Journal of Petrology Advance Access originally published online on April 25, 2007
Journal of Petrology 2007 48(5):1021-1039; doi:10.1093/petrology/egm009
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Crustal Contamination of Picritic Magmas During Transport Through Dikes: the Expo Intrusive Suite, Cape Smith Fold Belt, New Quebec
Department of Geology, University of Toronto, 22 Russell Street, Toronto, on, M5s 3b1, Canada
RECEIVED SEPTEMBER 29, 2005; ACCEPTED FEBRUARY 21, 2007
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
The Proterozoic Expo Intrusive Suite comprises a series of mafic to ultramafic intrusions crosscutting the Povungnituk Group of the Cape Smith Fold Belt in New Quebec. The intrusions are mainly in the form of blade-shaped dikes that penetrate a sediment-rich horizon in the middle of the Beauparlant Formation and terminate downward against massive basalts of the lower Beauparlant Formation. Significant accumulations of magmatic sulfide occur at the basal terminations of the dikes. At stratigraphic levels above the Beauparlant Formation the intrusions appear as broad dikes or sills within the Nuvilik Formation, below the mineralized lava flows and subvolcanic intrusions of the Raglan Formation. The Expo Intrusive Suite and the mineralized bodies of the Raglan Formation are probably coeval and comagmatic with the overlying Chukotat Group. Post-emplacement folding has exposed the Expo Intrusive Suite over about 5 km of structural relief, revealing the basal sulfide concentrations where dike segments terminate on the flanks of anticlines. The parent magma as preserved in chilled margins and narrow dikes was a picrite containing
17 wt % MgO (i.e. komatiitic basalt) and slightly depleted in Th, U and Nb relative to middle and heavy rare earth elements. The compositions of ultramafic cumulate rocks within the intrusions are strongly enriched in Th, U and Nb relative to heavy rare earth elements, reflecting assimilation of the enclosing basalts and metasediments. Modeling of the assimilation process suggests that the picritic magma was capable of assimilating masses of basalt or sediment up to 50% of the original mass of magma. Assimilation of 10% of a mixture of basalt and sediment caused the magma to become sulfide-saturated, and was accompanied by the crystallization of masses of ultramafic cumulates approximately equal to the mass of rock assimilated. The presence of dikes whose chilled margins resemble uncontaminated primary magmas but that contain abundant cumulates recording wholesale assimilation of host-rocks indicates that the process of assimilation and fractional crystallization required to produce continental tholeiites from picritic parent magmas may not require the presence of long-lived magma chambers, but can occur during transport along dikes and reaction with wall-rocks. KEY WORDS: komatiite; Expo Intrusive Suite; assimilation; fractional crystallization; sulfide mineralization
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
The processes by which primitive high-MgO basalts undergo assimilation of crustal rocks and concomitant fractional crystallization (i.e. AFC) are generally inferred from studies of the compositions of lavas (e.g. Wooden et al., 1993
; Lightfoot & Hawkesworth, 1997), and sometimes also from studies of shallow subvolcanic intrusive rocks (Arndt et al, 2003). The AFC process is commonly assumed to occur in a mid-crustal staging chamber, which is then tapped to feed eruption of lava flows (e.g. O’Hara & Mathews, 1981), although in many cases the inferred AFC chamber is not actually observed. In this study the structure and petrology of a series of upper crustal intrusions collectively termed the Expo Intrusive Suite (after the Expo–Ungava Ni–Cu–Pt–Pd–Au deposit at their centre) are interpreted to represent a complex system of upper crustal conduits in which AFC processes took place. These observations provide important constraints on the processes that modify the compositions of mafic magmas as they might migrate through the crust, including those that lead to the generation of economically significant concentrations of magmatic sulfide. |
REGIONAL GEOLOGY |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
The Expo Intrusive Suite was emplaced into Paleoproterozoic supracrustal rocks of the Cape Smith Fold Belt of the Trans-Hudson Orogen (Fig. 1) in northern Quebec at 1882·7 ± 1·3 Ma (Randall, 2005
). The following interpretation of the regional geological history is based on the results of recent mapping by the author and previously published reviews of available geochronological and structural data (Picard et al., 1990; Lucas & St-Onge, 1992; St-Onge et al., 1992; Picard, 1995; Randall, 2005).
|
The Cape Smith Fold Belt can be subdivided into allochthonous northern and parautochthonous southern domains. These two domains had contrasting metamorphic and tectonic histories, until they were amalgamated during the climax of the Trans-Hudson Orogeny at c. 1838 Ma. The southern domain, comprising the Povungnituk and Chukotat Groups, appears to represent a long-lived passive continental margin that laps onto the Archean Superior Province which existed for over 150 Myr. The northern domain, comprising the Spartan, Parent, and Watts Groups, has been interpreted as a series of fragments of an island arc and its underlying oceanic basement (Scott et al., 1989
, 1992) that were accreted during terminal collision of the Superior Province with the Archean Rae Province to the north. The Narsajuaq arc to the north is interpreted to represent the plutonic roots of the arc and post-collisional felsic intrusions. The geology of the studied portion of the Southern domain is shown in Fig. 2. This map was compiled at a scale of 1:20 000 using derivative products from a high-resolution airborne geophysical survey (Aerotem total magnetic field, conductance; July 2003), a 1:5000 scale map of the Expo deposit area by the author, and numerous traverses conducted by the author in 2002, 2003 and 2004 while mapping onto 1:20 000 scale air photographs or Ikonos satellite images, using global positioning satellite (GPS) data for navigation. Features of interest such as dikes or sills of the Expo Intrusive Suite were followed along their entire lengths, and all areas of outcrop in these areas were walked and prospected by the author.
|
Major formational conductors were drawn from the Aerotem survey results and verified in many instances in the field as graphitic argillites commonly associated with carbonate-, silicate- or sulfide-facies iron formations. Using the formational conductors as key marker horizons, the regional stratigraphy was divided into several units that can be correlated with the recognized stratigraphic column for the region (Fig. 3).
|
At the base of the succession, the Dumas Formation (also referred to separately by some as the Lamarche Group) of the Povungnituk Group represents a clastic wedge with a maximum thickness of 3400 m. The Dumas Formation is an autochthonous to parautochthonous assemblage of ferruginous sandstone, conglomerate, quartzite and iron formation, overlain by progressively finer-grained wackes and rare carbonate metasediments, which unconformably overlies the Superior Province basement. Portions of the unconformable contact with Archean basement rocks are faulted, presumably as a result of thrust imbrication during the closure of the ocean basin during the Trans-Hudson Orogeny.
The Dumas Formation is overlain by the predominantly basaltic Beauparlant Formation and the sedimentary Nuvilik Formation. The Beauparlant Formation comprises 2–3 km of tholeiitic basalts with ocean island basalt (OIB) affinity (Modeland et al., 2003) intercalated with volcaniclastic and locally graphitic siltsone horizons and minor carbonate metasediments. A mafic sill in the upper Dumas Formation, dated at 2038 + 4/–2 Ma (Machado et al., 1993), is considered to be comagmatic with the volcanic rocks of the Beauparlant Fm. A diorite intrusion crosscutting pillow basalt near the top of the Beauparlant Fm has been dated at 1991 ± 2 Ma (Machado et al. 1993). The volcanic activity that produced the Beauparlant Formation is, therefore, constrained to have occurred between 2038 and 1991 Ma. The presence of a clastic sedimentary sequence >3 km thick beneath the lavas of the Povungituk Group indicates that subsidence occurred before volcanic activity commenced, allowing this to be classified as a non-volcanic rifted margin (McKenzie & Bickle, 1988; White & McKenzie, 1989) rather than a volcanic rifted margin as has previously been suggested (Hynes & Francis, 1982; Francis et al., 1983). The Beauparlant Fm is overlain by a succession of deep-water sediments of the Nuvilik Fm, including greywackes, cherts, and graphitic argillites and rare pebbly quartz arenites. Elsewhere in the Cape Smith belt, the Nuvilik Formation overlies alkaline volcanic rocks of the Cecilia Formation, which have been dated at 1958·6 + 3·1/–2·7 Ma (Parrish, 1989).
The Beauparlant Fm is interpreted here to be a single sequence 3 km in true original thickness, and to comprise three distinct members that are preserved in their original stratigraphic sequence everywhere in the map area. This stratigraphic column can also be applied to the Povungnituk Fm for distances of more than 30 km east and west of the map area. The lower Beauparlant Fm consists of sheets and pillowed flows of basalt. In much of the map area this unit is nearly flat-lying and consequently has little topographic expression. The middle Beauparlant Fm is distinguished by the presence of regionally extensive formational conductors associated with distinctive magnetic highs at its lower and upper contacts with the lower and upper Beauparlant Fm. Outcrop is scarce within the area of the middle Beauparlant Fm, and it was mapped as metasediment in the past. Where it is encountered in outcrops along glacial outwash channels and in diamond drill cores it is found to be largely composed of basaltic sheet flows and pillowed basalts with abundant but volumetrically subsidiary intercalated metasediments. The metasediments include carbonates, carbonate-cemented sandstones, siltstones, graphitic argillites, silicate iron formation, chert, and turbiditic beds of quartz arenite. There is no indication anywhere that that the middle Beauparlant Fm is in fault contact with the underlying lower Beauparlant Fm. The magnetic and conductive metasedimentary horizon used to separate the lower and middle Beauparlant Fm is continuous throughout the map area, and where it is observed in outcrop it shows no sign of the intense deformation expected along a crustal-scale shear zone.
The upper Beauparlant Fm consists of sheet and pillowed basalt flows with only very minor quantities of graphitic interflow sediment. Much of what has been mapped in the past as gabbro sills within the Beauparlant Fm is actually a series of thick basaltic lava flows with vesicular chilled flow tops and highly distinctive columnar joints produced during cooling at the sea floor. As a result of the scarcity of interflow sediments, the upper Beauparlant Fm forms highly resistant ridges throughout the map area.
The Nuvilik Fm is poorly exposed, occurring in a basinal structure in the north of map area. The predominant lithology in the Nuvilik Fm is laminated to thin-bedded siltstone. Other units identified are minor carbonates, conglomeratic quartzite and two horizons of graphitic argillite that form strong conductors easily traced in the airborne electromagnetic survey data. Where it could be observed, the contact between the uppermost pillow basalts of the Beauparlant Fm and the base of the Nuvilik Fm was invariably seen to be conformable. There is no evidence for a thrust fault at or near the base of what is here classified as the Nuvilik Fm, and no reason to believe that it is tectonically transported metasediment of the Dumas Formation as has been proposed by Lucas & St-Onge (1992).
The Nuvilik Formation is overlain north of the map area by the predominantly volcanic deep-water assemblage of the Chukotat Group. The Chukotat Group is about 5 km thick, and comprises several cyclic volcanic sequences with compositions ranging from komatiitic basalt with MgO contents between 19 and 11%, through pyroxene-phyric tholeiitic basalts, to plagioclase-phyric basalts that show some compositional resemblance to normal mid-ocean ridge basalt (N-MORB). The volcanic rocks of the Chukotat Group are intercalated with fine-grained siliciclastic sediments including graphitic argillites. The uppermost Nuvilik Formation and the lower Chukotat Group contain ultramafic bodies of the Raglan Formation, which host the Raglan Ni–Cu–PGE (platinum group element) deposits and are generally interpreted to represent feeders and incised lava channels coeval with the Chukotat Group (Giovenazzo et al., 1989; Lévesque & Lesher, 2003). The contact between the upper Nuvilik Formation and the base of the Raglan formation is locally faulted but generally conformable, and has been interpreted as being essentially in place (C. M. Lesher, personal communication, 2005). Intrusive rocks near the top of the Nuvilik Formation, considered to be comagmatic with the Chukotat Group, have been dated at 1918 ± 9 Ma by Parrish (1989). However, the Chukotat suite is intruded near its top by a gabbroic sill dated at 1870 Ma (St-Onge et al., 1992) and a gabbro intrusion belonging to the Raglan Formation has been dated at 1887 + 39/–11 Ma (Wodicka et al., 2002). It has been argued that the Raglan intrusions served as feeders for the komatiitic basalts of the Chukotat Group, and that the ultramafic bodies of the Raglan Formation might be largely extrusive (e.g. Barnes et al., 1982; Bédard et al., 1984; Barnes & Barnes, 1990).
The strata of the Povungnituk Group were intruded at 1882·7 ± 1·3 Ma (Randall, 2005) by a series of mafic to ultramafic bodies of the Expo Intrusive Suite, which are described in detail below. The Raglan Formation, Chukotat Group, and Expo Intrusive Suite are all of similar age and might be considered a single magmatic suite.
Continental collision between the Rae and Superior Provinces occurred at c. 1838 Ma (Lucas & St Onge, 1992; St Onge & Lucas, 1992; Randall, 2005), causing the development first of 1–2 km wavelength tight to isoclinal upright folds with WSW–ENE-trending axes. The early folds were refolded into 10 km wavelength open folds along NW–SE-trending axes. The earlier folding event was largely confined to the area north of the outcrop area of the Expo Intrusive Suite, whereas the final folding event affected both the Expo Suite and its host supracrustal rocks, leading to gentle warping of the dike-like bodies of the Expo Suite. Local folding of the dikes around vertical axes is well developed in the area immediately to the east of the Expo–Ungava deposit, in the centre of the synclinorium that surrounds it (Fig. 2).
The map in Fig. 2 is somewhat different from what has been reported previously (e.g. St Onge & Lucas 1993), possibly because of the greatly increased resolution at which mapping was done in the present study, aided by the availability of high-quality and high-resolution airborne geophysical data. The principal difference between the present stratigraphic interpretation and those previously published is that there appears to be no evidence whatsoever in the field for the existence of the thrust faults proposed by St Onge and coworkers to cause structural repetitions of the Beauparlant Fm basaltic pile. Indeed, it would be exceedingly difficult to justify the need for a thrust fault anywhere in the map area on purely structural grounds. A key consideration in any attempt to place thrust faults with the area of Fig. 2 is that horizons along which the faults have been proposed, corresponding to the tops of the Lower and Upper members of the Beauparlant Formation as recognized here, are crosscut in several places by dikes of the Expo Intrusive Suite that are demonstrably not offset in any way. It is impossible to reconcile this observation with the proposition that faulting along these interfaces occurred during continental collision several tens of million years after the emplacement of the Expo Suite at 1882·7 Ma. There is no evidence for the existence of any significant thrust faults anywhere in the current map area, and as there may not be any such faults at the base of the Chukotat Group, the entire sequence from the base of the Beauparlant Formation to the top of the Chukotat Group may actually be preserved in place.
There are several places in the map area where metabasalts and metasediments of the Beauparlant Formation are intensely deformed along structural corridors up to 100 m wide, and in early stages of mapping these were interpreted as major thrust faults. When mapping was complete, these areas were found to correspond to the hinge zones of isoclinal folds. A good example occurs in the fold just north of the Hilltop–TK–Mesamax NW segment of the Expo Intrusive Suite. The intense deformation associated with this synclinal structure does not extend past the closure of the fold in which it occurs, and it is therefore not considered to be a fault of regional significance. A shear zone such as this, if it was encountered during a north–south mapping traverse, would be incorrectly identified by any geologist as a major fault; it can only be correctly interpreted in the context of detailed mapping supported by high-quality geophysical data.
|
EXPO INTRUSIVE SUITE |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Localities
The Expo Intrusive Suite has been the target of exploration for Ni–Cu–Co–PGE mineralization since 1957, and has seen intense activity since 2001. The locations of several mineral occurrences or deposits are shown in Fig. 2. The Expo–Ungava deposit contains net-textured to massive magmatic sulfide hosted by peridotite in a large trough-like body that cuts metasediments of the Nuvilik Formation. The Mesamax NW, TK, Tootoo, Mequillon, and Vaillant deposits comprise net-textured to massive sulfides hosted by pyroxenite and peridotite at the bases of shallowly plunging trough-shaped bodies at the terminations of dikes. Other localities mentioned in the text and illustrated in Fig. 2 are the Hilltop dike, a relatively narrow apophysis of the main Expo dike system between TK and Expo–Ungava, and the Snow Owl pipe, which is an ovoid outcrop of dunitic serpentinite SE of the main Expo–Ungava deposit area.
Lithologies
In the following discussion, the metamorphosed igneous rocks of the Expo Intrusive Suite are referred to using names that would be appropriate terms for their protoliths. The addition of the prefix ‘meta’ would make the discussion awkward, but its use is implied throughout whenever reference is made to the intrusive rocks in their original states. The application of protolith names to metamorphosed ultramafic rocks becomes especially problematic for the term pyroxenite, as discussed below; the use of the field term pyroxenite for amphibolites, however, is so deeply entrenched in the local terminology that it is difficult to avoid.
Gabbronorite
Much of the Expo Suite is composed of dikes of medium-grained olivine melagabbronorite spanning widths ranging from several meters up to several hundred meters. Locally this rock is relatively unaltered, with white saussuritized feldspar laths surrounded by partially uralitized orthopyroxene and clinopyroxene. Where the lower greenschist-facies metamorphic effects are more pervasive, the gabbronorite consists of tremolite after clinopyroxene, talc, anthophyllite and serpentine after orthopyroxene and olivine, and chlorite, clinozoisite or a fine-grained nearly opaque mixture of albite, hydrogrossular, zoisite and sericite replacing plagioclase. Minor chromite appears as submillimeter-sized euhedra with magnetite overgrowths, commonly within chlorite–serpentine pseudomorphs of olivine. Minor amounts of apatite, magnetite and mica are observed in pockets of late-crystallized melt. There is widespread development of late, poikilitic primary magmatic hornblende, which gives the rock numerous brown blotches that are easy to see in drill core and lend it the field name hornblende-gabbro. At chilled margins the gabbronorite fines down to an aphanitic brownish-black rock resembling basalt but showing a darker colour than the basalts of the Beauparlant Fm. In a few rare examples where the gabbronorite forms dikes tens of meters wide, these show columnar jointing perpendicular to their contacts. In the interiors of very wide dikes (hundreds of meters wide) the gabbronorite can be very coarse-grained, with pyroxene oikocrysts up to 2 cm across (commonly pseudomorphed by tremolite) enclosing serpentinized olivine chadocrysts.
Within the thickest part of the intrusive complex near the Expo–Ungava deposit, the gabbronorite is locally differentiated to form a leucogabbro that commonly shows variable texture from mafic pegmatite to fine-grained gabbro over distances of several centimeters. The pegmatitic patches may contain blebby sulfide and have been found in thin section to contain rare zircon euhedra [dated at 1882·7 ± 1·3 Ma by Randall (2005)].
Pyroxenite
In the study area, any rock predominantly composed of tremolite–actinolite has traditionally been called pyroxenite in the field. The term ‘pyroxenite’ is used to describe fine-grained rocks resembling quenched liquids and also to describe adcumulate rocks originally composed almost entirely of olivine and pyroxene with minor interstitial melt or hornblende and plagioclase; that is, olivine melagabbronorites, olivine websterites, and websterites all tend to be grouped together. On freshly broken surfaces these rocks are generally light green and show many reflective cleavage surfaces from the abundant tremolite crystals. Primary microtextures are effectively erased by the uralitization of the primary pyroxenes and the growth of extensive fibrous epitaxial tremolite that replaces intercumulus material. Primary plagioclase contents are easy to underestimate because of pervasive replacement of pyroxene and plagioclase by tremolite and chlorite.
Peridotite
Peridotite is completely serpentinized, to form a massive, highly resistant rock that weathers deep maroon to brown. The metamorphosed peridotite is composed of fine intergrowths of serpentine, talc, anthophyllite and chlorite forming pseudomorphs after olivine and orthopyroxene, dotted with grey tremolite pseudomorphs after clinopyroxene. Poikilitic primary hornblende, very fine-grained chlorite, abundant chromite, and traces of sulfides are commonly present in the intercumulus spaces of what generally appears to be an orthocumulate to heteradcumulate texture. Oikocrysts of fresh clinopyroxene are occasionally present. The peridotites are generally strongly magnetic because of the presence of abundant microscopic magnetite crystals formed during serpentinization of iron-bearing olivine. In some places the peridotites have been extensively altered to a talc- and carbonate-rich assemblage, that weathers to a rusty brick-brown gravelly surface and has very low magnetic susceptibility. As a result, there are locally very complex patterns of magnetic susceptibility in areas that are entirely underlain by peridotite, making field checking essential for the correct delineation of ultramafic bodies using geophysical responses.
Dunite
The interiors of the ultramafic bodies of the Expo Suite commonly consist of a serpentine–chlorite–magnetite rock that is classified as dunite. Dunite weathers to a yellowish-green or brown patina and lacks the hackly surface produced by resistant tremolite crystals in the peridotites. On fresh surfaces it is fine-grained and deep green to black, with a flinty, almost conchoidal fracture. Relict primary orthocumulate texture is commonly visible in thin section, as a result of the presence of minor quantities of primary intercumulus hornblende and intercumulus chlorite possibly replacing plagioclase. Dunite has a high magnetic susceptibility, and forms prominent highs in the total magnetic field.
Distribution
Map scale
The entire map area is traversed by numerous mafic and ultramafic intrusions of the Expo Suite (Fig. 2), most of which are sharply discordant, more or less tabular bodies and are therefore classified as dikes. Locally these bodies are highly irregular in form or appear to be more concordant in nature (e.g. at the Expo–Ungava deposit), and in at least one locality (Snow Owl occurrence) they are ovoid in plan view, entirely surrounded by metasedimentary host-rocks, and are therefore interpreted to represent pipe-like conduits. All known base- and precious-metal sulfide mineralization in the region is associated with these intrusions, usually along their margins or internal autointrusive contacts. The majority of Expo Suite intrusions have trends approximately ENE–WSW, apparently cutting across the folded Povungnituk Group. However, the Expo dike in the area of the Expo–Ungava deposit itself appears to have been folded about vertical axes during NE–SW-directed compression.
The schematic cross-section in Fig. 2b shows the likely subsurface configuration of the Expo Suite, based on interpretation of the 5 km structural relief afforded by the presence of the late NW–SE-trending folds. The reasoning behind this interpretation is given at more length in the discussion below.
Contact relations
The Expo Suite forms a complex of cross-cutting gabbronoritic dikes, ultramafic dikes, dunite pipes, and sill-like peridotite massifs within the Nuvilik Fm in the area surrounding the Expo–Ungava deposit (Fig. 2). The magma was, therefore, able to intrude the metasedimentary rocks easily in a variety of orientations. A large sill-like peridotite body extends southeastward from the Cominga area along the base of the Nuvilik Fm, whereas a pair of sub-parallel dikes of gabbronorite and ultramafic rock hundreds of meters wide trends east–west across the entire doubly plunging synclinorium occupied by the Nuvilik Fm. The Snow Owl dunite pipe appears to be the surface expression of a steeply inclined or vertical cylindrical mass of dunite. A very deep vertical extent is implied by a broad, high-amplitude magnetic anomaly associated with this relatively narrow structure.
The trough-like form of the mineralized peridotite body at Expo–Ungava is illustrated in Fig. 4. Here the Expo Suite forms at least two and possibly more stacked trough-like intrusions that are roughly concordant with the enclosing sediments. Immediately outside the Expo–Ungava deposit there is no known base to the intrusions, which may truly be dikes with kilometer-scale vertical extents; the mineralized sills hosting the deposit therefore appear to be narrow horizontal offshoots from a generally dike-like ultramafic intrusion. On the other hand, the mineralized troughs could be bounded by steeply dipping reverse faults along which the centre of the synclinorium has been elevated as a result of tectonic extrusion during NE–SW-directed compression. In this case the Expo–Ungava deposit might be expected to continue at greater depth along the bases of the ultramafic bodies to its east and west.
|
Where the Expo Suite intrudes the basalts of the upper Beauparlant Fm, the contacts between the gabbronorite dikes and the enclosing basalts are invariably sharp and planar, and where both sides of the intrusion can be observed they are parallel. In these examples there is a chilled margin on the gabbronorite, and the basalt shows textural evidence for metamorphism to the pyroxene hornfels facies over short distances (< 1 m).
Where the Expo Suite intrudes the basalt and metasediments of the middle Beauparlant Fm, the wall-rocks are generally strongly metamorphosed. The host-rocks commonly show signs of incipient melting, such as the presence of pockets of pegmatite. In these examples the intrusive rock is a coarse-grained pyroxenite that has evidently not been chilled against the enclosing wall-rocks, implying that the magma was passing by the contact at a sufficient rate to heat it to temperatures well above the solidus temperature of wall-rock. Contacts with carbonate sediments are occupied by calc-silicate hornfels zones. Where the intrusions encountered beds of quartzite in the predominantly semipelitic siltstone strata they were less able to assimilate it because of its refractory, monomineralic nature; in these situations, intrusive breccias were formed or individual beds of quartzite were left intact in their original attitudes but surrounded by ultramafic rock.
Where the Expo Suite intrudes basalt of the lower Beauparlant Fm at the Tootoo, Mesamax, TK, and Mequillon deposits it forms dikes whose trough-like basal terminations are lined with gabbronorite that locally shows a gradational contact with the enclosing basalt. The exact point of contact is commonly difficult or impossible to determine, because the wall-rock basalt has been heated and metamorphosed to a coarse-grained pyroxene hornfels or sanidinite-facies assemblage closely resembling the medium-grained gabbro of the intrusive body. This hybrid gabbroic marginal facies of the dikes grades into pyroxenite followed inward toward the core of the dike by peridotite and dunite. In some instances dunite clasts tens of centimeters across are suspended within the peridotite matrix. The general form of these structures is illustrated in Fig. 5, which shows a perspective cutaway view of the mineralized ultramafic dike associated with the Mequillon deposit; this is representative of the Mesamax, TK, Tootoo, Mequillon and Vaillant deposits.
|
It is noteworthy that the lithology and contact relations of the Expo Suite are very strongly controlled by the nature of the enclosing rocks (Fig. 2). As individual dike segments are traced across the region, their lithologies vary systematically with the stratigraphic level at which they are observed. The intrusions are exclusively gabbroic within the upper Beauparlant Fm, but are predominantly peridotite sheathed by pyroxenite within the middle Beauparlant Fm. The dikes are zoned from pyroxenite to peridotite or dunite, occasionally with a discontinuous sheath of gabbro in the lowermost Beauparlant Fm, and never pass below the base of the lower member of the Beauparlant Fm. In the Nuvilik Fm the intrusions are predominantly composed of peridotite and dunite with subsidiary gabbronorite. Some explanations of these distinctive contact relations are offered in a subsequent section discussing the controls on assimilation of host-rocks by the Expo magma.
|
LITHOGEOCHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Data collection
A total of 118 rock samples from both outcrops and diamond drill cores from the map area were analyzed by X-ray fluorescence spectroscopy (XRF) and solution inductively coupled mass spectrometry (solution ICP-MS) for a suite of major and trace elements at the Geoscience Laboratory of the Ontario Geological Survey. Samples came from the narrow dike in the Hilltop area, and from diamond drill holes in the Mesamax NW, Tootoo, Expo and TK areas (Fig. 2). Representative data are shown in Table 1 (the complete dataset is available as an Electronic Appendix at http://petrology.oxfordjournals.org). In subsequent figures these data are compared with the compositions of basalts and ultramafic rocks of the Raglan Formation (Burnham et al., 1999
). Major element concentrations were determined by XRF on fused beads; selected minor element concentrations were determined by XRF on pressed powder pellets, and the remainder of the trace element concentrations were determined by solution ICP-MS after complete sample digestion in closed Teflon beakers. Data for major elements are reported after recalculation to anhydrous equivalent compositions. Detection limits for each element in the whole-rock analyses are shown in the Electronic Appendix at the top of each column.
|
Classification of Expo Intrusive Suite
As a starting point, it is interesting to consider the compositions of the melagabbronorite and chilled margins or fine-grained dike rocks assigned to the Expo Intrusive Suite in Table 1. The compositions of all of these rocks are similar, and in view of the very fine-grained nature of some of them they can probably all be considered to represent the compositions of liquids that have not accumulated any olivine. Taking the Hilltop dike, for example, the average of about 17 wt% MgO, 11% Al2O3 and only 0·7 wt% TiO2 allows them to be classified as picrites at the boundary between basaltic komatiite and komatiite, of the Al-undepleted variety. Considering the ubiquitous association of the melagabbronorite with ultramafic members of the Expo Suite, it thus appears likely that the entire suite is derived from a komatiitic parental liquid.
Comparison of the Expo Suite with other magma types
In Fig. 6 the trace element concentrations in a variety of rocks interpreted to represent the parental magmatic liquids of the Expo and other suites are shown normalized to concentrations in the primitive mantle (McDonough & Sun, 1995). The samples of the Beauparlant Fm in this study are identical to those of Francis et al. (1983), which were sampled in the Kenty Lake area west of the map in Fig. 2.
|
The similarity between the olivine-phyric basalts of the Chukotat Group (Burnham et al., 1999
) and the proposed primitive magma of the Expo Suite is striking, as also is the difference between the Expo rocks and the enclosing basalts of the Beauparlant Fm (Mesamax basalt). The Expo Suite primitive magma (chilled margin) and the olivine-phyric basalts of the Chukotat Group are compositionally almost indistinguishable. The main features of the Expo Suite trace element signature are a flat heavy rare earth element (HREE) pattern and slight depletion in the light REE (LREE), a depletion that becomes increasingly evident in the concentrations of the large ion lithophile elements (LILE) U, Th and Nb. On the other hand, the basalts of the Beauparlant Fm, both from the present study and from published data, are characterized by strongly LILE-enriched patterns transitional between enriched MORB (E-MORB) and OIB and conspicuously lacking the negative Nb anomalies that are characteristic of continental tholeiites (see Modeland et al., 2003). |
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Evolution of the Expo Intrusive Suite
The principal controls on the evolution of the Expo Suite are likely to have been the removal or accumulation of crystals, and the assimilation of host-rocks. Figure 7 shows cation proportions of Al and Mg, showing the compositions of members of the Expo Suite compared with those of basalts of the Chukotat and Povungnituk Groups, ultramafic rocks from the Raglan Formation (Katinniq area) and sediments from the vicinity of the Mesamax NW deposit. The cation plot is used to simplify the projection of stoichiometric mineral compositions. Also shown are the compositions of olivine calculated to be in equilibrium with the average Hilltop dike magma (Roedder & Emslie, 1970
), and the compositions of pyroxenes from the gabbronorite near the Expo–Ungava deposit. Although norm calculations (not shown) indicate the presence of several per cent of plagioclase in most of the ultramafic rocks, much of this normative plagioclase may be represented by the minor hornblende observed in the mode. The average Hilltop dike plots in the center of the small scatter of chilled margin compositions, slightly more Mg-rich than the olivine-phyric Chukotat basalt (Burnham et al., 1999). The ultramafic rocks from the Expo suite fall along a trend between the chilled margins and the olivine composition, indicating a major role for olivine accumulation in the genesis of these rocks. Fine-grained gabbroic rocks from the Expo Suite trend towards basaltic compositions similar to the pyroxene-phyric basalts of the Chukotat Group. The trend shown by hybrid (i.e. strongly contaminated) melts is off toward the metasediments, below the main trend.
|
Assimilation of host-rocks
Assimilation of host-rocks by the parental magma is probably a necessary precondition for the formation of base metal sulfide deposits, because a primitive mantle-derived magma will be far from saturation with a sulfide liquid until it has undergone a substantial amount of cooling, contamination, and crystallization (e.g. Wendlandt, 1982
; Mavrogenes & O’Neill, 1999). Apart from the textural evidence discussed above for assimilation of sediments, there is also strong evidence for this process in the compositions of the ultramafic rocks of the Expo Intrusive Suite. To assess possible contaminants several examples of host lithologies were sampled from Mesamax NW (Fig. 8). The host-rocks comprise basalt, mafic tuff, iron formation and clastic sediments (siltstones) of the middle Beauparlant Fm. All of these rocks share steep LILE-enriched trace element patterns that would lead to distinctive changes in trace element ratios if they were assimilated by the Expo Suite magma with its nearly flat to slightly LILE-depleted character.
|
Numerical models of magmatic evolution
Assimilation–fractional crystallization
The evolution of the Expo Intrusive Suite during assimilation of host-rocks and concomitant crystallization has been modeled using the thermodynamic modelling software PELE (Boudreau, 1999
). The PELE program uses the algorithms and thermodynamic parameters incorporated in MELTS (Ghiorso & Sack, 1995) in a Windows-based utility to calculate the compositions of minerals and melts given a bulk composition, temperature and pressure. PELE was used to calculate a liquid line of descent for the parental magma of the Expo Intrusive Suite, and this matches the predictions of MELTS very well. On the other hand, whereas MELTS does not conveniently treat isenthalpic assimilation into a magma, the same problem has been incorporated very well into PELE. The consequences of the primitive Expo magma assimilating two possible contaminants (Table 2) were evaluated; one is the average Mesamax basalt, the other is a typical graphitic schist from the Mesamax area (i.e. middle Beauparlant metasediment). PELE was run starting with 100 g of the Expo magma at its liquidus temperature, adding 5 g increments of either assimilant as solid minerals at 300°C, and recalculating the temperature and phase equilbria in the bulk system at each step. The model explicitly accounts for the heats of fusion of minerals during their dissolution or precipitation reactions with the melt.
|
The metasedimentary assimilant was modelled as an assemblage of 64.38% plagioclase (An29), 31.95% cummingtonite (grunerite50), 1.28% biotite, 1.39% titanite, 0.74% magnetite, and 0.25% pyrrhotite. It was assumed to have been fully consolidated at the time of intrusion of the Expo Suite, which occurred at least 100 Myr after the deposition of the Beauparlant Fm (see preceding discussion of geological history). The basalt was modeled as an assemblage of 41.5% plagioclase, 36.99% cummingtonite, 9.26% clinopyroxene, 5.33% biotite, 0.34% apatite, 4.76% magnetite, and 1.83% titanite.
As can be seen from the results in Table 2, the magma is capable of assimilating at least 50% of its own mass of either basalt or semipelite. Numerous other runs using a variety of assimilants produced similar results. As the process of incorporation of cold host-rock continues, the magma cools considerably, but it actually increases in volume; that is, the mass of ultramafic cumulate produced to compensate for the heat of fusion of the metasediments or basalt is slightly less than the mass of assimilated material. This surprising result is a consequence of the fact that the heat of fusion of the metasediment is compensated for by the heat of crystallization of the cumulates, and the depression in temperature of the mixture is compensated for by the radical decrease in liquidus temperature accompanying the removal of magnesian olivine and the addition of fluxing components such as silica, alkalis and water. The key reason why komatiitic magmas have such a large capacity to assimilate their host-rocks is that they are not multiply saturated, and the olivine liquidus surface descends steeply to lower temperatures without producing large masses of olivine crystals.
The modeling results indicate that the interaction of large volumes of Expo magma with the Povungnituk Fm could lead to the removal of the host sediments by assimilation, simultaneously with the deposition of a similar mass of ultramafic cumulates in the space left by the melted sediments. One can envision this at the kilometer scale as a replacement process, but the model says nothing about the physical mechanism by which this might be accomplished. Despite the uncertainty regarding the details of the process, the presence of the contaminated ultramafic cumulates of the Expo Intrusive Suite indicates that this has indeed taken place in some way. These arguments indicate that we should not expect to see evidence of widespread deformation around voluminous ultramafic intrusions, as a result of wall-rock assimilation during their emplacement, and they obviate the common ‘room problem’ faced in an interpretation of their origins.
The modal proportions of cumulus phases resulting from the addition of a semipelitic assimilant at various proportions of assimilant to initial liquid are shown in Fig. 9. Over a wide range of amounts of assimilation, the cumulate rock would be a dunitic orthocumulate. Choices of other assimilants produced similar results, but with different cotectic phases at very large amounts of assimilation (e.g. wehrlite, troctolite, harzburgite, etc.).
|
The PELE output was used to predict the compositions of contaminated liquids and coexisting cumulus phases. A model cumulate rock composition was calculated by combining the solids with a small amount of liquid to represent a trapped liquid fraction of 20%. The results of the model after assimilation of 10 g of semipelite per 100 g of initial liquid are shown in Fig. 10, compared with the compositions of peridotite at the Mesamax NW deposit, the putative liquid starting composition based on the ultramafic dike at Hilltop, and the composition of a contaminated pyroxene-phyric basalt from the Chukotat Group (O. M. Burnham, personal communication, 2006). The correspondence between the model compositions and the actual compositions of ultramafic cumulates and apparently comagmatic basaltic liquids lends very strong support to the hypothesis that the ultramafic rocks of the Expo Suite formed as a result of assimilation of their host-rocks by a primitive komatiitic basalt magma. The end result was the eruption of a strongly cooled and contaminated mafic liquid resembling continental flood basalt, and the formation of ultramafic rock with fairly strongly contaminated trace element signatures, as evidenced by the ubiquity of these characteristics in the present sample suite.
|
Figure 11 shows the ratio La/SmN vs Th/NbN for samples including the cumulate rocks and liquids of the Expo Intrusive Suite, lavas and sediments of the Beauparlant Formation, compared with the range of compositions of ultramafic rocks from the Raglan Formation (Burnham et al., 1999
). The model compositions resulting from assimilation of basalt and semipelite are shown with tick marks indicating the amount of assimilant. The element ratios of liquids and their cumulates are indistinguishable on this diagram because of the very small partition coefficients of all four elements plotted. The ratio La/SmN is a measure of the steepness of the slope of the primitive mantle normalized LREE pattern, which is generally high in continental crust but equal to about one in primitive tholeiites and komatiites (compare with Figs 6, 8 and 10). The Beauparlant Fm basalts have significantly higher La/SmN than the Expo Suite and Chukotat basalts, in keeping with their classification as having affinities to ocean island basalts. The ratio Th/NbN is influenced by the assimilation of Th-rich sedimentary rocks of the upper continental crust. Whereas Th/NbN is similar for the Beauparlant Fm, Chukotat Group and Expo magma, the ultramafic rocks of the Expo Intrusive Suite show a broad scatter toward the relatively high values shown by the metasediments hosting the Mesamax NW deposit. The compositions of the ultramafic cumulates of the Expo Intrusive Suite do not coincide exactly with either trend, but could reflect combined assimilation of a mixture of up to 10% of semipelite and 10% basalt, dominated in most cases by sediment.
|
The primitive Expo magma is assumed to have been sulfide-undersaturated when it left its mantle source region. This assumption is based upon the large degree of partial melting required to generate a komatiitic basalt and upon the observation that the magma was enriched in the highly chalcophile elements such as Pt (Table 1). Given these constraints, there cannot have been more than about 900 ppm S in the initial melt (e.g. Mavrogenes & O’Neill, 1999
). The assimilation model shows that digestion of basalt would not trigger sulfide saturation in the contaminated magma unless at least 1000 ppm of S was present initially. On the other hand, assimilation of the metasediment, itself containing 0.6% FeS, caused the model magma to reach sulfide saturation after it had assimilated about 10% of its weight in host-rock (Fig. 9).
The physical controls on the assimilation of host-rocks by flowing magma were explored quantitatively by Huppert & Sparks (1989) and Williams et al. (2001). What follows is a qualitative discussion of the fundamental controls on the process in the study area. It might appear surprising that basalt could have been assimilated as easily as sediment by the Expo magma, in light of the very different contact relations observed where the Expo Intrusive Suite cuts these different lithologies. The reason for such different behaviour probably lies in the phase relations of the two possible contaminants. Although both could be assimilated in amounts as great as 50 wt % of the invading melt, this would occur only if the assimilant and the magma could be thoroughly mixed mechanically before reaching thermal equilibrium. In real systems, mixing can occur only if the physical states of the magma and the assimilant permit it. When magma at 1400°C comes into contact with sediment or basalt at 300°C, the temperature at the contact initially will be 850°C. This temperature is well below the solidus of komatiitic basalt, so that in both cases the magma will form an aphanitic chilled margin against its host-rock. The contact temperature is also below the solidus of metabasalt, so that, where the wall-rocks are basaltic, the contact will begin in a solid state and remain so indefinitely. In a closed system, as cooling of the intrusion continues, the cooled and crystallized margin will propagate inward and the heated thermal aureole will propagate outwards, but no part of the host-rock (basaltic) will ever be at its melting point. Intrusion of the Expo magma into basalt host-rocks will therefore not result in assimilation of the basalt unless some other process (e.g. stoping) intervenes to mechanically mix the basalt into the magma, or unless the magma flows continuously past the contact, heating it sufficiently to induce melting. On the other hand, the contact temperature of 850°C is above the likely solidus temperature of the metasediments. The contact zone would therefore be occupied by a thin crust of chilled magma in a vertical orientation with partially melted sediment on one side and still-molten intrusive rock on the other. This arrangement is mechanically unstable on a vertical dike wall and would break up the chilled crust, allowing the partially molten metasediment to mingle with the magma and therefore promote efficient assimilation of the host-rocks. Abundant fine-grained pyroxenite clasts form a locally clast-supported breccia with a peridotitic matrix in the lower portions of the Mequillon and Tootoo intrusions and could be interpreted as remnants of the foundered chilled margin in such a scenario. In summary, it is entirely consistent that the magma formed sharp chilled contacts with the host basalts rather than assimilating them, but was able to digest large quantities of the metasediments.
In either case, if the magma within the dike continues to flow, the temperature at the contact will increase and the chilled margin will tend to be resorbed. At this point the wall-rocks will be in contact with hot magma and assimilation (thermal erosion) will ensue regardless of the composition of the host-rock.
The preceding discussion of assimilation, fractional crystallization, and sulfide segregation leads to some interesting conclusions. The first is that a komatiitic basalt intruded into dikes and sills in the upper crust can assimilate up to 50% of its mass of host-rock, although in many cases it may only reach values nearer to 15%. The second is the observation that whereas the Expo magma reached the level of the Beauparlant Fm in a pristine, uncontaminated state as preserved in the chilled margins, it erupted at the surface (Chukotat Group) with a highly contaminated composition. In other words, the assimilation–fractional crystallization process happened at the present level of exposure within the dikes and sills that fed the overlying flows of the Chukotat Group, at a depth of less than 5 km below the paleosurface, rather than deep in the crust in some large but unexposed ‘staging’ magma chamber. Wherever ultramafic cumulate rocks are exposed in dikes or sills that show compositional evidence for contamination, a similar process might be inferred to have occurred.
|
GEOLOGICAL MODEL |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Structure and emplacement
The Expo Suite shows a distinct intrusive style in each of the three members of the Beauparlant Fm and in the Nuvilik Fm. Intrusions are essentially absent from the lower Beauparlant Fm, are broad and tend to show ultramafic character in the middle member, and are gabbroic in the upper member. Within the Nuvilik Fm, the intrusions are almost exclusively peridotitic and tend to form sills as well as dikes, whereas all of the intrusions hosted by the Beauparlant Fm appear to be dikes.
The preferred explanation for these observations is illustrated in the inset block diagram in the upper part of Fig. 2. The interpreted longitudinal section in Fig. 2 was drawn schematically along vertical panels following the trace of the intrusions through the mineralized Vaillant Lake, Mequillon, Cominga, Expo, Hilltop, TK and Mesamax NW portions of the system. Because the line of section was chosen to follow the axes of the dikes, they all appear in the longitudinal section as continuous bands; however, they extend into or out of the page only a few tens or hundreds of meters. The Expo Intrusive Suite appears to have been emplaced into a nearly flat-lying package of Povungnituk Group basalts and was subsequently deformed during the Trans-Hudson Orogeny along with the host basalts into the prominent NW–SE fold structures that dominate the present map pattern. The emplacement of the Expo Intrusive Suite was guided by the mechanical response of the layered rocks it intruded and by the buoyancy contrast between magma and host-rock, such that it was unable to penetrate far into the lower Beauparlant, but formed initially vertical, blade-shaped dikes through the middle and upper Beauparlant members and up into the overlying Nuvilik Fm. Where these dikes were surrounded by basalt of the upper Beauparlant Fm, they were unable to melt their host-rocks significantly and they therefore tended to freeze in place. Where the dikes passed through the volcanic–sedimentary pile of the middle Beauparlant Fm, they were able to melt the metasedimentary horizons and then pluck and assimilate basaltic components of the wall-rocks, opening themselves out into broad, long-lived magma conduits. Solidification of the upper portions of the gabbroic dikes in the upper Beauparlant sealed the mid-Beauparlant dike segments off from the surface and aided their development as long-lived conduits for lateral magma transport over distances of tens of kilometers. Lateral transport permitted the thermal erosion of the wall-rock sediment, and, to a more limited extent, of basalt within the middle Beauparlant Fm. Assimilation of wall-rock metasediment and basalt in the middle Beauparlant Fm caused the magma to become sulfide-saturated, a process that is recorded by ubiquitous minor sulfide mineralization along the contacts. As the sulfide melt generated in this setting settled, it sank to the bottom of the dikes and collected along the length of their basal terminations.
Where the Expo Suite magma broke through the upper Beauparlant Fm into the Nuvilik Fm, it either formed sills along the top of the contact, as in the Cominga area, or continued upward as dikes or cylindrical conduits like that at Snow Owl. Because of the ease of assimilation of the semipelitic host-rocks of the Nuvilik Fm, the intrusions developed into a complex of cross-cutting dikes and sills filled almost entirely with ultramafic rocks that effectively replace at constant volume the sediments they intruded.
The present distribution of the Expo Intrusive Suite in outcrop is dictated by the interplay between the original stratigraphic control on magma migration pathways and the subsequent deformation. Where the dikes cross the hinges of major anticlines, there are gaps in their outcrop because these fold hinges expose lower Beauparlant Formation or even Dumas Formation, below the lowest extent of the dikes. Where the intrusions cross the hinges of major synclines, the basal portions are deeply buried several kilometers below the present erosional surface and all that can be seen at surface are the gabbroic upper portions of the dike. The preservation of a typical exposure of a dike at the current erosional level in a syncline is illustrated in Fig. 12, which shows the configuration at the time of emplacement, after folding, and finally after erosion. Sulfide deposits appear at the surface in those places where dike segments terminate at the current erosional surface, but the sulfide bodies appear to be continuous down plunge along the entire lengths of dikes (Fig. 5).
|
The Expo–Ungava deposit and its surrounding ultramafic intrusions are exposed in the center of a 10 km scale doubly plunging synclinorium, or basin, exposing the Nuvilik Fm. These represent the highest structural levels preserved at present in the Expo Intrusive Suite, with the basal portions at present probably 4–5 km below the surface. The focusing of the basin structure on the Expo area may be a consequence of the presence of the dense ultramafic rocks themselves, as this large mass could have loaded the Beauparlant Fm and nucleated a synform during lateral compression. On the other hand, the type of intrusion visible in the Expo–Ungava deposit area may originally have been present along the entire length of the dike system and the segment found at Expo–Ungava may have been preserved purely by chance during regional folding.
Relationship between Expo, Chukotat and Raglan suites
The Expo magma shows remarkable similarities both to the olivine-phyric basalts of the Chukotat Group and to the putative initial liquid of the Raglan suite, which generated the important Ni–Cu–PGE deposits of the Raglan Mine. It is generally assumed that the Raglan suite constitutes a series of shallow subvolcanic intrusions and associated channelized lava flows related to the emplacement of the Chukotat Group volcanic rocks (Bédard et al., 1984; Barnes & Barnes, 1990; Burnham et al, 1999), and furthermore that the Expo Intrusive Suite formed contemporaneously as the upper-crustal feeder system to the Chukotat Group (e.g. Giovenazzo et al., 1989). The available radiometric ages and compositions are consistent with this interpretation. Large igneous provinces involving voluminous picritic magmas such as those of the Expo and Raglan suites probably undergo an episode of peak magma production lasting about 1–3 Myr, and effusion is generally entirely completed within a span of about 10 Myr (e.g. Courtillot & Renne, 2003). The estimated 30 Myr time span between the emplacement of the Expo and Raglan suites and the 1918 Ma age of a gabbroic intrusion below the Chukotat Group is too long to allow an interpretation that all three are comagmatic; hence, the earlier intrusion may not bear any relationship to the main magmatic event that produced the Expo, Raglan, and Chukotat suites at c. 1887–1870 Ma.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
The parent magma of the Paleoproterozoic Expo Intrusive Suite was a komatiitic basalt with a weakly LILE-depleted composition. Neither the basalts of the Povungnituk Group nor the intrusions of the Expo Intrusive Suite were emplaced in a rifting environment. The Povungnituk Group was emplaced onto a long-lived passive continental margin, followed at least 108 Myr later by the simultaneous intrusion of the sills and dikes that form the Expo Suite and eruption of the lavas of the Raglan Formation. A very strong control on the intrusive style and differentiation of the EIS was exerted by its country rocks. Where the intrusions were emplaced into metasediments, they formed large, sill-like and relatively irregular ultramafic cumulate bodies accommodated by assimilation of the wall-rocks. Where the intrusions were emplaced into predominantly basaltic host-rocks they were confined by their relatively infusible hosts and formed sharp-walled dikes. Assimilation of metasediments in the middle member of the Beauparlant Fm caused the magma to reach sulfide saturation. Sulfide melt percolated down the dike walls and accumulated in the basal terminations of the blade-shaped dikes, where they taper out within the lowest basalts of the Beauparlant Fm.
The very strong crustal contamination signature acquired by the magmas residual to the formation of the ultramafic cumulates gave them a classical continental tholeiite trace element signature, which is recorded at present in the compositions of some of the overlying basalts of the Chukotat Group. The mechanism and structural aspect of the assimilation–fractional crystallization scenario proposed here is different from the mid-crustal staging chamber commonly proposed as the site of AFC processes in other suites currently represented only as basalts and shallow intrusions, such as the Noril'sk–Talnakh intrusions. Many other continental or continental margin suites may have been emplaced through similar conduit systems, producing similar dike-like or sill-like bodies, which were the sites of extensive interaction with the continental crust. A very similar mineralized body is being mined at Jinchuan, China (Chai & Naldrett, 1992), and might have formed in much the same manner.
|
SUPPLEMENTARY DATA |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Supplementary data for this paper are available at Journal of Petrology online.
|
ACKNOWLEDGEMENTS |
|---|
This work has been made possible entirely through the generous support of Canadian Royalties Inc., who are exploring the Expo Intrusive Suite for Ni–Cu–PGE mineralization. Helicopter support, camp accommodation, and support for lithogeochemical analysis and petrography has been unstintingly offered over four field seasons. CRI geologists Todd Keast, Bruce Durham, and William Randall have offered invaluable advice and discussion of the ideas presented here. The present manuscript has benefited enormously from careful and thoughtful reviews by Peter Lightfoot, Mike Lesher, and associate editor Nick Arndt.
*Corresponding author. Telephone: 1 416 978 2975. Fax: 1 416 978 3938. E-mail: mungall{at}geology.utoronto.ca
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONREGIONAL GEOLOGYEXPO INTRUSIVE SUITELITHOGEOCHEMISTRYDISCUSSIONGEOLOGICAL MODELCONCLUSIONSSUPPLEMENTARY DATAREFERENCES |
|---|
Arndt NT, Czamanske GK, Walker RJ, Chauvel C, Fedorenko VA. Geochemistry and origin of the intrusive hosts of the Nori'sk–Talnakh Cu–Ni–PGE sulfide deposits. Economic Geology (2003) 98:495–515.
Barnes S-J, Barnes SJ. A new interpretation of the Katiniq Nickel Deposit, Ungava, northern Quebec. Economic Geology (1990) 85:1269–1272.
Barnes S-J, Coats CJA, Naldrett AJ. Petrogenesis of a Proterozoic nickel-sulfide–komatiite association, the Katiniq sill, Ungava, Quebec. Economic Geology (1982) 77:413–429.
Bédard JH, Francis DM, Hynes AJ, Nadeau S. Fractionation in the feeder system at a Proterozoic rifted margin. Canadian Journal of Earth Sciences (1984) 21:489–499.
Boudreau AE. PELE—a version of the MELTS software program for the PC platform. Computers and Geosciences (1999) 25:201–203.[CrossRef]
Burnham OM, Lesher CM, Keays RR. Geochemistry and petrogenesis of mafic–ultramafic complexes and associated basalts in the Raglan Block. In: Komatiitic Peridotite-Hosted Ni–Cu–(PGE) Deposits of the Raglan Area, Cape Smith Belt, New Quebec. Guidebook Series, Volume 2.—Lesher CM, ed. (1999) Sudbury, ON: Mineral Exploration Research Centre, Laurentian University. 159–173.
Chai G, Naldrett AJ. The Jinchuan ultramafic intrusion: cumulate of a high-Mg basaltic magma. Journal of Petrology (1992) 33:277–303.
Courtillot VE, Renne PR. On the ages of flood basalt events. Comptes Rendus Géosciences (2003) 335:113–140.[CrossRef]
Francis D, Ludden J, Hynes A. Magma evolution in a Proterozoic rifting environment. Journal of Petrology (1983) 24:556–582.[Web of Science]
Ghiorso MS, Sack RO. Chemical mass transfer in magmatic processes. 4. A revised and internally consistent thermodynamic model for the interpolation and extrapolation of liquid–solid equilibria in magmatic systems at elevated temperatures and pressures. Contributions to Mineralogy and Petrology (1995) 199:197–212.
Giovenazzo D, Picard C, Guha J. Tectonic setting of the Ni–Cu–PGE deposits of the central part of the Ungava trough. Geoscience Canada (1989) 16:134–137.[Web of Science]
Huppert HE, Sparks RSJ. Chilled margins in igneous rocks. Earth and Planetary Science Letters (1989) 92:397–405.[CrossRef][Web of Science]
Hynes A, Francis DM. A transect of the early Proterozoic Cape Smith foldbelt, New Quebec. Tectonophysics (1982) 88:23–59.[CrossRef][Web of Science]
Lévesque M, Lesher CM. Invasive features of mafic–ultramafic rocks at the Zone 3, Zone 2, and Katinniq Ni–Cu–(PGE) deposits. (2003) 9th International Platinum Symposium, Billings, Montana, 2003. Cape Smith Belt, Nouveau Quebec: Raglan Formation.
Mahoney John J, Coffin Millard F. Flood basalts and magmatic Ni, Cu, and PGE sulphide mineralization: comparative geochemistry of the Noril'sk (Siberian traps) and West Greenland sequences. Large Igneous Provinces: Continental, Oceanic and Planetary Flood Volcanism, Geophysical Monograph, American Geophysical Union (1997) 100:357–380.
Lucas SB, St-Onge MR. Terrane accretion in the internal zone of the Ungava orogen, northern Quebec. Part 2: Structural and metamorphic history. Canadian Journal of Earth Sciences (1992) 29:765–782.
Machado N, David J, Scott DJ, Lamothe D, Philippe S, Gariépy C. U–Pb geochronology of the western Cape Smith Belt, Canada: new insights into the age of initial rifting and arc magmatism. Precambrian Research (1993) 63:211–223.[CrossRef][Web of Science]
Mavrogenes J, O’Neill HStC. The relative effects of pressure, temperature and oxygen fugacity on the solubility of sulfide in mafic magmas. Geochimica et Cosmochimica Acta (1999) 63:1173–1180.[CrossRef][Web of Science]
McDonough WF, Sun S-s. The composition of the Earth. Chemical Geology (1995) 120:223–253.[CrossRef][Web of Science]
McKenzie D, Bickle MJ. The volume and composition of melt generated by extension of the lithosphere. Journal of Petrology (1988) 29:625–679.
Modeland S, Francis D, Hynes A. Enriched mantle components in Proterozoic continental-flood basalts of the Cape Smith foldbelt, northern Quebec. Lithos (2003) 71:1–17.[CrossRef][Web of Science]
O’Hara MJ, Mathews RE. Geochemical evolution in an advancing, periodically replenished, periodically tapped, continuously fractionated magma chamber. Journal of the Geological Society, London (1981) 138:237–277.
Parrish RR. U–Pb geochronology of the Cape Smith Belt and Sugluk block, northern Quebec. Geoscience Canada (1989) 16:126–130.[Web of Science]
Picard C. Synthèse pétrogéochimique des roches volcaniques protérozoiques de la ceinture orogénique de l’Ungava: évolution géologique des Groupes de Povungnituk, de Chukotat et de Parent. (1995) MB95-01. Ministère des ressources naturelles, Gouvernement de Québec. 264.
Picard C, Lamothe D, Piboule M, Oliver R. Magmatic and geotectonic evolution of a Proterozoic oceanic basin system: the Cape Smith Thrust–Fold Belt (New-Quebec). Precambrian Research (1990) 47:223–249.[CrossRef][Web of Science]
Randall W. U–Pb geochronology of the Expo Intrusive Suite, Cape Smith Belt, and the Kyak Bay intrusion, New Quebec Orogen: implications for the tectonic evolution of the northeastern Trans-Hudson Orogen. In: M.Sc. thesis (2005) University of Toronto.
Roedder PL, Emslie RF. Olivine–liquid equilibrium. Contributions to Mineralogy and Petrology (1970) 29:275–289.[CrossRef][Web of Science]
Scott DJ, St-Onge MR, Lucas SB, Helmstaedt H. The 1998 Purtuniq ophiolite: imbricated and metamorphosed oceanic crust in the Cape Smith Thrust Belt, northern Quebec. Geoscience Canada (1989) 16:144–147.[Web of Science]
Scott DJ, Helmstaedt H, Bickle MJ. Purtuniq ophiolite, Cape Smith belt, northern Quebec, Canada: a reconstructed section of early Proterozoic oceanic crust. Geology (1992) 20:173–176.
St-Onge MR, Lucas SB. Geology of the Eastern Cape Smith Belt: parts of the Kangiqsujuaq, Cratère du nouveau-Québec, and Lacs Nuvilik map areas, Quebec. Geological Survey of Canada Memoir (1993) 438:110.
St-Onge MR, Lucas SB, Parrish RR. Terrane accretion in the internal zone of the Ungava orogen, northern Quebec. 1: Tectonostratigraphic assemblages and their tectonic implications. Canadian Journal of Earth Sciences (1992) 29:746–764.
Wendlandt RF. Sulfide saturation of basalt and andesite melts at high pressures and temperatures. American Mineralogist (1982) 67:877–885.[Abstract]
White R, MacKenzie E. Magmatism at rift zones: the generation of volcanic continental margins and flood basalts. Journal of Geophysical Research (1989) 94:7685–7729.
Williams DA, Kerr RC, Lesher CM, Barnes SJ. Analytical/numerical modeling of komatiite lava emplacement and thermal erosion at Perseverance, Western Australia. Journal of Volcanology and Geothermal Research (2001) 110:27–55.[CrossRef][Web of Science]
Wodicka N, Madore L, Larbi Y, Vicker P. Géochronologie U–Pb de filons-couches mafiques de la Ceinture de Cape Smith et de la Fosse du Labrador. L’exploration minérale au Québec: notre savoir, vos découvertes. Séminaire d’information sur la recherche géologique, Programme et résumés 2002, Ministère des Ressources Naturelles, Québec (2002) V 2002-10:48.
Wooden JL, Czamanske GK, Fedorenko VA, Arndt NT, Chauvel C, Bouse RM, King B-SW, Knight RJ, Siems DF. Isotopic and trace-element constraints on mantle and crustal contributions to Siberian continental flood basalts, Noril'sk area, Siberia. Geochimica et Cosmochimica Acta (1993) 57:3677–3704.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||