Widespread and Short-Lived 1870 Ma Mafic Magmatism Along the Northern Superior Craton Margin
An enormous volume of mafic-ultramafic magmatism affected the margins of the Superior craton during the relatively brief interval ca.1890-1870 Ma, including the Molson dykes and the Fox River Sill of the northwestern Superior craton, the mineralized Setting Lake sill of the Thompson nickel belt, Ni-bearing gabbro sills of the Cape Smith Belt (Ungava region), and mafic sills in the Labrador Trough. Recent published work has also shown that carbonatite magmatism and extensive mafic dyke swarms also bisected the central and western interior of the craton during this time, implying that this expansive and broadly coeval magmatic event comprises a major, Paleoproterozoic large igneous province ("pan-Superior" LIP). New, high-precision U-Pb baddeleyite geochronology shows that a widespread subset of this magmatism, focused in the north-central circum-Superior region, was intruded at 1870 Ma, within a span of 1 m.y. or less. On the Belcher and Sleeper Islands in SE Hudson Bay, plagioclase-phyric gabbro sills (Haig intrusions) intrude upper levels of Eskimo Formation basalt flows, and overlying sediments to the base of the Flaherty Formation basalt flows, and yield ages of 1870.3 +/- 0.7 Ma, and 1870.1 +/- 1.1 Ma, respectively. On the mainland (James Bay lowlands), a gabbro sheet intruded Paleoproterozoic sediments of the Sutton inlier at 1869.9 +/- 1.7 Ma, possibly fed by a swarm of NW-trending Fort Albany diabase dykes, one of which is dated here at 1870.7 +/- 1.1 Ma. The 1884-1877 Ma Molson dykes, the 1876-1871 Ma (Ar-Ar) Pickle Crow and the 1870 Ma Fort Albany dyke collectively define a radiating fan of ca. 1885-1870 Ma magmatism with approximately 100° of arc, in which the precise geochronology resolves a pattern of emplacement that youngs as it swings to the east. The 1883 Ma Fox River and 1880 Ma Setting Lake sills lie near the focus of the radiating swarm immediately north of western Superior craton promontory. Together all these units are consistent with a model in which a mantle plume was located beneath the focal point of the radiating swarm. A primary paleomagnetic pole has been reported from the Molson dykes, a pre-fold pole from the Haig intrusives and Flaherty volcanics, and a virtual geomagnetic pole of uncertain age from the Sutton inlier gabbro. All these poles fall in the eastern Pacific Ocean, with the Molson pole displaced about 35° northwest of those from the other three units. It is unclear whether this difference is the result of rapid apparent polar wander, secondary overprinting, tectonic rotation, or incomplete averaging of secular variation.
Volcanism Associated With the Proterozoic Wiborg Rapakivi Granite Batholith
The ∼1650-1620 Ma locus classicus rapakivi granites of southeastern Finland and vicinity (the >18,000 km2 Wiborg batholith and its satellite intrusions) are relatively high-level, epizonal plutons that were emplaced into the Paleoproterozoic (∼1.9 Ga) Svecofennian metamorphic crust in an extensional tectonic setting. They are associated with relatively thin crust (with a highly reflective lower crustal section), swarms of basaltic and silicic dikes, as well as rare volcanic rocks. The current erosional level of the Wiborg batholith corresponds to a paleodepth of 1-5 kbar. Two spectacular supracrustal lithologic units associated with the classic Wiborg batholith have recently been recovered - one in a downfaulted crustal block southwest of the batholith in the Gulf of Finland area (the Island of Suursaari) and one on the northeastern flank of the batholith (in the Taalikkala area) in a megaxenolith that sank into the Wiborg magma chamber during its emplacement. These bodies are both >10 km long along strike and reveal a similar stratigraphy with a quartz arenitic conglomerate (thermally metamorphosed quartzite in the Taalikkala megaxenolith) atop a Paleoproterozoic nonconformity, followed upsection by basalts and silicic lavas and pyroclastic rocks. The (preserved) sequence on the Island of Suursaari is ∼200 m thick and almost in its original position (gently tilted to the northeast). That in the Taalikkala megaxenolith is ∼1.5 km thick and almost vertical, as the xenolith obviously rotated about its long axis (parallel to the strike) as it sank into the rapakivi granite magma chamber. High-precision ID-TIMS U-Pb zircon data on the pyroclastic rhyolitic units from Suursaari and Taalikkala imply upper intercept crystallization ages of 1633 +- 2 Ma and 1638 +- 3 Ma, respectively. The whole-rock ID-TIMS initial epsilon-Nd values of the lavas and pyroclastic rocks from Suursaari are slightly negative (∼ -0.5 for the basalts, ∼ -2 for the silicic rocks). The zircon ages comply with those of the early and main intrusive phases (coarse, rapakivi-textured fayalite-hastingsite and annite-hastingsite granites, leucogabbroids) of the Wiborg batholith and show that concomitant, quite extensive bimodal volcanism was associated with the emplacement of the Wiborg. The Wiborg batholith probably represents a major, relatively long-term caldera event that occurred in a mid-Proterozoic intra-plate extensional setting in equatorial Baltica.
Paleomagnetism of 1780-1770 Ma Mafic and Composite Intrusions of Småland,Sweden
The Late Palaeoproterozoic wide-spread orogeny resulted in assembly of many cratons - e.g., Laurentia, Baltica, Siberia, North China. Some researchers suggest that this world-wide orogeny resulted in the assembly of Mesorpoterozoic supercontinent - Columbia (also known as Nuna and Hudsonland.) Its configuration, building blocks, times of assemblage and break-up are widely debated. The variety of opinions is largely caused by a lack of reliable paleomagnetic poles. There are so few of well-dated and supported by rigorous field tests Paleo-Mezoproterozoic poles that a generation of Apparent Polar Wander Paths (APWPs) is impossible for the time being. However, new robust and well-dated paleomagnetic data can be used for testing of existing reconstructions of Columbia. Mafic and felsic plutons and dykes are widespread in the Småland-Värmland belt. In the eastern parts of the Småland province a special type of composite dykes has been found. These dykes are mainly composed of a marginal mafic and a central felsic part. To a varying degree one can also observe magma mixing features between the two parts. In the literature they are described under the general name of "Småland dyke porphyries" and they were found when the first geological mapping in the 19th century started in the area. Radiometric dating (U-Pb, zircons) of the felsic component in the composite dykes has been reported by Nilsson and Wikman (1997.) Samples from Påskallavik and Alsterbro intrusions have been analyzed, resulting in 1780 ± 3 Ma for the former and 1776 +8/-7 Ma for the latter. Mafic parts in the composite dykes have a similar chemistry as some mafic dykes without a felsic component, but with the same NW-trending direction in the northwestern part and mainly ENE-trending direction for the rest of the investigated area. This dyke swarm called the "Oskarshamn dykes" is therefore suggested to be of the same age as the composite dykes. We sampled four composite dykes, one mafic dyke and a large gabbro intrusion of presumably the same age. Paleomagnetic study revealed one stable bipolar magnetisation. One composite intrusion in Alsterbro is cut by thin mafic dykes of uncertain age. The inverse baked contact test is positive: the remanence directions on exacontacts of mafic dykes is close to the dyke's direction, but significantly different from the remanence direction of unbaked porphyries. This means that the ChRM of the composite dyke is older than the ChRM of cross-cutting dykes. The paleomagnetic pole calculated from the ChRM directions of composite intrusions and Oskarhamn dykes being combined with the palaeomagnetic data from 1780 Ma Sholsha Formation in Karelia (Russia) provide a 1780 Ma 'key-pole' for Fennoscandia. This pole together with coeval poles from Laurentia, Kalahari, North China, and Australia were used for testing the Columbia reconstruction of Zhao et al. (2004.) Paleomagnetic data do not support this reconstruction. However, they support the recent suggestion of Laurentia-Baltica connection between ~1780 Ma and ~1200 Ma (Salminen and Pesonen, 2007). These two continents probably formed a core of the Mesoproterozoic supercontinent. Recent paleomagnetic data (Wingate et al., in press) suggest that Siberia could also be part of this core at least since ~1470 Ma.
1.15 Ga Metamorphism at the Northern end of a Grenville-Aged Tectono-Thermal Belt in Western Laurentia
Thick, deformed and weakly metamorphosed successions of sedimentary rocks deposited during several extensional, basin-forming events between ca. 1.64 Ga and ca. 540 Ma, record Proterozoic tectonism and hydrothermal activity in east-central Yukon. A SHRIMP study of small (20-80 μm), complexly zoned zircon grains from 5 xenoliths in an Early Cambrian lamprophyre indicates a cryptic ca. 1.15 Ga metamorphic event, which is not recorded in the exposed stratigraphy. The studied xenoliths show variable grades of metamorphism and metasomatic alteration, but in general retain prograde mineral assemblages indicative of upper amphibolite to granulite-facies metamorphism that are well in excess of the highest grade of metamorphism (greenschist) in surface exposures. The analyzed zircons record three metamorphic events at ca. 1.60, 1.27 and 1.15 Ga. The first two are rationalized by metasomatism during Wernecke Breccia development and magmatism of the MacKenzie igneous event. Metamorphic zircon rims or whole crystals from sillimanite-bearing paragneissic xenolitihs yield ages of 1149±21 Ma and 1149±19, respectively. Similarly, zircon from an extensively retrograded/metasomatically-altered xenolith record recrystallization and rim growth at 1168±22 Ma. Importantly, textures and U-Pb systematics of zircon from another xenolith are consistent with igneous crystallization, followed shortly by sub-solidus recrystallization at ca. 1150 Ma. The ca. 1150 Ma event is coeval with the Grenville orogeny of the eastern and southeastern North America, the Sibao orogen of South China and a number of scattered igneous and metamorphic ages from the northern and central North American Cordillera. The "Grenvillian" event in western Laurentia is intriguing, as direct evidence of deformation or igneous activity is lacking. However, indirect means such as dating of granitic xenoliths and zircon xenocrysts from mafic intrusions, dating of metamorphic zircon, titanite and garnet, dating of hydrothermal deposits and dating of detrital zircon and clasts provide a picture of a "Grenvillian" belt spanning most of the length of the North American Cordillera. This belt may provide an important constraint for paleocontinental reconstructions involving the Mesoproterozoic supercontinent Rodinia.
Closing Clymene and Opening Iapetus: Paleomagnetic Evidence from the Paraguay Belt
The final assembly of Western Gondwana was accomplished by the closure of an ocean between the Amazon and Sao Francisco cratons in central Brazil - the Clymene ocean - through the Pampean, Paraguay and Araguaia fold belts. The closing of Clymene would be roughly synchronous to the opening of Iapetus, taking apart Amazonia and Laurentia. Geochronological data along the Pampean-Paraguay-Araguaia belt suggest these events have started at ca. 550 Ma (age of oldest syncollisional granites in Pampean and Araguaia belts) and ended by 500 Ma (age of post-collisional granites in Paraguay belt). Paleomagnetic evidence from Ediacaran successions sampled along a transect from the Amazon craton to the folded Paraguay belt indicate that the Amazon craton was not part of Gondwana by ca. 635 Ma, joining the proto-Gondwana only in the Cambrian. A secondary component of magnetization that post-dates the folds of the Paraguay belt matches Gondwanan reference poles for ca. 520 Ma, but the declination of this secondary magnetization varies along strike, suggesting that the 100 degree curvature of the belt was generated subsequent to an initial phase of folding and thrusting in the early Cambrian. We propose that a vertical axis rotation accounts for the E-W trends of the Paraguai belt, which served as a transform zone during the late Cambrian collision between the West Gondwanan elements Amazonia-West Africa and the proto-Gondwana, formed by Congo-São Francisco, Rio de Plata, and Kalahari.
An overview of Ediacaran magmatism in eastern Laurentia and adjacent regions
Almost forty years ago Ron Doig used K-Ar ages of alkali rocks in North America, Greenland and Europe to show that they were part of a single large alkali province dated around 575 Ma (age recalculated with modern constants). Since then the duration and extent of the Iapetan Large Igneous Province (I-LIP) has expanded so that a new overview is necessary. The earliest magmatism that can be associated with this event was the emplacement at 630 Ma of two groups of kimberlites in central Quebec. This was followed by dyke swarms in North America and Scandinavia at 615 to 590 Ma, which are usually considered to mark start of the event and the arrival of a plume head. Magmatism continued from this time until about 540 Ma. Basalts are preserved in many parts of the Appalachian Mountains with ages from 570 to 550 Ma. Basalts may have also been present north of the St Lawrence, overlying the 565 Ma giant Sept Iles Mafic Suite. I-LIP plutons started earlier and continued later than the basalts. They have a wide range in composition - gabbro, syenite, granite, alkali syenite, carbonatite, kimberlite and ultramafic lamprophyre. There does not seem to be any clear movement of the centre of the magmatism or change in compositional character with time. The duration of I-LIP appears to be much longer than other plume events, and indeed many people have divided it into two or three separate events, but the evidence for sub-division is not strong. If I-LIP was produced by a single long-lived plume then the mantle must have been almost stationary with respect to the crust. Paleomagnetic measurements on the I- LIP rocks have been difficult to understand. However, one interpretation is for rapid plate movements, which seems incompatible with a stationary mantle. Another possibility is one or many episodes of true polar wander. In this case an almost stationary mantle may have allowed the I-LIP plume to produce a long-lived mass asymmetry that destabilized the rotation axis of the earth.
Enigmatic Paleomagnetic Results From the Western end of the Grenville Dyke Swarm
Samples collected from 23 paleomagnetic sites from the western half of the ca. 590 Ma Grenville dyke swarm , after detailed alternating field demagnetization, show strong evidence of a pervasive overprint (steep positive inclination to the SE) which is similar to that found in the nearby 577 Ma Callander Bay alkaline complex (Symons and Chiasson, 1990). The unusually high LOI in chilled margins (up to 8 weight percent), especially in those sites which show the overprint to exclusion of other components, suggest that the remagnetization is caused by oxidizing fluids from the alkaline complex. At two sites (GD15, 16), about a kilometer apart in the same dyke, a component (A) with steep upward direction, also to the SE, is recorded. This component is isolated in two other dykes, and in several more the remanence vector moves upon demagnetization towards A from the overprint direction. These observations suggest that A is common throughout the swarm. A is antipodal to the characteristic component obtained for a Grenville dyke by Hyodo and Dunlop (1993), where a positive baked contact test was obtained, so that A is interpreted to be primary. At one site (GD2) a component B with a shallow down direction to the SE was found to the exclusion of other components, and 20 km to the east, a second site (GD7), probably in the same dyke, gave a similar remanence direction. U-Pb dating on baddeleyite from sites GD15 and GD2 (589.4 ± 4.7 and 604 ± 36 Ma respectively) both gave ages consistent with the dykes being part of the Grenville swarm. A similar paleomagnetic direction to B has been obtained from so-called McGregor Bay dykes (Palmer et al. 1977) one of which has given a U-Pb age on baddeleyite of 1233 ± 1.4 Ma (Dudas et al. 1994) and is therefore a member of the larger Sudbury swarm. These dykes are less than 20 km to the west of GD2 and their similarity in paleomagnetic direction and large difference in age compared to GD2 would suggest a regional overprint , but two dykes of the Sudbury swarm intervene and give typical Sudbury primary remanence directions with westerly declination and shallow inclination. A puzzle therefore remains: magnetizations A and B appear to be primary, they are approximately 90 degrees apart in direction and yet have a similar age. They could be the result of unresolved remanence components, secular variation or a consequence of TPW. The answer must await more precise age data and further paleomagnetic results from other dykes of the same age in the Grenville swarm.