Avalonia

The evolution of Avalonia has seven main elements (e.g. Nance et al., in press); (1) the development of juvenile crust at ca. 1.2 to 1.0 Ga, (2) an early arc phase (pre 650 Ma) , (3) accretion to Gondwana at ca. 650 Ma, (4) a main arc phase (640-570 Ma), (5) its transition to a platform (570-540 Ma), (6) the rifting of Avalonia from Gondwana (ca. 515 Ma), and (7) its accretion to Laurentia (ca. 440 Ma).

Although Avalonia developed along a continental margin subduction zone, the basement upon which the main Avalonian arc was developed is nowhere unequivocally exposed. Fragments of this basement may occur in northwestern Cape Breton Island (Keppie and Dostal, 1991) and in the Goochland terrane of the southern Appalachians (Hibbard and Samson, 1995), but the provenance of these regions is controversal (e.g., Farrar, 1984; Barr et al., 1998).

As a result, the nature of Avalonian basement has been characterized indirectly from the neodymium isotopic composition of crustally derived felsic igneous rocks ranging in age from 740 to 370 Ma. These igneous rocks have elemental Sm/Nd ratios typical of intracrustal melts (Sm/Nd ~0.19; Allegre and Ben Othman, 1980) and similar initial eNd values that range between Ð2.5 and +5.0 (Thorogood, 1990; Barr and Hegner, 1992; Whalen et al., 1994; Kerr et al., 1995; Murphy et al., 1996a; Keppie et al., 1997; Murphy et al., 2000; Samson et al., 2000). Extrapolated to the depleted mantle curve, eNd growth lines consistently yield overlapping model ages (TDM) of 0.8-1.1 Ga in Atlantic Canada and 1.0-1.3 Ga in southern Britain (Thorogood, 1990; Murphy et al., 2000).

These depleted mantle model ages closely coincide with the timing of Grenville (c. 1.25-0.95 Ga) orogenesis considered to be responsible for the amalgamation of Rodinia (e.g., Hoffman, 1991). However, igneous rocks in Grenville-aged orogens are characterized by higher model ages, implying significant recycling of older crust. In contrast, the quite strongly positive distribution of Avalonian eNd values suggests a largely juvenile basement dominated by c. 1.0 Ga mantle-derived material with only minor elements of older Proterozoic crust.

The interpretation of depleted mantle model ages is controversial (Arndt and Goldstein, 1987). Detailed arguments for the interpretation of the Sm-Nd data are presented elsewhere (e.g. Nance and Murphy, 1994, 1996; Murphy et al., 2000). For magma produced by recycling of a single crustal source, the model age represents the time at which the crustal basement was itself extracted from the mantle. More commonly, however, magmas contain mixtures of juvenile, mantle-derived material and older crustal components. In these situations, the model age has no geologic significance. That the model ages for Avalonia represent mantle extraction ages is suggested by the very similar model ages in Avalonian magmatism ranging from 740 Ma to 370 Ma (Murphy et al., 1996b; Murphy et al., 2000). The generation of such similar model ages over a time interval of this length is unlikely to be the outcome of mixing, but instead, indicates the predominant influence of a single basement source.

Hence, the neodymium isotopic data imply that successive generations of Avalonian felsic magma were produced largely as a result of recycling ca. 1.0-1.2 crust. Lead isotopic data (Ayuso et al., 1996) similarly suggest that the igneous rocks represent mixtures of juvenile, c. 1 Ga basement and typical Avalonian crust. The depleted mantle model ages are therefore thought to record a genuine tectonothermal event during which the bulk of Avalonian basement was itself extracted from the mantle. The formation of Avalonian basement is therefore considered to have been broadly coeval with Grenvillian orogenesis. However, the primitive isotopic signature of Avalonia relative to that of the Grenville Belt suggests that the basement formed, not as part of a collisional orogeny, but in one or more largely juvenile oceanic island arcs (Murphy et al., 2000). These data can be reconciled if the basement developed within the Panthalassa-type ocean that would have surrounded Rodinia following its amalgamation. Other remnants of primitive island arcs developed within this peri-Rodinian ocean may be preserved in the ca. 900-700 Ma island arc rocks of the Arabian-Nubian Shield (e.g., Blasband et al., 2000), and in the ca. 950-900 Ma calc-alkalic granitoid orthogneisses and metarhyolites of the Tocantins province in central Brazil which yield a similar envelope of eNd growth lines and almost identical (ca. 0.9-1.2 Ga) depleted mantle model ages to those of Avalonia (Pimental and Fuck, 1992). On our reconstructions therefore, this juvenile Avalonian crust, called proto-Avalonia, is positioned within the Panthalassa-type peri-Rodinian ocean.

Fragmentary evidence for initial subduction in Avalonia dates from at least 730 Ma to 650 Ma and is termed the early arc phase. In Atlantic Canada, examples of this activity include the ca. 734 Ma calc-alkalic Economy River Gneiss in mainland Nova Scotia (Doig et al., 1993), the ca. 681 Ma arc-related Stirling Belt (Bevier et al., 1993) in Cape Breton Island, and the calc alkalic ca. 683 Ma Tickle Point Formation and ca. 673 Ma Furby's Cove Intrusive Suite (Swinden and Hunt, 1991; OÕBrien et al., 1996) in southern Newfoundland. The rift ophiolite volcanics of the Burin Group in Newfoundland (Strong et al., 1978) may extend this early Avalonian magmatic activity to ca. 763 Ma (Krogh et al, 1988).

In Britain, evidence for early arc-related activity is represented by the c. 700 Ma calc-alkalic Stanner-Hanter Complex of central Wales (Patchett et al., 1980) and the c. 677 Ma calc-alkalic Malverns Plutonic Complex of the British Midlands (Tucker and Pharoah, 1991). 40Ar/39Ar mineral ages of ca. 650 Ma in the Malverns Complex are interpreted to date cooling following upper greenschist to amphibolite facies metamorphism (Strachan et al., 1996). Early arc activity may also be represented in the undated gneisses of the Rosslare Complex in southeastern Ireland and the Coedana Complex in North Wales (Gibbons and Hor‡k, 1996).

A short period of high grade metamorphism is recorded at ca. 650 Ma in various parts of Avalonia, including coastal Maine (Stewart and Tucker, 1998) and the Malvern Plutonic Complex (Strachan et al., 1996). Amphibolite facies metamorphism of pre-630 Ma age may also be present in central Cape Breton Island (Keppie et al., 1998) and southern Newfoundland (O'Brien et al., 1996), and some form of accretion must likewise be recorded in the emplacement of ophiolitic rocks of the ca. 760 Ma Burin Group (Keppie et al., 1991). This metamorphism is interpreted to reflect the accretion of Avalonia to the Gondwanan continental margin prior to the beginning of the main phase of Avalonian magmatism at ca. 635 Ma and coincides with a temporary cessation (ca. 650-635 Ma) in subduction-related magmatism. In our reconstructions, therefore, we show the outboard arc terranes of Avalonia colliding with the northern Gondwanan margin at 650 Ma.

The main phase of Avalonian magmatism is recorded in voluminous late Neoproterozoic magmatic arc-related volcanic and cogenetic plutonic rocks with crystallization ages of 635 to 570 Ma (e.g., Nance et al., 1991). Coeval sedimentary successions that are dominated by volcanogenic turbidites are locally associated with these arc-related magmatic rocks, and have been attributed to deposition in a variety of intra-arc, interarc and back arc basins (e.g., Pe-Piper and Murphy, 1989; Pe-Piper and Piper, 1989; Pauley, 1990; Smith and Socci, 1990; OÕBrien et al., 1996; Murphy et al., 1999). This magmatic activity and the generation of arc-related basins is interpreted to reflect oblique subduction beneath the northern Gondwanan margin.

The timing of the onset of this main phase activity was broadly similar throughout much of Avalonia. However, its cessation was diachronous, terminating at ca. 590 Ma in New England (Kaye and Zartman, 1980; Hermes and Zartman, 1985, 1992; Thompson et al., 1996; Thompson and Bowring, 2000), 600 Ma in southern New Brunswick (Bevier and Barr, 1990; Barr et al., 1994; Currie and McNicoll, 1999), 605 Ma in mainland Nova Scotia (Doig et al., 1991; Murphy et al., 1997; Keppie et al., 1998), 575 Ma in southern Cape Breton (Barr et al., 1990; Bevier et al., 1993), 585 Ma in Newfoundland (Krogh et al., 1988; OÕBrien et al., 1996) and 600 Ma in the British Isles (e.g., Tucker and Pharaoh, 1991; Hor‡k, 1993; Noble et al., 1993).

Cessation of main-phase subduction is accompanied by a transition to intracontinental extension, marked by the onset of bimodal magmatism. As with the onset of main phase arc arctivity, the transition occurred at different times along the belt; at c. 595 Ma in New England (Mancusco et al., 1996), at c. 560 Ma in southern New Brunswick (Bevier and Barr, 1990; Barr et al., 1994; Currie and McNicoll, 1999), at c. 605 Ma in mainland Nova Scotia (Murphy et al., 1997), between 575 Ma and 560 Ma in southern Cape Breton Island (Bevier et al., 1993), at c. 570 Ma in Newfoundland (O'Brien et al., 1996), and in the interval 570-560 Ma in Britain (Tucker and Pharoah, 1991). Although this stage is locally accompanied by deformation and metamorphism, no evidence exists for the regional orogenesis, crustal shortening, and crustal thickening and uplift characteristic of continental collision zones. Instead, deformation is usually localized and resulted in the inversion of the earlier volcanic arc basin successions.

To account for such a tectonic transition in the apparent absence of a major collisional event, Murphy and Nance (1989) proposed that Avalonian subduction was terminated as a result of transform activity. In their model, the main phase of Avalonian magmatism at c. 635-570 Ma occurred as the result of oblique subduction, leading to the development of an extensional magmatic arc and a variety of volcanic arc basins. Subsequently, the interaction of a continental margin transform system with the subduction zone resulted in the termination of subduction, the structural inversion of a number of volcanic arc basins, and the formation of new rift and wrench-related basins in the interval c. 590-540 Ma. Murphy et al. (1999) and Keppie et al. (2000) later postulated ridge-trench collision as a mechanism for the transition in order to account for the diachronous cessation of arc volcanism and the apparent reversal of kinematics on major basin-bounding faults.

Avalonia likely remaining linked to Gondwana until the Early Ordovician. During the Ordovician, however, faunal provinciality and paleomagnetic data indicate increasing separation from Gondwana concurrent with a decrease in the separation between Avalonia and Laurentia (Cowie, 1974; Boucot, 1975; Pickering et al., 1988; Cocks and Fortey, 1990; Trench and Torsvik, 1992; Dalziel et al., 1994; Cocks, 2000). The Arenig Stiperstone Quartzite in Britain and the correlative Armorican Quartzite of Cadomia and Iberia (e.g., Noblet and Lefort, 1990), is thought to reflect the subsidence associated with this separation. Minor bimodal rift volcanism in Avalonia is predominantly of Cambrian age and may reflect rifting prior to separation (e.g., Murphy et al., 1985; Greenough and Papezik, 1986).