and areal extent related to intrusive events (e.g., Guereschi and Martino, 2002; Martino and Guereschi, 2005).

Many published ages from this area and the adjacent southernmost part of the Santiago del Estero Province in the north are based on K-Ar and some Rb-Sr data that span the Neoproterozoic–Silurian interval (e.g., Castellote, 1982, 1985a, 1985b; Rapela et al., 1991; Koukharsky et al., 1999; Massabie et al., 2002; Millone et al., 2003; overview in Leal et al., 2003). Several U-Pb zircon ages from the Sierra Norte de Córdoba, however, give more precise time constraints (e.g., Rapela et al., 1998a, 1998b; Stuart-Smith et al., 1999; Leal et al., 2003; Llambías et al., 2003; Miró et al., 2005; Schwartz et al., 2008; Siegesmund et al., 2010; Iannizzotto et al., 2011, 2013) and point to a short-lived deformational and magmatic history. Iannizzotto et al. (2013) established an Early Cambrian age for dextral mylonitization in the southernmost part of the Sierra Norte through U-Pb zircon (SIMS) dating of the pre-/synmylonitic Juan García granodiorite (537 ± 4 Ma) and postmylonitic Villa Albertina granite (530 ± 4 Ma).

We present U-Pb zircon data from eight magmatic rocks collected in the Sierra Norte de Córdoba, north of the Iannizzotto et al. (2013) study area, in the context of their differing structural relationships with adjacent units. In the text, we use the terms “Pampean” (latest Neoproterozoic–mid-Cambrian) and “Famatinian” (early-mid-Ordovician) for the two main orogenic events (Dahlquist et al., 2008, with references therein). Concerning the assignments of the isotopic dates, we have used the time scale of Walker et al. (2012).

Samples for this study were collected at localities identified during structural studies of clastic metasedimentary rocks and their contacts with extrusive and intrusive igneous rocks in the Sierra Norte area and its continuation in the southernmost part of the Santiago del Estero Province (Fig. 2; von Gosen and Prozzi, 2009, 2010). The Lower Unit, exposed in the western strip of the Agua del Río occurrence (Fig. 3A), includes vertical to overturned metaclastic rocks intercalated with several layers of foliated metarhyolite to metarhyodacite. The volcanic layers are interpreted as flows and pyroclastic deposits rather than sills. A sample of massive red metarhyolite with well-developed quartz phenocrysts was taken from the northern portion of the occurrence (CE-54D; Fig. 3A).

In the south, the Talayaco metaconglomerate of the Lower Unit is intruded by a granite that is widely exposed in the east (Fig. 3B). Both the granite and the clastic succession to the west experienced compressive deformation. The metaconglomerate and enclosing quartz schist are foliated and show an increase in deformation toward the granite contact to the east. The granite at the contact is mylonitized and thrust westward over the metaconglomerate, with a local thin layer of mylonitized quartz schist within the thrust zone. West-directed reverse faulting was followed by dextral shearing, recorded by granite mylonite also east of the thrust contact (von Gosen and Prozzi, 2010). A sample was collected from the dextral mylonitic granite for U-Pb analysis (CE-56D; Fig. 3B).

The wide, NE-SW–trending strip of dextral granite mylonites south of San Pedro Norte (Fig. 2) is part of a long mylonite zone, which was named “faja de deformación Sauce Punco” by Martino et al. (1999a, 1999b) and described by Martino et al. (1999a, 1999b), Miró et al. (1999), Martino (2003), and von Gosen and Prozzi (2010). The deformation zone also contains thin bands of mylonitized schist that are interpreted as parts of the La Lidia–Simbol Huasi Formation (von Gosen and Prozzi, 2009). South of San Pedro Norte (southwest of Santa Cruz), the granite and schist mylonites are intruded by a granite porphyry (Fig. 4; von Gosen and Prozzi, 2010). At its western margin, the granite porphyry is affected by several subvertical, meter-thick mylonite zones. Northeastern exposures of the intrusion are either undeformed or locally affected by a conjugate set of thin dextral and sinistral shear zones. The shear zones die out over a short distance along strike into magmatic fabrics. The eastern margin of the intrusion contains no evidence of ductile deformation. These observations suggest that the granite porphyry intruded the mylonite zone during dextral shearing and was affected only by the final stage of the overall dextral mylonitization (von Gosen and Prozzi, 2010). Prekinematic emplacement for the intrusion is ruled out because it is not penetratively foliated and is discordant to the mylonite foliation in the country rocks at its northern margin. A sample of the synkinematic intrusion was collected east of the mylonite zone that marks the western contact of the intrusion against the mylonitized schist (CR-49D; Fig. 4).

Subvertical to slightly overturned metaclastic rocks of the Lower Unit intercalated with acid meta-igneous layers in the Agua del Río area are intruded by a coarse-grained, amphibole-bearing dacite porphyry (Fig. 3A). The country rocks were foliated and rotated into the subvertical position before emplacement of the intrusion, and the porphyry is locally dextrally sheared at its eastern intrusive margin (von Gosen and Prozzi, 2010). Thus, the emplacement probably took place during the final stage of Pampean compressive deformation. Minor deformation at its eastern contact indicates that intrusion of the porphyry was late synkinematic with shearing active during cooling. A sample of the dacite porphyry was taken in the northern part of the Agua del Río occurrence (CE-55D; Fig. 3A), west of the sheared intrusive margin where the rock was affected by only a weak deformation.

Deformed metaclastic rocks and intercalated acid meta-igneous rocks in the eastern portion of the large exposure in the El Tío area (Fig. 5B) are intruded by a porphyritic biotite-amphibole granite (El Tío granite) that is included with the Cambrian Ambargasta granitoids (e.g., Miró and Sapp, 2000). The granite contains angular xenoliths of the foliated clastic country rocks with different sizes. Chilled margins occur at the xenolith contacts, suggesting a high level of granite intrusion with brittle fracturing of the crust (von Gosen and Prozzi, 2010). The intrusion produced contact metamorphism of the deformed country rocks. The granite was not ductilely deformed. A sample was collected from the intrusion near its southern contact with the metaclastic country rocks (CE-66D; Fig. 5B).

Folded and cleaved rocks of the approximately NNE-SSW–striking metaclastic succession in the Rodeito area (Fig. 2) are intruded by a hypabyssal rhyolite to dacite. The intrusion also cuts the thin layer of the Rodeito metaconglomerate. A sample of the intrusion was collected just west of the metaconglomerate (CD-22D; Fig. 5A).

The small El Escondido occurrence of clastic rocks in the southernmost Santiago del Estero Province (Fig. 1) lies in a morphological depression between the Sierra de Ambargasta to the west and Sierra de Sumampa to the east (Fig. 6). The El Escondido Formation sandstones and conglomerates record synsedimentary folding but are not ductilely deformed (Martino, 2004; von Gosen et al., 2005). They are entirely different from the low-grade clastic rocks in the Sierra Norte de Córdoba, hornfelses in the La Clemira area, and medium-grade metamorphic rocks of the Pozo del Macho Formation (Castellote, 1985b) and equivalents in the east and southeast. The clastic rocks depositionally overlie a basal rhyolite in the El Escondido area. The rhyolite is intruded by a granite that does not cut but appears as angular clasts in the overlying sedimentary rocks of the El Escondido Formation (Fig. 5C; Martino, 2004; von Gosen et al., 2005). A sample was taken from the undeformed rhyolite beneath the El Escondido Formation sedimentary rocks (CD-7D). A sample of the El Escondido granite was analyzed to obtain a maximum age for the El Escondido clastic deposits and a minimum age for the Pampean igneous history (CD-8D; Fig. 5C).

Zircons for this study were analyzed as fractions of one to three grains using the thermal ionization mass spectrometry (TIMS) method and/or by the secondary ionization mass spectrometry (SIMS) method. SIMS analyses were employed on samples analyzed by the TIMS method to help resolve problems associated with inheritance and Pb loss. The sample preparation and analytical procedures for each method are summarized in  Appendixes 1–3. The results of the SIMS analyses are shown in Table 1. The results of the TIMS analyses are presented in Table DR1 and Figures DR1–DR3.¹ Although the TIMS results are not extensively discussed in the text, the data are presented because they provide insight into the complex zircons observed in these samples, demonstrate some of the undetected problems with other published data, and provide a rationale for integrated SIMS and chemical-abrasion TIMS (Mattinson, 2005) analysis in future work. Rare earth element (REE) data collected simultaneously with the SIMS U-Pb analyses help to distinguish disturbed and undisturbed zircon populations and are presented in Table 2. The age results provide time markers for the sedimentary, tectonic, and igneous histories in this easternmost part of the Eastern Sierras Pampeanas unit.

The zircon crystals analyzed from the metarhyolite of the Agua del Río occurrence (CE-54D) are middle to long prismatic, colorless clear, or have a slight honey-like color. Grains show clear oscillatory growth zoning with some inherited cores, opaque inclusions, and zones of resorption/recrystallization (Fig. 7A). SIMS analysis of oscillatory zoned domains gave ²⁰⁷Pb-corrected ²⁰⁶Pb/²³⁸U ages ranging from 495 to 546 Ma. Assuming the younger ages reflect Pb loss, as is demonstrated by the TIMS data for this sample (Fig. DR1 [see ^{footnote 1}]), 7 of 12 zircon analyses give a concordia age of 535 ± 5 Ma (Fig. 7B), which is interpreted as the crystallization age of the metarhyolite. The light REE elements (LREEs) are variable for three of the analyses (Fig. 8A), but there is little correlation of REE disturbance with observed U-Pb age.

The analyzed zircons from the granite mylonite at Talayaco (CE-56D) are short prismatic, partly isometric, and colorless to pale white. In cathodoluminescence (CL) images, the grains are oscillatory zoned with occasional cores (Fig. 9A). The SIMS results gave ²⁰⁷Pb-corrected ²⁰⁶Pb/²³⁸U ages ranging from 478 to 547 Ma. The LREEs in roughly half of the grains are elevated (Fig. 8B), indicating modification of the original igneous signature by alteration. The disturbed REE patterns are associated with younger ²⁰⁶Pb/²³⁸U ages. Assuming discordance is due to Pb loss, the oldest nine analyses give a concordia age of 533 ± 4 Ma (Fig. 9B).

The granite porphyry sample CR-49D has colorless short prismatic to elongate zircons with oscillatory zoned mantles that surround either similarly zoned cores or obvious cores with irregular zoning patterns (Fig. 10A) that might be related to rapid growth in the melt. Thin (5–20 μm) CL-dark, higher-U rims are present on roughly half of the grains. Single-grain TIMS analyses demonstrate significant Pb loss (Fig. DR2 [see ^{footnote 1}]), whereas the SIMS data show complexities of combined inheritance and Pb loss (Fig. 10B). SIMS analyses from domains that have suffered Pb loss are generally associated with elevated LREE patterns (Fig. 8C). Excluding older inherited cores and younger ages associated with disturbed REE concentrations, 6 of 16 analyses provide a weighted mean ²⁰⁶Pb/²³⁸U age of 534 ± 5 Ma (Fig. 10B). We interpret this date as the age of zircon crystallization.

Zircons from the Agua del Río dacitic porphyry (CE-55D) yielded stubby to elongate, prismatic, colorless zircons with opaque and colorless inclusions common and no obvious cores (Fig. 11A). The grains have well-developed oscillatory zoning, with thick (10–30 μm) CL-dark, higher-U rims common. SIMS analyses targeting uniform oscillatory zoned domains, including the higher-U rims, have uniform REE patterns (Fig. 8D), and 15 of 16 U-Pb analyses give a concordia age of 531 ± 4 Ma (Fig. 11B). One younger SIMS analysis and the scatter observed in the TIMS data (Fig. DR3 [see ^{footnote 1}]) are attributed to Pb loss.

The colorless clear zircons from the El Tío granite (CE-66D) are subequant to elongate, euhedral, and oscillatory zoned grains (Fig. 12A). Most grains have CL-dark, higher-U rims, and a minor subset of the population shows evidence for inherited cores. TIMS analyses of single grains indicate clear Pb loss. LREE patterns for three grains analyzed by SIMS show variability (Fig. 8E), indicating disturbance of the systematics. Excluding the four oldest analyses and one obviously disturbed analysis, the remaining seven give a concordia age of 530 ± 4 Ma (Fig. 12B), which is interpreted as the crystallization age of the zircons. The older rejected analyses are interpreted to represent slightly older xenocrystic cores, assuming the U-Pb systematics have not been disturbed.

Subequant to elongate oscillatory zoned zircons from the Rodeito rhyolite to dacite (CD-22D; Fig. 13A) give a range of discordant to concordant analyses that result from the combined effects of inheritance and Pb loss. Several analyses have elevated LREE patterns (Fig. 8F), but the analyses are overall consistent with one another and do not vary as a function of U-Pb age. Excluding older ages assumed to reflect inheritance and younger ages attributed to Pb loss, 10 of 18 analyses give a concordia age of 523 ± 5 Ma (Fig. 13B), which is interpreted to reflect the age of crystallization of the zircons.

The sample from the El Escondido rhyolite (CD-7D) contains clear colorless, oscillatory zoned zircons with minor inclusions (Fig. 14A). Five analyses of higher-U core and rim portions of the grains give discordant results with younger ²⁰⁶Pb/²³⁸U ages and elevated LREE patterns (Fig. 8G). Rejecting an additional three analyses with slightly younger ²⁰⁶Pb/²³⁸U ages but relatively undisturbed REE spectra compared to the concordant analyses, the remaining seven analyses define a concordia age of 519 ± 4 Ma (Fig. 14B).

The El Escondido granite (CD-8D) contains a heterogeneous population of colorless to yellow-brown, elongate to equant, euhedral zircons. The grains typically have cores mantled by oscillatory zoned zircon (Fig. 15A). SIMS analyses yield ²⁰⁶Pb/²³⁸U ages ranging from 470 to 544 Ma. The four oldest ages are from domains that differ in REE pattern with slightly lower middle REE abundances (Fig. 8H). Excluding these four analyses as inherited xenocrysts and the two youngest analyses that record Pb loss, the remaining 10 give a ²⁰⁶Pb/²³⁸U weighted mean age of 521 ± 4 Ma (Fig. 15B).

The crystallization age of 535 ± 5 Ma for the rhyolite at Agua del Río (CE-54D) suggests an Early Cambrian (Terreneuvian) age (time scale of Walker et al., 2012) for this part of the clastic succession with intercalated rhyolite to rhyodacite layers. The new age is within uncertainty of a K-Ar whole-rock date of 527 ± 21 Ma from a rhyolitic porphyry in the La Lidia area (Massabie et al., 2002). Combining the SIMS age with the older conventional U-Pb zircon age from the basal rhyolite that is overlain by the clastic succession in the La Lidia area to the south (584 +22/–14 Ma; Llambías et al., 2003) and the 560 Ma detrital zircon age (U-Pb, SIMS), indicating the maximum age of deposition in the southernmost part of the Sierra Norte (Iannizzotto et al., 2013), suggests that sedimentary deposits along with the underlying and intercalated acid volcanic rocks are Neoproterozoic to Early Cambrian in age.

The SIMS age of 534 ± 5 Ma for the synkinematic granite porphyry that intruded the dextral mylonite zone south of San Pedro Norte (CR-49D), interpreted as the age of crystallization, provides a time marker for the age of dextral shearing and mylonitization in this part of the Sierra Norte area related to the second stage of the Pampean deformation (D₂). It falls in the time span of dextral mylonitization in the southernmost part of the Sierra Norte found by Iannizzotto et al. (2011, 2013).

The 533 ± 4 Ma SIMS age from the granite mylonite of Talayaco (CE-56D) is interpreted as the crystallization age of the protolith. This date also provides a lower age limit of sedimentation in this part of the Sierra Norte, since the granite intruded the section, and an upper age limit of D₂ mylonitization, since the granite is deformed. This age is older than the 531 ± 4 Ma SIMS date from the dacitic porphyry of Agua del Río (CE-55D), which we interpret as the age of crystallization for this sample. It provides the approximate timing of the final stage of compressive Pampean D₂ deformation.

The SIMS zircon date of 530 ± 4 Ma from the postkinematic El Tío granite (CE-66D) suggests that the intrusion is not part of the undeformed Cambrian Ambargasta granite pluton (Figs. 5B and 6), the age of which was broadly estimated in the La Clemira area by K-Ar biotite dates of 517 ± 15 and 500 ± 15 Ma (Castellote, 1982). The SIMS date suggests that compressive deformation and regional metamorphism of the clastic rocks in the Sierra Norte area (La Lidia–Simbol Huasi Formation) should be Early Cambrian in age. The results presented here broadly point to a time span between 535 ± 5 and 530 ± 4 Ma for the first and second Pampean compressive deformation events (D₁, D₂), accompanying regional metamorphism, and subsequent uplift in the Sierra Norte de Córdoba. This time span is similar to that obtained by Iannizzotto et al. (2013) from the southernmost part of the Sierra Norte. The narrow age range is consistent with the SIMS dates of 519 ± 4 Ma from the undeformed El Escondido rhyolite (CD-7D) and 523 ± 5 Ma from the Rodeito rhyolite to dacite (CD-22D), both interpreted as reflecting the age of crystallization of igneous rocks that record acid magmatism after clastic sedimentation and cessation of the D₂ deformation in the Sierra Norte de Córdoba.

The El Escondido granite (CD-8D), with an interpreted emplacement age of 521 ± 4 Ma (SIMS), records late-stage Pampean magmatism. Overlying clastic rocks of the El Escondido Formation can be estimated to be mid-Cambrian or younger in age. The lack of deformation of the El Escondido granite indicates that no ductile deformation has taken place during Middle Cambrian–Ordovician and later times. Thus, Pampean deformation and metamorphism terminated before the Middle Cambrian, and this area was not affected by later Famatinian compression and regional metamorphism. The slightly deformed Ojo de Agua granite to the east, however, yielded a late Early Cambrian age (515 ± 4 Ma, U-Pb, zircon, SIMS; Stuart-Smith et al., 1999), suggesting that Pampean deformation may have locally persisted into Middle Cambrian time.

The SIMS date from the Rodeito rhyolite to dacite (CD-22D) is older than the age of the Los Burros dacite that intruded the eastern part of the Rodeito occurrence of clastic metasedimentary rocks (Fig. 5A). In the Totorillas area, the undeformed Los Burros dacite has been shown to be lowermost Middle Cambrian in age (512.6 ± 3.5 Ma; U-Pb, zircon, SIMS; Leal et al., 2003), which gives a further constraint for the lower time limit of the Pampean deformational history.

The isotopic dates provide additional time constraints for the different stages of the sedimentary, tectonic, and magmatic evolution in this easternmost segment of the Eastern Sierras Pampeanas and provide a more robust interpretation of the tectonic evolution of the pre-Andean Gondwana margin. Our present interpretation partly modifies those given by Rapela et al. (2007), von Gosen and Prozzi (2010), Casquet et al. (2012), and Iannizzotto et al. (2013). The isotopic dates presented here partly overlap in the range of errors, reflecting the complex systematics of the zircons as defined by both the TIMS and SIMS data. Use of both techniques enhances the characterization of zircon systematics and strengthens the reliability of the interpreted crystallization ages. Additional chemical-abrasion-TIMS analysis (Mattinson, 2005) of grains screened for simple systematics is an obvious next step in further refining the timing of magmatism and deformation in this setting. Nevertheless, we interpret the chronological succession of different events as follows (compare interpretative chart of Fig. 16 and diagrams of Fig. 17). The geographic orientations cited refer to present coordinates.

The time interval of deposition of the clastic sediments in the Sierra Norte de Córdoba is constrained by two isotopic dates. Detrital zircons from metasedimentary rocks in the southernmost Sierra Norte have a minor peak at 560 Ma defined by individual ages ranging from 537 ± 9 Ma to 570 ± 7 Ma (Iannizzotto et al., 2013). Considering the unresolved potential for Pb loss to skew the detrital zircon ages to younger ages, these results broadly limit the time of deposition to younger than ca. 560 Ma. Given this uncertainty, the detrital data are consistent with the greater age of 584 +22/–14 Ma reported by Llambías et al. (2003) from the basal metarhyolite (ignimbrite) beneath the clastic succession in the La Lidia area. Additional high-precision data are needed to further refine limits on the depositional age of these units.

The new crystallization (SIMS) age of 535 ± 5 Ma for the intercalated and deformed metarhyolite at Agua del Río suggests that cessation of sedimentation and the onset of D₁ deformation took place sometime after this date. The intrusion of the later mylonitized Talayaco granite (533 ± 4 Ma, SIMS, crystallization age) into the clastic deposits provides an additional approximate time marker in the Early Cambrian. However, the precise lower age limit of clastic sedimentation remains uncertain. These ages, combined with the loose constraint provided by Iannizzotto et al. (2013), suggest that deposition of the clastic sediments and acid magmatism in the Sierra Norte area took place in the Ediacaran–Early Cambrian interval. Thus, the metasedimentary rocks are time equivalents and represent the southern continuation of the Puncoviscana Formation (sensu lato) of northwest Argentina (see introductory section).

The clastic deposits in the Sierra Norte are characterized by metaconglomerates and fine-grained metaclastic rocks with intercalated metarhyolites and metadacites. The clastic deposits include locally derived detritus of clastic, magmatic, and metamorphic rocks with paleocurrent indicators indicating an approximately west-directed sedimentary transport (von Gosen and Prozzi, 2009, with further information therein). Geochemical analysis of clastic metasedimentary rocks shows that those from the Sierra Norte de Córdoba were deposited in a passive-margin setting, whereas those from the Sierras de Córdoba and Sierra de Guasayán point to an active arc setting (von Gosen et al., 2009).

Observations from the clastic succession in the Sierra Norte combined with the presence of acid magmatism suggest a postrift passive-margin setting for the sequence (von Gosen and Prozzi, 2009, 2010; von Gosen et al., 2009). Although geochemical data from the intercalated magmatic rocks do not exist, we interpret the magmatism as extension-related prior to emplacement of the arc intrusions (see following). The precise configuration of the margin remains uncertain.

It is unclear whether and to what extent the clastic sequence is underlain by basement rocks because rocks with a clear Proterozoic age are still not known from the Sierra Norte area. The only possible candidate is the Pozo del Macho Formation (compare Martino and Guereschi, 2004). At least one deformation event and regional metamorphism of these mica schists and gneisses and equivalent rocks in the southeast (Fig. 6) may be Neoproterozoic in age. This is consistent with the interpretation of Martino and Guereschi (2004) and two K-Ar whole-rock dates of Castellote (1985b). If such an age is confirmed by later studies, then these rocks may represent part of the Proterozoic “basement” below the La Lidia–Simbol Huasi Formation (Fig. 16).

Our interpretation of a west-facing passive margin (Fig. 17A) during the Ediacaran–Early Cambrian interval is comparable with that of Schwartz et al. (2008), although they proposed a Neoproterozoic age. This contrasts to models of Rapela et al. (2007) and Iannizzotto et al. (2013), who postulated a forearc setting and to that of Siegesmund et al. (2010), who proposed deposition in arc-related basins. The passive-margin interpretation, however, is similar to the geotectonic model of Casquet et al. (2012). These authors do not exclude a passive-margin setting for the older part of the Puncoviscana Formation on the western margin of the Kalahari craton with the Clymene Ocean in the west (their Fig. 4). Based on Cordani et al. (2013), we think that the Sierra Norte passive margin faced westward to the Puncoviscana Ocean of Escayola et al. (2011), and clastic sediments were shed from the east into this depocenter (Puncoviscana tract of Escayola et al., 2011).

In conjunction with the other isotopic dates (see following), the previously presented Early Cambrian ages are time markers for the compressive D₁ deformation and accompanying regional metamorphism of the clastic metasedimentary succession in the Sierra Norte de Córdoba. Deformation led to the formation of large-scale anticlines and synclines accompanied by an S₁ cleavage.

The D₁ deformation in the Early Cambrian (Fig. 16), interpreted as the result of partitioned dextral transpression (von Gosen and Prozzi, 2010), probably was related to east- to northeast-directed subduction with the Sierra Norte de Córdoba representing the upper plate on the Gondwanan side (von Gosen and Prozzi, 2010) and the accretionary complex developing in the southwestern and western footwall (e.g., Northrup et al., 1998; Fig. 17B herein). Oblique subduction was also proposed by Rapela et al. (2007) and Iannizzotto et al. (2013). According to Rapela et al. (1998a), it seems that compressive deformation led to the destruction of the passive margin. The mechanism operating during this change from passive to active margin, however, is unclear. It is possible that subduction initiation along the margin was preceded and localized by a strike-slip fault.

Geochemical analyses in the central part of the Sierra Norte area have shown that the granitoid intrusions of the Sierra Norte–Ambargasta batholith are products of a precollisional active continental margin forming a calc-alkaline magmatic arc (e.g., Lira et al., 1997). This is supported by geochemical analyses in the southern part of the Sierra Norte (Iannizzotto et al., 2013). We propose that the magmatic arc formed in the former passive-margin segment now representing the upper plate (compare Figs. 17A and 17B).

The timing of the onset of subduction is not well constrained. U-Pb zircon dates (TIMS) of two intrusions in the Sierra Norte reported by Schwartz et al. (2008; Fig. 16 herein) suggest subduction was on going by ca. 555 Ma. In this scenario, clastic sedimentation in the developing accretionary complex was coeval with intrusive activity in the eastern upper plate. Compressive D₁ deformation in the newly formed arc then may have begun prior to the Early Cambrian and impacted pluton emplacement and continuing clastic sedimentation. However, evidence for these older relationships was not found during our field studies in the Sierra Norte or by the isotopic dates reported here.

The SIMS date of the granite porphyry (534 ± 5 Ma), intruded in the dextral mylonite zone south of San Pedro Norte (part of the “faja de deformación Sauce Punco” of Martino et al., 1999a, 1999b) and locally affected by dextral mylonitization, provides an indication for the timing of this shearing, interpreted as the D₂ event (von Gosen and Prozzi, 2010). It shows that this stage of compressive deformation took place in the Early Cambrian just after cessation of the D₁ event (Fig. 16). This is supported by the SIMS date of the Talayaco granite mylonite (533 ± 4 Ma). It seems possible that the San Miguel gneiss, with its 533 ± 12 Ma SIMS age (Siegesmund et al., 2010), belongs to this stage, although the date is affected by a larger error.

The 531 ± 4 Ma SIMS date of the intrusion of the Agua del Río dacitic porphyry, affected by local dextral shearing at its eastern margin during cooling of the magmatic body, is interpreted as the age of crystallization. It points to the approximate timing of the latest stage of ductile dextral shearing, which was followed by uplift and subsequent intrusion of the high-level El Tío granite, dated at 530 ± 4 Ma (U-Pb, SIMS; Fig. 16). Our results fall in the 537 ± 4 Ma to 530 ± 4 Ma time span for dextral mylonitization constrained by U-Pb zircon (SIMS) dates in the southernmost part of the Sierra Norte (Iannizzotto et al., 2013). It should be noted that the younger age reported by Iannizzotto et al. (2013) from the postkinematic Villa Albertina granite is identical with the age of the El Tío granite given herein, suggesting a stage of uplift after dextral mylonitization also in the southernmost part of the Sierra Norte (see also following discussion).

Rapela et al. (1998a, 1998b) mentioned U-Pb zircon ages from granitoids of the Sierra Chica and Sierra Norte de Córdoba of 530 ± 3 Ma and suggested that the intrusion was partially coeval with, or closely followed, sedimentation of the Puncoviscana Formation (Rapela et al., 1998a). During our studies, however, we did not find field evidence for sedimentation during compressive D₂ deformation and intrusive activity in the Sierra Norte area.

Overall, the dates presented herein show that the D₂ deformation in the Sierra Norte area was also related to the Pampean orogeny in the Early Cambrian. This is according to the interpretations of Martino et al. (1999a, 1999b), Miró et al. (1999), and Rapela et al. (2007) for dextral shearing and mylonite formation of granites. In particular, our time estimates for dextral shearing and mylonitization confirm the interpretation of von Gosen and Prozzi (2010), who showed that the affected granitoids were synkinematically emplaced. The dates confirm the isotopic results and similar interpretation of Iannizzotto et al. (2013). An emplacement of the foliated granitoids in the Sierra Norte within shear zones, as proposed by Miró et al. (1999), seems probable. It is possible that pluton emplacement was enabled in areas between right-stepping, en échelon and overlapping dextral strike-slip faults or shear zones, consistent with pluton emplacement in a transpressional setting (Iannizzotto et al., 2013).

The isotopic dates furthermore show that the Pampean D₂ event in the Sierra Norte area, recorded by dextral shearing and mylonitization, in time overlaps with the D₁ deformation. It seems that the D₂ event heterogeneously overprinted some areas of the Sierra Norte, while distinct parts remained unaffected. This part of the orogen, east of the accretionary prism in the Sierras de Córdoba and to the west (e.g., Northrup et al., 1998; Rapela et al., 1998a), was part of the upper (overriding) plate affected by a different style of compressive deformation in the Pampean history (von Gosen and Prozzi, 2010). Overall, both the initial D₁ deformation and the dextral mylonitization (D₂) seem to represent a continuous event that can be related to oblique (dextral) convergence between the overriding plate in the east and the subducting and finally colliding plate in the west (von Gosen and Prozzi, 2005a, 2010; Rapela et al., 2007; Iannizzotto et al., 2013). In their model, Iannizzotto et al. (2013) pointed to the possibility that deformation in the Pampean belt was diachronous and became progressively younger toward the north due to oblique dextral collision with a continental terrane. They suggested a large dextral displacement of the Pampean belt relative to the Río de la Plata craton during D₂ mylonitization. In our interpretation, ridge subduction (e.g., Schwartz et al., 2008) and collision might have been followed by collision of a continental terrane (Fig. 17C). This interpretation is similar to that proposed by Siegesmund et al. (2010).

Our time dates for the Pampean deformational and metamorphic events slightly contrast with the timing of the main events in the Sierras de Córdoba in the southwest, which took place at ca. 525 Ma and were followed by uplift and widespread anatexis at ca. 520 Ma (Rapela et al., 1998a). The deformational events in the Sierra Norte area, however, seem to coincide with the ca. 540 to ca. 520 Ma interval reported by Sims et al. (1998) for deformation and high-grade metamorphism during the Pampean orogeny. New SIMS U-Pb ages have shown, however, that the metamorphic history in the main ranges of the Sierras de Córdoba started at 553 Ma (Siegesmund et al., 2010).

Intrusive activity in the Sierra Norte is represented by granitoids emplaced during the D₂ deformation (synkinematic) and after cessation of the deformational history (postkinematic; von Gosen and Prozzi, 2010; Iannizzotto et al., 2013). The El Tío granite is a high-level postkinematic intrusion with chilled margins, abundant angular xenoliths of the deformed country rocks, and well-developed contact metamorphic aureole (von Gosen and Prozzi, 2010). These observations suggest that both Pampean deformational events were followed by an important phase of uplift. The 530 ± 4 Ma crystallization (SIMS) age of the undeformed El Tío granite indicates that D₁ and D₂ deformation ceased and uplift in the Sierra Norte area was ongoing in the Early Cambrian and continued through the Middle Cambrian (Fig. 16; see also following). The timing of this uplift event was regionally variable, since it occurred prior to intrusion and contact metamorphism at ca. 530 Ma in the Sierra Norte, whereas deformation and metamorphism in the Sierras de Córdoba continued up to ca. 520 Ma (Rapela et al., 1998a).

The SIMS zircon dates from the Rodeito rhyolite to dacite and El Escondido rhyolite (523 ± 5 Ma and 519 ± 4 Ma, respectively) also argue for uplift after cessation of Pampean deformation because neither rock shows ductile deformation. The younger ages in the Sierra Norte area seem to represent the final stage of the overall Pampean history. The slight deformation of the Ojo del Agua granite, with its 514 ± 4 Ma crystallization age (Stuart-Smith et al., 1999), may represent local late-stage Pampean deformation, deformation related to post-Pampean pluton emplacement, or a Famatinian (Ordovician) overprint. On the basis of the undeformed El Escondido rhyolite, however, it can be assumed that the Famatinian deformation did not affect the northern and northeastern areas of the Sierra Norte.

Continuing uplift in the Sierra Norte area was accompanied by the intrusion of the Los Burros dacite in the lowermost Middle Cambrian (512.6 ± 3.5 Ma, U-Pb, zircon, SIMS; Leal et al., 2003; Fig. 16). A roughly comparable age was found by Koukharsky et al. (2003) for a dacitic dike (514 ± 15 Ma, K-Ar, whole rock) cutting across the El Escondido granite, with its late Early Cambrian zircon SIMS date of 521 ± 4 Ma (see previous). The undeformed Ambargasta granite appears to belong to the final stage of the Pampean cycle as well, and on the basis of K-Ar biotite ages (Castellote, 1982), we assume a Middle Cambrian age. All the age data suggest that a combination of granitoid intrusions and rhyolitic to dacitic extrusions were emplaced after cessation of Pampean deformation in the Middle Cambrian.

Our results, combined with additional dates from Iannizzotto et al. (2013) and structural observations of von Gosen and Prozzi (2010), demonstrate that many parts of the Sierra Norte–Ambargasta batholith consist of post-D₂ intrusions with different ages. The same is observed for extrusive units ranging from the Rodeito rhyolite to dacite at 523 ± 5 Ma (U-Pb, zircon, SIMS) to the Los Burros dacite at 512.6 ± 3.5 Ma (U-Pb, zircon, SIMS; Leal et al., 2003). An older U-Pb zircon age (TIMS) of 535 ± 8 Ma described by Miró et al. (2005) and Schwartz et al. (2008) from a porphyritic dacite in the area of the Los Burros dacite shows that the wide area of the Los Burros dacite does not consist of only lowermost Middle Cambrian magmatic rocks but also includes older components.

Intrusive igneous rocks with younger isotopic dates reported from the Sierra Norte de Córdoba (e.g., Rapela et al., 1991; Correa, 2003; present Fig. 16) and Sierras de Córdoba (e.g., Rapela et al., 1998a; Gromet and Simpson, 1999) are related to the Famatinian cycle and are interpreted as an effect of the Famatinian (Ordovician) subduction beneath the continental magmatic arc at the western margin of the Eastern Sierras Pampeanas unit.

New isotopic dates from the Sierra Norte de Córdoba, in combination with structural data and previously published isotopic dates and interpretations, suggest four successive stages of the sedimentary, magmatic, and tectonic evolution in the easternmost segment of the Eastern Sierras Pampeanas at the western pre-Andean Gondwana margin as follows.

• (1) Deposition of the clastic sediments, which are interpreted as time equivalents of the Puncoviscana Formation (sensu lato), took place in the Ediacaran–Early Cambrian interval. Clastic sedimentation, accompanied by acid magmatism, is interpreted to be related to a passive-margin setting, probably on marginal west-Gondwanan crust (Fig. 17A).

• (2) Sedimentation was directly followed by eastward-directed subduction beneath the west Gondwana margin and the generation of arc intrusions. In the developing upper plate (former passive margin), partitioned dextral transpressive D₁ deformation and regional metamorphism are assigned to a very short time interval during the Early Cambrian (Fig. 17B). An onset of subduction in the latest Neoproterozoic, however, cannot be excluded.

• (3) Local D₂ deformation and metamorphism in the hanging-wall plate were accompanied by synkinematic intrusive activity. This stage of dextral shearing and mylonitization between ca. 534 and 531 Ma can be related to intraplate compression due to ridge subduction and/or collision, which might have been followed by accretion of a continental terrane (Fig. 17C).

• (4) Both short periods of Pampean (Early Cambrian) deformational events at the western pre-Andean Gondwana margin in the Sierra Norte de Córdoba were followed by uplift prior to 530 ± 4 Ma. Subsequent emplacement of different postkinematic intrusions and extrusive units suggests an end of the Pampean magmatic activity in the Middle Cambrian (Fig. 17C).

The undeformed igneous rocks in the Sierra Norte show that this segment of the easternmost Sierras Pampeanas was not significantly affected by Famatinian (Ordovician) deformation and metamorphism, which dominated along the western margin of the Eastern Sierras Pampeanas complex.

Discussions with V.A. Ramos (Buenos Aires) helped to clarify the history of the Córdoba area. We are also grateful to L. DiNardo, J. Dristas, D. Gregori (Bahía Blanca), and A.A. Bonalumi (Córdoba) for their support during field work. Many thanks go to J. Alvarez (Córdoba), G. Escorza (Puerto Madryn), and J.I. Falco (Bahía Blanca) for all their help during field work and many discussions. We thank the staff from the Central Laboratory for Geochronology at the University of Münster for support and use of the facilities, especially K. Mezger (now University of Bern) and T. Grund. We thank J. Wooden for help and discussion of trace-element and U-Pb data from the U.S. Geological Survey–Stanford University SHRIMP-RG (sensitive high-resolution ion microprobe–reverse geometry) laboratory. Many thanks go to K. Ludwig (Berkeley Geochronology Center) for permission to use his unpublished software ISOPLOT version 3.50 (21 June 2006). We are grateful to R.J. Pankhurst, S.O. Verdecchia, and an anonymous reviewer for constructive comments, suggestions, and corrections that helped improve the manuscript. This study was enabled by grants from the German Research Foundation (DFG) to W. von Gosen (projects Go 405/5–1 and 5–2), which are gratefully acknowledged.

Samples were cleaned and crushed to a <10 mm grain size. Further pulverizing to a grain size of <1 mm was done by a disk mill. A heavy mineral separate from the sieved fraction of <0.25 mm was concentrated using a Wilfley-type wet shaking table. Further concentration of the zircons was achieved by heavy liquids and a Frantz isodynamic separator. The high-quality zircons for single-grain TIMS and SIMS analyses were selected from the bulk sample. Most of the zircons analyzed by TIMS were not air abraded; only the zircons marked in Table DR1 (sample CE-54D; see ^{footnote 1}) were abraded to ∼90% of the original size.

The handpicked crystals were washed with 3.5 N HNO₃ for 30 min at 80 °C and then rinsed with distilled water and dried with acetone. Individual crystals and fractions were decomposed with 2 μl of 24 N HF in steel-cased multi- or single-hole Teflon containers for 3–4 days at 180 °C. The decomposed zircon was spiked with an appropriate amount (2 μl) of a ²⁰⁵Pb-²³⁵U tracer solution. After drying, the sample was converted to the chloride form with 2 μl of 6 N HCl overnight. The dried sample was loaded with a mixture of silica gel, HCl, and phosphoric acid on a Re single filament. Measurement of U and Pb was performed on a VG 354 mass spectrometer equipped with a Daly multiplier and an ion counting device by peak hopping, with Pb at 1250–1350 °C and U at 1400–1450 °C. Mass fractionation was controlled by the measurement of the standards NBS 982 (0.13%) and U 500 (0.06%). Samples with high concentrations of ²⁰⁵Tl were quickly heated to 1400 °C (within 10 min) and then kept at this temperature for a maximum of 2–3 min until the ²⁰³Tl signal fell to 1%–2% of the estimated spike intensity (given by the 205 signal). The worst measured ²⁰⁶Pb/²⁰⁴Pb ratio was 111 (for a sample dissolved in a new Teflon container); usually, the ratios were scattered around 500–1500, and the maximum measured ²⁰⁶Pb/²⁰⁴Pb ratio during the analyses was >5000, which is indicative of a constant and low procedural blank. Calculations were corrected with a maximum Pb blank of 0.01 pg with the following composition: 6/4 = 17.72, 7/4 = 15.52, 8/4 = 37.7, and a U blank of 0.5 pg. The initial Pb composition was calculated according to Stacey and Kramers’ (1975) Pb evolution model.

Some of the performed analyses had to be rejected from concordia calculation due to their position above the concordia line. This effect appeared preferentially on zircons that had a high content of Tl. We presume that mainly an erroneous calculation of the Pb concentration led to the variable degree of discordance. The individually valid mass fractionation of Pb and U, according to the variable premeasurement conditions (in the course of getting rid of the Tl), could not be correctly verified by standard measurements at comparable conditions (flashing time and temperature).

The errors in the Table DR1 (see ^{footnote 1}) are quoted at the 1σ level, whereas the errors of the calculated discordias are given for the 95% confidence level. For data processing, programs were used as follows: PBDAT version 1.24, 30 July 1993, and ISOPLOT version 2.49 (2001) and 3.50, 21 June 2006, with kind permission of K. Ludwig (Berkeley Geochronology Center).

Zircons for SIMS analysis were handpicked under alcohol, mounted in a 2.5 cm epoxy round, and polished to expose the grain centers. The grains were imaged with transmitted light, reflected light, and CL prior to analysis. U-Pb isotopes and trace-element compositions were collected simultaneously on the sensitive high-resolution ion microprobe–reverse geometry (SHRIMP-RG) instrument at the U.S. Geological Survey–Stanford University Ion Probe Laboratory, Stanford, California, using a 25–30-μm-diameter spot size. The U-Pb analytical routine followed Barth and Wooden (2006), and data reduction utilized the SQUID program of Ludwig (2005). U-Pb isotopic composition was calibrated by replicate analyses of zircon standard R33 (419 Ma; Black et al., 2004), and U concentration was calibrated using zircon standard Madagascar Green (MAD, 4196 ppm U; Barth and Wooden, 2010). Calibration errors for ²⁰⁶Pb/²³⁸U ratios of R33 for the two analytical sessions were 0.62% and 0.39% (2σ). Ages were calculated as concordia ages or weighted mean ²⁰⁶Pb/²³⁸U ages using the Isoplot/Ex program of Ludwig (2003). Common Pb compositions were estimated from Stacey and Kramers (1975).

Trace-element data collection was simultaneous with U-Pb analysis, following Mazdab and Wooden (2006) and Mattinson et al. (2009). The REE routine measured ¹³⁹La, ¹⁴⁰Ce, ¹⁴⁶Nd, ¹⁴⁷Sm, ¹⁵³Eu, ¹⁵⁷Gd¹⁶O, ¹⁶³Dy¹⁶O, ¹⁶⁶Er¹⁶O, ¹⁷²Yb¹⁶O, and ¹⁸⁰Hf¹⁶O (Table 2). The routine for samples CE-66D and CE-54D included analysis of ¹⁷⁵Lu¹⁶O. Concentration calibrations used trace-element values of zircon standards CZ3 and MAD (Mazdab and Wooden, 2006). Estimated errors based on repeated analysis of CZ3 are 5%–10% for Y, Hf, Th, U, and the REEs except for La (30%) and include both analytical reproducibility and real abundance variation in the gem-quality zircon (cf. Mattinson et al., 2009). Chondrite-normalized REE plots (Fig. 8) use the chondrite REE abundances of Anders and Grevesse (1989) multiplied by a factor of 1.36 (Korotev, 1996). Chondrite-normalized values for Pr were calculated by interpolation (Pr_(N) = La_(N)^0.33 × Nd_(N)^0.67). Discussion of Eu anomalies is based on Eu_(N)/Eu* and Ce_(N)/Ce*, with Eu* and Ce* calculated as geometric means (e.g., Eu* = [Sm_(N) × Gd_(N)]^0.5).