Journal of Petrology Advance Access originally published online on January 30, 2008
Journal of Petrology 2008 49(3):493-522; doi:10.1093/petrology/egm090
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Petrogenesis of Cogenetic Silica-Oversaturated and -Undersaturated Syenites by Periodic Recharge in a Crustally Contaminated Magma Chamber: the Kangerlussuaq Intrusion, East Greenland
1Department of Earth Sciences, University of Aarhus, 8000 Århus C, Denmark
2Department of Geoscience, University of Iowa, 121 Trowbridge Hall, Iowa City, Ia 52242, USA
3Geological Institute, University of Copenhagen, ØSter Voldgade 10, 1350 Copenhagen K, Denmark
RECEIVED DECEMBER 2, 2004; ACCEPTED DECEMBER 27, 2007
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONThe Palaeogene Kangerlussuaq Intrusion (
50 Ma) of East Greenland displays concentric zonation from quartz-rich nordmarkite (quartz syenite) at the margin, through pulaskite, to foyaite (nepheline syenite) in the centre; modal layering and igneous lamination are locally developed but there are no internal intrusive contacts. This is an apparent violation of the phase relations in Petrogeny's Residua System. We propose that this intrusion is layered, grading from quartz syenite at the bottom to nepheline syenite at the top. Mineral and whole-rock major and trace element data and Sr–Nd–Hf–Pb isotope data are presented that provide constraints on the petrogenesis of the intrusion. Radiogenic isotope data indicate a continuously decreasing crustal component from the quartz nordmarkites (87Sr/86Sr = 0·7061; 
Ndi = 2·3; Hfi = 5·2; 206Pb/204Pbmeas = 16·98) to the foyaites (87Sr/86Sr = 0·7043–0·7044; Ndi = 3·8–4·9; Hfi = 10·7–11·1; 206Pb/204Pbmeas = 17·78–17·88); the foyaites are dominated by a mantle isotopic signature. The average Mg-number of amphibole cores becomes increasingly primitive, varying from 26·4 in the nordmarkites to 57·4 in the pulaskites. Modal layering, feldspar lamination and the presence of huge basaltic xenoliths derived from the chamber roof, now resting on the transient chamber floor, demonstrate bottom-upwards crystallization. The intrusion cannot, therefore, have formed in a system closed to magmatic recharge. The lack of gneissic xenoliths in the nordmarkites suggests that most contamination took place deeper in the crust. In the proposed model, the nordmarkitic magma formed during crustal assimilation in the roof zone of a large, silica-undersaturated alkali basaltic/basanitic, stratified magma chamber, prior to emplacement in the uppermost crust. The more primitive syenites, terminating with foyaite at the top of the intrusion, formed as a consequence of repeated recharge of the Kangerlussuaq Intrusion magma chamber by tapping less contaminated, more primitive phonolitic melt from deeper parts of the underlying chamber during progressive armouring of the plumbing system. KEY WORDS: Kangerlussuaq; East Greenland; syenite; crustal contamination; magma mixing
|
INTRODUCTION |
|---|
Processes leading to the formation of cogenetic quartz and nepheline syenites have been a long-standing issue in petrological research because of the presence of a thermal barrier (Ab–Or join) between them in Petrogeny's Residua System (Ne–Ks–Q–H2O). Liquidus phase relations in this simple system imply that a melt will fractionate either towards the granite minimum and give rise to silica-oversaturated rocks, or towards the nepheline syenite minimum and give rise to silica-undersaturated rocks (Schairer, 1950
; Tuttle & Bowen, 1958; Fudali, 1963; Hamilton & MacKenzie, 1965).
Mechanisms suggested to allow for generation of cogenetic over- and undersaturated rocks are: (1) processes that overcome or remove the thermal barrier, involving volatiles (Kogarko, 1974), increasing water pressure (Pankhurst et al., 1976), or crystal fractionation (Foland & Henderson, 1976; Giret et al., 1980); (2) open-system processes in which undersaturated magmas undergo assimilation of silica-rich crustal material and fractional crystallization (AFC) or produce oversaturated anatectic crustal melts (Brooks & Gill, 1982; Fitton, 1987). Increasing support has recently been given to open-system processes, as first proposed by Brooks & Gill (1982) (e.g. Foland et al., 1993; Harris, 1995; Harris et al., 1999; Marks et al., 2003). Foland et al. (1993) presented data from the complexes of Marangudzi, Zimbabwe, and Mt. Brome, southern Quebec, characterized by cogenetic quartz and nepheline syenites, which showed that the undersaturated rocks have low Sr isotopic ratios and high Nd isotopic ratios, whereas the oversaturated syenites have high Sr isotopic ratios and low Nd isotopic ratios, indicating that these rocks originated from a felsic, undersaturated magma subjected to open-system crustal contamination processes. Using phase equilibrium constraints, and assuming that the required heat for assimilation is balanced by the latent heat of crystallization with no cooling of the magma, Foland et al. (1993) calculated model liquid AFC and fractional crystallization paths from an undersaturated melt. They concluded that closed-system fractional crystallization will produce only silica-undersaturated melts, whereas silica-oversaturated melts require crustal contamination processes.
The Kangerlussuaq Intrusion is an important example of syenitic magma that appears to have differentiated across the thermal barrier between silica-oversaturated and -undersaturated compositions (Wager, 1965), in apparent violation of phase relations in Petrogeny's Residua System (Schairer, 1950). Unlike most alkaline complexes with coexisting quartz and nepheline syenites (e.g. the Ilímaussaq, Marangudzi and Mt. Brome complexes), the Kangerlussuaq Intrusion has no internal intrusive contacts, but displays a gradual transition from quartz syenite to nepheline syenite. In this paper we present comprehensive mineral, major and trace element whole-rock and Sr–Nd–Hf–Pb isotopic data for a suite of samples from the Kangerlussuaq Intrusion and evaluate the magmatic processes that can produce a close association of silica-oversaturated and -undersaturated syenites.
|
GEOLOGICAL BACKGROUND |
|---|
Regional geology and geochronology
The Kangerlussuaq Intrusion (KI) is situated within the East Greenland volcanic rifted margin and belongs to the Palaeogene North Atlantic Igneous Province (e.g. Saunders et al., 1997
) (Fig. 1). The East Greenland tholeiitic lavas and gabbroic intrusions can be divided into two discrete phases: (1) the earliest volcanics, including the Lower Basalts (Nielsen et al., 1981), Nansen Fjord (Larsen et al., 1989) and Urbjerget formations (Hansen et al., 2002) at 61–58 Ma, which have been related to plume impact under Central Greenland; (2) the voluminous tholeiitic Plateau Basalts (e.g. Larsen et al., 1989; Pedersen et al., 1997; Tegner et al., 1998b) at 56–55 Ma (Storey et al., 2007), possibly erupted in just 300 kyr (Larsen & Tegner, 2006), which represent decompressional melting during continental break-up (Tegner et al., 1998a). This activity is followed by younger events, dominantly at 50–47 Ma but extending to 37–35 Ma (Tegner et al., 2008) and 14–13 Ma (Storey et al., 2004), with production of mainly gabbroic and syenitic intrusions (e.g. Nielsen, 1987; Bernstein et al., 1998) that post-date the initiation of sea-floor spreading, and are inferred to be caused by the passage of the rifted margin over the Iceland plume axis (Tegner et al., 1998a, 2008).
|
The alkaline magmatism of the Kangerlussuaq area is volumetrically dominated by the Kangerlussuaq Alkaline Complex, which includes the KI (Wager, 1965
; Kempe & Deer, 1970, 1976; Kempe et al., 1970; Pankhurst et al., 1976; Brooks & Gill, 1982) plus a series of older and younger satellite intrusions (e.g. Deer & Kempe, 1976; Holm & Prægel, 1988, 2006; Nielsen, 2002; Riishuus et al., 2005, 2006) (Fig. 1b). The KI has been dated by several techniques. Beckinsale et al. (1970) dated a transitional pulaskite by the K–Ar method to 50·4 ± 1·2 Ma. Pankhurst et al. (1976) reported two Rb–Sr ages: 49·9 ± 1·0 Ma (biotite/feldspar/titanite) and 50·0 ± 1·9 Ma (whole-rock). Gleadow & Brooks (1979) reported zircon (50·8 ± 1·8 Ma) and titanite (51·1 ± 2·8 Ma) fission-track ages. Tegner et al. (2008) dated biotite from a pulaskite to 50·8 ± 1·1 Ma using 40Ar–39Ar.
The Kangerlussuaq Intrusion
The KI was discovered by L. R. Wager during the British Arctic Air Route Expedition in 1930. The intrusion was subsequently described in three papers that focused on its form and structure (Wager, 1965), petrology (Kempe et al., 1970) and mineralogy (Kempe & Deer, 1970).
The KI is roughly circular with a diameter of 30–35 km (800 km2), which makes it the largest exposed Palaeogene intrusion in East Greenland and one of the largest syenitic intrusions in world. To the authors’ knowledge the Brandberg Complex (Schmitt et al., 2000), and Khibina and Lovozero centres (e.g. Kramm & Kogarko, 1994) are the only other evolved intrusions of comparable size and similar setting. The KI grades from foyaites (>5% feldspathoids) in the centre, through main pulaskites (<5% feldspathoids) and transitional pulaskites (neither quartz nor feldspathoids), into nordmarkites (<10% quartz) and outermost quartz nordmarkites (10% quartz) (Wager, 1965) (Fig. 1b). Wager (1965) argued that the gradual transition between the units indicates only a single pulse of magma injection. Based on field relations, especially the presence of trains of basaltic xenoliths, from the quartz nordmarkites through the outer part of the main pulaskites, dipping 30–60° towards the centre (Fig. 2a) and igneous lamination in the form of platy feldspars (Fig. 3c) dipping similarly inwards, Wager (1965) visualized the three-dimensional shape of the intrusion as resembling a pile of saucers with a decreasing diameter inwards and upwards through the succession and an assumed horizontal base, which gave an estimated volume of 6500 km3 (Fig. 1c). He showed that the intrusion was emplaced at the unconformity between the Archaean basement and the overlying flood basalts, and had stoped into the lava pile. We propose a modification of Wager's model shape by suggesting an asymmetric, saucer-shaped base (Fig. 1c), not unlike the ‘classical’ funnel shape suggested for many large intrusions, which reduces the volume to 4000 km3. The field evidence of basaltic xenoliths and platy feldspars along with the relationship of nordmarkites being cut by nepheline-bearing pegmatites whereas the foyaites are not cut by quartz-bearing aplites and pegmatites, led Wager (1965) to regard the KI as a layered intrusion that solidified from the sides, inwards, and from the bottom, upwards. However, he could not explain the transition from over- to undersaturated syenites and assumed that the parent magma was nordmarkitic in composition and formed by melting and alkali metasomatism of the local granitic basement.
|
|
Kempe & Deer (1976
) and Pankhurst et al. (1976) argued that the KI was derived from an oversaturated quartz trachyte magma. Pankhurst et al. (1976) reported 87Sr/86Sri values of 0·7045 for the undersaturated rocks and 0·7045–0·7095 for the nordmarkites sensu lato. This led Kempe & Deer (1976) to suggest that the quartz trachytic magma differentiated from a mantle-derived alkali olivine basaltic melt, and they speculated that lowering of the liquidus in the Ne–Ks–Q–H2O system by elevated vapour pressure allowed crossing of the thermal barrier.
Pankhurst et al. (1976) observed that the whole-rock oxygen isotope compositions of the foyaites (
18O = 3·7–4·3
) are lower than those of the nordmarkites sensu lato (18O = 3·7–5·5). These 18O values are below purely mantle-derived magma values (18O = 5·7–6·5) (e.g. Bindeman et al., 2004). To produce the low 18O values of the foyaites, Pankhurst et al. (1976) suggested that meteoric water entered the system after 60% fractionation. Neither Kempe & Deer (1976) nor Pankhurst et al. (1976), however, were able to explain fully how the melt was driven towards the nepheline syenite minimum, but discussed possible mechanisms of silica loss and suggested vapour transfer of silica or assimilation of basalt as possible mechanisms.
In conflict with the models described above, Brooks & Gill (1982) found that the most magnesian alkali pyroxenes were present in the foyaites and proposed that the foyaites represent the products of fractional crystallization of alkali basalt or nephelinite. In the model of Brooks & Gill (1982), the pulaskites and nordmarkites formed by increasing contamination of the magma with country rock gneiss and basalt. The high 87Sr/86Sri values of the quartz nordmarkites reported by Pankhurst et al. (1976) are consistent with this model and show that the KI did not form as a response to simple closed-system fractional crystallization. Brooks & Gill (1982) pointed to the problem of having enough thermal energy available to heat and assimilate large amounts of wall-rock material, but argued that the crust was probably at an elevated temperature as the Kangerlussuaq area had undergone long-lasting magmatic activity.
|
FIELD RELATIONS AND SAMPLES |
|---|
A distinctive feature of the quartz nordmarkites is the presence of trains of huge, elongated, basaltic xenoliths (up to several hundreds of metres across) dipping 30–60° towards the centre of the intrusion (Fig. 2a). A few basaltic xenoliths also occur in the outer parts of the main pulaskites, but otherwise they are absent from the inner part (Wager, 1965
). The basaltic xenoliths are fractured and veined by syenite (Fig. 2b). The shape of the contact between basaltic xenoliths and penetrating syenite veins changes from being curved in the marginal part (outer 2–3 cm) of the xenoliths to straight further inwards. When detached in smaller fragments (<20 cm), the xenoliths are rounded and moderately crenulated against a 0–5 cm wide zone of syenite enriched in titanite and FeTi-oxides. Modal layering, defined by the planar concentration of mafic minerals, is locally developed in the quartz nordmarkites (Fig. 2c). Both the nordmarkites and pulaskites locally display igneous lamination defined by platy feldspars (Fig. 3c) dipping 30–60° towards the centre of the intrusion (Wager, 1965). The contact between the KI and the surrounding gneisses dips steeply (70–80°) away from the centre (Wager, 1965). Gneissic xenoliths are rare in the KI. The alpine topography imposes serious challenges for sampling, as solid syenite outcrops are usually frost-shattered and weathered. Sampling was carried out with helicopter support, which provided an opportunity to retrieve large fresh samples, thereby minimizing the effects of sample heterogeneity. A total of 31 syenite samples were collected from the innermost to outermost part of the KI, together with additional samples of the basaltic xenoliths (n = 3), mafic dykes cutting the KI (n = 3) and local Archaean basement (n = 3). Sample locations are shown in Fig. 1b and global positioning system data are given in Table 1.
|
|
STRUCTURE AND STRATIGRAPHY |
|---|
To present the compositional variations within the intrusion in a simple, stratigraphically meaningful manner, the KI samples have been projected onto a schematic stratigraphic section through the centre of the intrusion. The asymmetric, saucer-shaped structure (Fig. 1c) and the relative radial position of sample locations between the centre and margin of the intrusion (Fig. 1b), taking the petrographic transitions into account, have been used for the projection. The central-most foyaite samples (454088–89) are set to 0 m stratigraphic height, the base of the stratigraphic section (gneiss–quartz nordmarkite contact) is set to –7500 m, and the petrographic transitions foyaite–main pulaskite, main pulaskite–transitional pulaskite, transitional pulaskite–nordmarkite, and nordmarkite–quartz nordmarkite are set to –700 m, –1500 m, –2000 m, and –5000 m, respectively. As an example, main pulaskite sample 454048 is located about two-thirds of the distance from the transitional–main pulaskite transition towards the main pulaskite–foyaite transition (location 20, Fig. 1b), which equals a stratigraphic level of roughly –1000 m.
|
CLASSIFICATION AND PETROGRAPHY |
|---|
A summary of petrographic observations (this study; Kempe & Deer, 1970
; Kempe et al., 1970) is presented below and a stratigraphic section summarizes the approximate modal variations (Fig. 4a). Except for two samples, they are all peralkaline with peralkalinity indexes [PI, (Na2O + K2O)/Al2O3, molar ratio] of 0·98–1·17 (Table 1).
|
, analyses of pyroxene and amphibole inclusions in an alkali feldspar phenocryst from a quartz nordmarkite (454079). Mineral analyses represent means for each sample; horizontal error bars are ±1 SD (±1
), Pl, plagioclase; Px, pyroxene; Gt, garnet.Quartz nordmarkites
The quartz nordmarkites are classified as quartz alkali feldspar syenites after IUGS (Le Bas & Streckeisen, 1991
). They are generally porphyritic, white–yellow to light brown rocks with 0·5–3·0 cm phenocrysts of zoned grey, euhedral, fine microperthitic alkali feldspar laths (Fig. 3a) in a medium-grained groundmass of sub- to anhedral coarse microperthitic alkali feldspars (75–80% total feldspar) and interstitial quartz (10–13%) in clusters with sodic–calcic amphiboles (5–7%), aegirine–augite (1–3%) and accessory magnetite, ilmenite, titanite, apatite, euhedral zircon and chevkinite (Fig. 3b). Albitic rims on groundmass and phenocryst alkali feldspars are common in all rock types of the intrusion. Corona textures with aegirine–augite rimmed by amphibole are common, with varying core–rim proportions and a ‘disintegrated’ appearance to the aegirine–augite suggesting replacement. The amphiboles are commonly zoned from light green–brown ferrorichterite in the core to darker green–brown arfvedsonite at the margin (Fig. 3b). Rings of minute inclusions outline the zoning pattern in the alkali feldspar phenocrysts; these include biotite, magnetite, ilmenite and magnesiokatophorite-mantled aegirine–augite in the central parts (Fig. 3a).
Nordmarkites
The nordmarkites are classified as (quartz) alkali feldspar syenites after IUGS (Le Bas & Streckeisen, 1991). These rocks are medium-grained and porphyritic with 0·5–3·0 cm phenocrysts of zoned dark grey, euhedral to subhedral, fine microperthitic alkali feldspar laths in a groundmass of sub- to anhedral microperthitic feldspars (85–90% total feldspar), interstitial quartz (1–5%), commonly zoned sodic–calcic to sodic amphiboles (3–8%), and green pleochroic aegirine–augite (2%). Accessories are magnetite, ilmenite, plagioclase, biotite, titanite (in one sample up to 3%), apatite, and euhedral zircon. The mafic silicates occur in interstitial clusters with quartz and accessory phases. The alkali feldspar phenocrysts host a variety of inclusions, as in the quartz nordmarkites.
Transitional pulaskites
The transitional pulaskites are coarse-grained inequigranular alkali feldspar syenites with neither quartz nor nepheline. They occasionally contain 1·0–2·5 cm euhedral, fine microperthitic alkali feldspar phenocrysts in a groundmass of euhedral to anhedral coarse microperthitic alkali feldspar (85%), interstitial albitic plagioclase (5%), sodic–calcic to sodic amphiboles (2–7%), aegirine–augite (3–5%) and biotite (3%). The mafic silicates occur in clusters with accessory titanite, magnetite, ilmenite, and apatite. A variety of amphiboles are present, from unzoned magnesiokatophorite, richterite and ferrorichterite to zoned varieties with magnesiokatophorite (core) to ferrorichterite (rim) and richterite (core) to magnesioarfvedsonite (rim).
Main pulaskites
These coarse-grained grey, foid-bearing alkali feldspar syenites consist of euhedral to anhedral coarse microperthitic feldspars (75–80%), interstitial, zoned amphibole with a light green–yellow magnesiokatophorite (core) to brownish green–yellow katophorite (rim) (5–8%), biotite (4–5%), light green aegirine–augite (1–2%), and euhedral titanite (up to 3–4 mm). Aegirine–augite is present both as an interstitial phase and as inclusions in feldspars. Clusters of interstitial phases include mafic silicates, magnetite, titanite, apatite and occasional melanite garnet. An intergrowth of sodalite, nepheline and cancrinite occurs in interstitial pockets separate from the other interstitial phases. There is no sign of replacement of pyroxene by amphibole, but biotite appears to replace katophorite. Some samples show a distinct igneous lamination defined by oriented alkali feldspar laths (Fig. 3c).
Foyaites
These nepheline syenites have PI of 0·98–1·04 (Table 1) indicating that they are not agpaitic (PI 
1·2), but miaskitic (PI <1) to intermediate. They are medium-grained inequigranular grey rocks with 0·5–1·0 cm euhedral to subhedral, coarse microperthitic alkali feldspar (60%) with many minute inclusions, euhedral to subhedral nepheline (25–30%), sodalite (0–8%), aegirine–augite (5–10%) and melanite garnet (0–4%). The aegirine–augite is commonly strongly zoned; the core is weakly pleochroic light yellow to light green, whereas the rim is strongly pleochroic from yellow–green to dark green (Fig. 3d). Although the transition from core to rim is very sharp, the shape of the core is commonly irregular and patchy (Fig. 3d). The mafic phases occur in interstitial clusters and are commonly associated with the feldspathoid minerals (Fig. 3d). The foyaites contain no amphibole. When present, orange–brown subhedral to anhedral melanite garnet occasionally shows colour zoning (Fig. 3d). Sodalite is present as an intergrowth with haüyne, analcite and magnetite.
Basaltic xenoliths
The xenoliths vary texturally from aphyric to pyroxene glomeroporphyritic. The main constituents are plagioclase (50–70%), FeTi-oxides (10–30%), biotite (5–20%), ortho- and clinopyroxene (0–15%), chlorite (0–5%) and carbonate (0–5%). Former vesicles are filled with carbonate. These xenoliths are classified according to the total alkali–silica (TAS) diagram of Le Maitre (2005) as basalts to trachybasalts.
Dykes
The mafic dykes vary from aphyric to plagioclase microporphyritic and pyroxene glomeroporphyritic. The main constituents are plagioclase (45–60%), FeTi-oxide (10–20%), biotite (5–20%), augite (0–30%), sericite (0–15%), chlorite (0–5%) and carbonate (0–5%). Carbonate fills vesicular cavities. Based on the TAS classification scheme, they are trachybasalts.
|
ANALYTICAL METHODS |
|---|
A total of 40 whole-rock samples (Table 1) were analysed for major and trace elements by X-ray fluorescence (XRF) on glass discs and pressed powder pellets, respectively, using a Phillips PW-2400 spectrometer at the Department of Earth Sciences, University of Aarhus. Altered surfaces were removed prior to whole-rock powder preparation.
Mineral compositions were determined with a JEOL JXA-8600 Superprobe at the Department of Earth Sciences, University of Aarhus, using a beam current of 15 nA, accelerating voltage of 15 kV and 5 µm beam diameter. Pyroxenes were analysed for Al, Ti, Cr, Mn, Ca and Na (wavelength-dispersive spectrometry; WDS), and Si, Fe and Mg (energy-dispersive spectrometry; EDS). Amphiboles were analysed for Cr, Mn, Ca, Na and K (WDS), and Si, Al, Ti, Fe and Mg (EDS). Counting times for WDS were 40 s or until the standard deviation on total counts was less than 1%. Counting times for EDS were 200 s. The mineral chemistry was commonly based on three analysis points in each of several grains. Data given in Tables 2 and 3 represent average compositions of pyroxenes and amphiboles, respectively.
|
|
Titanite (n = 7) and alkali feldspar (n = 12) minor and trace element analyses were carried out by inductively coupled plasma mass spectrometry (ICP-MS) at Geocenter Copenhagen using a Perkin Elmer Sciex system equipped with a 266 nm Nd–YAG laser ablation facility (Tables 4 and 5). The phases were targeted in situ using slabs from thin-section preparations. The laser was operated at beam sizes of 100 µm (external glass standard NIST SRM 610; Pearce et al., 1997
) and 200 µm (samples). 30Si was used as internal standard for both titanites and alkali feldspars using known concentrations from microprobe analyses. Time-resolved replicate analyses of single crystals were carefully screened for the presence of inclusions (e.g. zircon and apatite) along the line scans. Selected trace and minor elements with standard errors less than 20% for titanite and less than 27% for alkali feldspar are presented. Pb, Th and U concentrations for alkali feldspars are included for purpose of age-correcting the Pb isotope data, although the errors are larger.
|
|
Sr–Nd–Hf–Pb isotope analyses were carried out on 19 representative samples at Geocenter Copenhagen (Table 6). Sr isotopes were measured by thermal ionization mass spectrometry (TIMS), and Nd, Hf and Pb isotopes were measured using the VG Elemental Axiom multi-collector (MC)-ICP-MS system at the Danish Lithosphere Centre. Separates of Sr for isotopic analysis were prepared from whole-rock powders by standard ion-exchange procedures. The 87Sr/86Sr data were measured in one analytical session and the NBS987 standard gave 87Sr/86Sr = 0·710249 ± 26 (2 SD, n = 5). Whole-rock samples for Sm–Nd and Lu–Hf isotope determinations underwent flux fusion digestion and chemical separation procedures described by Bizzarro et al. (2003
) and Ulfbeck et al. (2003). The Nd and Hf data were acquired in two analytical sessions, in which the Ames metal standard gave 143Nd/144Nd = 0·512136 ± 13 (25 ppm, 2 SD, n = 11) and 0·512129 ± 7 (13 ppm, 2 SD, n = 4), and the DLC Hf standard gave 176Hf/177Hf = 0·281847 ± 9 (2 SD, n = 11) and 0·281852 ± 11 (2 SD, n = 6). Sr, Nd and Hf ratios are normalized relative to accepted values for NBS987 (0·71025), Ames metal (0·512125) and DLC Hf (0·281890) standards. Pb isotopes were determined on either alkali feldspar separates (syenites), hand-picked 0·5–1·0 mm rock chips (basalts or dykes) or whole-rock powders (basement) that were leached in hot 6M HCl for 1 h prior to sample digestion and Pb separation. The Pb samples were measured in two analytical sessions, in which the SRM981 standard gave 206Pb/204Pb = 16·9402 ± 17, 207Pb/204Pb = 15·4984 ± 25 and 208Pb/204Pb = 36·7229 ± 64 (2 SD, n = 2) and 206Pb/204Pb = 16·9404 ± 2, 207Pb/204Pb = 15·4978 ± 7 and 208Pb/204Pb = 36·7210 ± 16 (2 SD, n = 2), using Pb double spike to correct for mass bias following the technique of Baker et al. (2004). Initial Sr, Nd, Hf and Pb isotope ratios are calculated back to an age of 50 Ma using the Rb and Sr concentrations measured on glass discs by XRF, Sm, Nd, Lu and Hf concentrations measured by isotope dilution, and Pb, U and Th concentrations measured by ICP-MS and laser ablation (LA)-ICP-MS.
|
|
MINERAL CHEMISTRY |
|---|
Mineral chemical data for pyroxenes and amphiboles are given in Tables 2 and 3. The variation of Mg-number with stratigraphic height in alkali pyroxenes and sodic–calcic to sodic amphiboles is presented in Fig. 4b and c. In the stratigraphically lowest quartz nordmarkites (–7500 m), interstitial aegirine–augite with ferrorichterite rims has Mg-number 29·5, whereas an aegirine–augite without amphibole rim has Mg-number 7·1. Inclusions of pyroxene in alkali feldspar phenocrysts have a much more primitive sodian augite composition with Mg-number 53·4. Interstitial aegirine–augites from the overlying quartz nordmarkites, nordmarkites and pulaskites have magnesium contents below Mg-number 5·3 and typically less than 1·0. Mg-number px increases markedly from the main pulaskites into the foyaites. Aegirine–augite in the foyaites have Mg-numbers of 52·7–59·1 (cores) and 21·9–27.0 (rims). As shown by Brooks & Gill (1982
) and in Table 2, the pyroxene cores in the foyaites and pyroxene inclusions in quartz nordmarkite sample 454067 are clearly less acmitic (En24–27Fs21–25 Wo43–45Ac5–6) than their corresponding rims and interstitial pyroxenes (En0–10Fs37–50Wo6–39Ac12–47). Excluding the stratigraphically lowest sample (454067, –7500 m), all pyroxenes in nordmarkites sensu lato and transitional pulaskite are relatively sodic and have extremely low Mg contents. In Fig. 4c we document for the first time systematic variations in amphibole compositions from the KI. From the stratigraphically lowest quartz nordmarkite to the quartz nordmarkites at –5500 m, the core compositions of interstitial amphiboles change from Mg-number 37·2 to 23·0. Stratigraphically upwards from –5500 m, amphibole cores become more magnesian, increasing to Mg-number 56·9–60·4 in the pulaskites. The rim Mg-number of zoned amphiboles is consistently lower than the cores, but mimics the up-section increase in Mg-number of the cores. Amphibole rims (Mg-number 58·1) on aegirine–augite included in alkali feldspar phenocrysts from the stratigraphically lowest quartz nordmarkites are similar in composition to amphiboles in the pulaskites.
|
WHOLE-ROCK CHEMISTRY |
|---|
Major elements
Concentrations of SiO2, Al2O3, Na2O + K2O and Mg-number from the KI and associated samples from dykes, basaltic xenoliths and local basement gneisses (Table 1) are plotted against stratigraphic height in Fig. 5a–d. SiO2 first increases from
65 wt % in the stratigraphically lowest quartz nordmarkites (–7500 m) to 70 wt % in the quartz nordmarkites at –5500 m, then displays a gradual decrease through the nordmarkites to 63 wt % in the main pulaskites (Fig. 5a), followed by an abrupt drop into the foyaites (56 wt %). The Al2O3 trend is opposite to that of SiO2 (Fig. 5b). The total alkalis increase gradually from the quartz nordmarkites (11 wt %) through the pulaskites (13 wt %), followed by an abrupt increase into the foyaites (16 wt %). Whole-rock Mg-number (Fig. 5d) decreases from 25 in quartz nordmarkites at –7500 m to a low of 14 at –5500 m, then increases through the nordmarkites and pulaskites to 30. Whole-rock Mg-number varies from 5 to 20 in the foyaites.
|
Trace elements
Trace element data for 40 whole-rock samples are presented in Table 1 and Fig. 5e–h. The data generally display large variations for samples from the same inferred stratigraphic height as a result of variable accumulation of accessory phases such as titanite, zircon and chevkinite, but some systematic trends can be recognized. V increases from a fairly constant level of 5–12 ppm in the nordmarkites sensu lato to
40 ppm in the foyaites (Fig. 5e). Zr increases from 400–700 ppm in the quartz nordmarkites to 1200–1650 ppm in the nordmarkites at –4000 m, followed by a decrease through the uppermost nordmarkites to 200–400 ppm in the foyaites (Fig. 5f). Rb also shows a high of 150–170 ppm in the middle part of the silica-oversaturated stratigraphy (–4000 m to –5500 m) and lows of 50–100 ppm below and above. Sr increases from a fairly constant low level of 200 ppm in the quartz nordmarkites through pulaskites to a high of 800 ppm in the foyaites (Fig. 5h). |
IN SITU TITANITE AND FELDSPAR MINOR AND TRACE ELEMENTS |
|---|
Titanite
Minor and trace element data from seven samples are given in Table 4. Variations in titanite rare earth element (REE)concentrations are presented in a chondrite-normalized REE diagram (Fig. 6). Titanite can accommodate large quantities of REE and high field strength elements (HFSE) and all the samples are extremely enriched in these elements. REE abundances in titanite increase upwards from the quartz nordmarkites into the nordmarkites followed by a decrease through the uppermost foyaites. Titanites from all samples are enriched in light REE (LREE) compared with heavy REE (HREE). The foyaites show the steepest trends as they have relatively lower HREE. The quartz nordmarkites to main pulaskites have convex-upwards REE patterns, whereas the uppermost foyaites display convex-downwards patterns. The two nordmarkite samples and one foyaite display negative Eu anomalies. The Ta content of titanite increases from
350 ppm at –7500 m in the quartz nordmarkites to 500 ppm in the upper nordmarkites, followed by a decrease through the foyaites to 140–220 ppm (Table 4).
|
Alkali feldspar
Minor and trace element data from 12 feldspar samples are given in Table 5. Fe in matrix alkali feldspars displays a steady increase from
3600 ppm in the quartz nordmarkites to 6500 ppm in the nordmarkites, followed by a sharp decrease through the pulaskites to 1900 ppm in the foyaites. One quartz nordmarkite outlier has more than 10 000 ppm Fe but also a large standard deviation. The trend of alkali feldspar Fe enrichment or depletion is roughly similar to that of whole-rock SiO2, Zr and Ce (REE) variation (Fig. 5a, f and g). The Sr concentrations of matrix alkali feldspars principally show the same variation as the whole-rock contents, increasing from a constant low level of 50 ppm in the quartz nordmarkites through pulaskites to a high of 400–800 ppm in the foyaites (Table 5). Alkali feldspar phenocrysts from a quartz nordmarkite (454067) show much higher Sr concentrations (600 ppm) than the coexisting matrix feldspar (50 ppm), and the phenocryst content is similar to that of feldspar in the foyaites. Higher in the nordmarkite stratigraphy the phenocrysts are not significantly different from the matrix. Rb in alkali feldspar varies in a similar way to that in whole-rock (Tables 3 and 5). |
Sr–Nd–Hf–Pb ISOTOPES |
|---|
Sr, Nd, Hf and Pb isotope data on 19 whole-rock and alkali feldspar samples are presented in Table 6. 87Sr/86Sri (50 Ma) decreases markedly from
0·706 in the lowermost quartz nordmarkites to 0·705 at –6800 m, followed by a gradual decrease through the stratigraphy to 0·7043–0·7044 in the foyaites (Fig. 7a). Ndi and Hfi both increase through the stratigraphy from the lowermost quartz nordmarkites (Ndi = 2·3, Hfi = 5·2), over a constant level in the nordmarkites (Ndi = 3·1–3·5, Hfi = 6·5–6·6), to the foyaites (Ndi = 3·8–4·9, Hfi = 10·7–11·1) (Fig. 7b and c). The Archaean basement samples have very radiogenic 87Sr/86Sr50 Ma compositions (0·715–0·735) and very unradiogenic Nd, 50 Ma (–18 to –33) and Hf, 50 Ma (–37 to –56) compositions. The mafic dyke and basaltic xenoliths have 87Sr/86Sri (50 Ma) = 0·7045–0·7055, Ndi = 0·4–1·5 and Hfi = –4·3 to 6·2.
|
The alkali feldspar 207Pb/204Pbmeas–206Pb/204Pbmeas variation is shown in Fig. 8. Alkali feldspars have more radiogenic Pb isotopic compositions (206Pb/204Pbmeas = 16·98–17·88; 207Pb/204Pbmeas = 15·16–15·37) compared with the most unradiogenic (contaminated) East Greenland Plateau Basalts. The stratigraphic variation from the relatively unradiogenic alkali feldspars in the quartz nordmarkites to more radiogenic alkali feldspars in the foyaites is consistent with the Sr, Nd and Hf isotopic variation towards a weaker crustal signature stratigraphically upwards within the intrusion. Analyses of both phenocryst and matrix alkali feldspar in two nordmarkite samples (454096 and 454097) consistently show that the matrix phase is more radiogenic than the phenocryst phase, beyond the effect of age correction. The dyke and basaltic xenoliths are far less radiogenic than the foyaites (206Pb/204Pbmeas = 15·89–17·01; 207Pb/204Pbmeas = 15·01–15·26) (Fig. 8 inset).
|
|
DISCUSSION |
|---|
Parental magma
Mafic alkaline magmatism in central East Greenland (Nielsen, 1987
) includes: (1) inland alkaline basalts of the Prinsen af Wales Bjerge Formation (55–52 Ma, Peate et al., 2003); (2) the Gardiner complex (56–54 Ma, Waight et al., 2002; Tegner et al., 2008), hosting ultramafic cumulates and a ring-dyke system of evolved alkaline rocks (melilitolites, agpaitic syenites and carbonatites) (Nielsen, 1980, 1981); (3) two suites of mildly alkaline coastal dykes (55–50 Ma and 50–33 Ma, Gleadow & Brooks, 1979) of saturated to undersaturated composition (predominantly alkali basalt and hawaiite) (Nielsen, 1978); (4) the gabbroic Lilloise intrusion of alkali basaltic or picritic parentage (50 Ma, Brown, 1973; Chambers & Brown, 1995), basanite dykes (37–36 Ma) and nephelinite diatremes (50 Ma) from the Wiedemann Fjord–Kronborg Gletscher lineament (Nielsen et al. 2001; Tegner et al., 2008); (5) a mildly alkaline 47 Ma diorite intrusion and trachyandesite pillowed intrusive rocks in the Astrophyllite Bay Complex (2·5% ne to 1·1% qz normative) (Riishuus et al., 2005).
Kempe & Deer (1976) suggested that the KI parental magma was a mantle-derived alkali olivine basalt, whereas Brooks & Gill (1982) favoured fractionation of a nephelinitic parent to produce the phonolitic melt from which the foyaites crystallized. Unlike the inland Gardiner complex, there is no evidence from the Kangerlussuaq Alkaline Complex for the presence of extremely alkaline and primitive mafic rocks that might indicate differentiation directly from nephelinitic magma to phonolite. Conversely, the presence of alkali olivine basaltic to basanitic dykes and intermediate diorites and trachyandesites within the Kangerlussuaq Alkaline Complex (Riishuus et al., 2005) suggests differentiation from an alkali olivine basalt or basanite parent magma. The composition of aegirine–augite cores in the foyaites (En24–27Fs21–25Wo43–45Ac5–6, Table 2) are similar to those in oceanic phonolites from Rarotonga (En23–25Fs23–24Wo42–44Ac9–10) inferred to be formed by fractionation of alkali basaltic to basanitic magmas (Thompson et al., 2001). The highly peralkaline rocks of the Ilímaussaq Complex, South Greenland, are taken to represent residues from extreme fractional crystallization in an alkali basaltic parent magma chamber situated deep in the crust (e.g. Larsen & Sørensen, 1987; Stevenson et al., 1997; Markl et al., 2001; Marks & Markl, 2001), giving rise to distinct negative Eu anomalies in the agpaitic rocks (Bailey et al., 1978). Kramm & Kogarko (1994) suggested that the lack of significant Eu anomalies in the rocks of the Khibina and Lovozero alkaline centres of the Kola Peninsula indicate that they formed as residues from nephelinite magmas at mantle pressures. The titanites from the KI display significant negative Eu anomalies in the nordmarkites and either absent or negative Eu anomalies in the pulaskites and foyaites (Fig. 6). The negative Eu anomalies of the nordmarkites indicate a ‘plagioclase-effect’, as do the generally low Sr contents of the KI (11–852 ppm, Table 1 and Fig. 5h) relative to nepheline syenites from the Khibina centre (255–2800 ppm; Kramm & Kogarko, 1994). The low contents of Cr (<5 ppm) and Ni (<7 ppm) in all rock types of the KI suggest olivine and clinopyroxene fractionation (Table 1). Ultimately, a nephelinitic parent appears less likely than a basanitic or alkali basaltic parent magma for the KI.
Structure and volume of plumbing system
The occurrence of kaersutite gabbro inclusions in dykes cross-cutting the KI led Brooks & Platt (1975) to suggest that an alkaline mafic intrusion underlies the Kangerlussuaq Alkaline Complex. This is supported by the total magnetic intensity field in the Kangerlussuaq area (Fig. 9) based on aeromagnetic data from Verhoef et al. (1996). The location of the complex coincides with a large positive magnetic anomaly with a maximum intensity of 600 nT and a width of 60 km. Normally, the depth of a magnetic anomaly roughly corresponds to half its width. As the East Greenland crust is 30 km thick (Korenaga et al., 2000; Holbrook et al., 2001; Dahl-Jensen et al., 2003), this magnetic anomaly could represent one or several large mafic bodies situated at various crustal levels below the KI. The magnetic data indicate that the KI is only the surface expression of a much larger magmatic complex at depth in the crust. If fractionation models suggesting that phonolites represent 13–40% residuals of the parent basanite (Kyle, 1981; le Roex et al., 1990; Ablay et al., 1998; Thompson et al., 2001) are considered, then a large volume (10 000–30 000 km3) of solidified parent magma is required in the crust below to produce the estimated volume (4000 km3) of evolved alkaline rocks in the KI (Fig. 10a).
|
|
The basaltic xenoliths from the chamber roof
The lavas of the Lower Basalts range from picrite to basaltic andesite with minor occurences of ne-normative lavas, olivine basalt as the most abundant rock type and they vary texturally from aphyric to olivine- and pyroxene-phyric (Nielsen et al., 1981
; Fram & Lesher, 1997; Hansen & Nielsen, 1999). The Plateau Basalts are dominantly olivine- and plagioclase-phyric tholeiitic basalts (qtz- or ol-normative) and there are no picrites or basaltic andesites (Larsen & Watt, 1985; Larsen et al., 1989). Comparison with the petrology of the established East Greenland lava sequence suggests that the silica-saturated to -undersaturated, aphyric to pyroxene-phyric basaltic xenoliths found in the KI belong to the Lower Basalts. Field observations of the rounded shapes of the basaltic xenoliths and the presence of reaction zones and small (<10 cm) diffuse mafic bodies in the syenite matrix, but sharp contacts between syenite veins and the interior of the xenoliths indicate minor digestion of the basaltic xenoliths by the host trachyte magma. Although basaltic xenoliths in gabbros usually show evidence of partial melting, such as LREE depletion and the presence of trapped pegmatitic pods (Brandriss et al., 1996), such evidence is not found in this study; this feature argues against significant anatexis of the basaltic xenoliths by trachyte magma.
Magma chamber model
The field relations, mineralogy and geochemical data presented provide evidence of a series of complex magma chamber processes that require revision of previous models (Wager, 1965; Kempe & Deer, 1976; Pankhurst et al., 1976; Brooks & Gill, 1982) and have wider implications for the formation of coexisting silica-oversaturated and -undersaturated syenites.
Following the field observations by Wager (1965) and later workers of inward-sloping basaltic xenoliths and platy feldspars, and nepheline-bearing pegmatites cutting the nordmarkites and not vice versa, we consider that the KI is a layered intrusion that solidified from the bottom upwards. Although the main heat loss occurred at the top and the sides, cumulus alkali feldspar must have formed or accumulated at the bottom of the chamber, which implies convective circulation of the magma. Had convection not taken place the magma would have solidified from the top downwards; however, the prominent trains of basaltic xenoliths originating from collapses of the roof, and present from the quartz nordmarkites through to the pulaskites, rule this out. One could envisage that between the repeated roof collapses, periods of recurrent top-downwards crystallization might have taken place, but unfortunately the roof zone is not preserved. Layered syenitic intrusions are not uncommon. Magmatic convection coupled with crystal settling has been proposed for the Lovozero alkaline complex (e.g. Kogarko & Khapaev, 1987; Kogarko et al., 2002, 2006), and there is evidence that cumulates formed on both the roof and floor of the Ilímaussaq alkaline intrusion (e.g. Larsen & Sørensen, 1987; Sørensen & Larsen, 1987; Bailey et al., 2006) and the Klokken intrusion (e.g. Parsons, 1979; Parsons & Becker, 1987).
The radiogenic isotopic compositions of the KI clearly testify to an increasing crustal component from the foyaites through the quartz nordmarkites (Figs 7 and 8) and require involvement of a crustal contamination process in the petrogenesis of the magmas. In a study of the temporal evolution of the Kangerlussuaq Alkaline Complex, Riishuus et al. (2006) suggested that the evolved alkaline rocks formed as products of AFC (DePaolo, 1981) between a magmatic end-member intermediate to the Plateau Basalts (Peate & Stecher, 2003; Andreasen et al., 2004) and the alkalic Prinsen af Wales Bjerge Formation basalts (Peate et al., 2003) (87Sr/86Sri = 0·7035, Sr = 300 ppm; Ndi = 6·5, Nd = 30 ppm; Hfi = 13·5, Hf = 5 ppm) and varying proportions of two different local contaminants, granulite-facies gneiss (87Sr/86Sr50 Ma = 0·706, Sr = 500 ppm; Nd, 50 Ma = –38, Nd = 20 ppm; Hf, 50 Ma = –65, Hf = 2·5 ppm) and amphibolite-facies gneiss (87Sr/86Sr50 Ma = 0·750, Sr = 260 ppm; Nd, 50 Ma = –36, Nd = 5 ppm; Hf, 50 Ma = –65, Hf = 2·5 ppm) (Taylor et al., 1992; Riishuus et al., 2005; this study). Using these end-members and an r value (rate of assimilation/rate of crystallization) of 0·3, the observed large variations in the Sr–Nd–Hf isotope compositions in the KI can be produced when 15–45% of the initial liquid has solidified (Riishuus et al., 2006). The consistent relationship of more radiogenic Pb isotope compositions in matrix alkali feldspars compared with coexisting alkali feldspar phenocrysts (Fig. 8) suggests involvement of magma recharge. We propose a model for the petrological and geochemical evolution of the magma chamber that includes fractional crystallization, magma recharge and crustal assimilation. The model is discussed below and illustrated in Fig. 10.
The outer part of the KI, hosting the quartz nordmarkites, represents the earliest known cumulates formed after emplacement of magma into the chamber at the contact between the Archaean basement gneisses and the overlying Palaeogene flood basalts. The lower quartz nordmarkites (e.g. 454067, –7500 m) clearly host two populations of pyroxene and amphibole as inclusions in alkali feldspar phenocrysts and in interstitial pockets, and two alkali feldspar populations as large zoned phenocrysts and in the groundmass (Fig. 3a and b). The close resemblance between alkali feldspar-hosted aegirine–augite and amphibole inclusions in the quartz nordmarkites (454067) and aegirine–augites in the foyaites and amphiboles in the pulaskites could suggest crystallization from similar melt compositions (Figs 4b and c, Tables 1 and 2). This is in strong contrast with the more evolved groundmass pyroxene (Fig. 4b), amphibole (Fig. 4c) and alkali feldspar matrix compositions (Table 5), higher whole-rock SiO2 (Fig. 5a), REE and Ta in titanite (Fig. 6 and Table 4), Fe in alkali feldspar (Table 5), and lower V contents (Fig. 5e) of the same quartz nordmarkite sample when compared with the pulaskites and foyaites. Furthermore, the Sr–Nd–Hf–Pb isotope data clearly show that the quartz nordmarkite (454067) has a much larger crustal component than the foyaites (Figs 7 and 8) and indicate that crustal contamination played an important role. Contaminated layered gabbroic intrusions with AFC trends (Fig. 11) often contain abundant xenoliths of metabasalts, metasediments and gneisses (Sørensen & Wilson, 1995; Nielsen et al., 1996; Tegner et al., 1999), whereas syenites with xenoliths of local wall-rock show evidence of in situ assimilation during final emplacement that led to significant but only local effects (Marks & Markl, 2001). The thermal budget between trachyte magma and its country rock is less favourable for assimilation to take place than when basaltic magma is considered. Hence, the lack of gneissic xenoliths in the quartz nordmarkites indicates that little or no in situ assimilation of gneiss took place at the final emplacement level. As gneiss is the only available crustal contaminant that is able to produce the radiogenic isotope composition of the quartz nordmarkites, we argue that most contamination must have taken place at a deeper crustal level. The AFC modelling by Riishuus et al. (2006) showed that the KI is dominated by contamination with lower crustal granulite-facies gneisses. We suggest that silica-oversaturated magma formed as a consequence of contamination of an initially silica-undersaturated melt, not unlike the melt from which the foyaites crystallized, by silica-rich gneisses in the roof zone of a large, stratified, deep-seated magma chamber below the present KI (Fig. 10a). By tapping from the roof zone and downwards into this chamber, the KI developed as an initially silica-oversaturated, lopolithic, sill-like body emplaced in the uppermost crust (Fig. 10a and b). The outermost quartz nordmarkites crystallized as the earliest cumulates in this lopolith, after any hidden cumulates (Fig. 10b and c).
|
The up-section increase in whole-rock SiO2 content (Fig. 5a) and decrease in Al2O3 content (Fig. 5b), whole-rock Mg-number (Fig. 5d) and amphibole Mg-number (Fig. 4c) from –7500 m to –5500 m suggest initial fractionation of the residual magma towards the silica-oversaturated minimum. Despite crystallization of accessory titanite and zircon, acting as REE and HFSE sinks, the concentrations of incompatible trace elements increase upwards into the nordmarkites (Figs 5f, g and 6) and could indicate that the magma became more evolved. In contrast, decreasing 87Sr/86Sri (Fig. 7a) and increasing
Ndi, Hfi, 206Pb/204Pbmeas and 207Pb/204Pbmeas (Figs 7b, c and 8) show that, during the same interval, the magma also became less contaminated. This implies mixing between the resident magma and a less contaminated melt, which we suggest was derived from a lower level of the underlying chamber (Fig. 10a). The variation of Mg-number in amphibole cores with whole-rock Sr isotope composition up through the stratigraphy is opposite to the trend developed in layered intrusions that have undergone AFC (Fig. 11). If the stratigraphically lowest quartz nordmarkite crystallized directly from an evolved melt tapped from an underlying, stratified magma chamber it should plot on the AFC trend of this sub-KI chamber. Periodic recharge (with less evolved and less contaminated melt from the sub-KI chamber) and fractional crystallization (RFC) followed by FC alone could explain the stratigraphic variations towards the lowest Mg-number of amphibole cores. Injection of fresh magma would result in expansion and/or tapping of the KI magma chamber. The numerous basaltic xenoliths (Fig. 2a) in the nordmarkites sensu lato suggest expansion of the chamber by stoping into the lava sequence forming the chamber roof (Wager, 1965). We have no evidence for eruptions of trachytic to phonolitic material, but intuitively it is hard to envisage that such a large, shallow chamber was not periodically tapped.
The stratigraphic sequence from the base of the nordmarkites through the main pulaskites (–1000 m) is characterized by upwards decreasing whole-rock SiO2, Zr and Ce (Fig. 5a, f and g), titanite REE and Ta (Fig. 6 and Table 4) and alkali feldspar Fe (Table 5) contents, and increasing whole-rock Al2O3, V contents (Fig. 5b and e), whole-rock Mg-number (Fig. 5d) and amphibole Mg-number (Fig. 4c). We believe that this indicates crystallization from an increasingly primitive and silica-poor magma. Upsection decreasing 87Sr/86Sri and increasing Ndi, Hfi (Fig. 7) and Pb isotope compositions (Fig. 8) imply periodic recharge by less contaminated, silica-undersaturated melt (Fig. 11). The Pb isotope compositions of coexisting phenocryst and matrix alkali feldspars show that the matrix phase is more radiogenic than the phenocryst phase (Fig. 8), beyond the effect of age-correction, thereby supporting a periodic recharge–mixing model. The common porphyritic texture of the nordmarkites sensu lato may, in fact, be a consequence of multiple magma influxes and intervening stages of cooling, during which crystallization of feldspar was repeatedly halted and renewed, leading to at least two feldspar populations within the nordmarkites. From the level where the basaltic xenoliths cease to be present (lower part of the main pulaskites) we believe that the magma chamber, now dramatically reduced in size, was effectively shielded from its country rock envelope as a result of crystallization downwards from the roof (Fig. 10d).
The transition from the main pulaskites into the foyaites in the uppermost part of the succession shows a marked decrease in SiO2 (63–64 to 55–58 wt %) and marked increases in Al2O3 (18 to 21–23 wt %), total alkalis (13 to 16 wt %) and pyroxene Mg-number (1–2 to 18–59). As no intrusive contacts have been found, the marked chemical changes could be explained by a major magma recharge relative to the resident melt. As the lowest 87Sr/86Sri values and highest Ndi, Hfi, 206Pb/204Pbmeas and 207Pb/204Pbmeas values are found in the foyaites, the last part of the KI to crystallize formed from the most primitive, least crustally contaminated melt. The final recharge magma tapped from the underlying chamber must have been essentially uncontaminated mafic phonolite.
Magma mixing model
To evaluate how the sample suite conforms to the proposed magma mixing model, we have constructed simple mixing curves for the variation in Sr, Hf and Pb isotopes between a crustally contaminated quartz trachytic melt in equilibrium with the first formed quartz nordmarkites and a silica-undersaturated melt in equilibrium with the foyaites (Figs 8 and 12). The exact composition of the recharge magma is unknown, as the foyaites are likely to have crystallized from mixed magmas themselves. Nevertheless, based on their isotopic composition, the foyaites represent the most primitive and uncontaminated syenites of the intrusion. The Sr, Hf and Pb concentrations of the KI magma at different stratigraphic levels can be estimated using titanite and alkali feldspar trace element concentrations (Tables 4 and 5) and appropriate mineral/melt partition coefficients (Villemant, 1988; Lynton et al., 1993). The end-member melt compositions are based on the most contaminated and stratigraphically lowest quartz nordmarkite (454067) (87Sr/86Sri = 0·7061, Sr = 8·5 ppm; Hfi = +5·2, Hf = 3·1 ppm; 206Pb/204Pb = 16·98, 207Pb/204Pb = 15·16, Pb = 1 ppm), and the least contaminated and stratigraphically highest foyaite (454089) with a slight sway from the other foyaite samples (87Sr/86Sri = 0·7044, Sr = 185 ppm, Hfi=+11, Hf = 9·8 ppm; 206Pb/204Pb = 17·88, 207Pb/204Pb = 15·37, Pb = 3 ppm). The KI samples fall very close to the calculated mixing lines for both Pb (Fig. 8) and Sr–Hf (Fig. 12), supporting the view that the silica-oversaturated and -undersaturated KI rocks are related through periodic magma recharge. The isotopic composition of single samples is produced by similar mixing ratios in Sr–Hf and Pb isotope space, which adds further support for the mixing model and validity of the end-member parameters.
|
Phase equilibrium evolution—the thermal barrier
Any petrogenetic model attempting to explain the evolutionary trend of the KI should be consistent with phase equilibrium constraints. The possibilities for changing the magma from a silica-oversaturated to a silica-undersaturated composition by isothermal magma mixing can be evaluated using isothermal sections in Petrogeny's Residua System at 1 kbar P(H2O) (Fig. 13). Alkali feldspar compositions F1 (Or38), F2 (Or36), F3' (Or41) and F4' (Or55) in Fig. 13a and b correspond to groundmass alkali feldspar compositions based on
01 spacing X-ray analyses of nordmarkites, transitional pulaskites, main pulaskites and foyaites, respectively, from Kempe & Deer (1970). Let us consider a hypothetical initial bulk composition X (Q6Ab60Or34) consisting of alkali feldspar composition Or38 (F1) and liquid L1 (Fig. 13b), and two hypothetical silica-undersaturated recharge magma compositions, C and C', that can give rise to the observed alkali feldspar compositions when mixed with the initial bulk X. At 850°C (Fig. 13a–c), the L1 temperature is below the alkali feldspar saddle minimum [865°C, 1 kbar P(H2O)] with solid solution crossing the Ab–Or syenite divide. By recharge with C, the alkali feldspar composition will change from F1 to F2, and the initial liquid L1 will change composition and eventually disappear, leaving F2 as the only phase present. If more C could be mixed with the completely solid F2, a silica-undersaturated melt L2 would be produced at first, followed by a superheated liquid. By recharge with C', L1 will be consumed, leaving alkali feldspar (with a slightly more Or-rich composition than F1) as the only phase present. Again, if more C' could be mixed with the alkali feldspar phase, a silica-undersaturated melt L3' would be developed and coexist with alkali feldspar F3 at bulk composition B3'. Increasingly Or-rich alkali feldspar will develop with further addition of C', and when the bulk composition is identical to L4', alkali feldspar F4' disappears.
|
If the temperature is raised only 20°C to 870°C (Fig. 13d), slightly above the alkali feldspar minimum (865°C), the silica-oversaturated and -undersaturated liquid fields are joined. Recharge of the initial bulk composition with C or C' at this temperature would let the bulk composition cross the Ab–Or join, without intersecting the solidus. Once the bulk composition crosses the Ab–Or join it will eventually terminate at the nepheline syenite minimum (Fig. 13e). At these conditions the recharge magma would be superheated and probably at a higher temperature than the resident magma to raise the bulk temperature above the thermal barrier. The continuous normative compositional transition across the Ab–Or join from nordmarkite to pulaskite suggests that the bulk magma was either kept at the liquidus or elevated above the liquidus in small increments before returning to it and precipitating alkali feldspar again. Overcoming the thermal barrier may be aided by depression of the liquidus surface by elevation of the vapour pressure (e.g. Tuttle & Bowen, 1958
; Morse, 1969), as discussed in relation to the KI by Kempe & Deer (1976) and Pankhurst et al. (1976). This is supported by an oxygen and hydrogen isotope study of the KI indicating release of water with a meteoric origin into the syenite magma by dehydration of hydrothermally altered basaltic xenoliths (Riishuus et al., in preparation). The decreasing amphibole (Fig. 4c) and whole-rock (Fig. 5d) Mg-numbers and increasing whole-rock SiO2 content (Fig. 5a) up through the quartz nordmarkites indicate that the KI magma initially fractionated towards the granite minimum, despite being mixed with less crustally contaminated material as suggested by the Sr–Nd–Hf–Pb isotopic variation (Figs 7 and 8). This implies an initial relatively low recharge/resident magma ratio, followed by an increasing ratio after formation of the quartz nordmarkites, forcing the bulk magma back towards the Ab–Or join.
The exact liquid evolution of the KI is difficult to establish, as the compositions and temperatures of the initial residual and recharge magmas are unknown. Recharge with C will satisfy the alkali feldspar compositional change from the nordmarkites (F1 = Or38) to the transitional pulaskites (F2 = Or36), but will not be able to produce a liquid composition in equilibrium with F3 (Or41) of the main pulaskites (Fig. 13b). On the other hand, recharge with C' cannot produce the F2 in equilibrium with the transitional pulaskites, but can produce a bulk composition close to the normative composition of the main pulaskites and a liquid in equilibrium with F3' (Fig. 13b and c). This relationship suggests that the recharge magma changed composition between formation of the transitional and main pulaskites. The large compositional jump between the main pulaskites and the foyaites must result from a further increase in the recharge/residual magma ratio (Fig. 13c). The location of the normative compositions of the foyaites in extension of the F4'–L4' tie line shows that the foyaites from –500 m and 0 m can be related through fractionation of the remaining melt that crystallized alkali feldspar with Or55 (Fig. 13b and c).
Comparison with other alkaline complexes
Several major syenite complexes resemble the KI in showing temporal evolution from early (outer) quartz syenites or monzonites to later (inner) nepheline syenites.
The 178 Ma Marangudzi ring complex (80 km2), SE Zimbabwe, formed during the waning stage of Karoo flood basalt volcanism and consists, in order of intrusion, of a gabbro body, quartz syenite ring dykes and nepheline monzonite or syenite cone sheets (e.g. Foland & Henderson, 1976; Foland et al., 1993). The quartz syenites have high 87Sr/86Sr and low 143Nd/144Nd values, whereas the nepheline syenites have low 87Sr/86Sr and high 143Nd/144Nd values (Foland et al., 1993). The entire suite seems to converge at a common low 87Sr/86Sr and high 143Nd/144Nd composition; this feature led Foland et al. (1993) to suggest an origin from a common silica-undersaturated parental magma. Their model involves development of the quartz syenites from felsic, silica-undersaturated melts undergoing assimilation of felsic crust and fractional crystallization, whereas the nepheline syenites formed by closed-system fractional crystallization alone or with only minor contamination.
The alkaline core of the 131–127 Ma Messum complex in NW Namibia, part of the Paraná–Etendeka flood basalt province, consists of an outer quartz syenite suite and an inner silica-undersaturated suite dominated by nepheline syenite (e.g. Harris et al., 1999). The core is surrounded by older gabbros. The Messum felsic alkaline core is only 6 km in diameter and therefore much smaller than the KI. The quartz syenites (cut by undersaturated dykes) are presumably older than the undersaturated syenites but the contact between them is not exposed. Conversely, the centre of the complex suggests a gradation from nepheline syenite through nepheline-poor syenite to syenite and quartz syenite. Ultimately, it remains unclear whether the alkaline core formed from one or several bodies. Sr and O isotope data indicate a decreasing effect of crustal contamination with time, favouring progressive armouring of the magmatic plumbing system by less contaminated material (Harris et al., 1999).
The very large larvikite–lardalite complex (monzonite–nepheline syenite) (1000 km2) of the Permian rift-system in the Oslo district, Norway, displays at least eight ring structures, identified on the basis of topographic features, orientation fabrics (igneous lamination or modal layering), contact relations and aeromagnetic properties. They cut each other in a pattern suggesting a general shift of magmatic focus from east to west (Petersen, 1978). Compositional variations in the larvikite complex from silica-oversaturated, through larvikites with neither quartz nor nepheline, to silica-undersaturated larvikites follow the pattern of structural younging towards the west. The lardalites constitute the core of the complex and display conspicuous contact zones with dendritic growth of alkali feldspar, nepheline and pyroxene along internal contacts and against the earlier larvikites, supporting an origin of the entire complex by multiple injections (Petersen, 1978, 1985).
The 1130 Ma Ilímaussaq Complex (17 km x 8 km) consists of an early silica-undersaturated augite syenite intruded by an alkali granite sheet and followed by a later series dominated by layered, agpaitic nepheline syenites (Ferguson, 1964). The magmas leading to the intrusive events of Ilímaussaq have been related to fractional crystallization and crustal assimilation of a stratified alkali basaltic magma deeper in the crust, before ascending to their present level, where further assimilation and extreme fractionation took place (e.g. Larsen & Sørensen, 1987; Markl et al., 2001; Marks & Markl, 2001). Based on whole-rock Nd isotope data Stevenson et al. (1997) suggested that, during the evolution of the complex, the degree of crustal assimilation changed from 12% in the augite syenite, over 35–40% in the alkali granite, to 17% in the agpaitic rocks with respect to the initial mass of magma. Mineral Nd and O isotope compositions led Marks et al. (2004) to favour a model in which both the augite syenite and the agpaitic rocks of the Ilimaussaq complex evolved in a closed system without significant contamination, whereas the alkali granite developed from parts of the augite syenite magma that became contaminated with lower crust (13% bulk assimilation) during ascent.
The Marangudzi and Ilímaussaq complexes are clearly made up of discrete intrusive bodies. The Oslo Rift larvikite–lardalite complex also formed as a series of separate intrusions, whereas contact relations between the syenites of the Messum complex are unclear. Wager (1965) did not find any internal intrusive contacts in the KI and reported a gradual transition from the quartz nordmarkites to the foyaites. In comparison with the models proposed for the Marangudzi ring complex and the Messum alkaline core, where the suites of coexisting quartz and nepheline syenites formed as distinct intrusions from initially silica-undersaturated melts that underwent decreasing amounts of contamination, our favoured model for the KI involves replenishment of an initially silica-oversaturated, crustally contaminated and continuously fractionating magma by multiple injections of salic, silica-undersaturated melt from the roof zone of an underlying, evolving, stratified magma chamber in the lower crust. We suggest that the earliest known silica-oversaturated KI magma, from which the quartz nordmarkites crystallized, formed by a process similar to that suggested by Foland et al. (1993) and Harris et al. (1999). Our petrogenetic emplacement model has similarities to the favoured model for the Ilímaussaq Complex (e.g. Larsen & Sørensen, 1987), but the much larger KI formed from a single, long-lasting intrusive event with (1) repeated magmatic recharge leading to progressively less contaminated syenites, (2) nepheline syenites being subordinate in volume to quartz syenites, and (3) absence of differentiation to the extreme agpaitic compositions of Ilímaussaq. Like the KI (Fig. 9), the Messum (Bauer et al., 2003), Larvik (Ramberg & Smithson, 1971) and Ilímaussaq (Blundell, 1978; Forsberg & Rasmussen, 1978) complexes are associated with geophysical anomalies suggesting large, dense structures at depth. It therefore seems to be a general feature that many syenitic intrusions represent the upper crustal expressions of larger, dominantly mafic, underlying igneous complexes.
The temporal evolution in silica content and Ndi of a series of alkaline complexes with cogenetic silica-oversaturated and -undersaturated syenites, including the whole Kangerlussuaq Alkaline Complex (e.g. Riishuus et al., 2006), is summarized in Fig. 14. The complexes all demonstrate a clear negative correlation of Ndi with the degree of silica saturation, and it is evident that they developed from early, crustally contaminated quartz syenites to late, less contaminated nepheline syenites. This is generally interpreted as assimilation of fusible crustal material during establishment of the plumbing system. It appears to be a general rule that following tapping of contaminated, trachytic to rhyolitic melts to higher crustal levels or eruption, an increasingly shielded plumbing system will then be able to develop uncontaminated, phonolitic melts. Alkaline complexes such as those included in Fig. 14 are typically characterized by marginal quartz syenites and central nepheline syenites, and the KI is a classic example of such a structure (Fig. 1b). The Kangerlussuaq Alkaline Complex as a whole, however, differs from this pattern, as the youngest satellite intrusions are silica-oversaturated (Fig. 14). These intrusions are located SE of the Kangerlussuaq Intrusion (Fig. 1b) and therefore indicate that relocation of the plumbing system took place with the possibility for renewed contamination (Fig. 14), and that prolonged magmatic activity at a fixed location appears to be a requirement for phonolitic magma to reach the upper continental crust (Riishuus et al., 2006).
|
To the authors’ knowledge no examples of layered intrusions with a complete compositional span from gabbro to nepheline syenite have been documented. However, indirect evidence for such compositional variation is found in the Proterozoic Gardar Province in southern Greenland, where composite giant dykes with mafic margins and no internal chilled contacts to granitic, quartz syenitic or nepheline syenitic centres testify to the presence of contemporaneous magmas of contrasting compositions prior to emplacement in the upper crust, implying the presence of an underlying, compositionally zoned magma chamber (e.g. Upton et al., 1985
, 2003; Upton, 1987; Halama et al., 2004). Furthermore, successions of increasingly primitive intrusions for both undersaturated (South Qôroq and Motzfeldt) and oversaturated (Kûngnât) complexes (Stephenson & Upton, 1982) are interpreted as representing the periodic tapping of deeper levels of compositionally stratified magma chambers. The tephra succession of the Quaternary Laacher See volcano in Germany, part of the alkaline province of Central Europe, extends from differentiated phonolite to later mafic phonolites, implying an eruptive mechanism whereby magma was tapped from successively deeper parts of a compositionally zoned, phonolitic magma chamber (e.g. Bogaard & Schmincke, 1984; Wörner & Schmincke, 1984; Harms & Schmincke, 2000; Harms et al., 2004). The last eruptive products were heterogeneous hybrids of phonolite and basanite, the latter having been injected into the base of the magma chamber (Wörner & Wright, 1984). These occurrences of both plutonic and volcanic rocks formed by the tapping of progressively deeper parts of compositionally stratified, alkaline magma chambers add further validation for similar processes to have operated in the petrogenesis of the Kangerlussuaq Intrusion. |
CONCLUSIONS |
|---|
(1) The Kangerlussuaq Intrusion is a large, felsic, alkaline layered intrusion that becomes progressively less contaminated and more primitive from the lower quartz nordmarkites to the upper foyaites. Aeromagnetic data indicate that the Kangerlussuaq Intrusion is only the surface expression of a very large, underlying alkaline mafic–felsic complex. We favour a model in which assimilation of Archaean crust in the roof zone of a large, stratified, alkaline olivine basaltic to basanitic magma chamber, emplaced in the crust below the Kangerlussuaq Intrusion, led to the development of a silica-oversaturated nordmarkitic liquid. The heat required to assimilate the gneisses was provided by the latent heat of crystallization of underlying mafic cumulates. The first melts to be emplaced in the Kangerlussuaq Intrusion magma chamber, from which the quartz nordmarkites and any hidden silica-oversaturated cumulates formed, were tapped from the uppermost roof zone of the underlying chamber. The undersaturated, primitive, basic magma of this chamber, unable to assimilate gneiss because of armouring, fractionated towards a phonolitic composition.
(2) The driving force producing the transition from silica-oversaturated to -undersaturated syenites in the Kangerlussuaq Intrusion chamber was the periodic recharge of uncontaminated, or only slightly contaminated, phonolitic melt that was tapped from the underlying stratified chamber. Mixing of the resident magma in the Kangerlussuaq Intrusion with phonolitic recharge melts resulted in transitional liquid compositions and gradually forced the bulk composition to become undersaturated. The recharge magma had to be hot enough to superheat the bulk magma to overcome the thermal barrier in Petrogeny's Residua System. Alternatively, or operating simultaneously, increasing water pressure, produced by dehydration of basaltic xenoliths as the magma stoped into the roof, superheated the magma to allow the resident and recharge liquids to mix and solidify forming a range of compositions. The last liquids to reach the Kangerlussuaq Intrusion chamber were pristine phonolites produced directly by differentiation of the parent alkali basalt or basanite that gave rise to the foyaites.
(3) Decreasing amphibole and whole-rock Mg-numbers, and increasing SiO2 content, up through the quartz nordmarkites indicate that the Kangerlussuaq Intrusion magma initially fractionated towards the granite minimum in Petrogeny's Residua System. Increasing recharge/resident magma ratio after formation of the quartz nordmarkites forced the bulk magma back towards the Ab–Or join. The large compositional jump between the main pulaskites and the foyaites may result from a further increase in the recharge/resident magma ratio.
(4) Unlike other occurrences of silica-oversaturated and -undersaturated syenites, in which the different lithologies formed as discrete intrusions, the Kangerlussuaq Intrusion appears to have formed in an open-system, periodically recharged and fractionating magma chamber.
|
ACKNOWLEDGEMENTS |
|---|
We thank Tod E. Waight, Joel A. Baker and David G. Ulfbeck for help and technical support with the radiogenic isotope analyses, and Thorkild Rasmussen and Bo Møller Nielsen for access to geophysical data. Sidsel Grundvig, Ingrid Aaes and Jette Villesen are thanked for their help and technical support with the electron microprobe analyses and thin-section preparation. This work was carried out as part of M. S. Riishuus’ Ph.D. project, financed by the Science Faculty at the University of Aarhus, with additional financial support from the defunct Danish Lithosphere Centre (financed by the Danish National Research Foundation). Finally, we thank Gregor Markl, Chris Harris, Stefan Bernstein and editor Marjorie Wilson for their thorough review of the manuscript.
*Corresponding author. Present address: Department of Geological and Environmental Sciences, Stanford University, Stanford, CA 94305, USA. Telephone: +1 650 723 0841. Fax: +1 650 723 2199. E-mail: riishuus{at}stanford.edu
|
REFERENCES |
|---|
Ablay GJ, Carroll MR, Palmer MR, Mart J, Sparks RSJ. Basanite–phonolite lineages of the Teide–Pico Viejo volcanic complex, Tenerife, Canary Islands. Journal of Petrology (1998) 39:905–936.[CrossRef][Web of Science]
Andreasen R, Peate DW, Brooks CK. Magma plumbing systems in large igneous provinces: Inferences from cyclical variations in Palaeogene East Greenland basalts. Contributions to Mineralogy and Petrology (2004) 147:438–452.[CrossRef][Web of Science]
Bailey JC, Gwozdz R, Rose-Hansen J, Sørensen H. Preliminary geochemical work on the Ilímaussaq alkaline intrusion, South Greenland. Rapport Grønlands Geologiske Undersøgelse (1978) 90:75–79.
Bailey JC, Sørensen H, Andersen T, Kogarko LN, Rose-Hansen J. On the origin of microrhythmic layering in arfvedsonite lujavrite from the Ilímaussaq alkaline complex, South Greenland. Lithos (2006) 91:301–318.[CrossRef][Web of Science]
Baker J, Peate D, Waight T, Meyzen C. Pb isotopic analysis of standards and samples using a 207Pb–204Pb double spike and thallium to correct for mass bias with a double-focusing MC-ICP-MS. Chemical Geology (2004) 211:275–303.[CrossRef][Web of Science]
Bauer K, Trumbull RB, Vietor T. Geophysical images and a crustal model of intrusive structures beneath the Messum ring complex, Namibia. Earth and Planetary Science Letters (2003) 216:65–80.[CrossRef][Web of Science]
Beckinsale RD, Brooks CK, Rex DC. K–Ar ages for the Tertiary of East Greenland. Bulletin of the Geological Society of Denmark (1970) 20:27–37.
Bernstein S, Kelemen PB, Tegner C, Kurz MD, Blusztajn J, Brooks CK. Post-breakup basaltic magmatism along the East Greenland Tertiary rifted margin. Earth and Planetary Science Letters (1998) 160:845–862.[CrossRef][Web of Science]
Bindeman IN, Ponomareva VV, Bailey JC, Valley JW. Volcanic arc of Kamchatka; a province with high-delta 18O magma sources and large-scale 18O/16O depletion of the upper crust. Geochimica et Cosmochimica Acta (2004) 68:841–865.[CrossRef][Web of Science]
Bizzarro M, Baker JA, Ulfbeck D. A new digestion and chemical separation technique for rapid and highly reproducible determination of Lu/Hf and Hf isotope ratios in geological materials by MC-ICP-MS. Geostandards Newsletter (2003) 27:133–145.[CrossRef][Web of Science]
Blundell DJ. A gravity survey across the Gardar Igneous Province, SW Greenland. Journal of the Geological Society, London (1978) 135:545–554.
Bogaard P, Schmincke H-U. The eruptive center of the late Quaternary Laacher See tephra. Geologische Rundschau (1984) 73:935–982.
Brandriss ME, Bird DK, O’Neil JR, Cullers RL. Dehydration, partial melting, and assimilation of metabasaltic xenoliths in gabbros of the Kap Edvard Holm Complex, East Greenland. American Journal of Science (1996) 296:333–393.
Brooks CK, Gill RCO. Compositional variation in the pyroxenes and amphiboles of the Kangerdlugssuaq intrusion, East Greenland: further evidence for the crustal contamination of syenite magma. Mineralogical Magazine (1982) 45:1–9.[CrossRef][Web of Science]
Brooks CK, Platt RG. Kaersutite-bearing gabbroic inclusions and the late dike swarm of Kangerdlugssuaq, East Greenland. Mineralogical Magazine (1975) 40:259–283.[CrossRef][Web of Science]
Brown PE. A layered plutonic complex of alkali basalt parentage: the Lilloise intrusion, east Greenland. Journal of the Geological Society, London (1973) 129:405–418.
Chambers AD, Brown PE. The Lilloise Intrusion, East Greenland—fractionation of a hydrous alkali picritic magma. Journal of Petrology (1995) 36:933–963.
Chen J, Henderson CMB, Foland KA. Open-system, sub-volcanic magmatic evolution: constraints on the petrogenesis of the Mount Brome Alkaline Complex, Canada. Journal of Petrology (1994) 35:1127–1153.
Cohen RS, O’Nions RK. The lead, neodymium and strontium isotopic structure of ocean ridge basalts. Journal of Petrology (1982) 23:299–324.
Dahl-Jensen T, Larsen TB, Woelbern I, Bach T, Hanka W, Kind R, Gregersen S, Mosegaard K, Voss P, Gudmundsson O. Depth to Moho in Greenland: receiver-function analysis suggests two Protorozoic blocks in Greenland. Earth and Planetary Science Letters (2003) 205:379–393.[CrossRef][Web of Science]
Deer WA, Kempe DRC. Geological investigations in East Greenland, Part XI, The minor peripheral intrusions, Kangerdlugssuaq, East Greenland. Meddelelser om Grønland (1976) 197:1–25.
DePaolo DJ. Trace element and isotopic effects of combined wallrock assimilation and fractional crystallisation. Earth and Planetary Science Letters (1981) 53:189–202.[CrossRef][Web of Science]
Droop GTR. A general equation for estimating Fe3+ concentrations in ferromagnesian silicates and oxides from microprobe analyses, using stoichiometric criteria. Mineralogical Magazine (1987) 51:431–435.[Web of Science]
Elliott TR, Hawkesworth CJ, Grönvold K. Dynamic melting of the Iceland plume. Nature (1991) 351:201–206.[CrossRef]
Ferguson J. Geology of the Ilímaussaq alkaline intrusion, south Greenland. Bulletin Grønlands Geologiske Undersøgelse (1964) 39:82.
Fitton JG. The Cameroon line, West Africa: a comparison between oceanic and continental alkaline volcanism. In: Alkaline Igneous Rocks—Fitton JG, Upton BGJ, eds. (1987) Oxford: Blackwell. 273–291.
Foland KA, Henderson CMB. Application of age and Sr isotope data to the petrogenesis of the Marangudzi ring complex, Rhodesia. Earth and Planetary Science Letters (1976) 29:291–301.[CrossRef][Web of Science]
Foland KA, Landoll JD, Henderson CMB, Chen JF. Formation of cogenetic quartz and nepheline syenites. Geochimica et Cosmochimica Acta (1993) 57:697–704.[CrossRef][Web of Science]
Forsberg R, Rasmussen KL. Gravity and rock densities in the Ilímaussaq area, South Greenland. Rapport Grønlands Geologiske Undersøgelse (1978) 90:81–84.
Fram MS, Lesher CE. Generation and polybaric differentiation of East Greenland early Tertiary flood basalts. Journal of Petrology (1997) 38:231–275.[CrossRef][Web of Science]
Fudali RF. Experimental studies bearing on the origin of pseudoleucite and associated problems of alkalic rock systems. Geological Society of America Bulletin (1963) 74:1101–1126.
Furman T, Frey FA, Park K-H. Chemical constraint on the petrogenesis of mildly alkaline lavas from Vestmannaeyjar, Iceland: the Eldfell (1973) and Surtsey (1963–1967) eruptions. Contributions to Mineralogy and Petrology (1991) 109:19–37.[CrossRef][Web of Science]
Giret A, Bonin B, Leger J. Amphibole compositional trends in oversaturated alkaline plutonic ring-complexes. Canadian Mineralogist (1980) 18:481–495.
Gleadow AJW, Brooks CK. Fission track dating, thermal histories and tectonics of igneous intrusions in East Greenland. Contributions to Mineralogy and Petrology (1979) 71:45–60.[CrossRef][Web of Science]
Halama R, Marks M, Brugmann G, Siebel W, Wenzel T, Markl G. Crustal contamination of mafic magmas: evidence from a petrological, geochemical and Sr–Nd–Os–O isotopic study of the Proterozoic Isortoq dike swarm, South Greenland. Lithos (2004) 74:199–232.[CrossRef][Web of Science]
Hamilton DL, MacKenzie WS. Phase-equilibrium studies in the system NaAlSiO4 (nepheline)–KAlSiO4 (kalsilite)–SiO2–H2O. Mineralogical Magazine (1965) 34:214–231.[CrossRef][Web of Science]
Hanan BB, Schilling J.-G. The dynamic evolution of the Iceland mantle plume: the lead isotope perspective. Earth and Planetary Science Letters (1997) 151:43–60.[CrossRef][Web of Science]
Hansen H, Nielsen TFD. Crustal contamination in Palaeogene East Greenland flood basalts: plumbing system evolution during continental rifting. Chemical Geology (1999) 157:89–118.[CrossRef][Web of Science]
Hansen H, Pedersen AK, Duncan RA, Bird DK, Brooks CK, Fawcett JJ, Gittins J, Gorton MP, O’Day P. Volcanic stratigraphy of the southern Prinsen af Wales Bjerge region, East Greenland. In:. The North Atlantic Igneous Province: Stratigraphy, Tectonic, Volcanic and Magmatic Processes. Geological Society, London, Special Publications—Bell BR, ed. (2002) 197:183–218.
Hardarson BS, Fitton JG, Ellam RM, Pringle MS. Rift relocation—a geochemical and geochronological investigation of a palaeo-rift in northwest Iceland. Earth and Planetary Science Letters (1997) 153:181–196.[CrossRef][Web of Science]
Hards VL, Kempton PD, Thompson RN. The heterogeneous Iceland plume: new insights from the alkaline basalts of the Snaefell volcanic centre. Journal of the Geological Society, London (1995) 152:1003–1009.
Harms E, Schmincke H-U. Volatile composition of the phonolitic Laacher See magma (12,900 yr BP): Implications for syneruptive degassing of S, F, Cl and H2O. Contributions to Mineralogy and Petrology (2000) 138:84–98.[CrossRef][Web of Science]
Harms E, Gardner JE, Schmincke H-U. Phase equilibria of the Lower Laacher See Tephra (East Eifel, Germany): Constraints on pre-eruptive storage conditions of a phonolitic magma reservoir. Journal of Volcanology and Geothermal Research (2004) 134:125–138.[CrossRef][Web of Science]
Harris C. Oxygen isotope geochemistry of the Mesozoic anorogenic complexes of Damaraland, northwest Namibia: evidence for crustal contamination and its effect on silica saturation. Contributions to Mineralogy and Petrology (1995) 122:308–321.[CrossRef][Web of Science]
Harris C, Marsh JS, Milner SC. Petrology of the alkaline core of the Messum Igneous Complex, Namibia: Evidence for the progressively decreasing effect of crustal contamination. Journal of Petrology (1999) 40:1377–1397.[CrossRef][Web of Science]
Holbrook WS, Larsen HC, Korenaga J, Dahl-Jensen T, Reid ID, Kelemen PB, Hopper JR, Kent GM, Lizarralde D, Bernstein S, Detrick RS. Mantle thermal structure and active upwelling during continental breakup in the North Atlantic. Earth and Planetary Science Letters (2001) 190:251–266.[CrossRef][Web of Science]
Holm PM, Prægel N-O. The Tertiary Kærven Syenite Complex, Kangerdlugssuaq, East Greenland—mineral chemistry and geochemistry. Mineralogical Magazine (1988) 52:435–450.[CrossRef][Web of Science]
Holm PM, Prægel N-O. Cumulates from primitive rifting-related East Greenland Palaeogene magmas: petrological and isotopic evidence from the ultramafic complexes at Kælvegletscher and near Kærven. Lithos (2006) 92:251–275.[CrossRef][Web of Science]
Kempe DRC, Deer WA. Geological investigations in East Greenland, part IX, The mineralogy of the Kangerdlugssuaq alkaline intrusion, East Greenland. Meddelelser om Grønland (1970) 190:1–97.
Kempe DRC, Deer WA. The petrogenesis of the Kangerdlugssuaq alkaline intrusion, East Greenland. Lithos (1976) 9:111–123.[CrossRef][Web of Science]
Kempe DRC, Deer WA, Wager LR. Geological investigations in East Greenland, part VIII, The petrology of the Kangerdlugssuaq alkaline intrusion, East Greenland. Meddelelser om Grønland (1970) 190:1–52.
Kogarko LN. Role of volatiles. In: The Alkaline Rocks—Sørensen H, ed. (1974) Chichester: John Wiley. 474–487.
Kogarko LN, Khapaev V. The modelling of formation of apatite deposits of the Khibina massif (Kola Peninsula). In:. Origins of Igneous Layering. NATO ASI Series. Series C: Mathematical and Physical Sciences—Parsons I, ed. (1987) 196:589–611.
Kogarko LN, Williams CT, Woolley AR. Chemical evolution and petrogenetic implications of loparite in the layered, agpaitic Lovozero complex, Kola Peninsula, Russia. Mineralogy and Petrology (2002) 74:1–24.[CrossRef][Web of Science]
Kogarko LN, Williams CT, Woolley AR. Compositional evolution and cryptic variation in pyroxenes of the peralkaline Lovozero intrusion, Kola Peninsula, Russia. Mineralogical Magazine (2006) 70:347–359.
Korenaga J, Holbrook WS, Kent GM, Kelemen PB, Detrick RS, Larsen HC, Hopper JR, Dahl-Jensen T. Crustal structure of the southeast Greenland margin from joint refraction and reflection seismic tomography. Journal of Geophysical Research (2000) 105:21591–21614.[CrossRef]
Kramm U, Kogarko LN. Nd and Sr isotope signatures of the Khibina and Lovozero agpaitic centres, Kola Alkaline Province, Russia. Lithos (1994) 32:225–242.[CrossRef][Web of Science]
Kyle PR. Mineralogy and geochemistry of a basanite to phonolite sequence at Hut Point peninsula, Antarctica, based on core from Dry Valley Drilling Project drillholes 1, 2 and 3. Journal of Petrology (1981) 22:451–500.
Landoll JD, Foland KA, Henderson CMB. Nd isotopes demonstrate the role of contamination in the formation of coexisting quartz and nepheline syenites at the Abu-Khruq complex, Egypt. Contributions to Mineralogy and Petrology (1994) 117:305–329.[CrossRef][Web of Science]
Larsen LM, Sørensen H. The Ilímaussaq intrusion—progressive crystallization and formation of layering in an agpaitic magma. In:. Alkaline Igneous Rocks. Geological Society, London, Special Publications—Fitton JG, Upton BGJ, eds. (1987) 30:489–515.
Larsen LM, Watt WS. Episodic volcanism during break-up of the North Atlantic: evidence from the East Greenland plateau basalts. Earth and Planetary Science Letters (1985) 73:105–116.[CrossRef][Web of Science]
Larsen LM, Watt WS. Geology and petrology of the lower Tertiary plateau basalts of the Scoresby Sund region, East Greenland. Geological Survey of Greenland Bulletin (1989) 157:1–164.
Larsen RB, Tegner C. Pressure conditions for the solidification of the Skaergaard: Eruption of East Greenland flood basalts in less than 300,000 years. Lithos (2006) 92:181–197.[CrossRef][Web of Science]
Leake BE, Woolley AR, Arps CES, et al. Nomenclature of amphiboles: report of the subcommittee on amphiboles of the international mineralogical association, commission on new minerals and mineral names. Canadian Mineralogist (1997) 35:219–246.[Web of Science]
Le Bas MJ, Streckeisen AL. The IUGS systematics of igneous rocks. Journal of the Geological Society, London (1991) 148:825–833.
Leeman WP, Dasch EJ, Kays MA. 207Pb/206Pb whole-rock ages of gneisses from the Kangerdlugssuaq area, eastern Greenland. Nature (1976) 263:469–471.[CrossRef]
Le Maitre RW, ed. A Classification of Igneous Rocks and Glossary of Terms (2005) Oxford: Blackwell. 256.
leRoex AP, Cliff RA, Adair BJI. Tristan da Cunha, South Atlantic: geochemistry and petrogenesis of a basanite–phonolite lava series. Journal of Petrology (1990) 31:779–812.
Lynton SJ, Candela PA, Piccoli PM. An experimental study of the partitioning of copper between pyrhotite and a high silica rhyolitic melt. Economic Geology and the Bulletin of the Society of Economic Geologists (1993) 88:901–915.
Markl G, Marks M, Schwinn G, Sommer H. Phase equilibrium constraints on intensive crystallization parameters of the Ilímaussaq complex, South Greenland. Journal of Petrology (2001) 42:2231–2258.
Marks M, Markl G. Fractionation and assimilation processes in the alkaline augite syenite unit of the Ilímaussaq intrusion, South Greenland, as deduced from phase equilibria. Journal of Petrology (2001) 42:1947–1969.
Marks M, Vennemann T, Siebel W, Markl G. Quantification of magmatic and hydrothermal processes in a peralkaline syenite–alkali granite complex based on textures, phase equilibria, and stable and radiogenic isotopes. Journal of Petrology (2003) 44:1247–1280.
Marks M, Vennemann T, Siebel W, Markl G. Nd-, O-, and H-isotopic evidence for complex, closed-system fluid evolution of the peralkaline Ilímaussaq intrusion, South Greenland. Geochimica et Cosmochimica Acta (2004) 68:3379–3395.[CrossRef][Web of Science]
McDonough WF, Sun S-s. The composition of the Earth. Chemical Geology (1995) 120:223–253.[CrossRef][Web of Science]
Morse SA. Syenites. Carnegie Institution of Washington Annual Report Geophysical Laboratory (1969) 67:112–120.
Nielsen FM, Campbell IH, McCulloch M, Wilson JR. A strontium isotopic investigation of the Bjerkreim–Sokndal layered intrusion, southwest Norway. Journal of Petrology (1996) 37:171–193.
Nielsen TFD. The Tertiary dike swarms of the Kangerdlugssuaq area, East Greenland. An example of magmatic development during continental break-up. Contributions to Mineralogy and Petrology (1978) 67:63–78.[CrossRef][Web of Science]
Nielsen TFD. The petrology of a melilitolite, melteigite, carbonatite and syenite ring dike system in the Gardiner Complex, East Greenland. Lithos (1980) 13:181–197.[CrossRef][Web of Science]
Nielsen TFD. The ultramafic cumulate series, Gardiner Complex, East Greenland—cumulates in a shallow level magma chamber of a nephelinitic volcano. Contributions to Mineralogy and Petrology (1981) 76:60–72.[CrossRef][Web of Science]
Nielsen TFD. Tertiary alkaline magmatism in East Greenland: a review. In:. Alkaline Igneous Rocks. Geological Society, London, Special Publications—Fitton JG, Upton BGJ, eds. (1987) 30:489–515.
Nielsen TFD. Palaeogene intrusions and magmatic complexes in East Greenland, 66 to 75° N. Geological Survey of Denmark and Greenland Report (2002) 2002/113:249. pp.
Nielsen TFD, Soper NJ, Brooks CK, Faller AM, Higgins AC, Matthews DW. The pre-basaltic sediments and the Lower Basalts at Kangerdlugssuaq, East Greenland: their stratigraphy, lithology, palaeomagnetism and petrology. Meddelelser om Grønland, Geoscience (1981) 6:1–25.[Medline]
Nielsen TFD, Hansen H, Brooks CK, Lesher CE, field parties. The East Greenland continental margin, the Prinsen af Wales Bjerge and new Skaergaard intrusion initiatives. Geology of Greenland Survey Bulletin (2001) 189:83–98.
Pankhurst RJ, Beckinsale RD, Brooks CK. Strontium and oxygen isotope evidence relating to the petrogenesis of the Kangerdlugssuaq alkaline intrusion, East Greenland. Contributions to Mineralogy and Petrology (1976) 54:17–42.[CrossRef][Web of Science]
Park KH. Sr, Nd and Pb Isotope Studies of Ocean Island Basalts: Constraints on their Origin and Evolution (1990) New York: Columbia University. 252.
Parsons I. Klokken gabbro–syenite complex, South Greenland—cryptic variation and origin of inversely graded layering. Journal of Petrology (1979) 20:653–694.
Parsons I, Becker SM. Layering, compaction and post-magmatic processes in the Klokken intrusion. In:. Origins of Igneous Layering. NATO ASI Series. Series C: Mathematical and Physical Sciences—Parsons I, ed. (1987) 196:29–92.
Pearce NJG, Perkins WT, Westgate JA, Gorton MP, Jackson SE, Neal CR, Chenery SP. A compilation of new and published major and trace element data for NIST SRM 610 and NIST SRM 612 glass reference materials. Geostandards Newsletter (1997) 21:115–144.[CrossRef][Web of Science]
Peate DW, Stecher O. Pb isotope evidence for contributions from different Iceland mantle components to Palaeogene East Greenland flood basalts. Lithos (2003) 67:39–52.[CrossRef][Web of Science]
Peate DW, Baker JA, Blichert-Toft J, Hilton DR, Storey M, Kent AJR, Brooks CK, Hansen H, Pedersen AK, Duncan RA. The Prinsen af Wales Bjerge Formation lavas, East Greenland: the transition from tholeiitic to alkalic magmatism during Palaeogene continental break-up. Journal of Petrology (2003) 44:279–304.
Pedersen AK, Watt M, Watt WS, Larsen LM. Structure and stratigraphy of the early Tertiary basalts of the Blosseville Kyst, East Greenland. Journal of the Geological Society, London (1997) 154:565–570.
Petersen JS. Structure of the larvikite–ladarlite complex, Oslo-region, Norway, and its evolution. Geologische Rundschau (1978) 67:330–342.
Petersen JS. Columnar–dendritic feldspars in the Lardalite Intrusion, Oslo region, Norway: 1. Implications for unilateral solidification of a stagnant boundary layer. Journal of Petrology (1985) 26:223–252.
Ramberg IB, Smithson SB. Gravity interpretation of the southern Oslo Graben and adjacent Precambrian rocks, Norway. Tectonophysics (1971) 11:419–431.[CrossRef][Web of Science]
Riishuus MS, Peate DW, Tegner C, Wilson JR, Brooks CK, Waight TE. Petrogenesis of syenites at a rifted continental margin: origin, contamination and interaction of alkaline mafic and felsic magmas in the Astrophyllite Bay Complex, East Greenland. Contributions to Mineralogy and Petrology (2005) 149:350–371.[CrossRef][Web of Science]
Riishuus MS, Peate DW, Tegner C, Wilson JR, Brooks CK, Harris C. Temporal evolution of a long-lived syenitic centre: the Kangerlussuaq Alkaline Complex, East Greenland. Lithos (2006) 92:276–299.[CrossRef][Web of Science]
Saunders AD, Fitton JG, Kerr AC, Norry MJ, Kent R. The North Atlantic Igneous Province. In:. Large Igneous Provinces: Continental, Oceanic, and Planetary Flood Volcanism. Geophysical Monograph, American Geophysical Union—Mahoney JJ, Coffin MF, eds. (1997) 100:45–93.
Schairer JF. The alkali–feldspar join in the system NaAlSiO4–KAlSiO4–SiO2. Journal of Geology (1950) 58:512–517.[Web of Science]
Schmitt AK, Emmermann R, Trumbull RB, Bühn B, Henjes-Kunst F. Petrogenesis and 40Ar/39Ar geochronology of the Brandberg Complex, Namibia: Evidence for a major mantle contribution in metaluminous and peralkaline granites. Journal of Petrology (2000) 41:1207–1239.
Stephenson D, Upton BGJ. Ferromagnesian silicates from a differentiated alkaline complex: Kûngnât Fjeld, South Greenland. Mineralogical Magazine (1982) 46:283–300.[Web of Science]
Stevenson R, Upton BGJ, Steenfelt A. Crust–mantle interaction in evolution of the Ilímaussaq Complex, South Greenland: Nd isotopic studies. Lithos (1997) 40:189–202.[CrossRef][Web of Science]
Sørensen H, Larsen LM. Layering in the Ilímaussaq alkaline intrusion, South Greenland. In:. Origins of Igneous Layering. NATO ASI Series. Series C: Mathematical and Physical Sciences—Parsons I, ed. (1987) 196:1–28.
Sørensen HS, Wilson JR. A strontium and neodymium isotopic investigation of the Fongen–Hyllingen layered intrusion, Norway. Journal of Petrology (1995) 36:161–187.
Stecher O, Carlson RW, Gunnarsson B. Torfajokull: a radiogenic end-member of the Iceland Pb-isotopic array. Earth and Planetary Science Letters (1999) 165:117–127.[CrossRef][Web of Science]
Storey M, Pedersen AK, Stecher O, Bernstein S, Larsen HC, Larsen LM, Baker JA, Duncan RA. Long-lived postbreakup magmatism along the East Greenland margin: evidence for shallow-mantle metasomatism by the Iceland plume. Geology (2004) 32:173–176.
Storey M, Duncan RA, Tegner C. Timing and duration of volcanism in the North Atlantic Igneous Province: implications for geodynamics and links to the Iceland hotspot. Chemical Geology (2007) 241:264–281.[CrossRef][Web of Science]
Sun S-S, Jahn B-M. Lead and strontium isotopes in post-glacial basalts from Iceland. Nature (1975) 255:527–530.[CrossRef]
Sun S-S, Tatsumoto M, Schilling J-G. Mantle plume mixing along the Reykjanes Ridge axis: lead isotope evidence. Science (1975) 190:143–147.[Web of Science]
Taylor PN, Kalsbeek F, Bridgwater D. Discrepancies between neodymiun, lead and strontium model ages from the Precambrian of southern East Greenland: evidence for a Proterozoic granulite-facies event affecting Archaean gneisses. Chemical Geology (Isotope Geoscience Section) (1992) 94:281–291.[CrossRef]
Tegner C, Duncan RA, Bernstein S, Brooks CK, Bird DK, Storey M. 40Ar–39Ar geochronology of Tertiary mafic intrusions along the East Greenland rifted margin: Relation to flood basalts and the Iceland hotspot track. Earth and Planetary Science Letters (1998a) 156:75–88.[CrossRef][Web of Science]
Tegner C, Lesher CE, Larsen LM, Watt WS. Evidence from the rare-earth-element record of mantle melting for cooling of the Tertiary Iceland plume. Nature (1998b) 395:591–594.[CrossRef]
Tegner C, Robins B, Reginiussen H, Grundvig S. Assimilation of crustal xenoliths in a basaltic magma chamber: Sr and Nd isotopic constraints from the Hasvik Layered Intrusion, Norway. Journal of Petrology (1999) 40:363–380.[CrossRef][Web of Science]
Tegner C, Brooks CK, Duncan RA, Heister LE, Bernstein S. 40Ar–39Ar ages of intrusions in East Greenland: Rift-to-drift transition over the Iceland hotspot. Lithos (2008) doi:10.1016/j.lithos.2007.09.001.
Thompson GM, Smith IEM, Malpas JG. Origin of oceanic phonolites by crystal fractionation and the problem of the Daly gap: an example from Rarotonga. Contributions to Mineralogy and Petrology (2001) 142:336–346.[Web of Science]
Tuttle OF, Bowen NL. Origin of Granite in the Light of Experimental Studies in the System NaAlSiO4–KAlSiO4–SiO2–H2O. Geological Society of America, Memoirs (1958) 74:1–153.
Ulfbeck D, Baker J, Waight T, Krogstad E. Rapid sample digestion by fusion and chemical separation of Hf for isotopic analysis by MC-ICPMS. Talanta (2003) 59:365–373.
Upton BGJ. Gabbroic, syenogabbroic and syenitic cumulates of the Tugtutôq Younger Giant Dyke Complex, South Greenland. In:. Origins of Igneous Layering. NATO ASI Series. Series C: Mathematical and Physical Sciences—Parsons I, ed. (1987) 196:93–123.
Upton BGJ, Stephenson D, Martin AR. The Tututoq Older Giant Dyke Complex: Mineralogy and geochemistry of an alkali gabbro–augite syenite–foyaite association in the Gardar Province of South Greenland. Mineralogical Magazine (1985) 49:623–642.[CrossRef][Web of Science]
Upton BGJ, Emeleus CH, Heaman LM, Goodenough KM, Finch AA. Magmatism of the mid-Proterozoic Gardar Province, South Greenland: chronology, petrogenesis and geological setting. Lithos (2003) 68:43–65.[CrossRef][Web of Science]
Verhoef J, Roest WR, Macnab R. Magnetic anomalies of the Arctic and North Atlantic oceans and adjacent land areas. Geological Survey of Canada, Open File (1996) 3125.
Villemant B. Trace element evolution in the Plegrean Fields, Central Italy: fractional crystallization and selective enrichment. Contributions to Mineralogy and Petrology (1988) 98:169–183.[CrossRef][Web of Science]
Wager LR. The form and internal structure of the alkaline Kangerdlugssuaq intrusion, East Greenland. Mineralogical Magazine (1965) 34:487–497.[CrossRef][Web of Science]
Waight T, Baker J, Willigers B. Rb isotope dilution analyses by MC-ICPMS using Zr to correct for mass fractionation: towards improved Rb–Sr geochronology? Chemical Geology (2002) 186:99–116.[CrossRef][Web of Science]
Wörner G, Schmincke H-U. Petrogenesis of the zoned Laacher See Tephra. Journal of Petrology (1984) 25:836–851.
Wörner G, Wright TL. Evidence for magma mixing within the Laacher See magma chamber (East Eifel, Germany). Journal of Volcanology and Geothermal Research (1984) 22:301–327.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. G. Barnes, T. Prestvik, Y. Li, L. McCulloch, A. S. Yoshinobu, and C. D. Frost Growth and zoning of the Hortavaer intrusive complex, a layered alkaline pluton in the Norwegian Caledonides Geosphere, June 1, 2009; 5(3): 286 - 301. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. R. Frost and C. D. Frost A Geochemical Classification for Feldspathic Igneous Rocks J. Petrology, November 20, 2008; (2008) egn054v1. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||