Available online at www.sciencedirect.com
Lithos 100 (2008) 293 – 321
www.elsevier.com/locate/lithos
The origin and compositions of Mesoarchean oceanic crust: Evidence
from the 3075 Ma Ivisaartoq greenstone belt, SW Greenland
A. Polat a,⁎, R. Frei b,c , P.W.U. Appel d , Y. Dilek e , B. Fryer a,f ,
J.C. Ordóñez-Calderón a , Z. Yang f
a
f
Department of Earth Sciences, University of Windsor, Windsor, ON, Canada N9B 3P4
b
Geological Institute, University of Copenhagen, 1350-Cope nhagen, Denmark
c
NordCEE, Nordic Center for Earth Evolution, Geological Institute, Denmark
d
Geological Survey of Denmark and Greenland, 1350-Copenhagen, Denmark
e
Department of Geology, Miami University, Oxford, OH 45056, USA
Great Lakes Institute for Environmental Research, University of Windsor, Windsor, ON, Canada N9B 3P4
Received 12 August 2006; accepted 8 June 2007
Available online 2 August 2007
Abstract
The Mesoarchean (ca. 3075 Ma) Ivisaartoq greenstone belt contains well-preserved primary magmatic structures, such as pillow
lavas, volcanic breccias, and clinopyroxene cumulate layers (picrites), despite the isoclinal folding and amphibolite facies
metamorphism. The belt also includes variably deformed gabbroic to dioritic dykes and sills, actinolite schists, and serpentinites. The
Ivisaartoq rocks underwent at least two stages of post-magmatic metamorphic alteration, including seafloor hydrothermal alteration and
syn- to post-tectonic calc-silicate metasomatism, between 3075 and 2961 Ma. These alteration processes resulted in the mobilization of
many major and trace elements. The trace element characteristics of the least altered rocks are consistent with a supra-subduction zone
geodynamic setting and shallow mantle sources. On the basis of geological similarities between the Ivisaartoq greenstone belt and
Phanerozoic forearc ophiolites, and intra-oceanic island arcs, we suggest that the Ivisaartoq greenstone belt represents a relic of
dismembered Mesoarchean supra-subduction zone oceanic crust. This crust might originally have been composed of a lower layer of
leucogabbros and anorthosites, and an upper layer of pillow lavas, picritic flows, gabbroic to dioritic dykes and sills, and dunitic to
wehrlitic sills.
The Sm–Nd and U–Pb isotope systems have been disturbed in strongly altered actinolite schists. In addition, the U–Pb isotope
system in pillow basalts appears to have been partially open during seafloor hydrothermal alteration. Gabbros and diorites have the least
disturbed Pb isotopic compositions. In contrast, the Sm–Nd isotope system appears to have remained relatively undisturbed in picrites,
pillow lavas, gabbros, and diorites. As a group, picrites have more depleted initial Nd isotopic signatures (εNd = +4.23 to +4.97) than
pillow lavas, gabbros, and diorites (εNd = +0.30 to +3.04), consistent with a variably depleted, heterogeneous mantle source.
In some areas gabbros include up to 15 cm long white inclusions (xenoliths). These inclusions are composed primarily (N 90%) of
Ca-rich plagioclase and are interpreted as anorthositic cumulates brought to the surface by upwelling gabbroic magmas. The
anorthositic cumulates have significantly higher initial εNd (+ 4.8 to + 6.0) values than the surrounding gabbroic matrix (+ 2.3 to + 2.8),
consistent with different mantle sources for the two rock types.
© 2007 Elsevier B.V. All rights reserved.
Keywords: Archean; Greenstone belt; Oceanic crust; Pillow basalt; Anorthosite; Ocelli; Isotope
⁎ Corresponding author.
E-mail address: [email protected] (A. Polat).
0024-4937/$ - see front matter © 2007 Elsevier B.V. All rights reserved.
doi:10.1016/j.lithos.2007.06.021
294
A. Polat et al. / Lithos 100 (2008) 293–321
1. Introduction
Archean greenstone belts are composed predominantly
of variably metamorphosed and deformed mafic to felsic
volcanic and siliciclastic sedimentary rocks (Condie, 1981;
Goodwin, 1991; Eriksson et al., 1994; Condie, 2005).
There are also volumetrically minor banded iron formations (BIF), komatiites, gabbros, anorthosites, serpentinites, cherts, and carbonates (locally stromatolitic) in some
Archean greenstone belts (Condie, 1981; Goodwin, 1991;
Condie, 2005). Geochemical data derived from the study of
Archean greenstone belts over the last three decades show
the occurrence of diverse volcanic rock types, suggesting
diverse magmatic processes, such as plume and arc
magmatism, in oceanic or continental settings for their
origin (Dostal and Mueller, 1997; Polat et al., 1998; Kusky
and Polat, 1999; Polat and Hofmann, 2003; Dostal et al.,
2004; Smithies et al., 2005a, b; Kerrich and Polat, 2006). In
addition, greenstone belts from 3.8 to 2.5 Ga include
volcanic rock types reported from Phanerozoic convergent
margins, such as boninites, picrites, adakites, Mg-ande-
sites, and Nb-enriched basalts (Polat and Kerrich, 2006;
and references therein). The distribution of rock types and
the internal structure of many Archean greenstone belts
suggest that they are the products of multiple geological
processes, such as tectonism, magmatism, sedimentation,
and metamorphism, operating over different spatial and
temporal scales (Corcoran and Dostal, 2000; Sandeman
et al., 2004; Kerrich and Polat, 2006). Collectively, the
geological characteristics of many Archean greenstone
belts are comparable to those of lithotectonic assemblages
occurring in Phanerozoic convergent plate boundaries
(Kusky and Polat, 1999; Şengör and Natal’in, 2004;
Kerrich and Polat, 2006; and references therein).
Thermal and geodynamic models and geochemical
and isotopic constraints derived from Archaean mafic–
ultramafic rocks suggest that oceanic crust formation
must also have occurred in the Archean. From thermal
modeling, Abbott et al. (1994) inferred that Neoarchean
mid-ocean ridge crust was ∼ 11 km-thick, in contrast to
∼ 7 km-thick in-situ oceanic crust developed at modern
mid-ocean ridges. Numerical and modeling studies infer
Fig. 1. (a) A simplified geological map of the northeastern Nuuk region, showing the Eoarchean to Neoarchean tectonic terranes and location of the Ivisaartoq
belt. Modified from Friend and Nutman (2005). (b) Geological map of the Ivisaartoq and surrounding area. Modified from Chadwick and Coe (1988).
A. Polat et al. / Lithos 100 (2008) 293–321
that the negative buoyancy of ancient oceanic lithosphere is responsible for b1% of its preservation in the
Phanerozoic rock record, whereas relatively younger
and hotter backarc and forearc crusts have been more
readily accreted to orogenic belts (Cloos, 1993; Van
Hunen et al., 2002; Şengör and Natal’in, 2004).
The question of whether Archean greenstone belts
represent the fragments of ancient oceanic crust or
alternatively are the remnants of continental flood basalts
remains controversial (Bickle et al., 1994; Kusky, 2004).
Based on the criteria of xenocrystic zircons and
geochemical contamination trends of their mafic–
ultramafic lavas, some greenstone belts may be considered intra-oceanic in origin, whereas some others may
have formed from mantle magmas erupted through
continental crust (Nisbet and Fowler, 1983; Polat et al.,
1998; Arndt et al., 2001; Bleeker, 2002; Condie, 2005).
The recognition of Archean oceanic crust can be done
most effectively through comparative studies of the wellestablished lithological and geochemical characteristics
for oceanic crustal fragments formed in Phanerozoic
supra-subduction zone environments. Phanerozoic
supra-subduction oceanic crust includes the presence
of: (1) contemporaneous mafic–ultramafic intrusive and
extrusive units; (2) commonly picritic and boninitic
units; (3) chemical sedimentary rocks and sporadically
volcanogenic sediments; (4) diorite–plagiogranite intrusives; (5) diagnostic metasomatic style temporally
associated with the intrusions; (6) geochemical signatures of extrusive rock assemblages formed in intraoceanic versus continental settings; (7) geochemical and
isotopic signatures of a depleted mantle source; and (8)
convergent margin geochemical signatures of magmas in
terms of REE/HFSE fractionations (see Dewey, 2003;
Dilek, 2003; Hawkins, 2003; Schuth et al., 2004; Python
et al., 2007).
Pillow structures, volcanic breccia, cumulate and
ocellar (eye-shaped) textures have been well preserved
in low-strain domains of the Mesoarchean Ivisaartoq
greenstone belt in SW Greenland, despite the two major
phases of deformation and amphibolite-facies metamorphism (Friend et al., 1981; Hall, 1981; Chadwick, 1985,
1986; Hall et al., 1987; Chadwick, 1990; Appel, 1997;
Polat et al., 2007). Preservation of these primary
structures and textures provides a unique opportunity
to study Mesoarchean petrogenetic and geodynamic
processes. The lithological, trace element, and hydrothermal alteration characteristics of the Ivisaartoq
greenstone belt are comparable to those of Phanerozoic
forearc ophiolites (Polat et al., 2007). Given the fact that
all rocks in the Ivisaartoq greenstone belt have been
metamorphosed, the prefix ‘meta’ will be taken implicit.
295
In this study, we report new high-precision major and
trace element data (34 samples), and Nd (35 samples)
and Pb (32 samples) isotope data obtained from
actinolite schists, ultramafic cumulates, gabbros, diorites, and anorthositic inclusions (xenoliths) in the
Ivisaartoq greenstone belt. Accordingly, the objective
of this study is threefold: (1) to assess the effect of postmagmatic alteration on element mobility and of isotopic
composition; (2) to understand the petrologic and
geodynamic origin of Mesoarchean oceanic crust
preserved in the Ivisaartoq greenstone belt; and (3) reevaluate the existing geodynamic models proposed for
the origin of Archean anorthosites.
2. Regional geology and field characteristics
The Ivisaartoq greenstone belt contains the largest
Mesoarchean supra-crustal assemblage in southern West
Greenland (Fig. 1; Hall and Friend, 1979; Brewer et al.,
1984; Chadwick, 1985, 1986, 1990; Friend and Nutman, 2005). It is located in the central part of the inner
Godthåbsfjord region (Fig. 1a). The belt occurs within the
recently recognized Mesoarchean (∼3075–2950 Ma)
Kapisilik tectonic terrane (Friend and Nutman, 2005),
which is tectonically bounded by the Eoarchean Isukasia
Fig. 2. A simplified tectonostratigraphic column of the Ivisaartoq belt.
Modified from Chadwick (1986, 1990).
296
A. Polat et al. / Lithos 100 (2008) 293–321
terrane (3600–3800 Ma) to the north, and the Eoarchean
Færingehavn and the Neoarchean Tre Brødre terranes to
the south and west, respectively (Fig. 1a; Friend and
Nutman, 2005). The Kapisilik and Isukasia terranes were
juxtaposed and metamorphosed by 2950 Ma. It appears
that the collision between the southern Færingehavn and
the Kapisilik terranes occurred at about 2800 Ma (Friend
and Nutman, 2005). Field relationships indicate that the
Isukasia terrane is structurally overlain by the Kapisilik
terrane to the south; and the Kapisilik terrane is in turn
Fig. 3. Field photographs of the Ivisaartoq pillow basalts, gabbros, cumulates and serpentinites. (a) Compositionally zones pillow basalts, recording
the formation of stage I metasomatic assemblage during seafloor hydrothermal alteration. (b) Pillow basalts with concentric cooling fractures, filled
with plagioclase and quartz. (c) Pillow cores, rims and interstitial filled replaced by stage I metasomatic assemblage. (d) Gabbro with well-preserved
fine- and coarse-grained layers. (e) Clinopyroxene (cpx)-bearing cumulate. (f) Deformed serpentinite in amphibolites (deformed pillow basalts).
A. Polat et al. / Lithos 100 (2008) 293–321
structurally overlain by the Færingehavn and Tre Brødre
terranes to the south–southwest.
The precise age of the volcanic and intrusive (gabbro,
diorite) rocks in the Ivisaartoq greenstone belt is unknown. Siliceous volcaniclastic sedimentary rocks have
yielded an average U–Pb zircon age of 3075 ±15 Ma
297
(Friend and Nutman, 2005; Polat et al., 2007), constraining the maximum age of the belt. The Ivisaartoq
greenstone belt is intruded by weakly deformed 2961 ±
12 Ma granites to the north, constraining the minimum
age of the belt (Chadwick, 1990; Friend and Nutman,
2005). The Ivisaartoq sequence is truncated by an up to
Fig. 4. Field photographs of the Ivisaartoq rocks. (a) Strongly deformed rock with a possible siliciclastic sedimentary origin. (b) Pyrite-bearing
siliceous (metacherts) sedimentary layer within amphibolites. Contacts are typically sharp. (c) Tectonite with stage II calc-silicate assemblage near the
lower and upper amphibolite contact. (d) Pillow basalts with drainage cavity-filling quartz and ocelli in the outer core. (e) Ocelli in an outer pillow
core partly replaced by stage I calc-silicate assemblage. (f) Flattened centimeter-sized anorthositic inclusions in gabbros.
298
A. Polat et al. / Lithos 100 (2008) 293–321
2 m-thick mylonite zone to the south, separating the belt
from an association of leucogabbros and anorthosites.
These leucogabbros and anorthosites are intruded by
2963 ± 8 Ma old tonalites and granodiorites (now
gneisses). On the basis of field observations and zircon
ages, Friend and Nutman (2005) interpreted the mylonite
zone as a post-2960 Ma structure deforming the Kapisilik
terrane. The leucogabbro and anorthosite association is
lithologically and structurally similar to those found in the
Fiskenaesset region of southern West Greenland, and is
interpreted as intrusive into the Ivisaartoq greenstone belt
(Chadwick, 1990).
Fig. 5. Field photographs of the Ivisaartoq rocks. (a) Flattened centimeter-sized anorthositic inclusions (xenoliths) in gabbros. (b) Stage I calc-silicate
alteration at a pillow basalt gabbro contact and in pillow cores. (c) Stage II calc-silicate metasomatic assemblage, replacing actinolite schists.
(d) Actinolite schist (dark) replaced by a massive layered stage II calc-silicate rock assemblage. (e) A diopside + garnet + hornblende + quartz ± epidote
vein. (f) Boudins of stage II calc-silicate assemblage in banded amphibolite.
A. Polat et al. / Lithos 100 (2008) 293–321
The Ivisaartoq greenstone belt is composed mainly of
mafic to ultramafic volcanic rocks, gabbros, minor diorites,
and serpentinites (Figs. 2–4; Hall, 1981; Chadwick, 1985,
1986, 1990; Polat et al., 2007). Volcanic rocks consist
dominantly of deformed pillow basalts and ultramafic lava
flows (Fig. 3; Chadwick, 1990; Polat et al., 2007). Sedimentary rocks constitute a volumetrically minor component of the belt (Figs. 1, 2, 4). Chadwick has subdivided the
Ivisaartoq greenstone belt into a lower and an upper amphibolite unit (Fig. 2; Chadwick, 1985, 1986, 1990). These
units are separated by a thin layer (up to 50 m-thick) of
magnetite-rich ultramafic schists, called the ‘magnetic
marker’ (Fig. 2; Chadwick, 1986, 1990). Hydrothermal
alteration of the ‘magnetic marker’ and volcanic rocks in its
vicinity resulted in the formation of calc-silicate rocks
hosting strata-bound scheelite mineralization (Appel, 1994,
1997). The intensity of deformation appears to increase
towards the boundary between the two amphibolite units
(Fig. 4c), suggesting that they are tectonically juxtaposed.
Volcanic rocks in the lower amphibolite unit are more
intensely deformed than those in their upper counterpart.
They display a well-developed foliation characterized by
amphibole- and plagioclase-rich domains. Pillow structures are rare. Volcanic breccias are composed of pillow
fragments locally with possible ocelli and hyaloclastites
(Polat et al., 2007). Up to 50 m wide and 5 km long rusty
layers of pyrite-rich siliceous rocks of probable felsic
volcaniclastic origin are exposed discontinuously in the
lower amphibolite unit (Fig. 1b).
The upper amphibolite unit is composed mainly of
variably deformed pillow basalts, actinolite schists, gabbros, diorites, ultramafic cumulates, and serpentinites
(Figs. 1–3). Serpentinites (ultramafic layers) are exposed
discontinuously as three major layers throughout the sequence (Figs. 1b, 2, 3f; Chadwick, 1986, 1990). Chadwick
(1986) reports the presence of fresh olivine, likely of a
299
metamorphic origin, in serpentinites. On the basis of field
relationships, Chadwick (1986, 1990) suggested that the
protoliths of the serpentinites intruded as sills into submarine lavas. In many outcrops they are in tectonic contact
with pillow basalts and gabbros (Fig. 3f).
Pillow basalts are characterized by well-preserved core
and rim structures (Fig. 3a). The least deformed pillow
basalts have concentric cooling cracks filled mainly
with quartz, and display way-up directions (Fig. 3b).
Pillow cores are mineralogically zoned (Fig. 3a, c). Many
inner pillow cores display drainage cavities at the center,
which are either empty or filled with quartz (Fig. 3c). The
pillow cores often display ocellar texture consisting
chiefly of white ellipsoidal (eye-shaped) millimeter- to
centimeter-sized ocelli set in a dark green fine-grained
mafic matrix (Fig. 4d, e). The contacts between ocelli and
matrix are sharp. In many cores the ocelli-matrix texture
has been partly to completely replaced by a calc-silicate
metasomatic assemblage (Fig. 4). Pillow rims often
display silica alteration; some pillows have been
completely silicified. Some pillows are composed predominantly of actinolite, consistent with an ultramafic
protolith. Primary magmatic textures, such as clinopyroxene cumulates, are locally preserved in ultramafic flows
of low-strain domains (Fig. 3e). With increasing intensity
of deformation clinopyroxene cumulates grade into
actinolite schists.
Gabbros and minor diorites occur as one to several
tens of meter-thick sills and dykes in pillow basalts
(Fig. 3d). Gabbroic and dioritic sills also occur
sporadically between pillow basalts and ultramafic
flows. Chilled margins between pillow basalts and
gabbroic dykes are preserved in a few locations. Primary
igneous textures and minerals are locally preserved in lowstrain domains (Fig. 3). Some gabbros contain deformed
anorthositic inclusions up to 15 cm long (Figs. 4f, 5a). Like
Table 1
Mineralogical compositions of the Ivisaartoq rocks
Lithology
Mineral assemblage
Cumulate
Actinolite schist
Inner pillow core
Outer pillow core
Pillow rim
Gabbro
Diorite
Amphibolite
Inclusion in gabbros
Ocelli in pillows
Calc-silicate stage I
Calc-silicate stage II
Magnetic marker
Actinolite + clinopyroxene ± plagioclase ± quartz
Actinolite + diopside ± plagioclase ± quartz
Diopside + plagiocalse + quartz + epidote ± amphibole ± sulphide ± titanite
Hornblende + plagioclase + quartz ± diopside ± epidote ± titanite
Hornblende + quartz + plagioclase + epidote ± biotite ± titanite
Hornblende + plagioclase ± epidote ± quartz
Hornblende + plagioclase ± epidote ± quartz ± biotite
Hornblende + plagioclase + quartz ± diopside ± epidote ± titanite ± sulphide
Plagioclase + hornblende ± quartz
Plagioclase + quartz + amphibole ± epidote
Diopside + quartz + plagioclase + epidote ± hornblende ± scapolite
Diopside + garnet + amphibole + plagioclase + quartz ± vesuvianite ± scapolite ± epidote ± titanite ± calcite ± scheelite
Actinolite + olivine + diopside + magnetite + plagioclase ± epidote ± scapolite ± calcite ± titanite ± scheelite
300
A. Polat et al. / Lithos 100 (2008) 293–321
pillow basalts, gabbros and diorites underwent calc-silicate
metasomatic alteration mainly along fractures and pillow
basalt contacts (Fig. 5b). An alternating association of over
200 m-thick actinolite schist, pillow basalt, gabbro, and
pyrite-bearing siliceous volcaniclastic rocks is exposed in
the northern margin of the belt. This sequence is intensively
sheared and metasomatized mainly along the actinolite
layers (Fig. 5c). The intensity of calc-silicate metasomatic
alteration increases from gabbros through pillow basalts to
actinolite schists (Fig. 5). Siliceous pyrite-bearing rocks
were interpreted as metamorphosed cherts (Chadwick,
1990; Fig. 4b). There are minor, up to several meter-thick,
Fig. 6. Photomicrographs of the Ivisaartoq rocks. (a) Altered clinopyroxene phenocrysts surrounded by actinolite matrix (crossed polarized light).
(b) Actinolite schist (crossed polarized light). (c) Gabbro (crossed polarized light). (d) Ocelli texture in outer pillow core (plain polarized light).
(e) Magnetic marker (plain polarized light). Magnetite occurs mainly along the foliation planes. (f) Calc-silicate assemblage consisting mainly of
diopside and calcite (plain polarized light). (Cpx: clinopyroxene; act: actinolite; hornb: hornblende; plag: plagioclase).
A. Polat et al. / Lithos 100 (2008) 293–321
301
mineral assemblages are shown in Fig. 6. Cumulates are
composed primarily of altered clinopyroxene phenocrysts
(Figs. 4e, 6a; Table 1). Ultramafic schists are composed
dominantly of actinolite (Fig. 6b; Table 1). Gabbros and
diorites are composed mainly of hornblende + plagioclase ±
epidote ± quartz (Fig. 6c).
Ocelli in the pillow cores (Figs. 4d, e; 6d) consist
mainly of plagioclase (30–50%) + quartz (30–40%) +
amphibole (10–20%) ± epidote (0–5%). No internal
structure has been observed in the ocelli. The darker
matrix surrounding the ocelli is made of amphibole (50–
60%) + plagioclase (20–30%) + quartz (10–20%) ± epidote (0–5%) ± titanite (0–5%) (Fig. 6d; Table 1). The
pillow rims are composed of fine-grained hornblende +
quartz + plagioclase ± epidote ± biotite.
Magnetic marker is composed primarily of actinolite + olivine (metamorphic) + magnetite + diopside ±
Table 2
Measured and recommended values for the USGS standards BHVO-1
and BHVO-2
Element
Fig. 7. Photomicrographs of anorthositic inclusions in gabbros (see
Fig. 5a), showing recrystallized plagioclase.
lenses of siliciclastic sedimentary rocks in the upper unit
(Fig. 4a). Contacts between volcanic and sedimentary
rocks are sharp (Fig. 4b).
Field and petrographic observations suggest that the
Ivisaartoq greenstone belt underwent at least two stages of
calc-silicate metasomatic alteration prior to the intrusion
of 2961 Ma granitoids (Polat et al., 2007). Stage I
alteration assemblage typically occurs within the inner
pillow cores, pillow interstitials, and along the pillowgabbro contacts (Figs. 3–5; Table 1). This assemblage is
also found along the pillow-gabbro contacts and fractures
within gabbros (Fig. 5b). Stage II metasomatic assemblage
occurs as calc-silicate veins and boudins that are
concordant to discordant to the dominant foliation planes
in the replaced host rocks, consistent with a syn- to post
deformation origin (Fig. 5). Most of these veins are
spatially associated with shear zones.
3. Petrography
The mineralogical characteristics of different rock types
are summarized in Table 1 and photomicrographs of major
Li
V
Cr
Co
Ni
Cu
Zn
Rb
Sr
Y
Zr
Nb
Cs
Ba
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Hf
Ta
Pb
Th
U
BHVO-2
(n = 15)
BHVO-2
BHVO-1
(n = 10)
BHVO-1
Measured
Recommended
Measured
Recommended
4.5
331
286
45
116
135
102
9.0
390
24.1
165
15.29
0.10
131
14.93
37.24
5.27
24.08
5.95
2.00
6.17
0.92
5.19
0.95
2.54
0.33
1.94
0.27
4.39
1.01
1.77
1.47
0.36
5.0
317
280
45
119
127
103
9.8
389
26.0
172
18.00
4.63
398.1
340.3
44.55
148.69
143.11
170.43
9.12
390.5
23.2
163.8
15.3
0.10
127.1
14.7
37.2
5.17
23.6
5.99
1.97
6.21
0.90
5.13
0.94
2.49
0.32
1.91
0.27
3.91
0.88
1.85
1.36
0.27
28
179
19
0.13
139
16
39
5.4
25
6.4
2.06
6.4
0.96
5.2
0.99
130
15
38
25.0
6.20
6.30
0.90
1.04
2.00
0.28
4.10
1.40
1.20
0.33
2.00
0.29
4.4
1.2
2.6
1.1
302
A. Polat et al. / Lithos 100 (2008) 293–321
garnet ± plagioclase ± cheelite ± titanite (Fig. 6e). Stage
I metasomatic assemblage is composed predominantly
of epidote (now mostly diopside) + quartz + plagioclase ±
hornblende ± scapolite (Table 1). Stage II metasomatic
assemblage consists mainly of diopside + garnet + amphibole + plagioclase + quartz ± vesuvianite ± scapolite ± epidote ± titanite ± calcite ± scheelite (Table 1; Fig. 6f).
Amphibolites (foliated pillows) are composed mainly of
hornblende + plagioclase + quartz ± diopside ± epidote ±
titanite ± sulphide.
White inclusions in gabbros have an assemblage
of plagioclase (90%) + amphibole (5–10%) + quartz (0–
5%), consistent with an anorthositic composition
(Fig. 7; Table 1). Because of extensive recrystallization,
magmatic plagioclase is rarely present. Amphiboles in
the anorthositic inclusions typically have dark green to
blue green pleochroism and range from subhedral to
euhedral.
4. Analytical methods and data presentation
All whole-rock samples were powdered using an agate
mill in the Department of Earth Sciences, University of
Windsor, Canada. Major and some trace elements (Zr, Sc,
Ni) were determined by Thermo Jarrel-Ash ENVIRO II
ICP at ACTLABS in Ancaster, Canada. Samples were
mixed with a flux of lithium metaborate and lithium
tetraborate, and were fused in an induction furnace.
Molten sample was immediately poured into a solution of
5% nitric acid containing an internal standard, and was
mixed continuously until completely dissolved. Totals of
major element oxides are 100 ± 1 wt.% and the analytical
precisions are 1 to 2%.
Samples were analyzed for REE, HFSE, LILE, and
transition metals (Co, Cr, and V) by a high-sensitivity
Thermo Elemental X7 ICP-MS in the Great Lakes
Institute for Environmental Research (GLIER), University
Table 3
Summary of major (wt.%) and trace (ppm) element concentrations and significant element ratios for the Ivisaartoq rocks
SiO2 (wt.%)
TiO2
Al2O3
Fe2O3
MgO
CaO
Mg-number
Cr (ppm)
Co
Ni
Sc
V
Nb
Zr
Th
Y
La
Nd
Sm
Gd
Yb
La/Ybcn
La/Smcn
Gd/Ybcn
Eu/Eu⁎
Al2O3/TiO2
Nb/Ta
Y/Ho
Zr/Y
Ti/Zr
Zr/Zr⁎
Nb/Nb⁎
Ti/Ti⁎
Actinolite schist
Cumulates
Pillow lavas
Gabbros
Diorites
Anorthositic inclusions
Gabbroic matrix
44.2–55.4
0.10–0.67
4.0–15.2
6.8–12.4
15.5–25.6
4.5–12.4
72.7–86.7
1325–12700
68–107
430–1520
12–50
100–550
0.16–2.68
11–25
0.06–0.30
2.6–21.6
0.23–80.0
0.63–54.0
0.26–7.42
0.42–6.62
0.34–2.44
0.47–23.5
0.64–7.74
0.84–1.91
0.77–2.16
23–38
4.7–17.4
3.9 –72.0
1.1–6.0
59–163
0.1–1.9
0.01–0.96
0.25–1.78
48.7–50.3
0.27–0.28
6.3–7.1
9.2–10.4
22.3–23.5
9.6–10.0
81.7–83.2
1575–1670
80–84
730–830
24–26
100–170
0.69–0.77
12.5–15.6
0.22–0.36
7.0–8.4
1.59–1.83
2.35–2.85
0.69–0.87
0.98–1.22
0.80–1.00
1.27–1.60
1.50–1.90
0.97–1.02
0.61–0.81
23.5–25.6
11.5–16.9
25.5–28.5
1.8–2.1
103–128
0.66–0.78
0.31–0.41
0.64–0.81
47.7–55.7
0.37–0.75
8.7–15.2
7.7–13.9
4.5–18.8
8.5–17.2
53.6–76.9
62–5700
48–96
125–705
26.5–43.4
136–485
0.09–1.61
22.3–42.4
0.37–0.79
9.4–15.2
2.25–3.56
3.66–5.26
1.10–1.90
1.52–2.47
1.08–1.69
1.09–2.19
0.97–2.31
1.00–1.20
0.63–1.05
20.3–26.3
13.3–16.3
25.7–28.7
2.0–2.9
76–116
0.52–0.94
0.28–0.63
0.64–0.82
47.6–51.0
0.50–1.00
13.6–16.0
9.0–13.2
7.8–14.1
8.3–12.3
53.9–75.6
230–1060
45–61
87–230
33–41
190–475
0.10–1.77
28.2–48.5
0.22–0.74
13.6–19.8
2.21–7.98
4.12–9.47
1.26–2.47
1.83–2.94
1.49–2.20
0.80–3.30
0.90–2.30
1.00–1.40
0.70–1.00
15.8–27.0
8.7–16.7
24.5–27.0
1.2–2.7
94–123
0.50–0.90
0.30–0.80
0.60–0.93
55.2–57.1
0.64–1.14
15.0–16.9
6.2–8.0
3.8–7.6
9.2–12.0
50.6–70.1
180–1030
36–48
60–120
35–48
230–300
1.33–3.06
36.3–68.4
0.15–0.36
9.5–22.2
2.44–6.63
4.50–10.30
1.34–2.76
1.70–3.47
1.08–2.58
1.05–2.05
0.89–1.89
1.10–1.30
0.89–1.07
14.8–23.5
12.6–16.0
23.5–27.1
1.2–3.8
95–106
0.86–1.00
0.30–0.90
0.80–1.00
47.4–49.0
0.04–0.08
29.1–30.3
2.4–3.2
1.1–1.6
14.0–15.9
43.6–49.5
6–11
6–9
b2
1–3
14–21
0.10–0.21
3.6–21.1
0.04–0.12
1.3–2.2
0.68–0.98
0.61–1.00
0.15–0.24
0.19–0.32
0.16–0.21
2.80–3.32
2.60–3.35
0.94–1.28
4.1–6.5
390–673
46.9–50.4
0.73–1.08
15.8–16.4
10.7–13.0
8.1–8.6
10.6–11.1
56.6–60.6
210–240
45–50
90–140
38–42
240–284
1.45–2.15
48–57
0.33–0.38
18–20
2.67–3.09
6.38–7.25
2.04–2.36
2.90–3.21
2.07–2.29
0.92–1.00
0.92–0.97
1.12–1.16
0.84–0.94
15–22
12.7–14.6
24–26
2.6–2.9
17–23
0.86–0.95
0.53–0.82
0.69–0.91
29–32
1.8–9.8
13–90
0.50–3.05
0.11–0.16
0.40–0.64
A. Polat et al. / Lithos 100 (2008) 293–321
303
Table 4
Major (wt.%) and trace (ppm) element concentrations and significant element ratios for the Ivisaartoq actinolite schists, mafic to ultramafic pillows,
cumulates, gabbros, and diorites
Actinolite schists
SiO2 (wt.%)
TiO2
Al2O3
Fe2O3
MnO
MgO
CaO
K2O
Na2O
P2O5
LOI
Mg-number
Cr (ppm)
Co
Ni
Rb
Sr
Cs
Ba
Sc
V
Nb
Ta
Zr
Th
U
Y
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Cu
Zn
Ga
Pb
La/Ybcn
La/Smcn
Gd/Ybcn
Ce/Ce⁎
Eu/Eu⁎
Al2O3/TiO2
Nb/Ta
Y/Ho
Zr/Y
Ti/Zr
485426
485427
485430
485431
485433
485434
455435
485436
485437
52.35
0.21
5.67
9.11
0.13
21.83
10.12
0.02
0.54
0.01
3.11
82.6
12733
86
1249
0.2
29
0.1
5
20.6
199
0.38
0.02
12.4
0.16
0.26
4.8
2.179
4.132
0.447
1.802
0.440
0.201
0.641
0.127
0.846
0.194
0.613
0.092
0.630
0.100
38.1
70.1
14.1
1.0
2.48
3.55
0.84
1.03
1.16
26.6
15.2
24.8
2.6
103
52.67
0.21
5.59
8.96
0.14
21.28
10.54
0.01
0.59
0.01
2.76
82.5
12484
84
1192
0.2
30
0.1
5
18.5
207
0.41
0.03
11.3
0.12
0.44
4.9
2.842
6.216
0.693
2.538
0.522
0.268
0.684
0.137
0.859
0.196
0.604
0.095
0.576
0.098
36.7
59.6
13.7
0.9
3.54
3.90
0.98
1.09
1.37
27.3
13.6
24.7
2.3
108
46.39
0.44
12.68
11.52
0.22
15.52
11.44
0.30
1.46
0.03
1.78
72.7
1988
68
477
11.8
28
3.1
56
25.4
287
1.11
0.08
28.1
0.06
0.02
2.6
3.440
7.244
0.972
4.464
1.332
0.482
1.671
0.304
1.995
0.410
1.317
0.177
1.132
0.174
3.0
139.0
34.1
1.9
2.18
1.85
1.22
0.97
0.99
28.8
13.6
6.4
2.7
94
53.35
0.10
3.98
6.84
0.12
22.59
12.42
0.04
0.56
0.01
2.76
86.7
6715
84
1520
0.2
48
0.1
35
12.3
100
0.16
0.01
10.7
0.07
0.03
7.5
0.229
0.860
0.108
0.634
0.255
0.083
0.424
0.071
0.502
0.103
0.337
0.048
0.347
0.052
19.0
170.6
15.9
5.2
0.47
0.64
1.01
1.34
0.77
37.9
11.4
72.0
4.1
59
44.17
0.30
7.40
11.99
0.18
25.58
10.02
0.01
0.33
0.02
3.27
80.9
11438
99
1241
0.3
190
0.3
13
25.6
260
0.50
0.04
11.3
0.20
0.25
6.8
0.687
2.044
0.332
1.858
0.702
0.245
1.037
0.193
1.393
0.295
0.952
0.135
0.964
0.152
44.9
112.1
17.8
6.4
0.51
0.70
0.89
1.05
0.88
24.8
12.4
23.0
1.5
158
45.67
0.39
11.48
12.41
0.15
19.22
9.57
0.15
0.93
0.03
0.39
75.4
7298
102
1106
8.7
32
2.6
35
31.2
269
0.72
0.05
19.6
0.06
0.08
8.1
22.238
44.179
4.915
17.358
2.630
0.945
1.683
0.252
1.384
0.277
0.843
0.121
0.816
0.143
2.64
99.78
25.6
5.6
19.53
6.06
1.70
1.04
1.28
29.2
15.9
29.2
2.9
120
46.47
0.31
9.46
12.28
0.20
21.18
9.52
0.05
0.50
0.02
0.90
77.4
11896
94
767
0.5
89
0.1
17
32.4
338
2.68
0.57
20.5
0.30
0.37
21.6
2.659
6.741
0.928
3.967
1.029
0.543
1.241
0.211
1.562
0.326
0.994
0.152
1.059
0.167
5.06
181.90
23.7
2.6
1.80
1.85
0.97
1.05
1.47
30.3
4.7
66.1
2.5
91
45.38
0.67
15.22
9.63
0.23
24.09
4.51
0.01
0.24
0.02
0.75
83.2
1325
107
431
1.8
30
1.0
10
49.9
550
0.77
0.05
24.7
0.09
0.08
3.7
80.022
151.484
16.151
53.958
7.415
3.602
5.644
0.777
4.515
0.945
2.795
0.391
2.443
0.357
2.72
143.58
28.1
2.3
23.49
7.74
1.91
1.03
1.64
22.6
16.4
3.9
1.1
163
50.84
0.47
10.92
9.28
0.16
15.53
10.36
0.04
2.39
0.02
2.04
76.8
4794
87
877
1.2
226
0.3
38
35.7
414
0.78
0.04
22.3
0.09
0.08
3.7
0.912
2.220
0.357
1.802
0.572
0.439
0.669
0.128
0.795
0.159
0.515
0.069
0.419
0.068
2.4
78.0
25.7
6.0
1.56
1.14
1.32
0.95
2.16
23.4
17.4
23.3
6.0
125
(continued on next page)
(continued on next page)
304
A. Polat et al. / Lithos 100 (2008) 293–321
Table 4 (continued )
Actinolite schists
Zr/Zr⁎
Nb/Nb⁎
Ti/Ti⁎
North
West
485426
485427
485430
485431
485433
485434
455435
485436
485437
0.96
0.12
0.95
64°44.903′
049°53.261′
0.68
0.12
0.81
64°44.903′
049°53.261′
0.80
0.25
0.70
64°44.900′
049°53.466′
1.85
0.96
0.76
64°44.900′
049°53.466′
0.69
0.81
0.83
64°44.940′
049°53.402′
0.20
0.02
0.44
64°44.940′
049°53.402′
0.70
0.95
0.65
64°44.940′
049°53.402′
0.09
0.01
0.25
64°44.943′
049°53.379′
1.52
0.77
1.78
64°44.962′
049°53.313′
Cumulates
SiO2
TiO2
Al2O3
Fe2O3
MnO
MgO
CaO
K2O
Na2O
P2O5
LOI
Mg-number
Cr
Co
Ni
Rb
Sr
Cs
Ba
Sc
V
Nb
Ta
Zr
Th
U
Y
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Lu
Cu
Zn
Ga
Pb
La/Ybcn
La/Smcn
Gd/Ybcn
Ce/Ce⁎
Pillow lavas
485473 a
485474 a
485475 a
50.34
0.27
6.36
9.21
0.17
23.09
10.23
0.02
0.30
0.01
4.15
83.2
1575
84
845
0.8
33
0.2
1.9
24
104
0.77
0.05
12.5
0.34
0.293
7.0
1.69
4.10
0.52
2.35
0.73
0.23
0.98
0.18
1.20
0.26
0.85
0.13
0.84
0.13
10.8
79.2
15.0
1.6
1.45
1.65
0.97
1.07
48.73
0.28
7.05
10.42
0.17
23.51
9.56
0.01
0.25
0.02
4.64
81.7
1670
84
852
0.7
31
0.2
10.9
24
112
0.73
0.05
14.6
0.26
0.668
7.1
1.59
3.75
0.54
2.43
0.70
0.22
0.98
0.17
1.26
0.27
0.82
0.12
0.81
0.12
14.4
91.2
17.8
1.7
1.41
1.62
1.00
1.00
48.84
0.27
6.28
9.90
0.17
22.29
12.03
0.01
0.20
0.01
3.97
81.7
1654
83
735
0.7
33
0.1
6.1
26
107
0.69
0.06
15.6
0.36
0.183
8.4
1.82
4.26
0.63
2.89
0.87
0.21
1.22
0.22
1.52
0.33
0.99
0.14
1.03
0.15
6.7
80.6
16.1
1.8
1.27
1.50
0.98
0.97
485478
1617
80
857
0.6
36
0.2
2.45
172
0.67
0.05
0.23
0.271
7.04
1.83
3.95
0.53
2.44
0.69
0.22
1.01
0.17
1.19
0.25
0.79
0.12
0.82
0.12
18.3
81.7
16.2
0.33
1.60
1.89
1.02
0.98
485418 a
485420 a
485422
485424
485468
485469
485470
48.41
0.43
10.87
12.51
0.22
14.90
11.26
0.15
1.23
0.02
1.50
70.2
5713
93
605
12.3
43
4.7
45
35.5
178
0.94
0.07
22.3
0.40
0.101
11.0
2.25
4.95
0.76
3.66
1.11
0.44
1.64
0.29
1.96
0.42
1.30
0.19
1.24
0.18
15.5
93.5
30.7
2.2
1.30
1.45
1.09
0.93
52.06
0.51
12.01
11.17
0.19
10.12
12.66
0.01
1.23
0.04
0.82
64.2
2999
62
331
0.4
90
0.1
30
37.3
199
1.26
0.09
33.3
0.708
0.192
12.3
3.31
7.83
1.06
4.96
1.39
0.46
1.85
0.33
2.28
0.48
1.46
0.21
1.40
0.21
78.3
108.7
32.2
1.7
1.70
1.70
1.09
1.02
47.72
0.45
10.58
12.31
0.20
17.27
9.64
0.90
0.89
0.04
2.93
73.5
5282
83
647
148.5
16
29.4
137
35.0
173
1.09
0.07
27.5
0.55
0.167
10.9
2.39
5.60
0.77
3.69
1.11
0.38
1.59
0.28
1.89
0.41
1.27
0.18
1.19
0.18
8.0
98.9
46.0
1.3
1.44
1.55
1.11
1.01
55.661
0.516
12.489
7.724
0.227
4.513
17.204
0.050
1.585
0.030
0.776
53.6
2325
53
235
8.7
69
0.9
23
38.3
421
1.32
0.09
35.3
0.70
0.194
13.4
3.53
8.00
1.10
5.14
1.48
0.53
1.92
0.35
2.31
0.50
1.53
0.22
1.51
0.23
38.5
68.0
32.3
2.0
1.68
1.71
1.05
0.99
50.09
0.60
14.12
11.60
0.19
9.75
10.78
0.14
2.67
0.05
0.85
62.5
230
57
125
2.0
123
0.0
37
37.3
234
1.54
0.11
36.3
0.79
0.190
14.7
2.98
7.60
1.09
5.26
1.67
0.58
2.17
0.39
2.65
0.57
1.73
0.23
1.50
0.22
43.1
104.7
39.6
5.3
1.42
1.28
1.19
1.03
50.95
0.56
14.49
10.85
0.17
8.90
11.27
0.14
2.63
0.05
0.69
61.9
300
55
128
2.0
120
0.0
58
37.3
294
1.51
0.09
30.2
0.48
0.183
14.4
2.85
7.08
1.01
4.94
1.56
0.57
1.99
0.36
2.56
0.55
1.59
0.23
1.47
0.22
40.7
99.1
42.2
3.2
1.39
1.32
1.12
1.02
51.74
0.54
14.32
10.33
0.17
8.59
11.94
0.16
2.16
0.05
0.60
62.2
300
57
141
2.7
96
0.0
49
36.2
291
1.42
0.10
28.2
0.48
0.145
14.1
2.84
6.99
1.00
4.83
1.43
0.56
1.90
0.37
2.44
0.51
1.61
0.23
1.49
0.21
94.9
91.2
40.3
11.2
1.37
1.42
1.06
1.02
A. Polat et al. / Lithos 100 (2008) 293–321
305
Table 4 (continued )
Cumulates
485473
Eu/Eu⁎
Al2O3/TiO2
Nb/Ta
Y/Ho
Zr/Y
Ti/Zr
Zr/Zr⁎
Nb/Nb⁎
Ti/Ti⁎
North
West
Pillow lavas
a
485474
0.82
23.6
16.9
27.5
1.8
128
0.66
0.37
0.78
64°
44.763′
049°
51.318′
a
0.81
25.1
13.9
26.5
2.1
115
0.78
0.41
0.83
64°
44.763′
049°
51.318′
485475
0.61
23.5
11.5
25.5
1.9
103
0.68
0.31
0.64
64°
44.763′
049°
51.318′
a
485478
485418 a
485420 a
485422
485424
485468
485469
485470
0.78
0.99
25.1
14.4
25.9
2.0
116
0.77
0.36
0.79
64°
.375′
049°
.130′
0.88
23.5
14.1
25.6
2.7
92
0.88
0.30
0.78
64°
44.334′
049°
.183′
0.88
23.7
15.7
26.4
2.5
97
0.94
0.40
0.79
64°
44.361′
049°
56.678′
0.97
24.2
14.5
26.6
2.6
88
0.88
0.32
0.72
64°
44.325′
049°.
342′
0.93
23.5
14.2
25.7
2.5
99
0.85
0.49
0.75
64°
44.798′
049°
51.544′
0.98
25.9
16.3
26.2
2.1
110
0.75
0.49
0.75
64°
44.812′
049°
.477′
1.05
26.3
14.6
27.4
2.0
116
0.74
0.46
0.78
64°
44.812′
049°
51.477′
14.7
28.5
0.29
64°
44.761′
049°
51.332′
Pillow lavas
485481
SiO2
TiO2
Al2O3
Fe2O3
MnO
MgO
CaO
K2O
Na2O
P2O5
LOI
Mg-number
Cr
Co
Ni
Rb
Sr
Cs
Ba
Sc
V
Nb
Ta
Zr
Th
U
Y
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
50.39
0.35
8.69
11.08
0.20
18.61
8.46
1.33
0.86
0.03
2.05
76.9
1298
81
608
275.8
69
60.8
216
27.6
142
1.11
0.08
25.5
0.68
0.148
10.9
3.56
7.80
1.00
4.34
1.10
0.41
1.54
0.25
1.84
0.38
1.19
0.18
1.24
a
Gabbros
485482
49.75
0.37
9.02
10.65
0.18
17.10
11.50
0.03
1.38
0.02
1.80
76.1
1414
81
659
2.3
76
0.5
10
26.5
136
1.29
0.09
29.5
0.83
0.286
10.7
3.34
7.96
1.04
4.57
1.17
0.41
1.54
0.27
1.75
0.39
1.15
0.18
1.21
a
485486
48.31
0.37
8.70
13.91
0.25
18.77
8.62
0.08
0.96
0.03
2.46
72.8
1640
81
705
1.0
61
0.6
31
27.7
204
1.20
0.09
27.7
0.72
0.260
9.4
3.31
7.52
0.93
4.02
1.19
0.28
1.52
0.25
1.70
0.36
1.06
0.16
1.08
485423
3153
75
378
4.5
46
0.7
44
485
1.40
0.09
22.4
0.66
0.228
14.1
3.56
8.86
1.19
5.53
1.60
0.60
2.15
0.37
2.44
0.55
1.67
0.24
1.54
Diorites
485477
499730A
50.98
0.62
16.04
9.76
0.16
8.08
11.87
0.04
2.40
0.04
0.72
62.1
323
53
127
1.7
54
0.1
12
37
213
1.68
0.10
33.2
0.50
0.26
13.94
3.40
8.43
1.22
5.57
1.56
0.60
2.22
0.36
2.42
0.52
1.53
0.22
1.49
47.55
1.00
15.83
12.34
0.21
9.21
11.50
0.53
1.76
0.07
1.02
0.6
232
57
152
63.5
90
14.1
61
41
276
2.13
0.24
48.5
0.34
0.09
19.55
2.93
8.40
1.34
6.71
2.14
0.76
2.96
0.53
3.56
0.76
2.31
0.34
2.20
485432
842
45
87
1.5
51
0.1
29
n.d.
475
1.86
0.13
0.52
0.19
15.38
2.44
7.68
1.22
5.99
1.96
0.68
2.61
0.46
3.06
0.63
1.86
0.27
1.66
499739
485429
57.02
0.65
14.96
6.22
0.12
7.58
10.14
0.23
3.01
0.05
0.81
0.71
407
36
60
2.9
62.8
0.1
66.1
47.4
235
1.33
0.11
36.29
0.79
0.18
9.45
3.09
7.28
1.00
4.52
1.34
0.53
1.70
0.29
1.96
0.40
1.16
0.17
1.08
55.18
0.64
15.16
8.02
0.13
6.99
11.40
0.11
2.31
0.05
1.06
63.3
1029
42
93
1.6
107
0.1
59
48
247
1.56
0.11
40.4
0.69
0.15
11.64
4.32
10.04
1.31
6.04
1.74
0.59
2.14
0.37
2.39
0.49
1.40
0.18
1.33
(continued on next page)
306
A. Polat et al. / Lithos 100 (2008) 293–321
Table 4 (continued )
Pillow lavas
485481
Lu
Cu
Zn
Ga
Pb
La/Ybcn
La/Smcn
Gd/Ybcn
Ce/Ce⁎
Eu/Eu⁎
Al2O3/TiO2
Nb/Ta
Y/Ho
Zr/Y
Ti/Zr
Zr/Zr⁎
Nb/Nb⁎
Ti/Ti⁎
North
West
a
a
0.18
2.6
92.3
52.6
1.9
2.06
2.31
1.03
1.01
0.96
25.0
13.3
28.7
2.3
81
0.81
0.26
0.66
64°44.515′
049°53.299′
Gabbros
485482
a
0.19
5.2
78.3
23.9
1.3
1.98
2.05
1.05
1.05
0.93
24.1
14.2
27.6
2.8
76
0.88
0.28
0.68
64°.515′
049°53.361′
Diorites
485486
485423
485477
499730A
485432
0.16
25.7
110.7
24.9
2.8
2.19
1.99
1.16
1.05
0.63
23.8
13.5
25.9
2.9
79
0.88
0.31
0.64
64°.438′
049°54.399′
0.25
84.5
97.2
36.7
1.7
1.66
1.59
1.16
1.06
0.99
0.23
44.4
116.4
38
7.46
1.64
1.56
1.23
1.02
0.99
25.9
16.7
27.0
2.4
112
0.78
0.46
0.79
0.33
86.5
77.5
31
1.41
0.95
0.96
1.11
1.04
0.91
15.8
8.7
25.8
2.5
123
0.87
0.78
0.93
64°44.767′
049°51.576′
0.27
19.0
69.2
37
4.44
1.05
0.89
1.30
1.09
0.92
15.2
25.4
0.52
0.37
64°.355′
049°56.753′
499739
0.16
22.8
54.5
27.6
4.9
2.05
1.63
1.30
1.02
1.07
23.1
14.6
12.7
24.5
23.74
1.2
3.8
106
1.01
0.89
0.38
1.00
64 44.492
64°44.742
049°56.613 049°54.082
485429
0.19
263.2
66.8
45
2.54
2.33
1.79
1.33
1.03
0.94
23.5
14.4
23.5
3.5
95
0.86
0.31
0.79
64°44.776′
049°53.759′
Published in Polat et al. (2007).
of Windsor, Canada. Wet chemical procedures were
conducted under clean lab conditions, and all acids were
distilled twice. Approximately 100–130 mg of sample
powder was used for dissolution. Samples were dissolved
in a concentrated HF–HNO3 mixture at a temperature of
∼120 °C for four days, and then further attacked with
concentrated HNO3 until no residue was visible. BHVO-1
and BHVO-2 were used as international reference
materials to estimate precision and accuracy (Table 2).
Analytical precisions are estimated as follows: 1 to 10 %
for REE, Rb, Li, Cs, Sr, Ba, Y, Nb, Cu, Zn, and Pb; 10 to
20 % for Zr, V, Cr, Co, and U; and 20 to 30 % for Ta and Th
(Table 2).
Selected elements are normalized to primitive mantle
(pm) (Hofmann, 1988) and chondrite (cn) (Sun and McDonough, 1989). Nb/Nb⁎, Zr/Zr⁎, Ce/Ce⁎, and Eu/Eu⁎
ratios, representing anomalies, were calculated with respect
to the neighboring immobile elements, following the
method of Taylor and McLennan (1985). Samples were
recalculated to 100 % anhydrous for inter-comparisons.
Mg-numbers (%) were calculated as the molecular ratio of
Mg /(Mg+ Fe2+), where Fe2+ is assumed to be 90% of total
Fe.
Whole-rock Pb and Sm–Nd isotope analyses were
carried out on a VG Sector 54-IT TIMS in the Geological
Institute, University of Copenhagen, Denmark. Dissolution of the powder samples was achieved in two successive, but identical steps, which consisted of a strong 8N
HBr attack that has been shown to effectively dissolve
accessory phosphates (Frei et al., 1997; Schaller et al.,
1997), followed by a concentrated HF–14N HNO3 mixture and finally by strong 9N HCl. Independent dissolutions were performed for REE and Pb analyses. A mixed
150
Nd–149Sm spike was added to the REE aliquot beforehand. Chemical separation of REEs was carried out on
conventional cation exchange columns, followed by an
Sm–Nd separation using HDEHP-coated beads (BIORAD) charged in 6 ml quartz glass columns. Neodymium
ratios were normalized to 146Nd/144Nd =0.7219. The mean
value for our internal JM Nd standard (referenced against
La Jolla) during the period of measurement was 0.511098
for 143Nd/144Nd, with a 2σ external reproducibility of ±
0.000011 (seven measurements). Procedural blanks run
during the period of these analyses show insignificant blank
levels of ∼5 pg Sm and ∼12 pg Nd. Precisions for
concentration analysis are approximately 0.5% for Sm and
Nd. Initial εNd values were calculated at 3075 Ma U–Pb
zircon ages obtained from volcanoclastic rocks (Friend and
Nutman, 2005).
5. Geochemical results
5.1. Major and trace elements
5.1.1. Clinopyroxene cumulates (picrites) and actinolite
schists
Field relationships suggest that actinolite schists were
derived from clinopyroxene cumulates by increasing
A. Polat et al. / Lithos 100 (2008) 293–321
intensity of deformation and metamorphism. Cumulates
have more uniform major and trace element compositions than actinolite schists (Tables 3, 4; Figs. 8, 9).
Cumulates have sub-chondritic Nb/Ta (11.5–16.9) and
Zr/Y (1.8–2.1) ratios, and slightly super-chondritic
Al2O3/TiO2 (23–25) ratios (Tables 3, 4). On chondriteand primitive mantle-normalized diagrams, they have the
following trace element characteristics: (1) moderately
307
enriched LREE (La/Smcn = 1.50–1.90; La/Ybcn = 1.30–
1.60) patterns; (2) flat HREE (Gd/Ybcn = 0.97–1.02)
patterns; and (3) negative Eu (Eu/Eu⁎ = 0.61–0.82), Nb
(Nb/Nb⁎ = 0.29–0.41), Zr (Zr/Zr⁎ =0.66–0.78), and Ti (Ti/
Ti⁎ = 0.64–0.83) anomalies (Fig. 8a, e; Table 4).
Despite their simple mineralogical composition
(Fig. 6b; Table 1), actinolite schists display large
variations in Al2O3 (4.0–15.2 wt.%), Cr (1325–
Fig. 8. Chondrite-normalized REE and primitive mantle-normalized trace element patterns for cumulates, pillow lavas, gabbros, and diorites.
Chondrite normalization values are from Sun and McDonough (1989) and primitive mantle normalization values are from Hofmann (1988).
308
A. Polat et al. / Lithos 100 (2008) 293–321
12500 ppm), Ni (430–1250 ppm), V (100–550 ppm),
TiO2 (0.10–0.67 wt.%), Nb (0.16–2.70 ppm), and Y
(2.6–21.6 ppm) (Tables 3, 4). They have moderate
variations in SiO2 (44–53 wt.%), MgO (15.5–25.6 wt.%),
Fe2O3 (6.9–12.3 wt.%), Zr (11–25 ppm), Ga (14–34 ppm),
and Co (84–106 ppm). Abundances of REE (e.g.
La = 0.23–80 ppm; Ce = 0.9–152 ppm) are extremely
scattered (Table 4). On a chondrite-normalized diagram,
they have the following characteristics: (1) variably
depleted to strongly enriched LREE patterns (La/
Smcn = 0.6–7.7; La/Ybcn = 0.50–23.50); (2) slightly depleted to enriched HREE (Gd/Ybcn = 0.84–1.90) patterns; (3)
negative to positive Eu (Eu/Eu⁎ = 0.9–2.16) anomalies; and
(4) positive Ce (Ce/Ce⁎ =0.97–1.34) anomalies (Fig. 9).
On a primitive mantle-normalized diagram, they have
the following significant features: (1) variably negative
Nb (Nb/Nb⁎ = 0.02–0.96); and (2) negative to positive
Ti (Ti/Ti⁎ =0.2–1.8) and Zr (Zr/Zr⁎ =0.09–1.85) anomalies (Fig. 9). Al2O3/TiO2 (22–38) ratios are chondritic to
super-chondritic, whereas the ratios of Zr/Y (1.4–4.0) and
Ti/Zr (59–162) range from sub-chondritic to superchondritic values (Tables 3, 4).
5.1.2. Pillow lavas, gabbros, and diorites
Pillow lavas are basaltic, but have a variable composition (Tables 3, 4). Ti/Zr (76–116) and Zr/Y (2.03–2.94)
ratios range from sub-chondritic to slightly super-chondritic
values. Al2O3/TiO2 (24–25) ratios are slightly super-chondritic. The ratios of Nb/Ta (13.3–16.3) and Y/Ho (25.6–
28.7) tend to be sub-chondritic. In addition, they have the
following trace element characteristics: (1) flat to enriched
LREE (La/Smcn = 0.97–2.31; La/Ybcn = 1.10–2.20) patterns; (2) flat to slightly enriched HREE (Gd/Ybcn =1.03–
1.19) patterns; and (3) negative Nb (Nb/Nb⁎ = 0.26–0.63),
Zr (Zr/Zr⁎ = 0.52–0.94), and Ti (Ti/Ti⁎ = 0.64–0.82)
anomalies (Fig. 8).
The following compositional ranges in gabbros and
diorites represent the new and previously published
(Polat et al., 2007) data (Table 3, 4). Major element
compositions of these rocks are similar to those of pillow
lavas (Tables 3, 4). There are large variations in Ni (121–
234 ppm), Cr (246–1060 ppm), and REE (e.g., La = 2.2–
8.0 ppm), and moderate variations in Co (52–61 ppm), V
(186–264), Zr (28–44 pm), and Y (14–20 ppm) (Tables
3, 4). Mg-numbers vary between 54 and 76 (Tables 3, 4).
The ratios of Al2O3/TiO2 (16–27), Zr/Y (2.1–2.7), and
Ti/Zr (94–121) extend from sub-chondritic to superchondritic values (see Sun and McDonough, 1989). Nb/
Ta (8.7–16.7) and Y/Ho (23.5–27.0) ratios are subchondritic. In addition, they have the following trace
element characteristics: (1) slightly depleted to moderately enriched LREE (La/Sm cn = 0.90–2.30; La/
Ybcn = 0.80–3.20) patterns; (2) flat to slightly enriched
HREE (Gd/Ybcn =1.00–1.37) patterns; and (3) variably
large negative Nb (Nb/Nb⁎ = 0.29–0.90) and Ti (Ti/
Ti⁎ = 0.60–0.90) anomalies (Fig. 8; Table 3, 4). Diorites
have more evolved geochemical compositions (i.e.
higher SiO 2 , but lower MgO, Fe 2 O 3 , Ni and
Cr concentrations and Mg-numbers) than gabbros
(Tables 3, 4). The chondrite- and primitive mantle-normalized trace element patterns of diorites are similar to
those of gabbros (Fig. 8).
5.1.3. Anorthositic inclusions and surrounding gabbroic
matrix
The anorthositic inclusions have high concentrations of Al2O3 (29–30 wt.%) and CaO (14–16 wt.%)
(Table 5). They have extremely low Ni (b 2 ppm),
Fig. 9. Chondrite-normalized REE and primitive mantle-normalized trace element patterns for actinolite schists. Non-coherent patterns in primitive
mantle-normalized diagram reflect the mobility of LREE. Given that sample with very different REE patterns (and concentrations) have similar Th,
Nb, Zr, and Ti concentrations, we suggest that these elements were less mobile than REEs. Chondrite normalization values are from Sun and
McDonough (1989) and primitive mantle normalization values are from Hofmann (1988).
Table 5
Major (wt.%) and trace (ppm) compositions and significant element ratios of anorthositic inclusions and surrounding gabbroic matrix
Anorthositic inclusions
499728-A1
499729-A1
499729-B1
499731-A1
499731-B1
499731-C1
499728-A2
499729-A2
499729-B2
499731-A2
499731-B2
499731-C2
47.39
0.04
30.09
3.16
0.05
1.47
15.21
0.16
2.42
0.01
0.44
48.0
6.2
9.4
b2
8.4
393
1.88
62
1.00
15.3
n.d.
0.192
5.0
0.054
0.017
2.21
0.950
1.962
0.225
0.959
0.235
0.392
0.288
0.052
0.321
0.071
0.207
0.031
0.205
48.00
0.05
30.36
2.42
0.04
1.20
15.87
0.03
2.02
0.02
0.49
49.5
5.7
8.0
b2
1.9
399
0.42
25
2.01
17.1
n.d.
0.178
3.6
0.038
0.017
2.03
0.986
1.718
0.245
1.005
0.253
0.363
0.281
0.054
0.313
0.067
0.219
0.031
0.212
48.99
0.05
29.43
3.06
0.04
1.35
13.95
0.17
2.94
0.02
0.36
46.7
8.3
7.4
b2
18.4
436
4.28
64
2.01
17.4
n.d.
0.123
3.6
0.036
0.028
1.46
0.682
1.274
0.166
0.698
0.187
0.393
0.203
0.033
0.223
0.046
0.154
0.025
0.175
48.46
0.05
30.16
2.93
0.06
1.14
14.23
0.16
2.80
0.01
0.37
43.6
5.6
6.4
b2
16.8
429
4.57
54
b1
16.4
n.d.
0.124
5.0
0.128
0.022
1.32
0.720
1.291
0.157
0.619
0.153
0.367
0.190
0.029
0.186
0.042
0.146
0.022
0.167
48.84
0.05
29.10
3.24
0.05
1.56
14.09
0.18
2.89
0.01
0.31
48.7
5.8
8.8
b2
15.1
408
3.71
78
2.01
14.0
n.d.
0.103
4.2
0.042
0.031
1.72
0.779
1.727
0.190
0.776
0.196
0.355
0.242
0.038
0.246
0.054
0.164
0.025
0.173
48.44
0.08
29.52
3.17
0.05
1.38
14.68
0.06
2.62
0.01
0.01
46.3
10.9
7.8
b2
3.5
383
1.28
44
3.00
20.9
n.d.
0.208
7.5
0.038
0.020
2.03
0.946
1.735
0.213
0.931
0.241
0.365
0.322
0.049
0.344
0.071
0.227
0.032
0.208
50.19
0.91
15.81
11.10
0.19
8.43
10.55
0.55
2.19
0.08
0.98
60.1
209
45
121
74
133
15.2
207
39
244
0.13
1.87
47.5
0.37
0.11
17.9
2.67
8.39
1.25
6.38
2.04
0.68
2.81
0.50
3.30
0.73
2.15
0.30
2.07
49.85
0.73
15.81
11.77
0.19
8.19
11.11
0.26
2.05
0.04
0.80
58.0
222
47
131
18
134
3.1
38
38
265
0.11
1.45
52.4
0.38
0.06
18.9
2.94
8.56
1.29
6.56
2.15
0.77
2.90
0.51
3.52
0.74
2.22
0.32
2.08
48.30
0.98
16.44
11.95
0.20
8.56
10.75
0.66
2.09
0.07
1.09
58.7
236
49
142
100
140
21.5
253
42
275
0.16
2.12
56.6
0.38
0.08
19.8
3.09
8.96
1.41
7.14
2.29
0.73
3.13
0.55
3.67
0.77
2.38
0.34
2.25
49.88
0.94
15.98
11.46
0.19
8.12
10.89
0.22
2.23
0.07
0.63
58.4
240
46
141
9
142
1.2
44
40
260
0.15
1.89
50.3
0.35
0.07
19.1
2.85
8.66
1.32
6.71
2.17
0.73
3.01
0.53
3.53
0.77
2.25
0.33
2.18
46.89
1.08
16.40
13.02
0.21
8.56
10.87
0.78
2.10
0.09
1.12
56.6
227
50
101
112
128
24.0
292
42
284
0.16
2.15
52.6
0.43
0.07
20.0
3.08
9.47
1.44
7.25
2.36
0.76
3.21
0.56
3.75
0.77
2.38
0.34
2.29
50.41
0.99
15.79
10.66
0.19
8.27
11.18
0.30
2.14
0.07
1.33
60.6
231
45
91
19
142
3.3
80
41
265
0.16
2.04
52.7
0.33
0.07
19.3
2.96
8.75
1.36
6.84
2.25
0.73
3.03
0.54
3.61
0.77
2.33
0.34
2.23
309
(continued on next page)
A. Polat et al. / Lithos 100 (2008) 293–321
SiO2
TiO2
Al2O3
Fe2O3
MnO
MgO
CaO
K2 O
Na2O
P2O5
LOI
Mg-number
Cr (ppm)
Co
Ni
Rb
Sr
Cs
Ba
Sc
V
Ta
Nb
Zr
Th
U
Y
La
Ce
Pr
Nd
Sm
Eu
Gd
Tb
Dy
Ho
Er
Tm
Yb
Gabbroic matrix
310
Table 5 (continued )
Anorthositic inclusions
499728-A1
499729-A1
499729-B1
499731-A1
499731-B1
499731-C1
499728-A2
499729-A2
499729-B2
499731-A2
499731-B2
499731-C2
0.032
30.9
36.8
38.6
4.7
27.0
3.32
2.87
1.16
4.58
1.04
674
9.58
13
17
0.16
3.05
0.40
64° 44.492′
49° 56.613′
0.030
16.8
23.7
35.5
4.1
18.5
3.33
2.77
1.10
4.12
0.86
563
1.76
91
19
0.12
0.48
0.48
64° 44.492′
49° 56.613′
0.026
34.1
23.7
37.0
3.5
30.2
2.80
2.60
0.96
6.11
0.93
561
2.49
86
18
0.13
0.69
0.63
64° 44.492′
49° 56.613′
0.025
18.2
49.8
34.8
5.2
29.4
3.09
3.35
0.94
6.54
0.94
653
7.98
26
17
0.11
2.36
0.64
64° 44.492′
49° 56.613′
0.026
37.1
18.0
37.7
3.4
31.6
3.23
2.83
1.16
4.96
1.10
592
2.45
70
21
0.11
0.74
0.53
64° 44.492′
49° 56.613′
0.031
26.9
23.5
33.5
3.4
24.9
3.26
2.81
1.28
4.00
0.95
391
3.67
60
22
0.15
1.09
0.64
64° 44.492′
49° 56.613′
0.31
26
76
43
2.00
77
0.92
0.92
1.12
0.86
1.12
17.3
2.65
116
23
0.82
0.89
0.90
64° 44.492′
49° 56.613′
0.31
27
79
27
2.34
63
1.01
0.97
1.15
0.93
1.08
21.7
2.77
84
17
0.53
0.95
0.69
64° 44.492′
49° 56.613′
0.34
23
79
51
2.72
89
0.99
0.95
1.15
0.82
1.05
16.8
2.86
104
21
0.74
0.95
0.86
64° 44.492′
49° 56.613′
0.32
31
103
27
3.97
59
0.94
0.93
1.14
0.87
1.09
17.0
2.64
110
21
0.75
0.90
0.85
64° 44.492′
49° 56.613′
0.35
25
112
55
3.52
99
0.96
0.92
1.16
0.84
1.10
15.2
2.63
121
22
0.80
0.86
0.91
64° 44.492′
49° 56.613′
0.33
54
118
30
4.99
64
0.95
0.92
1.12
0.84
1.07
16.0
2.73
111
22
0.76
0.91
0.88
64° 44.492′
49° 56.613′
n.d.: not determined.
A. Polat et al. / Lithos 100 (2008) 293–321
Lu
Cu
Zn
Ga
Pb
Li
La/Ybcn
La/Smcn
Gd/Ybcn
Eu/Eu⁎
Ce/Ce⁎
Al2O3/TiO2
Zr/Y
Ti/Zr
Ti/V
Nb/Nb⁎
Zr/Zr⁎
Ti/Ti⁎
North
West
Gabbroic matrix
A. Polat et al. / Lithos 100 (2008) 293–321
Cr (5.6–10.9 ppm), Co (6.4–9.4 ppm), Sc (1–
3 ppm), V (14–21 ppm), and TiO2 (0.04–0.08 wt.
%) contents. Mg-numbers range between 44 and
50. In addition, they display very low HFSE (Nb =
0.10–0.21 ppm; Y = 1.3–2.2 ppm) and REE (La =
0.68–0.98 ppm; Yb = 0.17–0.22 ppm) concentrations (Tables 3, 5). Al2O3/TiO2 (390–670) ratios
are extremely high. On primitive mantle- and chondrite-normalized diagrams, they have the following
significant features: (1) moderately enriched LREE
(La/Smcn = 2.60–3.35; La/Ybcn = 2.8–3.3); (2) flat to
slightly enriched HREE (Gd/Ybcn = 0.94–1.28); (3)
large positive Eu (Eu/Eu⁎ = 4.1–6.5) anomalies; and
(4) negative Nb (Nb/Nb⁎ = 0.11–0.16) and Ti (Ti/
Ti⁎ = 0.4–0.6) anomalies (Fig. 10).
The gabbroic matrix has higher MgO, Fe2O3, TiO2,
Sc, Ni, Cr, and Co, but lower CaO, Al2O3, and Sr
contents than the anorthositic inclusions (Tables 3, 5).
In addition, the matrix is characterized by higher
absolute concentrations of REE and HFSE than the
inclusions (Tables 3, 5). In comparison to the
inclusions, the matrix has less fractionated LREE
(La/Smcn = 0.92–0.97 versus 2.60–3.35) patterns, and
smaller Nb (Nb/Nb⁎ = 0.53–0.82 versus 0.12–0.16),
and Ti (Ti/Ti⁎ = 0.69–0.91 versus 0.40–0.64) anomalies (Fig. 10; Tables 3, 5).
311
5.2. Nd isotopes
5.2.1. Clinopyroxene cumulates (picrites) and actinolite
schists
Regression of the Sm–Nd isotope data for clinopyroxene cumulates and their more deformed counterpart
actinolite schists yields an errorchron age of 3092 ±
260 Ma (MSWD = 97) (Fig. 11a; Table 6; excluding
extremely altered actinolite schist samples 485434 and
485436). This age, within uncertainties, is in good
agreement with the 3075 ± 15 Ma U–Pb zircon age of
the spatially associated siliceous volcaniclastic sedimentary rocks (see Friend and Nutman, 2005). Large
uncertainty in the errorchron age is likely due to large
scatter in the data. Cumulates have a narrow range of
initial εNd (+ 4.97 to + 4.23) values, whereas actinolite
schists display large variations (+ 2.54 to + 9.53)
(Table 6). Samples (e.g., 485434, 485436) with very
large εNd (+ 8.35 and + 9.53) values have much higher
LREE concentrations (Tables 4, 6; Nd = 16–43 ppm)
and more fractionated LREE (La/Ybcn = 20–24; La/
Smcn = 6.1–7.7) patterns, compared to the rest of the
samples in the group (Fig. 9; Table 4). In addition, these
samples display the lowest 147 Sm/144 Nd (0.088–0.089)
ratios in the group (Table 6).
5.2.2. Pillow lavas, gabbros, and diorites
Pillow lavas, gabbros, and diorites define an
errorchron age of 3069 ± 220 Ma (MSWD = 80)
(Fig. 11b). This age, within uncertainties, agrees well
with the 3075 ± 15 Ma U–Pb zircon age of siliceous
volcaniclastic sedimentary rocks (Friend and Nutman,
2005; Polat et al., 2006). Large uncertainty in the
errorchron ages reflects the narrow compositional
range and large scatter in the data points. The initial
εNd values in pillow lavas (+ 1.10 to + 3.10) overlap
with, but extend to higher values than, gabbros and
diorites (+ 0.30 to + 2.39) (Table 6). Samples with
higher MgO content tend to have greater initial εNd
values (e.g., 485475, 485477). All rock types have
similar range of 147 Sm/144 Nd (Table 6).
Fig. 10. Chondrite-normalized REE and primitive mantle-normalized
trace element patterns for inclusions (xenoliths) in gabbros (see
Fig. 5a). Chondrite normalization values are from Sun and
McDonough (1989) and primitive mantle normalization values are
from Hofmann (1988).
5.2.3. Anorthositic inclusions and surrounding gabbroic matrix
The Sm–Nd isotopic compositions of the anorthositic
inclusions and surrounding gabbroic matrix are presented in Table 7. The inclusions have much higher
initial εNd values than the matrix (εNd = + 4.8 to + 6.0
versus + 2.3 to +2.8). The Nd isotopic compositions of
the inclusions are comparable to those of clinopyroxene
cumulates (Tables 6, 7). The initial εNd (+ 2.3 to +2.8)
isotopic composition of the matrix overlaps with, but
312
A. Polat et al. / Lithos 100 (2008) 293–321
Fig. 11. 147Sm/144Nd verses 143Nd/144Nd and 206Pb/204Pb versus 207Pb/204Pb errorchron diagrams for cumulates, actinolite schists, pillow lavas,
gabbros, and diorites. The isoplot program of Ludwig (1988, 2003) was used for age and initial 143Nd/144Nd ratio calculations.
extends to higher values than, gabbros (devoid of
anorthositic inclusions) and diorites (+ 0.3 to + 2.4).
5.3. Pb isotopes
5.3.1. Clinopyroxene cumulates (picrites) and actinolite
schists
On a 207Pb/204Pb versus 206Pb/204Pb isotope diagram,
cumulates and actinolite schists define an errorchron age
of 2774 ± 180 (MSWD = 167) (Fig. 11c). This age is lower
than the U–Pb zircon age (3075 Ma) of the spatially
associated volcanoclastic sedimentary rocks, but agrees,
within uncertainties, with 2781 and 2847 Ma U–Pb zircon
ages yielded by granodioritic gneisses to the west of the
Ivisaartoq greenstone belt (Friend and Nutman, 2005).
Cumulates have higher and more limited rages of
206
Pb/204Pb, 207Pb/204Pb, and 208Pb/204Pb than actinolite
schists (Table 8).
5.3.2. Pillow lavas, gabbros, and diorites
Pillow lavas, gabbros, and diorites define an errorchron
age of 3058± 240 Ma (MSWD = 505) (Fig. 11d). Despite
the large errors, this age agrees with the Sm–Nd (3069 ±
220 Ma) errorchron and U–Pb zircon (3075 ± 15 Ma) ages.
Pillow lavas display larger variations in 206Pb/204Pb and
207
Pb/204Pb ratios than gabbros and diorites (Table 8).
Gabbros tend to have larger 206Pb/204Pb and 207Pb/204Pb
ratios than diorites, consistent with higher U/Pb ratios in the
latter group (Table 8).
6. Discussion
6.1. Post-magmatic alteration, element mobility and
modification of isotopic composition
The Ivisaartoq greenstone belt underwent at least two
stages of metamorphic alteration prior to the intrusion of
2961 ± 12 Ma granitoids, resulting in the formation of
widespread calc-silicate metasomatic mineral assemblages
(Figs. 3–6; Polat et al., 2007). Stage I metasomatic assemblage appears to have formed during seafloor hydrothermal alteration under greenschist to lower-amphibolite
facies metamorphic conditions (Polat et al., 2007). Stage
II metasomatic assemblage was formed during a regional
A. Polat et al. / Lithos 100 (2008) 293–321
313
Table 6
Sm–Nd isotope composition of the Ivisaartoq actinolite schists, cumulates, pillow lavas, gabbros, and diorites
Sample #
Rock type
143
±2σ
Nd (ppm)
147
Sm (ppm)
εNd (3075 Ma)
Sm/Nd
485426
485430
485431
485433
485434
485435
485436
485437
485473 a
485474 a
485475 a
485481 a
485482 a
485486
485414
485418 a
485420 a
485422
485468
485469
485428
485432
485438
485467
485472
485476
485477
499730-A
499739
485483
485429
485484
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Cpx cumulate
Cpx cumulate
Cpx cumulate
Pillow basalt
Pillow basalt
Pillow basalt
Pillow basalt
Pillow basalt
Pillow basalt
Pillow basalt
Pillow basalt
Pillow basalt
Gabbro
Gabbro
Gabbro
Gabbro
Gabbro
Gabbro
Gabbro
Gabbro
Diorite
Diorite
Diorite
Diorite
0.512027
0.512244
0.513339
0.511848
0.510883
0.511767
0.510923
0.512678
0.512504
0.512512
0.512479
0.512086
0.511967
0.512060
0.512617
0.512549
0.512226
0.512438
0.512453
0.512483
0.512922
0.512472
0.511830
0.512601
0.512449
0.512395
0.512248
0.512693
0.512219
0.512098
0.512158
0.512188
11
9
28
11
13
11
9
11
10
11
12
11
10
13
14
10
12
9
12
12
9
10
12
13
6
11
11
5
5
10
10
11
1.82
3.45
0.72
3.67
16.3
3.59
42.6
1.53
2.07
2.44
2.77
4.56
4.34
3.95
5.67
3.13
4.59
3.88
5.19
4.86
6.0
5.72
8.28
3.97
4.99
5.49
5.49
6.94
4.64
9.97
5.57
6.84
0.1557
0.1709
0.2236
0.1489
0.0894
0.1465
0.0884
0.1869
0.1795
0.1787
0.1764
0.1618
0.1584
0.1646
0.1906
0.1853
0.1732
0.1792
0.1848
0.1850
0.204
0.1878
0.1521
0.1925
0.1854
0.1813
0.1727
0.1935
0.1735
0.1663
0.1702
0.1713
0.469
0.974
0.267
0.902
2.413
0.869
6.218
0.471
0.615
0.719
0.807
1.219
1.135
1.075
1.786
0.958
1.313
1.149
1.586
1.486
2.025
1.776
2.080
1.263
1.528
1.643
1.565
2.218
1.332
2.739
1.566
1.937
4.37
2.54
3.06
3.56
8.35
2.91
9.53
4.71
4.23
4.72
4.97
3.10
2.10
1.44
2.03
2.81
1.29
3.04
1.10
1.65
2.86
0.30
1.91
0.94
0.79
1.39
1.94
2.39
1.01
1.55
1.15
1.28
0.26
0.28
0.37
0.25
0.15
0.24
0.15
0.31
0.30
0.30
0.29
0.27
0.26
0.27
0.31
0.31
0.29
0.30
0.31
0.31
0.34
0.31
0.25
0.32
0.31
0.30
0.29
0.32
0.29
0.27
0.28
0.28
Nd/144Nd
Sm/144Nd
All initial εNd ages calculated at 3075 Ma yielded by U–Pb zircon analyses (see Friend and Nutman, 2005).
a
Published in Polat et al. (2007).
tectonothermal metamorphic event under middle- to upperamphibolite facies metamorphic conditions (Appel, 1997;
Polat et al., 2007).
Many pillow basalts are mineralogically and chemically
zoned (Fig. 3a; Polat et al., 2007). The rims have higher
contents of Fe2O3, MgO, MnO, and K2O, whereas the
inner and outer cores possess higher concentrations of
CaO, and Na2O and SiO2, respectively, consistent with the
mobility of these elements during post-magmatic alteration. Similarly, large variations in Ba, Sr, Pb, Rb, Cs, Li, U,
Zn, and Cu contents between pillow cores and rims are
consistent with the mobility of these elements. Compared
with the less altered cores, the rims have lower LREE
abundances and La/Smcn ratios, indicating the loss of these
elements. In contrast to the above elements, Al2O3, TiO2,
Th, Zr, Y, Cr, Ni, Co, Ga, and HREE display minor variations between the cores and rims, suggesting that these
elements were relatively immobile during Mesoarchean
seafloor hydrothermal alteration. Similarly, REE, HFSE
(Ti, Nb, Ta, Zr, Y) in gabbros, diorites, pillow lavas, and
cumulates display coherent primitive mantle- and chondrite-normalized patterns (Fig. 8), indicating that these
elements were also relatively immobile during post-magmatic alteration.
Cumulates with relict clinopyroxene phenocrysts are
characterized by more coherent trace element patterns and
narrower ranges of many major and trace elements than
their more deformed actinolite schist counterparts
(Figs. 8, 9). In addition, they have a narrow range of
Sm/Nd (0.29–0.30) ratios and initial εNd (+4.23 to +4.97)
values (Table 6), consistent with the Sm–Nd system remaining closed.
Given the preservation of primary minerals and
texture in cumulates, the initial εNd values in these rocks
likely reflect the near-primary magmatic composition
(Fig. 3e; Table 6).
In contrast to those in cumulates, many elements in
actinolite schists display large variations (Fig. 9; Table 4).
314
A. Polat et al. / Lithos 100 (2008) 293–321
Table 7
Sm–Nd isotope composition of the anorthositic inclusions and surrounding gabbroic matrix in the Ivisaartoq belt
Sample #
Rock type
143
±2σ
Nd (ppm)
147
Sm (ppm)
εNd (3075 Ma)
499728-A1
499729-A1
499731-A1
499731-B1
499731-C1
499728-A2
499729-A2
499731-A2
499731-B2
499731-C2
Anorthositic inclusion
Anorthositic inclusion
Anorthositic inclusion
Anorthositic inclusion
Anorthositic inclusion
Gabbroic matrix
Gabbroic matrix
Gabbroic matrix
Gabbroic matrix
Gabbroic matrix
0.511941
0.511935
0.511802
0.511811
0.512015
0.512699
0.512687
0.512732
0.512713
0.512703
6
7
7
6
6
6
6
5
4
4
0.850
0.980
0.578
0.615
0.871
6.732
6.612
6.936
7.269
7.223
0.1505
0.1474
0.1406
0.1420
0.1532
0.1929
0.1929
0.1943
0.1934
0.1942
0.2114
0.2388
0.1343
0.1444
0.2204
2.146
2.107
2.227
2.323
2.317
4.77
5.85
5.99
5.57
5.14
2.69
2.49
2.79
2.81
2.30
Nd/144Nd
Sm/144Nd
All initial εNd ages calculated at 3075 Ma yielded by U–Pb zircon analyses (see Friend and Nutman, 2005).
Despite the fact that actinolite schist samples analyzed for
this study came from the least metasomatized outcrops,
they still display a large spread in U, Pb, Sm, Nd concentrations, and isotopic ratios (Tables 4, 6, 8). Additionally, these samples have large positive to negative Eu and
Ce anomalies (Tables 3, 4). Even the samples that were
collected from the same outcrop (e.g., 485426 and
485427; and 485430 and 485431; 485434 and 485435)
display significant variations in these isotopic ratios
(Tables 4, 6, 8). In summary, these geochemical characteristics are consistent with the mobility of U, Pb, Sm, and
Nd in actinolite schists during post-magmatic alteration.
The late Archean (2774 ± 180) Pb–Pb errorchron age
(Fig. 11c) yielded by actinolite schists may reflect the
∼ 2800 Ma tectonothermal event that affected the region
(Friend and Nutman, 2005). Despite the Mesoarchean
(3092 ± 260 Ma) Sm–Nd errorchron age (Fig. 11a), some
samples (e.g., 485434 and 485435) collected from the
same outcrop have significant variations in Sm/Nd ratios
(0.15–0.24) and initial εNd (+ 2.9 to + 8.35), indicating
that the Sm–Nd isotope system in these rocks was
partially open on a whole-rock scale during Mesoarchean hydrothermal alteration. Compared to their less
strained cumulate counterparts, actinolite schists appear
to have been significantly enriched in LREE (e.g., La/
Ybcn = 16–24) and LILE (e.g., Rb = 12 ppm) elements by
hydrothermal fluids. Accordingly, we interpret the large
variations in the initial εNd (+ 2.5 to +9.5) values and
206
Pb/204Pb and 207 Pb/204Pb isotope ratios in actinolite
schists as indication of the disturbance of the Sm–Nd and
U–Pb isotope systems, rather than recording Mesoarchean mantle source heterogeneity.
Lead isotopes in pillow lavas (206 Pb/204Pb = 13.075–
17.866; 207 Pb/204 Pb=14.025–15.425) display larger
scatter than those in gabbros and diorites (206Pb/204
Pb=13.100–15.877; 207 Pb/ 204 Pb=14.027–14.687)
(Table 8). This variation correlates well with the higher
degree of metasomatic alteration in the former group.
Although samples for this study were collected from the
least metasomatized pillow cores, some of these cores
contain concentric cooling cracks filled with quartz and
plagioclase (Fig. 3b). On the basis of Mesoarchean
(3058 ± 240 Ma) Pb–Pb errorchron age, we suggest that
mobilization of U and Pb took place during seafloor
hydrothermal alteration, as recorded by mineralogical
and chemical zonation in pillow basalts (Figs. 3, 11).
The initial εNd values in gabbros and diorites (+ 0.30
to +2.86) overlap with, but extend to lower values than,
the pillow lavas (+1.10 to +3.10). We suggest that these
initial εNd values likely reflect the near-primary magmatic
compositions for the following reasons: (1) both groups
plot co-linearly in a 143Nd/144Nd versus147Sm/144Nd
diagram, yielding a Mesoarchean age (3060 ± 220 Ma,
MSWD = 80); (2) there are no co-variations between the
initial εNd values and La/Smcn ratios within each group;
and (3) there are no correlations between εNd and mobile
elements (e.g. LILE) (Table 6).
6.2. Geodynamic setting
Stage I metasomatic assemblage in the Ivisaartoq
greenstone belt is analogous to that of Phanerozoic
supra-subduction ophiolites and intra-oceanic island
arcs (see Polat et al., 2007, Python et al., 2007, and
references therein). The LREE-enriched trace element
patterns of the Ivisaartoq volcanic and intrusive rocks,
and those of anorthositic inclusions in gabbros (Figs. 8–
10) are consistent with a subduction zone geochemical
signature (cf. Saunders et al., 1991; Hawkesworth et al.,
1993). However, a backarc origin for the belt cannot be
ruled out on the basis of geochemical data alone.
High MgO (14–24 wt.%), Ni (600–850 ppm) and Cr
(1600–1700 ppm) concentrations, and Mg-numbers
(70–83) in clinopyroxene cumulates and high-Mg
pillow lavas are consistent with an island arc picritic
composition. Tertiary island arc picrites have been
A. Polat et al. / Lithos 100 (2008) 293–321
315
Table 8
U–Th–Pb isotope compositions of the Ivisaartoq actinolite schists, cumulates, pillow lavas, gabbros, and diorites
Sample #
Rock type
206
Pb/204Pb ±2σ
485426
485430
485431
485433
485434
485435
485436
485437
485473
485474
485475
485481
485482
485486
485414
485418
485420
485422
485468
485469
485428
485432
485438
485467
485472
485476
485477
499730-A
499739
485483
485429
485484
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Actinolite schist
Cpx cumulate
Cpx cumulate
Cpx cumulate
Pillow lava
Pillow lava
Pillow lava
Pillow lava
Pillow lava
Pillow lava
Pillow lava
Pillow lava
Pillow lava
Gabbro
Gabbro
Gabbro
Gabbro
Gabbro
Gabbro
Gabbro
Gabbro
Diorite
Diorite
Diorite
Diorite
15.645
15.590
12.687
14.287
13.575
13.760
15.617
12.854
21.150
19.656
19.345
17.816
18.352
17.693
13.075
13.741
17.866
17.603
16.985
15.226
14.917
13.521
13.320
13.958
13.100
13.238
13.516
15.877
14.355
14.250
15.138
14.230
0.030
0.011
0.010
0.013
0.008
0.021
0.011
0.009
0.057
0.025
0.021
0.015
0.017
0.028
0.007
0.015
0.016
0.017
0.013
0.020
0.012
0.014
0.006
0.007
0.008
0.008
0.011
0.021
0.008
0.013
0.015
0.012
207
Pb/204Pb ±2σ
14.591
14.590
13.870
14.373
14.040
14.157
14.399
13.892
15.502
15.325
15.225
15.069
14.977
14.852
14.025
14.155
15.425
14.961
15.219
14.666
14.551
14.111
14.133
14.264
14.027
14.096
14.152
14.687
14.349
14.338
14.587
14.330
0.029
0.012
0.012
0.014
0.010
0.023
0.011
0.011
0.043
0.021
0.018
0.014
0.015
0.025
0.010
0.018
0.015
0.015
0.013
0.020
0.013
0.016
0.009
0.009
0.011
0.010
0.013
0.021
0.009
0.015
0.015
0.014
208
Pb/204Pb ±2σ
34.931
34.594
32.355
33.910
32.788
32.898
33.783
32.393
39.975
38.868
40.474
37.301
37.038
38.322
33.357
33.696
37.523
36.254
36.639
35.167
35.237
33.190
33.123
33.453
32.923
33.158
33.888
35.015
34.183
34.478
35.006
34.324
0.075
0.032
0.032
0.036
0.027
0.056
0.030
0.029
0.111
0.059
0.050
0.039
0.042
0.065
0.027
0.048
0.040
0.040
0.034
0.050
0.035
0.039
0.025
0.025
0.029
0.028
0.034
0.052
0.026
0.038
0.039
0.037
r1
r2
Th⁎
U⁎
Pb⁎
U/Pb
Th/Pb
0.98
0.96
0.96
0.97
0.96
0.98
0.96
0.96
0.98
0.97
0.97
0.95
0.97
0.98
0.96
0.91
0.97
0.97
0.97
0.98
0.97
0.97
0.97
0.97
0.95
0.96
0.96
0.97
0.95
0.96
0.98
0.96
0.91
0.94
0.93
0.94
0.92
0.95
0.93
0.93
0.98
0.88
0.95
0.91
0.93
0.97
0.93
0.84
0.93
0.95
0.94
0.97
0.94
0.96
0.93
0.94
0.91
0.92
0.94
0.95
0.93
0.95
0.95
0.91
0.155
0.121
0.369
0.060
0.064
0.193
0.085
0.085
0.339
0.259
0.356
0.683
0.825
0.722
0.370
0.402
0.708
0.552
0.787
0.478
0.217
0.515
0.532
0.726
0.465
0.628
0.503
0.34
0.793
1.161
0.686
0.942
0.264
0.444
0.134
0.023
0.079
0.035
0.084
0.084
0.293
0.668
0.183
0.148
0.286
0.260
0.064
0.101
0.192
0.167
0.190
0.183
0.035
0.188
0.100
0.160
0.149
0.176
0.257
0.09
0.185
0.355
0.150
0.298
1.023
0.924
1.862
5.162
5.595
8.567
2.325
5.970
1.554
1.652
1.833
1.897
1.280
2.776
1.809
2.229
1.712
1.257
5.319
3.196
1.046
4.441
6.033
3.463
4.399
6.433
7.461
1.411
4.934
5.357
2.543
3.831
0.258
0.481
0.072
0.004
0.014
0.004
0.036
0.014
0.188
0.405
0.100
0.078
0.223
0.094
0.035
0.045
0.112
0.133
0.036
0.057
0.033
0.042
0.017
0.046
0.034
0.027
0.034
0.065
0.037
0.066
0.059
0.078
0.152
0.130
0.198
0.012
0.011
0.023
0.037
0.014
0.218
0.157
0.194
0.360
0.645
0.260
0.205
0.180
0.414
0.439
0.148
0.149
0.208
0.116
0.088
0.210
0.106
0.098
0.067
0.244
0.161
0.217
0.270
0.246
r1 = 206Pb/204Pb vs. 207Pb/204Pb error correlation (Ludwig, 1988).
r2 = 206Pb/204Pb vs. 208Pb/204Pb error correlation (Ludwig, 1988).
Errors are two standard deviations absolute (Ludwig, 1988).
Fractionation of Pb was controlled by repeated analysis of the SRM NBS 981 standard (values of Todt et al., 1993) and amounted to 0.103 +/−
0.007%/a.m.u. (2 s, n = 5).
*U, Th, Pb concentrations represent measurements by high-resolution ICP-MS at Windsor.
documented in the Solomon and New Hebrides (Vanuatu
arc) oceanic island arcs (see Eggins, 1993; Schuth et al.,
2004). In the Solomon Islands, picrites occur only in
New Georgia Island above the subducting Woodlark
spreading centre. In the New Hebrides subduction
system, the overriding plate is currently undergoing
extension east of the Vanuatu arc, forming a suprasubduction oceanic crust within the North Fiji Basin
(Hawkins, 2003). As a corollary, given that the
geochemical characteristics and hydrothermal alteration
features of the Ivisaartoq rocks are comparable to those
of Phanerozoic ophiolites and intra-oceanic island arcs,
we suggest that the Ivisaartoq belt originated as
Mesoarchean supra-subduction oceanic crust (Fig. 12).
6.3. Petrogenesis and mantle source characteristics
The positive initial εNd values (e.g. +4.2 to +5.0 in
clinopyroxene cumulates; +4.8 to +6.0 in anorthositic
inclusions; +1.1 to +3.1 in pillow basalts) require longterm depleted upper mantle sources. Given near-flat
HREE patterns in these lithologies, melting occurred at
b 80 km (cf. Hirschmann and Stolper, 1996). Picritic
cumulates and anorthositic inclusions plot above the
estimated evolution curve of the depleted mantle (Fig. 13).
The depleted Nd isotopic signatures and low LREE and
HFSE (Nb, Ta, Zr, Ti, Y) abundances indicate that the
mantle source region had experienced significant melt
extraction prior to 3075 Ma.
316
A. Polat et al. / Lithos 100 (2008) 293–321
Fig. 12. A simplified geodynamic and petrologic model for the Ivisaartoq greenstone belt and Mesoarchean forearc oceanic crust.
The majority of pillow lavas, gabbros, and diorites
plot below the predicted depleted mantle evolution curve
(Fig. 13; cf. Henry et al., 2000; Bennett, 2003). Large
variations in the initial εNd (+ 0.30 to +3.10) values may
reflect either mantle source heterogeneity or crustal
contamination. Contamination of the Ivisaartoq rocks by
continental crust during magma ascent, rather than
contamination of their source regions by subducted
crustal material, can be ruled out on the basis of the
following observations: (1) the association of pillow
basalts, gabbros, and sulphide-rich siliceous volcaniclastic sedimentary rocks is expected to have formed in
an oceanic rather than a continental setting; (2) there is
no field evidence indicating that the Ivisaartoq greenstone belt was deposited on an older continental
basement; (3) the lack of co-variations between εNd
and abundances of contamination-sensitive elements or
their ratios (e.g. SiO2, Ni, Cr, Th, Zr, La/Smcn) within
each volcanic group (Tables 4, 6, 7); and (4) there are no
correlation between 147 Sm/144 Nd initial εNd values,
A. Polat et al. / Lithos 100 (2008) 293–321
317
Fig. 13. Age versus initial εNd variation diagram for the Ivisaartoq cumulates, pillow lavas, gabbros, and diorites, Wawa adakites and Mg-andesites
(Polat and Kerrich, 2002), Kostomuksha komatiites (Puchtel et al., 1998), and Winnipeg River granitoids (Henry et al., 2000). Modified after Henry
et al. (2000). (DM: Depleted mantle; CHUR: Chondrite uniform reservoir).
which, according to Vervoort and Blichert-Toft (1999), is
a robust criterion for identifying crustal contamination
(Table 6). Accordingly, the lower initial εNd values
(b+ 2.0) likely indicate a Nd-enriched component in the
source region, rather than crustal contamination. We
therefore suggest that an enriched component was added
to the mantle wedge in variable proportions by recycling
of older continental material, with super-chondritic Nd/
Sm ratios (cf. Shirey and Hanson, 1986; Polat and
Kerrich, 2002).
Light-REE-enriched patterns of the Ivisaartoq rocks,
however, imply that the depleted sub-arc mantle source
must have been metasomatized shortly before or during
partial melting that took place at about 3075 Ma (cf. Polat
and Münker, 2004). Hydrous fluids and/or melts derived
from either subducted altered oceanic crust or sediments,
with sub-chondritic Nb/La, Nb/Th, Sm/Nd, and Ti/Gd
ratios, were probably the main cause of the metasomatism,
generating LREE-enriched, HFSE-depleted trace element
patterns in the Ivisaartoq rocks (Figs. 8, 10).
6.4. Origin of the anorthositic inclusions in gabbros
Ocellar texture characterized by eye-shaped anorthosite
inclusions and gabbroic matrix (Figs. 4f, 5a) was previously interpreted as solidified immiscible liquids (Polat
et al., 2007). However, new petrographic evidence, such as
the presence of relic reaction rims and smaller pieces of
peeled anorthosites at the inclusion-matrix contacts, suggest that the inclusions had already been solidified before
they were enclosed in gabbroic magma. The smaller
anorthositic inclusions, aligned parallel to the contacts
(Fig. 5a), might have resulted from the thermal erosion of
the larger ones during their transportation to shallower
depths. Therefore, we suggest that the anorthositic inclusions were carried, as crustal xenoliths, from lower oceanic
crust to the shallower depths by upwelling magmas. Both
the inclusions and matrix were deformed and recrystallized
under amphibolite facies metamorphic conditions before
the intrusion of 2961 ± 12 Ma granitoids.
Low abundances of MgO, Ni, Cr, Co, and Sc in the
anorthositic inclusions are consistent with olivine, clinopyroxene and/or orthopyroxene fractionation (Fig. 10;
Table 5). Depletion of Nb, relative to Th and La, and nearflat HREE patterns are consistent with a subduction zone
geodynamic setting and a shallow mantle source (Figs. 10,
12; Table 5). The gabbroic matrix shares the negative Nb
anomalies (Fig. 10; Table 5). The initial εNd values of the
anorthositic inclusions are much larger than those of the
gabbroic matrix (+4.8 to +6.0 versus +2.3 to +2.8),
indicating two different mantle sources. On the other hand,
the anorthositic inclusions are isotopically comparable to
picrites (clinopyroxene cumulates) (Tables 6, 7), indicating
a petrogenetic link between the two rock types. It is likely
that picrites and anorthosites were derived from the same
parental magma through clinopyroxene and plagioclase
fractionation, respectively. Given their low viscosity,
picritic magmas could easily have reached the surface to
form the ultramafic sills and/or flows. In contrast,
anorthositic magmas might have been too viscous to
reach to the surface; instead, they might have crystallized
within the lower oceanic crust to form layered anorthosites.
7. Implications for the generation of Mesoarchean
oceanic crust
Field relationships and geochemical data indicate that
all volcanic and intrusive rock types in the Mesoarchean
(ca. 3075 Ma) Ivisaartoq greenstone belt are part of the
same lithotectonic assemblage, sharing a common history
of magmatism, deformation, and metamorphism. The
association of pillow basalt, gabbros and sulphide-rich
318
A. Polat et al. / Lithos 100 (2008) 293–321
siliceous volcaniclastic sedimentary rocks in the belt
suggests an intra-oceanic depositional environment
(Figs. 3–5).
The large initial εNd (e.g. +2 to +6) isotopic values in
the Mesoarchean Ivisaartoq rocks indicate a long-term
LREE-depleted (Sm/Ndcn N 1) mantle source(s), similar to
the source of modern N-MORB (see Hofmann, 2003).
However, the majority of the least altered samples have
LREE-enriched (La/SmcnN 1; Sm/Ndcnb 1) but Nb-depleted, relative to Th and La, trace element patterns (Figs. 9,
10), consistent with a subduction zone geodynamic setting
(Fig. 12). Accordingly, we propose a two-stage evolutionary geodynamic model to explain the geological characteristics of the Ivisaartoq greenstone belt. In the first stage, the
mantle source of the Ivisaartoq rocks had originated as a
sub-oceanic depleted upper mantle, like the source of
present-day N-MORB. The second stage marks the
development of an intra-oceanic subduction system.
Following the initiation of an intra-oceanic subduction
zone along either a mid-ocean ridge or a transform fault, the
mantle source of the Ivisaartoq rocks was converted to a
sub-arc mantle wedge (cf. Casey and Dewey, 1984; Dilek
and Flower, 2003). Hydrous fluids and/or melts originating
from the subducted slab metasomatized the sub-arc mantle
wedge, resulting in LREE-enriched and HFSE-depleted,
relative to Th and LREE, patterns (Figs. 8, 10).
The forearc region of the overriding plate may have
undergone a significant extension in response to slab
rollback, resulting in a large degree of partial melting of the
hydrated upper mantle wedge at shallow depths (cf. Dilek
and Flower, 2003; Fig. 12). Such a high degree of partial
melting is expected to have resulted in the formation of a
large magma chamber (Fig. 12). Extensive partial melting
beneath the Ivisaartoq forearc may have generated a thick
(N 20 km) oceanic crust (cf. Sleep and Windley, 1982).
Such an intact oceanic crust might have been composed of
two major crustal sections: (1) a lower layer of anorthosites
and leucogabbros; and (2) an upper layer of basaltic to
picritic flows, gabbroic to dioritic dykes, and dunitic to
wehrlitic sills (Figs. 2, 12).
The major and trace element characteristics of the
anorthositic inclusions in the Ivisaartoq are comparable
to those of Meso- to Neoarchean anorthosite complexes
in SW Greenland (Windley et al., 1973; Weaver et al.,
1981; Ashwal and Myers, 1994; Dymek and Owens,
2001). Like the Buksefjorden, Nordland, and Fiskenaesset anorthosites, the Ivisaartoq counterparts have LREE
enriched chondrite-normalized patterns with large positive Eu anomalies (see Weaver et al., 1981; Dymek and
Owens, 2001), suggesting a similar petrogenetic process.
However, the Ivisaartoq anorthositic inclusions have flat
to less fractionated HREE patterns compared to the
Buksefjorden, Nordland, and Fiskenaesset anorthosites,
indicating a shallower, garnet-free mantle source region
(see Dymek and Owens, 2001).
In the Ivisaartoq belt, anorthosites and leucogabbros
occur as volumetrically minor intrusions within the lower
amphibolites (Figs. 1, 2; Chadwick, 1990); however, they
might originally have been thicker. There are two main
reasons why this might be the case. First, given the record
of several generations of deformation in the region
(Chadwick, 1990; Friend and Nutman, 2005), finding an
intact, thicker leucogabbro–anorthosite association is
unlikely. Second, if Archean oceanic crust was thicker
due to potentially higher mantle temperatures (Sleep and
Windley, 1982; McKenzie and Bickle, 1988), then it is
possible that only the upper basaltic crustal section was
peeled off and accreted while the lower anorthosite–
leucogabbro section was subducted. Notwithstanding
these problems, partial sections of 2800–3000 Ma
anorthosite–leucogabbro associations have been identified throughout the Archean terranes of southern West
Greenland (Windley, 1970; Windley et al., 1973; Escher
and Myers, 1975; Windley et al., 1981; Myers, 1985;
Ashwal and Myers, 1994; Owens and Dymek, 1997).
The petrogenetic origin of Archean anorthosites
remains unresolved (Weaver et al., 1981, 1982; Pinney
et al., 1988; Ashwal and Myers, 1994; Owens and
Dymek, 1997; Dymek and Owens, 2001). In the beststudied Fiskenaesset anorthosite complex, the anorthosites and gabbros appear to have intruded into the
overlying greenstone sequences (Escher and Myers,
1975; Ashwal and Myers, 1994). Geochemical studies
suggest that the Fiskenaesset anorthosite complex is
genetically related, by fractional crystallization, to mafic
to ultramafic volcanic rocks into which they were
emplaced (Weaver et al., 1981, 1982; Peck and Valley,
1996) and were derived from a long-term depleted
mantle source (Ashwal et al., 1989). The anorthositic
inclusions in gabbros, and anorthosites layers intruding
the lower amphibolites, in the Ivisaartoq belt are likely
related to the ultramafic lithologies in the belt by
fractional crystallization.
Greenstone–anorthosite associations in SW Greenland
(e.g. Fiskenaesset Complex) are interpreted as remnant
Archean oceanic crust (ophiolite?) in several studies
(Windley et al., 1981; Weaver et al., 1982; Myers, 1985;
Ashwal and Myers, 1994). On the basis of geological
similarities between the Ivisaartoq and Fiskenaesset
greenstone belts, and geological and geochemical data
presented in this study, we interpret the Ivisaartoq
greenstone belt as a relic of an Archean forearc oceanic
crust. We do not propose that all Archean anorthosites
formed in a forearc tectonic setting. Like basaltic
A. Polat et al. / Lithos 100 (2008) 293–321
counterparts, Archean anorthosites, depending on their
geological and geochemical characteristics, might have
formed in diverse geodynamic settings, including in midocean ridges, forearcs, backarcs, plume-derived oceanic
plateaus, and intra-continental rifts.
Acknowledgements
We thank A. Trenhaile and R. Kerrich for reviewing
the initial draft of the manuscript. J.C. Barrette, and J.
Gagnon are acknowledged for their help during geochemical analyses. Reviewers J. Dostal, H. Smithies, P.
Spadea, and M.F. Zhou are acknowledged for their
constructive comments, which have resulted in significant
improvements to the paper. We are grateful to B.F.
Windley for helpful discussion on the geology of Archean
greenstone belts and anorthosite complexes. This is a
contribution of NSERC grants 250926 to AP and 83117 to
B. Fryer. R. Frei is supported by FNU (Forskningsrådet
for Natur of Univers) grant no. 21-01-0492 56493. Field
work was supported by the Bureau of Minerals and
Petroleum in Nuuk and the Geological Survey of
Denmark and Greenland (GEUS). A. Polat thanks Mr.
Tekin Demir for helping his family during fieldwork in
Greenland.
References
Abbott, D., Drury, R., Smith, W.H.F., 1994. Flat to steep transition in
subduction style. Geology 22, 937–940.
Appel, P.W.U., 1994. Stratabound scheelite in altered Archean
komatiites, West Greenland. Miner. Depos. 29, 341–352.
Appel, P.W.U., 1997. High bromine and low Cl/Br ratios in
hydrothermally altered Archean komatiitic rocks, West Greenland.
Precambrian Res. 82, 177–189.
Arndt, N.T, Bruzak, G., Reischmann, T., 2001. The oldest continental
and oceanic plateaus-geochemistry of basalts and komatiites of the
Pilbara Craton, Australia. In: Buchan, K.L, Ernst, R.E. (Eds.), Mantle
plumes: their identification through time. Geol. Soc. Am. Spec.
Paper, vol. 352, pp. 1–30.
Ashwal, L.D., Myers, J., 1994. Archean anorthosites. In: Condie,
K.C. (Ed.), Archean Crustal Evolution. Elsevier, Amsterdam,
pp. 315–355.
Ashwal, L.A., Jacobsen, S.B., Myers, J.S., Kalsbeek, F., Goldstein, S.J.,
1989. Sm–Nd age of the Fiskenaesset Anorthosite Complex, West
Greenland. Earth Planet. Sci. Lett. 91, 261–270.
Bennett, V.C., 2003. Compositional evolution of the mantle. In: Carlson,
R.V. (Ed.), Treatise on Geochemistry, vol. 3. Elsevier, Amsterdam,
pp. 493–519.
Bickle, M.J., Nisbet, E.G., Martin, A., 1994. Archean greenstone belts
are not oceanic crust. J. Geol. 102, 121–138.
Bleeker, W., 2002. Archean tectonics: a review, with illustrations
from the Slave craton. In: Fowler, M.R., Ebinger, C.J., Hawkesworth, C.J. (Eds.), The Early Earth: Physical, Chemical and
Biological Development. Geol. Soc. London Spec. Publ., vol. 199,
pp. 151–181.
319
Brewer, M.A., Chadwick, B., Coe, K., Park, J.F.W., 1984. Further field
investigations in the Ivisaartoq region of southern West Greenland.
Rapp. - Grùnl. Geol. Unders. 120, 55–67.
Casey, J.F., Dewey, J.F., 1984. Initiation of subduction zones along
transform and accreting plate boundaries, triple-junction evolution,
and forearc spreading centres — implications for ophiolitic geology
and obduction. In: Gass, I.G., Lippard, S.J., Shelton, A.W. (Eds.),
Ophiolites and Oceanic Lithosphere. Geol. Soc. London, Spec.
Publ., vol. 13, pp. 269–290.
Chadwick, B., 1985. Contrasting styles of tectonism and magmatism
in the late Archean crustal evolution of the northeastern part of the
Ivisaartoq region, inner Godthåbsfjord, southern West Greenland.
Precambrian Res. 27, 215–238.
Chadwick, B., 1986. Malene stratigraphy and late Archean structure:
new data from Ivisaartoq, inner Godthåbsfjord, southern West
Greenland. Rapp. - Grùnl. Geol. Unders. 130, 74–85.
Chadwick, B., 1990. The stratigraphy of a sheet of supracrustal rocks
within high-grade orthogneisses and its bearing on late Archean
structure in southern West Greenland. J. Geol. Soc. (Lond.) 147,
639–652.
Chadwick, B., Coe, K., 1988. Geological Map of Greenland,
1:100 000, Ivisaartoq Sheet 64 V. Nord. Copenhagen, Geological
Survey of Greenland.
Cloos, M., 1993. Lithospheric buoyancy and collisional orogenesis:
subduction of oceanic plateaus, continental margin, island arcs,
spreading ridges, and seamounts. GSA Bull. 105, 715–737.
Condie, K.C., 1981. Archean Greenstone Belts. Elsevier, Amsterdam.
435 pp.
Condie, K.C., 2005. Earth as an Evolving Planetary System. Elsevier,
Amsterdam. 447 pp.
Corcoran, P.L., Dostal, J., 2000. Development of an ancient-back arc
basin overlying continental crust: the Archean Peltier Formation,
Northwest Territories, Canada. J. Geol. 109, 329–348.
Dewey, J., 2003. Ophiolites and lost oceans: rifts, ridges, arcs, and/or
scraping. In: Dilek, Y., Newcomb, S. (Eds.), Ophiolite Concept and
the Evolution of Geological Thought. Geol. Soc. Am. Spec. Paper,
vol. 373, pp. 153–158.
Dilek, Y., 2003. Ophiolite concept and its evolution. In: Dilek, Y.,
Newcomb, S. (Eds.), Ophiolite Concept and the Evolution of
Geological Thought. Geol. Soc. Am. Spec. Paper, vol. 373, pp. 1–15.
Dilek, Y., Flower, M.F.J., 2003. Arc-trench rollback and forearc
accretion: 2. A model template for ophiolites in Albania, Cyprus,
and Oman. In: Dilek, Y., Robinson, P.T. (Eds.), Ophiolites in Earth
History. Geol. Soc. London Spec. Publ., vol. 218, pp. 43–68.
Dostal, J., Mueller, W.U., 1997. Komatiite flooding of a rifted Archean
rhyolite arc complex: geochemical signature and tectonic significance of Stoughton-Roquemaure Group, Abitibi greenstone belt,
Canada. J. Geol. 105, 545–563.
Dostal, J., Mueller, W.U., Murphy, J.B., 2004. Archean molasses basin
evolution and magmatism, Wabigoon subprovince, Canada. J. Geol.
112, 435–454.
Dymek, R.F., Owens, B.R., 2001. Chemical assembly of Archean
anorthosites from amphibolite- and granulite-facies terranes, SW
Greenland. Contrib. Mineral. Petrol. 141, 513–528.
Eggins, S.M., 1993. Origin and differentiation of picritic arc
magmas, Ambae [Aoba], Vanuatu. Contrib. Mineral. Petrol.
114, 79–100.
Eriksson, K.A., Krapez, B., Fralick, P.W., 1994. Sedimentology of
Archean greenstone belts: signatures and tectonic evolution. Earth
Sci. Rev. 37, 1–88.
Escher, J.C., Myers, J.S., 1975. New evidence concerning the original
relationships of early Precambrian volcanics and anorthosites in
320
A. Polat et al. / Lithos 100 (2008) 293–321
the Fiskenaesset region, southern West Greenland. Rapp. - Grùnl.
Geol. Unders. 75, 72–76.
Frei, R., Villa, I.M., Nägler, Th.F., Kramers, J.D., Pryzbylowicz, W.J.,
Prozesky, V.M., Hofmann, B.A., Kamber, B.S., 1997. Single
mineral dating by Pb–Pb step-leaching method; Assessing the
mechanisms. Geochim. Cosmochim. Acta 61, 393–414.
Friend, C.R.L., Nutman, A.P., 2005. New pieces to the Archean jigsaw
puzzle in the Nuuk region, southern West Greenland: steps in
transforming a simple insight into a complex regional tectonothermal model. J. Geol. Soc. (Lond.) 162, 147–162.
Friend, C.R.L., Hall, R.P., Hughes, D.J., 1981. The geochemistry of the
Malene (mid-Archean) ultramafic–mafic amphibolite suite, southern West Greenland. In: Glover, J.E., Groves, D.E. (Eds.), Archean
Geology. Spec. Pub. Geol. Soc. Australia, vol. 7, pp. 301–312.
Goodwin, M.A., 1991. Precambrian Geology, The Dynamic Evolution
of the Continental Crust. Academic press, London. 666 pp.
Hall, R.P., 1981. The tholeiitic and komatiitic affinity of Malene
metavolcanic amphibolites from Ivisaartoq, southern West
Greenland. Rapp. - Grùnl. Geol. Unders. 97 20 pp.
Hall, R.P., Friend, C.R.L., 1979. Structural evolution of the Archean
rocks in Ivisaartoq and neighbouring inner Godthåbsfjord region,
southern West Greenland. Geology 7, 311–315.
Hall, R.P., Hughes, D.J., Friend, C.R.L., 1987. Mid-Archean basic
magmatism of southern West Greenland. In: Park, R.G., Tarney, J.
(Eds.), Evolution of the Lewisian and Comparable Precambrian
High Grade Terrains. Geol. Soc. London Spec. Publ., vol. 27,
pp. 261–275.
Hawkesworth, C.J., Gallanger, K., Hergt, J.M., McDetmott, F., 1993.
Mantle and slab contributions in arc magmas. Annu. Rev. Earth
Planet. Sci. 21, 175–204.
Hawkins, J.W., 2000. Geology of supra-subduction zones—implications for the origin of ophiolites. In: Dilek, Y., Newcomb, S. (Eds.),
Ophiolite Concept and the Evolution of Geological Thought. Geol.
Soc. Am. Spec. Paper, vol. 373, pp. 227–268.
Henry, P., Stevenson, R.S., Larbi, Y., Gariépy, C., 2000. Nd isotopic
evidence for early to late Archean (3.4–2.7 Ga) crustal growth in
the Western Superior Province (Ontario, Canada). Tectonophys
322, 135–151.
Hirschmann, M.M., Stolper, E.M., 1996. A possible role for garnet
pyroxenite in the origin of “garnet signature” in MORB. Contrib.
Mineral. Petrol. 124, 185–208.
Hofmann, A.W., 1988. Chemical differentiation of the Earth: the
relationships between mantle, continental crust, and oceanic crust.
Earth Planet. Sci. Lett. 90, 297–314.
Hofmann, A.W., 2003. Sampling mantle heterogeneity through oceanic
basalts: isotopes and trace elements. In: Carson, R.W. (Ed.),
Treatise on Geochemistry, Th mantle and Core, vol. 2, pp. 61–101.
Kerrich, R., Polat, A., 2006. Archean greenstone–tonalite duality:
thermochemical mantle convection models or plate tectonics in the
early Earth global dynamics? Tectonophysics 415, 141–165.
Kusky, T.M., 2004. Introduction. In: Kusky, T.M. (Ed.), Precambrian
Ophiolites and Related Rocks. Elsevier, Amsterdam, pp. 1–34.
Kusky, T.M., Polat, A., 1999. Growth of granite–greenstone terranes at
convergent margins, and stabilization of Archean cratons. Tectonophysics 305, 43–73.
Ludwig, K.R., 1988. A computer program to convert raw U–Th–Pb
isotope ratios to blank corrected isotope ratios and concentrations
with associated error-correlations: United States Geological
Survey. Open File Report. OF-82-820.
Ludwig, K., 2003. User's Manual for Isoplot/Ex Version 3.0, A
Geochronological Toolkit for Microsoft Excel. Special Publication
1a, Berkeley Geochronology Center.
McKenzie, D., Bickle, M.J., 1988. The volume and composition of
melt generated by extension of the lithosphere. J. Petrol. 29,
625–679.
Myers, J.S., 1985. Stratigraphy and structure of the Fiskenaesset
Complex, West Greenland. Gronland Geol. Unders. Bull. 150. 72 pp.
Nisbet, E.G., Fowler, C.M.R., 1983. Model for Archean plate
tectonics. Geology 11, 376–379.
Owens, B.E., Dymek, R.F., 1997. Comparative petrology of Archean
anorthosites in amphibolite and granulite facies terranes, SW
Greenland. Contrib. Mineral. Petrol. 128, 371–384.
Peck, W.H., Valley, J.W., 1996. The Fiskenaesset Anorthosite
Complex: stable isotope evidence for shallow emplacement into
Archean oceanic crust. Geology 24, 523–526.
Pinney, W.C., Morrison, D.A., Maczuga, D.E., 1988. Anorthosites and
related megacrystic units in the evolution of Archean crust.
J. Petrol. 29, 1283–1323.
Polat, A., Hofmann, A.W., 2003. Alteration and geochemical patterns
in the 3.7–3.8 Ga Isua greenstone belt, West Greenland.
Precambrian Res. 126, 197–218.
Polat, A., Kerrich, R., 2002. Nd-isotope systematics of ∼ 2.7 Ga
adakites, magnesian andesites, and arc basalts, Superior Province,
Canada: evidence for shallow crustal recycling at Archean
subduction zones. Earth Planet. Sci. Lett. 202, 345–360.
Polat, A., Kerrich, R., 2006. Reading the geochemical fingerprints of
Archean hot subduction volcanic rocks: evidence for accretion and
crustal recycling in a mobile tectonic regime. In: Benn, K., Mareschal,
J.C., Condie, K.C. (Eds.), Archean Geodynamics and Environments.
AGU Geophys. Monograph Series, vol. 164, pp. 189–213.
Polat, A., Münker, C., 2004. Hf-Nd isotope evidence for contemporaneous subduction processes in the source of late Archean arc
lavas from the Superior Province, Canada. Chem. Geol. 213,
403–429.
Polat, A., Kerrich, R., Wyman, D.A., 1998. The late Archean SchreiberHemlo and White River-Dayohessarah greenstone belts, Superior
Province: collages of oceanic plateaus, oceanic arcs, and
subduction–accretion complexes. Tectonophysics 294, 295–326.
Polat, A., Appel, P.W.U., Frei, R., Pan, Y., Dilek, Y., OrdonezCaldeeron, J.C., Fryer, B., Raith, J.G., 2007. Field and
Geochemical Characteristics of the Mesoarchean (∼ 3075 Ma)
Ivisaartoq Greenstone Belt, Southern West Greenland: Evidence
for Seafloor Hydrothermal Alteration in a Supra-Subduction
Oceanic Crust. Gondwana Research 11, 69–91.
Puchtel, I.S., Hofmann, A.W., Mezger, K., Jochum, K.P., Shchipansky,
A.A., Samsonov, A.V., 1998. Oceanic plateau model for
continental crustal growth in the Archean: a case study from the
Kostomuksha greenstone belt, NW Baltic Shield. Earth. Planet.
Sci. Lett. 155, 57–74.
Python, M., Ceuleneer, G., Ishida, Y., Barrat, J.A., Arai, S., 2007. Oman
diopsides: a new lithology diagnostic of very high temperature
hydrothermal circulation in mantle peridotite below oceanic
spreading centres. Earth Planet. Sci. Lett. 255, 289–305.
Sandeman, H.A., Hanmer, S., Davis, W.J., Ryan, J.J., Peterson, T.D.,
2004. Neoarchean volcanic rocks, Central Hearne supracrustal
belt, Western Churchill Province, Canada: geochemical and
isotopic evidence supporting intra-oceanic, suprasubduction zone
extension. Precambrian Res. 134, 113–141.
Saunders, A.D., Norry, M.J., Tarney, J., 1991. Fluid influence on the
trace element compositions of subduction zone magmas. Phil.
Trans. Royal Soc. London A 335, 377–392.
Schaller, M., Steiner, O., Suder, I., Frei, R., Kramers, J.D., 1997. Pb
stepwise leaching (PbSL) dating of garnet-addressing the inclusion
problem. Schweiz. Mineral. Petrogr. Mitt. 77, 113–121.
A. Polat et al. / Lithos 100 (2008) 293–321
Schuth, S., Rohrbach, A., Münker, C., Ballhaus, C., Garbe-Schönberg,
D., Qopoto, C., 2004. Geochemical constraints on the petrogenesis
of arc picrites and basalts, New Georgia Group, Solomon Islands.
Contrib. Mineral. Petrol. 148, 288–304.
Şengör, A.M.C., Natal'in, B.A., 2004. Phanerozoic analogues of
Archean oceanic basement fragments: Altaid ophiolites and
ophirags. In: Kusky, T.M. (Ed.), Precambrian Ophiolites and
Related Rocks. Elsevier, Amsterdam, pp. 675–726.
Shirey, S.B., Hanson, G.N., 1986. Mantle heterogeneity and crustal
recycling in Archean granite–greenstone belts: Evidence from Ndisotopes and trace elements in the Rainy Lake area, Superior Province,
Ontario, Canada. Geochim. Cosmochim. Acta 50, 2631–2651.
Sleep, N.H., Windley, B.F., 1982. Archean plate tectonics: constrains
and inferences. J. Geol. 90, 363–379.
Smithies, R.H., Champion, D.C., van Kranendonk, M.J., Howard, H.M.,
Hickman, A.H., 2005a. Modern-style subduction processes in the
Mesoarchean: geochemical evidence from the 3.12 Ga Whundo
intra-oceanic arc. Earth Planet. Sci. Lett. 231, 221–237.
Smithies, R.H., van Kranendonk, M.J., Champion, D.C., 2005b. It
started with a plume — early Archean basaltic proto-continental
crust. Earth Planet. Sci. Lett. 238, 284–297.
Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic systematics of
oceanic basalts: implications for mantle composition and processes.
In: Saunders, A.D., Norry, M.J. (Eds.), Magmatism in the Ocean
Basins. Geol. Soc. London Spec. Publ., vol. 42, pp. 313–345.
Taylor, S.R., McLennan, S.M., 1985. The Continental Crust: Its
Composition and Evolution. Blackwell, Oxford. 312 pp.
321
Todt, W., Cliff, R.A., Hanser, A., Hofmann, A.W., 1993. Recalibration
of NBS lead standards using a 202Pb + 205Pb double spike. Terra
Abstr. 5 (Suppl. 1), 396.
van Hunen, J., van den Berg, A.P., Vlaar, N.J., 2002. On the role of
oceanic plateaus in the development of shallow flat subduction.
Tectonophysics 352, 317–333.
Vervoort, J.D., Blichert-Toft, J., 1999. Evolution of the depleted
mantle: Hf isotope evidence from juvenile rocks through time.
Geochim. Cosmochim. Acta 63, 533–556.
Weaver, B.L., Tarney, J., Windley, B., 1981. Geochemistry and
petrogenesis of the Fiskenaesset anorthosites complex, southern
West Greenland: Nature and parent magma. Geochim. Cosmochim.
Acta 45, 711–725.
Weaver, B.L., Tarney, J., Windley, B., Leake, B.E., 1982. Geochemistry and petrogenesis of Archean metavolcanic amphibolites from
Fiskenaesset, S.W. Greenland. Geochim. Cosmochim. Acta 46,
2203–2215.
Windley, B.F., 1970. The stratigraphy of the Fiskenaesset complex.
Rapp. - Grùnl. Geol. Unders. 35, 19–21.
Windley, B.F., Herd, R.K., Bowden, A.A., 1973. The Fiskenaesset
Complex, West Greenland, Part 1. Grønlands Geol. Unders. Bull.
106 80 pp.
Windley, B.F., Bishop, F.C., Smith, J.V., 1981. Metamorphosed
layered igneous complexes in Archean granulite–gneiss belts.
Annu. Rev. Earth Planet. Sci. 9, 175–198.