The Late Mesoproterozoic Sirdal Magmatic Belt, SW Norway: Relationships between magmatism and metamorphism and implications for Sveconorwegian orogenesis

TRi

Na

b

c

a

ARRAA

KSSGGPH

1

atGt

h0

Precambrian Research 265 (2015) 57–77

Contents lists available at ScienceDirect

Precambrian Research

jo ur nal homep ag e: www.elsev ier .com/ locate /precamres

he Late Mesoproterozoic Sirdal Magmatic Belt, SW Norway:elationships between magmatism and metamorphism and

mplications for Sveconorwegian orogenesis

. Cointa,∗, T. Slagstada, N.M.W. Robertsb, M. Markera, T. Røhra, Bjørn E. Sørensenc

Geological Survey of Norway, 7491 Trondheim, NorwayNERC Isotope Geosciences Laboratory, British Geological Survey, Nottingham NG12 5GG, UKDepartment of Geology and Mineral Resources Engineering, Norwegian University of Science and Technology, 7491 Trondheim, Norway

r t i c l e i n f o

rticle history:eceived 18 November 2014eceived in revised form 16 March 2015ccepted 6 May 2015vailable online 15 May 2015

eywords:veconorwegian orogenyW Norwayranitic batholitheochronologyetrographyigh-grade metamorphism

a b s t r a c t

The Late Mesoproterozoic Sveconorwegian Province is commonly correlated with the continent-collisionrelated Grenville Province in eastern Canada. Recently, however, the evolution of the SveconorwegianProvince in SW Norway has been strongly debated, casting doubt on a direct correlation between theseprovinces.

Metamorphism in SW Norway has traditionally been interpreted as representing a main collisionalevent between ca. 1030 and 970 Ma, followed by a contact metamorphic event at 930 Ma. Magmatismhas been grouped into a ‘syn-collisional’ suite at 1050–1035 Ma, a ‘post-collisional’ suite at 980–930 Ma,and an anorthosite–mangerite–charnockite–granite (AMCG) suite at 930 Ma. New detailed mapping andgeochronology in the area reveal a very different and much more complex evolution, and require re-evaluation of previously presented models. In this paper, we focus on the introduction and description of anewly discovered, ca. 200 km × 50 km magmatic belt, the Sirdal Magmatic Belt (SMB). Previously mappedas granitic gneisses in many areas, the existence of this large, commonly undeformed and unmetamor-phosed granitoid batholith was only recognized a few years ago (Slagstad et al., 2013a). Magmatismin this belt between 1060 and 1020 Ma precedes and overlaps the main Sveconorwegian metamorphicevent(s) that affected the region.

Our observations of cross-cutting relationships between previously metamorphosed gneisses and SMBrocks indicate that at least one episode of amphibolite- to granulite-facies metamorphism occurred inthe region during or prior to emplacement. A lack of widespread metamorphic overprinting and com-mon preservation of igneous textures in most of the SMB indicate that high-grade Sveconorwegian
metamorphism after ca. 1020 Ma was local rather than regional in SW Norway.
The orogenic evolution of SW Norway is characterized by emplacement of large volumes of graniticmagma and more localized UHT metamorphism, which is quite different from the widespread, long-lasting metamorphic evolution observed in the Grenville Province, and may point to different tectonicregimes for the two provinces.

© 2015 Elsevier B.V. All rights reserved.

. Introduction

The Late Mesoproterozoic features widespread orogenesiscross many continents that is generally ascribed to the forma-
ion of a supercontinent named Rodinia (e.g. Li et al., 2008). Therenville orogen in North America and Canada is the largest of
hese, and has traditionally been linked and correlated to the

∗ Corresponding author. Tel.: +47 73904329.E-mail address: [email protected] (N. Coint).

ttp://dx.doi.org/10.1016/j.precamres.2015.05.002301-9268/© 2015 Elsevier B.V. All rights reserved.

Sveconorwegian orogen of southwest Fennoscandia (Gower et al.,1990; Romer, 1996; Karlstrom et al., 2001; Johansson, 2009). Recentwork in the eastern Grenville Province by Gower et al. (2008) has,however, shown that this region was characterized by dextral,strike-slip movements, in contrast to the frontal-thrust ramp tec-tonics that characterize most of the Grenville Province farther west.Based on these observations, Gower et al. (2008) argue that the
Sveconorwegian Province may not be a direct continuation of theGrenville Province. The Sveconorwegian orogeny in S Norway andSW Sweden involved magmatic, metamorphic and deformationalprocesses of variable character that took place across different parts
dx.doi.org/10.1016/j.precamres.2015.05.002

http://www.sciencedirect.com/science/journal/03019268

http://www.elsevier.com/locate/precamres

http://crossmark.crossref.org/dialog/?doi=10.1016/j.precamres.2015.05.002&domain=pdf

mailto:[email protected]

dx.doi.org/10.1016/j.precamres.2015.05.002

5 ian Re

o(its

bbTgoTmwaohgip

2

ttbcTiBigaGfSaoPamptb(mtboswtmbece

rtwn1m

8 N. Coint et al. / Precambr

f the orogen at different times between ca. 1200 and 900 MaBingen et al., 2008a). Tying these events into a coherent tectonicnterpretation has proved difficult and controversial, and at presenthere is no consensus as to the large-scale tectonic processes thathaped the orogen (e.g., Möller et al., 2013; Slagstad et al., 2013a,b).

The Sirdal Magmatic Belt (SMB) is a recently discovered graniticatholith in the southwestern part of the orogen (Fig. 1a), emplacedetween ca. 1060 and 1020 Ma (Slagstad et al., 2013a; this work).he purpose of this paper is to present new field, petrological andeochronological data from the SMB, as well as from para- andrthogneiss xenoliths within and forming the host rocks of the SMB.hese data provide new constraints on the timing and character ofagmatic and metamorphic events in the SW part of the Sveconor-egian orogen. We focus on primary observations, i.e., field data

nd cross-cutting relationships, that provide first order constraintsn the orogenic evolution. The new data reveal a more complexistory for this part of the Sveconorwegian orogen, with severaleographically overlapping events of different character and tim-ng, that need to be understood in detail, to gain insight into therocesses that shaped the orogen.

. Regional geology

The study area is located in the southwesternmost part ofhe Sveconorwegian Province and has previously been interpretedo consist of high-grade ortho- and lesser amounts of para- andanded gneisses that were deformed and metamorphosed betweena. 1035 and 970 Ma (Bingen et al., 2008a and references therein).his orogenic event – the Sveconorwegian orogeny – is commonlynterpreted to reflect a continent–continent collision betweenaltica–Laurentia on one side and a third continent, typically

nferred to be Amazonia, on the other. Although the Sveconorwe-ian Province in S Norway and SW Sweden is commonly regardeds a continuation of the Grenville Province in eastern Canada (e.g.,ower et al., 1990; Karlstrom et al., 2001), recent work has argued

or significant differences in orogenic style (Bybee et al., 2014;lagstad et al., 2013a). In particular, the Sveconorwegian orogens a whole appears to have experienced significantly more syn-rogenic magmatism than the Grenville Province (Falkum andetersen, 1980; Slagstad et al., 2013a; Vander Auwera et al., 2014),nd also features localized ultra-high-temperature (UHT) meta-orphism (Drüppel et al., 2013). In contrast, work in the eastern

art of the orogen demonstrates the effects of large-scale contrac-ional processes and high-pressure metamorphism (as discussedy Möller et al., 2013). However, as shown by Slagstad et al.2013a,b) and Bybee et al. (2014), and expanded upon here, the

etamorphic and magmatic evolution in the southwestern part ofhe province does not appear compatible with indenter tectonism,ut is more in line with that expected in a long-lived accretionaryr Andean-type orogen related to subduction of oceanic litho-phere. The arguments for an Andean-type, accretionary settingere summarized by Slagstad et al. (2013b), and revolve around

he regional-scale, long-lived magmatism and high-T/low-P meta-orphism that apparently is rather local, and therefore unlikely to

e related to crustal thickening alone. The evolution in the east-rn part of the orogen is characterized by large-scale thrusting andontinental subduction; however, these processes are not unique toither continent–continent collision zones or accretionary orogens.

Given this apparent controversy, it is interesting to brieflyeview the evidence for continent–continent collision (indenterectonism) in the Sveconorwegian Province. Early work in the Nor-
egian part of the Province, in the 1970s and into the 1980s, didot argue for continent–continent collision at all. Torske (1976,977) argued that the S Norway Precambrian region was a seg-ent of a cordilleran-type orogen based mainly on lithological
search 265 (2015) 57–77

considerations. In this case, however, it appears that Torske wasreferring to older, ca. 1.5 Ga metavolcanic rocks that formed dur-ing the Telemarkian orogeny (Bingen et al., 2005; Roberts et al.,2013), as well as younger, granitic rocks that we now know formpart of the SMB. This confusion is nearly unavoidable given the lackof geochronological data at the time; he nevertheless concludedthat the period between ca. 1750 and 850 Ma was “referable to asingle cordilleran-type orogenic belt constructed along a largely sta-tionary andean-type continental margin”. A few years later, Falkumand Petersen (1980) and Falkum (1985) argued, based on work inSW Norway, that the evolution of the Sveconorwegian orogenicbelt was “somewhat analogous to the development of a cordilleran-type mountain belt”. These workers also realized that differentparts of the orogen showed rather different orogenic styles, fromwest to east. They defined a “core zone” in the west (area under-lain by the SMB) characterized by emplacement of voluminoussyn-orogenic granitoids, and argued that this zone represented a“true accretion zone with considerable crustal addition”. Farther east,they defined a “marginal orogenic zone” (in SW Sweden) charac-terized by thrust- and mylonite zones, and large-scale imbricatestructures.

More recent work (e.g., Stephens et al., 1996; Viola et al., 2011;Slagstad et al., 2013a,b; this work) has shown that Falkum’s subdivi-sion of the Sveconorwegian orogen is largely correct. Nevertheless,a continent–continent collision model has been invoked more orless indiscriminately to account for the Sveconorwegian orogenicevents since the mid-1990s (e.g., Romer, 1996; Stephens et al.,1996). It is, however, not entirely clear to us what geologicalfeatures require continent–continent collision per se. A potentialsource of confusion is Falkum’s use of the term “plate collision”,which has later been interpreted as meaning continent–continentcollision (e.g., Stephens et al., 1996). However, from Falkum’s dis-cussion and tectonic cross-section (Falkum and Petersen, 1980;Falkum, 1985) it seems quite clear that they were referring to aconvergent ocean-continent margin, with eastward subduction ofoceanic crust beneath SW Norway.

Starting in the late 1980s to early 1990s, the main reasonsfor invoking continent–continent collision for the Sveconorwe-gian Province appear to have been correlation with the GrenvilleProvince (e.g., Gower, 1985; Gower et al., 1990) and paleomagneticstudies (Park, 1992). The correlations with the Grenville Provincewere mainly based on work in the eastern parts of the Sveconorwe-gian Province, where the tectonic style is similar to that observed inthe Grenville, involving large-scale thrusting and crustal imbrica-tion. So from work in the east alone, such a correlation appearsreasonable. However, as discussed by Slagstad et al. (2013a,b),Bybee et al. (2014) and in this article, newly discovered (or, morecorrectly, rediscovered following Falkum and others) magmaticand metamorphic features in the western part of the orogen appearincompatible with an overall continent–continent collision model,and more in line with features expected in an accretionary orAndean-type orogen. As for the paleomagnetic studies, the range ofpossible Baltica-Laurentia-Amazonia configurations allowed by theavailable data render conclusions based on these data inconclusive(e.g., Pisarevsky and Bylund, 2006 and references therein).

To summarize, the evidence invoked for continent–continentcollision in the Sveconorwegian Province is either circumstantial(correlations with the Grenville and paleomagnetic studies) or notunique to such settings (large-scale thrusting and crustal imbri-cation). Nevertheless, this model has become an undisputed andself-reinforcing truth over the last two to three decades. Although ahighly axiomatic statement, we argue that insight into the nature ofthe Sveconorwegian orogeny can only come from a greater under-standing of the magmatic, metamorphic and structural evolution
of the rocks defining this orogen, rather than on unsubstantiatedcorrelations with other areas.

N. Coint et al. / Precambrian Research 265 (2015) 57–77 59

F outlin( and it

3

lwF1saffontgi“padW

asmtn

ig. 1. (a) Simplified geologic map of the Roglaland–Vest-Agder area, including the

c). (b) The Knaben area and xenolith-rich zones in this area. (c) The garnet granite

. Importance of rock nomenclature

A significant source of confusion regarding the geologic evo-ution of SW Norway is erroneous identification of rocks and/or

rong usage of rock nomenclature on regional-scale maps (e.g.,alkum, 1982) and in the literature (e.g., Bingen and van Breemen,998b). It is not our intention to place the blame for this confu-ion on a few persons or studies, but rather point out that thereppears to have been a common practice of labeling Precambrianelsic medium- to coarse-grained rocks a “gneiss”. Most of the rocksrom the SMB have previously been mapped as granitic gneissesr augen gneisses. Locally, these names are justified, however,ormally, the rocks are undeformed and preserve their igneous tex-ures (Fig. 2a–c), or are only relatively weakly foliated. Most “augenneisses” within the SMB are, for example, porphyritic granitesn which the K-feldspar phenocrysts have not been recrystallized.Banded gneisses” within the SMB are commonly found to be map-able zones that are rich in xenoliths, most likely of older, deformednd metamorphosed host rock, rather than zones where strongeformation has resulted in lithological transposition and banding.e will come back to that in the next section.An important contribution of this paper is to present new field

nd petrographic descriptions of rocks from a large area in the
outhwestern part of the province, that will serve as a basis forore up-to-date interpretations of the evolution of the area. It is,
herefore, necessary to briefly explain our usage of the relevant rockomenclature.

e of the Sirdal Magmatic Belt (SMB). The squares show the areas covered by (b) ands host rocks.

We use ‘gneiss’ to describe rocks that display evidence forsubsolidus deformation (i.e., granoblastic texture, extensive recrys-tallization, alignment of the metamorphic minerals) and mm- tocm-scale compositional layering resulting from metamorphic seg-regation (typically mm- to cm-spaced foliae rich in biotite). ‘Bandedgneiss’ refers to lithologically distinct rocks (e.g., mafic and felsic)that are interlayered, typically on a cm- to m-scale, as a result ofstrong deformation. Geologic maps (e.g., the Mandal map sheet ofFalkum, 1982) show numerous zones of “banded gneiss” in the areacovered by the SMB. However, as shown below, these zones aremainly comparatively narrow, sinuous zones in the SMB granitesthat are rich in xenoliths (Fig. 1b) and did not form as a result oftectonic transposition or interleaving of different lithologies. Thesezones are typically well located on the Mandal map sheet, but theirdescription is erroneous and misleading when discussing the geo-logic evolution of the area. We use ‘foliated granite’ for graniticrocks that have a planar fabric, commonly defined by orientedbiotite, but do not display evidence of metamorphic segregation.It is typically difficult to determine if this foliation was acquiredduring flow of the crystallizing magma, or later, during subsolidusdeformation, and it is also unclear to what extent these fabricsresult from emplacement of later magma batches/plutons or fromexternal tectonic forces. ‘Migmatites’ are rocks that preserve evi-
dence of in situ partial melting, typically in the form of leucosomes(Sawyer, 2008). For igneous rocks that do not show evidence ofsubsolidus deformation or metamorphic overprinting, we use mag-matic rock terminology based on the IUGS classification (Le Maitre

60 N. Coint et al. / Precambrian Research 265 (2015) 57–77

Fig. 2. (a) Undeformed non-foliated porphyritic biotite granite (Norwegian krone for scale). (b) Foliated porphyritic biotite granite, where the foliation is defined by thealignment of K-feldspar phenocrysts. (c) Contact between a hornblende biotite granite at the bottom right and an equigranular pinkish biotite granite on the top left cornero iotite

b in thi

etrrtcpcgwa

f the photo. (d) Xenolith in leucogranite. (e) Equigranular, undeformed garnet belonging to the metaSMB lithologies. (For interpretation of the references to color

t al., 1989). There is also some confusion related to usage of theerm “charnockite” in SW Norway. We reserve this term to igneousocks of a granitic (sensu stricto) composition, where orthopy-oxene (opx) is present as a primary igneous phase and wherehe magmatic texture of the rock is preserved. Granitic rocks thatontain metamorphic opx as a result of granulite-facies metamor-hism are not referred to as charnockitic. By this definition, there
an be no such rock as ‘migmatitic charnockite’ or ‘charnockiticneiss’ (cf., Drüppel et al., 2013). This distinction is important ase find both igneous and metamorphic opx in granitic rocks in the
rea.

granite containing fragments of xenolith. (f) Porphyroclastic opx-bearing gneiss,s figure legend, the reader is referred to the web version of the article.)

The metamorphic para- and orthogneisses that constitute thexenoliths and host rocks to the SMB are named according totheir mineral assemblages without reference to their commonlydifficult-to-identify protolith.

4. Field description and petrography of the SirdalMagmatic Belt

In the following section we describe the different lithologiesthat constitute the SMB, and also the nature of xenoliths and hostrocks found within and around the batholith. Determination of rock

N. Coint et al. / Precambrian Res

A P

Q

monzo grani te

syenogranite

granodiorite

tonalite

alk.

F g

rani

te

qtz syeni te

qtz rich granitic rocks

qtz monzoni teqtz

monzodio rite

Leu cograni te

Garnet grani te

Main SMB

MetaSMB

Fb

ncpS

mnmi‘ihw

4

bTbKvotnai2t

lKQtig(cmw

The nature of the xenoliths varies with the location in the belt.

ig. 3. QAP plot of samples from throughout the SMB. The modal proportions areased on XRD analyses.

ames for magmatic rocks is based on thin-section observationsombined with X-ray diffraction (XRD) data. The XRD method isresented in Electronic Supplement 1, and the data in Electronicupplement 2.

The geographic extent of the SMB, as shown in Fig. 1a, is based onapping by the Geological Survey of Norway since 2000, abundant

ew geochronologic data (presented in Fig. 1a), and the existingap by Falkum (1982). The lithologies in the SMB can be divided

nto four main groups, referred to as ‘porphyritic biotite granite’,leucogranite’, ‘garnet granite’, and ‘metaSMB’. The latter groupncludes granites in the SW part of the SMB that have undergoneigh-grade metamorphism and deformation, the significance ofhich is discussed below.

.1. Porphyritic biotite granite

By far the most common rock type in the SMB is porphyriticiotite granite to minor granodiorite, locally containing amphibole.hese rocks are characterized by a medium-grained, biotite-earing groundmass with 2–3 cm, prismatic to stubby, reddish-feldspar phenocrysts (Fig. 2a). The proportion of phenocrysts isariable, from non-existent or sparse to constituting more than 80%f the rock. Phenocryst sizes are also variable, ranging from ca. 1o 6–7 cm, and there seems to be a rough correlation between phe-ocryst abundance and size. Magmatic foliations defined by thelignment of K-feldspar phenocrysts are common (Fig. 2b), whichs also observed in other granitic batholiths (Vernon and Paterson,006; Paterson et al., 2008), whereas post-crystallization deforma-ion with development of gneissic textures is rare.

The SMB rocks can be classified as porphyritic to equigranu-ar biotite granite, and locally granodioritic (Fig. 3). Along with-feldspar, plagioclase also occurs as small phenocrysts locally.uartz is part of the matrix and is anhedral (Fig. 4a). Biotite is

he main ferro-magnesian phase and is commonly subhedral tonterstitial. Hornblende is rare, but is found in monzogranite andranodiorite, locally accompanied by pale green clinopyroxenecpx). Accessory minerals are apatite and zircon. Some rocks also
ontain sphene and strongly metamict allanite (Fig. 4a). Secondaryinerals such as chlorite, actinolite, epidote, sericite and minorhite micas are also present.
earch 265 (2015) 57–77 61

Schlieren defined by local concentrations of mafic minerals inthe granite are common. Preserved magmatic contacts between dif-ferent varieties of porphyritic biotite granite that can be mappedover several kilometers have been observed (Fig. 2c). These con-tacts are pristine and have not been overprinted by later subsolidusdeformation. Although we have yet to distinguish and map outindividual plutons within the SMB, recently obtained, unpublishedairborne radiometric data suggest that much of the belt consistsof elongate, N–S-oriented bodies. We are currently trying to definethese bodies and understand their geometry and geological signif-icance.

4.2. Leucogranite

A characteristic unit within the central part of the SMB is aleucocratic granite (Fig. 1a) with a characteristic texture rangingdiffusely and irregularly between coarse grained and pegmatitic(Fig. 2d). The contacts to surrounding porphyritic granite are sharp,intrusive and in one area, north of Lysefjorden, the two types ofgranite are separated by a layer of gneissic migmatite that is typi-cally only tens to a few hundred meters thick but can be followedfor several kilometers. The leucogranite locally contains abundantxenoliths of ca. 1.5 Ga metavolcanics and younger metasediments(Slagstad and Marker, unpublished data). These xenoliths have astrong preferred orientation, and form dm- to m-thick screens(sheet-like xenoliths) in the leucogranite. From a distance, thexenoliths give the rocks a layered appearance. However, this layer-ing is not related to subsolidus deformation.

The leucogranite is a medium-grained to pegmatitic equigran-ular granite with sparse biotite (Fig. 4b). The proportions ofK-feldspar (perthitic orthoclase) and plagioclase vary, so that therock composition varies between monzo- and syenogranite (Fig. 3).Biotite is the only ferromagnesian phase and is typically subhedral.Secondary minerals are chlorite, epidote and white mica, indicat-ing local retrogression. Mineral habits are anhedral to subhedraland present no apparent alignment. No evidence for recrystalliza-tion has been observed in the thin section. Quartz, which is presentas several mm-long crystals, only displays undulose extinction.

4.3. Xenolith-rich zones within the SMB

N–S-oriented zones rich in xenoliths are characteristic of theSMB (Fig. 1b), and were previously mapped as banded gneisseson the 1:250,000 Mandal mapsheet (Falkum, 1982). In the por-phyritic granite and leucogranite, xenoliths or larger screens ofmetamorphosed and deformed rocks tend to be concentrated inN–S-trending, up to hundreds-of-meter-wide zones that in manycases can be followed for several kilometers. The density of xeno-lith is highly variable. In some areas, individual xenoliths are onlyseparated by decimetre-scale tongues of granite, whereas in otherplaces, the xenoliths are sparse. The significance of these zones isunclear, but we speculate that they may represent contact zonesbetween different intrusions/plutons, i.e., are similar to ‘screens’ ofwall-rock described in more recent Cordilleran batholiths such asSierra Nevada (e.g. Bateman, 1983; Bartley et al., 2006; Patersonet al., 2012). Individual screens can be hundreds of meters long andseveral meters wide and typically dip shallowly (30–20◦) east- orwestwards. In Lysefjorden, where exposure is remarkable, the ori-entation of the xenoliths defines broad, open folds (Fig. 5a, b). Innumerous locations throughout the belt, the foliation in the xeno-liths is clearly cut by SMB granites (Fig. 5c, e, f). In the garnet granite,xenoliths are more randomly distributed (Fig. 1c).

Xenoliths in the southern part of the belt (Knaben, Åseral, Man-dal) are granitic gneisses (Fig. 5c), amphibolite fragments (Fig. 5d),migmatitic grey gneisses (Fig. 5e), and foliated granite.


F b) LeuW

t1mtFma2

g

4

xgtm(z

l(sptm

ig. 4. Photomicrograph of the main SMB granites. (a) Porphyritic biotite granite. (hitney and Evans (2010).

In the Suldal region, in the northern part of Rogaland (Fig. 1a),he weakly deformed to undeformed SMB rocks intruded into ca..5 Ga upper-greenschist to amphibolites-facies metaplutonics andetavolcanics (Bingen et al., 2005; Roberts et al., 2013). Xenoliths in

his region are commonly amphibolie-facies metavolcanics (Fig. 5f).arther north, in the Hardangervidda region, undated plutons thatay also be part of the SMB, intrude into both ca. 1.5 Ga rocks and

ca. 1.26–1.21 Ga supracrustal belt (Sæsvatn-Valldal; Bingen et al.,002; Brewer et al., 2004).

In the central and southern parts of the belt, three main litholo-ies prevail in the N–S-trending xenolith-rich zones.

.3.1. Fine-grained, grey migmatitic gneissFine-grained, felsic grey gneiss is quite commonly found in the

enolith-rich zones (Fig. 5e). They display a characteristic fine-rained granoblastic texture (Fig. 6a). The mineral assemblage inhese rocks is quartz, plagioclase, K-feldspar, biotite and opaque

inerals. Leucosomes of granitic composition are 0.3–5 cm thickFigs. 5e and 6a), and define the foliation in the rocks. Apatite andircon are present as accessory minerals.

Near Konsmo, in the southeastern part of the SMB, one of theargest xenolith screens found, has been studied in more detailFig. 1b). This xenolith is several hundred meters wide across
trike and of unknown length, but smaller, similar xenoliths inorphyritic SMB granite are found along strike as far as 10 kmo the north. The screen is composed of migmatitic, light-colored,
edium-grained orthogneiss with dm-thick layers of fine-grained

cogranite. (c) Garnet granite. (d) MetaSMB. Mineral abbreviations can be found in

amphibolite (Fig. 7). Contact relationships cannot be defined atKonsmo, but the area is surrounded by underformed porphyriticgranite, and similar, smaller xenoliths to the north have metamor-phic fabrics that are clearly cut by the granites (Fig. 5e).

4.3.2. Granitic flaser gneissThis type of xenolith (Fig. 5c) is common in the xenolith-

rich zone that runs through Knaben and represents a deformedmegacrystic granite. The rock displays variations in the intensityof deformation so that it is possible to recognize the protolith.The foliation is anastomozing (Fig. 5c) and defined by alignmentof biotite (Fig. 6b). K-feldspar megacrysts are commonly recrys-tallized into sub-millimeter grains (Fig. 6b). Relics of a medium- tocoarse-grained magmatic texture are rare, and large portions of therock are recrystallized with development of a polygonal texture.The mineral assemblage is quartz, K-feldspar, plagioclase, biotiteand opaque minerals. Zircon, apatite and metamict allanite areaccessory minerals.

4.3.3. AmphiboliteAmphibolite is sparse and, where found, displays a nemato-

blastic texture where hornblende and biotite define the foliation(Fig. 6c). Plagioclase is the main felsic mineral. Sphene is quite
abundant in these rocks, whereas apatite and opaque minerals areaccessory phases.
The second group of xenoliths also labeled amphibolite in thefield, derive from more felsic quartz monzonitic protoliths, as


Fig. 5. Outcrop photos of xenolith-rich zones and individual xenoliths. (a) View of a xenolith-rich zone in Lysebotn, along Lysefjorden, with zoom-in on a particular xenolithat the base of the fjord. (b) Zoom-in on the cliff in Lysebotn. (c) Flaser granitic gneiss cut by foliated porphyritic SMB granite at Knaben. (d) Amphibolite fragment in granitic-gneiss xenolith in SMB granite (Åseral area). (e) Xenoliths of migmatitic grey gneiss in undeformed SMB granite. Large xenolith at middle right of photo is ca. 1 m thick. (f)Porphyritic biotite granite with a xenolith of a ca. 1500 Ma metavolcanic rock. Hammer for scale.

64 N. Coint et al. / Precambrian Re

Fig. 6. Photomicrographs of xenoliths samples in the xenolith-rich zones. (a)Mim

Ktwpa

igmatitic grey gneiss. Note the granofels texture is cut by a leucosome display-ng an igneous texture. (b) Flaser granitic gneiss with recrystallized K-feldspar

egacrysts. (c) Amphibolite.

-feldspar and quartz are part of the assemblage. The texture of
hese meta qtz-monzonite rocks is nematoblastic to granoblasticith the foliation defined by hornblende and biotite. One sam-le contains pale green cpx. Sphene and epidote are present, andllanite, apatite and zircon are accessory minerals. Large, subhedral
search 265 (2015) 57–77

K-feldspar crystals containing euhedral inclusions of plagioclaseand biotite, possibly representing injected melt from the surround-ing porphyritic granite, are common.

4.4. Garnet granite

The garnet granite is located west of the main SMB, and unlikethe N–S-oriented trend of the main belt, the garnet granite forms anearly circular body (Fig. 1c). The garnet granite (Fig. 2e) intrudesdeformed, high-grade metamorphic ortho- and paragneisses, andthese rocks are also found as up to several hundred meter-longxenoliths or screens in the granite. As such, the undeformed gar-net granite is an important marker for the timing of deformationand high-grade metamorphism in this part of the Sveconorwegianorogen. Modal abundances of garnet in the garnet granite increasein the vicinity of garnet-rich paragneiss xenoliths suggesting thatthe garnet are somehow related to the presence of the latter,however, the origin of the garnet themselves is not clear. Theymay be inherited (xenocrysts) from assimilation of the migmatiticmetapelites, or be a peritectic reaction product from this assimi-lation. Garnet in the granite is typically larger than garnet in themetapelites, and probably has a different composition (based oncolor), suggesting they are new reaction products. It is also com-mon to find xenoliths of garnet-rich paragneiss that taper off intoschlieren that in turn disappear into the granite, leaving behindtrails of cm-size, red-brown garnet. The garnet granite is classifiedas syenogranite (Fig. 3), and the rock displays a medium- to coarse-grained equigranular texture. Plagioclase, K-feldspar (orthoclase)and quartz are typically anhedral. Garnet is present as anhedral tosubhedral crystals that locally contain rounded quartz inclusions(Fig. 4c).

Plagioclase is strongly sericitized, whereas K-feldspar displaysa brown dusty coat in plane light. Biotite, present as subhe-dral to anhedral crystals, is not always preserved. Accessoryminerals are zircon and monazite. Locally, the garnet granite under-went partial retrogression. Only rare relics of brown biotite ininclusions in feldspars or garnet are preserved. More commonly,ferromagnesian-mineral pseudomorphs have been replaced bychlorite, actinolite, epidote and minor calcite. Despite this retro-gression, garnet is still preserved throughout most of the granite.

Sericitized feldspars and altered mafic phases define a weakmagmatic foliation that is difficult to see in thin section, but canbe seen in hand specimen. The deformation recorded is weak, andevidence for high-temperature deformation and recrystallizationhas not been observed. The rock is affected locally by grain-sizereduction associated with veins containing epidote. These veinsare probably coeval with the retrogression observed in the maficminerals and related to fluid circulation. This retrogression in thegreenschist facies could be associated with the emplacement of theCaledonian nappes in the surrounding area (Fig. 1a).

4.5. Xenoliths and host rocks in and around the garnet granite

The garnet granite (Fig. 1c) was emplaced into high-gradeopx-bearing orthogneisses that are variably migmatitic as well asgarnet-bearing paragneisses.

4.5.1. Opx-bearing orthogneissThese rocks vary from opx granitoid to opx biotite gneiss. The

rocks yield interpreted protolith ages between 1210 and 1230 Ma(Slagstad and Marker, unpublished data). Where little deformed,the rocks are medium-grained, equigranular, allotriomorphic opx
granites. Both plagioclase and perthitic orthoclase are present. Pla-gioclase is commonly sericitized. Orthoclase can be poikilitic andcontain subhedral inclusions of plagioclase. Quartz is anhedral,commonly poikilitic, and can be more than 1 cm long and display


F the cm

spt

actSgfst

4

m

ig. 7. Field photograph, CL images of dated zircons with arrows showing whereigmatitic screen at Konsmo (see Fig. 1a, b for location).

light undulose extinction. Orthopyroxene is anhedral and rarelyreserved. Accessory minerals are zircon and apatite. The propor-ion of mafic minerals varies.

In the gneissic equivalent (Fig. 8a), the texture is granoblasticnd the foliation is defined by the alignment of biotite, opx, plagio-lase and quartz. Plagioclase displays the same characteristic as inhe less deformed granitoid, with higher relief and sericitization.econdary minerals are common and include epidote, chlorite andreenish amphibole. Opx is commonly altered. In some samples,eatures interpreted to represent melt films (see arrows in Fig. 8a)uggest that the rocks underwent low-degree partial melting orhat most of the melt was extracted from the rocks.

.5.2. Opx-bearing migmatitic orthogneissUnlike the opx-bearing orthogneisses, these rocks are clearly

igmatitic, which is best seen in outcrop where cm-thick granitic

rystals were analyzed, and Tera-Wasserburg plot of the U–Pb data from a large

leucosomes can be observed. Geochronologic data from theserocks suggest protolith ages around 1450 Ma (Slagstad and Marker,unpublished data). In thin section, the texture is granoblastic andquartz grains mostly display lobate boundaries (Fig. 8b). The foli-ation is defined by dark, opx-rich layers and, more leucocratic,quartzo-feldspathic layers. The rocks contain plagioclase, quartz,biotite, opx, and minor K-feldspar. Unlike in the opx-bearingorthogneisses described above, these migmatitic rocks contain verywell-preserved opx crystals that are blocky and anhedral. Biotite isalso present in the rock. Zircon, apatite and opaque minerals arepresent as accessory minerals.

4.5.3. Garnet-bearing paragneissPreviously mapped as the Gyadalen metapelite (Hermans et al.,

1975), this unit can be followed over large distances south ofthe garnet granite. In the field, this unit is quite distinct as it


F garneo aragn

cmicagAcAppttaq

g

ig. 8. Photomicrographs of xenoliths and host rocks sampled within and around thepx-bearing orthogneiss (ca. 1450 Ma protolith). (c and d) Migmatitic grt-bearing p

ontains abundant garnet. The association of garnet with alu-inosilicates, such as sillimanite or cordierite in some levels,

ndicate a pelitic or semipelitic protolith. At the scale of the out-rop, cm- to dm-thick layers of Al-rich migmatitic parageneisslternate with migmatitic meta-aluminous paragneiss (Fig. 8c),arnet-bearing leucosomes and two-pyroxene-bearing gneisses.l-rich migmatitic layers contain garnet, Zn-bearing spinel, plagio-lase, K-feldspar, biotite, quartz and either sillimanite or cordierite.ccessory minerals are monazite, zircon and ilmenite. The sam-le in Fig. 8d contains two generations of garnet, one subhedral,oikiloblastic with large quartz inclusions and the second genera-ion defining the foliation and riddled with sillimanite inclusions. Inhe matrix, the foliation is defined by prismatic sillimanite. Meta-
luminous layers contain opx, garnet, plagioclase, K-feldspar anduartz.
Layers of pyribolite are also found. The texture of the latter isranoblastic. There is no modal layering or any foliation observed

t biotite granite. (a) Opx-bearing orthogneiss (ca. 1230 Ma protolith), (b) migmatiticeiss.

in these layers. Plagioclase, opx and cpx exsolving pigeonite are themain mineral phases in the pyribolite. Accessory minerals includeopaques, brown hornblende and apatite.

A sample of the migmatitic paragneissic unit (ROG643A) wascollected close to the contact with the garnet granite in Oltedal,for geochronological analysis, discussed below (Figs. 1c and 8c, seeElectronic Supplement 3 for the precise location of the sample).

Other lithologies such as forsterite marble are also encoun-tered in the garnet granite. Xenoliths displaying similar high-gradeassemblages are sparse in the rest of the SMB.

4.6. Metamorphosed and deformed SMB granites (metaSMB)

In the SW part of the SMB, close to the city of Moi (Fig. 1a), therocks are overprinted by granulite-facies metamorphism and arelocally migmatitic. The porphyritic to megacrystic granite in thislocation is now a porphyroclastic augen gneiss with large crystals


Fig. 9. Schematic cross section for the Gyadalen transect (see location in Fig. 1a) illustrating the rock relationships. Note: The nature of the contact between the metapelitesand the migmatitic qtz-dioritic gneiss is unknown. The foliation trends were drawn based on the foliation measured in the field. Simplified cross-section along the Gyadalentransect (see Fig. 1 for location) showing the relationships between opx-bearing gneisses and the SMB.

Fig. 10. (a) Migmatitic quartz dioritic gneiss. (b) Foliated hornblende biotite granitic gneiss cut by weakly foliated SMB biotite granite. The hand lens used as a scale in bothp

od(tatttdotMmgt

ictures is 2.5 cm long.

f grey K-feldspar (Fig. 2f). These rocks have been previouslyescribed in the literature as part of the ‘Feda augen gneiss’Bingen and van Breemen, 1998b). Farther north, around Tonstad,he same (on earlier maps) granitic unit is only weakly foliatednd unmetamorphosed, and does not stand out from the rest ofhe SMB. We therefore consider the Feda augen gneiss as part ofhe SMB. The extent of the metamorphism and the deformation inhe SW part of the SMB is not yet well constrained, but evidence ofeformation can be observed at least 8 km east of Moi. In this partf the belt, it is also possible to observe gneissic screens similaro those observed farther north within the SMB, described above.

ineral assemblages are different however, with opx being part ofost of the assemblages. Cross-cutting relationships between the

neissic screens and the host metaSMB are less evident because ofhe superimposed deformation.

MetaSMB rocks tend to contain less modal quartz (20–30%) thanunmetamorphosed SMB rocks (Fig. 3) and plot in the lower partof the monzogranite field. Whether the protolith was less evolvedor quartz-rich melt left the system is unclear. In thin section, leu-cosomes display an igneous texture with large K-feldspars andquartz crystals that are unstrained (Fig. 4d). The rest of the rockdisplays a granoblastic texture. K-feldspar porphyroclasts are sur-rounded by a matrix of plagioclase, perthitic K-feldspar, quartz,two pyroxenes, brown hornblende, and biotite (Fig. 4d). Close tothe anorthosite, opx is the main ferro-magnesian mineral. Horn-blende appears locally as poikilitic crystals containing inclusions of
opaque minerals, plagioclase and apatite. Accessory minerals arezircon and apatite. Secondary minerals such as chlorite and epi-dote are common. Foliations in the metaSMB granites are variablydeveloped. Some samples display a strong foliation defined by the

6 ian Research 265 (2015) 57–77

aod

5

hascp

5

seleaiodmcsHnitrt(KbacmUmsmmo

gCegbdt

5

sosg(t1

Fig. 11. Photomicrographs of two xenolith/host-rock pendants sampled along the


lignment of most of the minerals present, whereas other samplesnly preserve evidence of high-temperature recrystallization withevelopment of a polygonal texture.

. Host rock characteristics and relationships to the SMB

Identifying the nature of the contact between the SMB and itsost rocks is crucial to understand the regional geologic evolution,s SMB granitoids can be used as time markers. In the followingection we describe in detail the nature of the host rocks and theirontacts with the SMB, focusing on the western and southwesternart of the SMB.

.1. Gyadalen transect

The nature of the western boundary of the SMB varies alongtrike as a result of variable overprinting by later metamorphicvents. Along the Gyadalen transect (Fig. 1a), which has been theocation of several previous studies (Hermans et al., 1975; Möllert al., 2002, 2003; Westphal et al., 2003), two main rock typesre found as screens and xenoliths in a weakly foliated gran-te, the latter interpreted as part of the SMB (Fig. 9). The firstne is a high-grade migmatitic hornblende two-pyroxene quartzioritic gneiss (Fig. 10a). The texture is medium grained, alliotro-orphic to locally granoblastic (Fig. 11a). The mafic minerals are

px that exsolved pigeonite, opx and hornblende. Opx and cpx areeveral mm long but also appear as smaller polygonal crystals.ornblende is present both as poikilitic subhedral and polygo-al smaller crystals and is pleochroic from pale to dark brown

n plane-polarized light. The rock is rich in mm-size apatite crys-als. Plagioclase, K-feldspar and quartz constitute the rest of theock. The second rock type is a foliated hornblende biotite grani-oid gneiss with local recrystallization of mm-size polygonal grainsFigs. 10b and 11b). The mineral assemblage is quartz, plagioclase-feldspar (perthitic orthoclase and microcline), hornblende, andiotite. Opaques, zircon and apatite are the main accessory miner-ls and are quite abundant, whereas sphene is sparser. Interstitialrystals of opaque minerals and K-feldspar are interpreted as thinelt films, suggesting that the rock underwent partial melting.nlike the migmatitic two-pyroxene gneiss, evidence of partialelting is restricted to microstructural observations. There is no

egregation of the melt phase. Chlorite and actinolite are secondaryinerals and are rare. The foliation is mainly defined by the align-ent of biotite, whereas more blocky minerals are only slightly

riented.The granite cutting the foliation in both gneisses is a medium-

rained, weakly deformed biotite granite (Figs. 10b and 11c).rystals are subhedral and often display lobate boundaries, how-ver, no evidence of extensive recrystallization associated withrain-size reduction or development of granoblastic texture haseen observed. Only a few myrmekites are visible. The foliation isefined by weakly oriented biotite crystals. Zircon and apatite arehe main accessory minerals in this rock.

.2. SW contact of the SMB

Near the Rogaland Igneous Complex, SW of Moi (Fig. 1a), aheet-like body of SMB-age porphyritic granite is deformed andverprinted by granulite-facies metamorphism (see the MetaSMBection above). Southeast of Moi, cpx-bearing porphyroclastic
ranitic gneisses (Fig. 2f) give ages of ca. 1050 and 1030 MaSlagstad et al., 2013a; this work), indicating that this part ofhe belt was overprinted by high-grade metamorphism after ca.030 Ma.
Gyadalen transect and the host granite (see Figs. 1a and 9 for location and Fig. 10for field photographs). (a) Migmatitic quartz diorite. (b) Foliated hornblende biotitegranitic gneiss. (c) Weakly foliated biotite granite that cuts the metamorphic fabricin the gneisses.


Table 1Geochronological summary of SMB granite samples.

Sample id Ea Na Rocktype Method used Age 2s

073178 404000 6480981 Sparesly prophyritic granite LA-ICP-MS 1020 15MM02230b 350409 6554133 Medium grained bt granite SIMS 1020 61MM026182b 335158 6587408 Porphyritic grainitoid LA-ICP-MSR,E 1021 10R060350 374975 6485675 Medium grained granitoid LA-ICP-MS 1022 13MM026183b 375014 6615718 Coarse grained granite LA-ICP-MSR,E 1022 8MM026302b 342702 6570573 Porpyritic granite SIMS 1022 6MM057918 350427 6592812 Porpyritic granite LA-ICP-MS 1024 6.5MM026191b 362621 6549102 Leucogranite SIMS 1025 25033152 365130 6479885 Medium grained grt LA-ICP-MS 1027 23MM02277 325424 6526915 Garnet granite ID-TIMS 1027 4MM026226 315203 6513355 Massive granite ID-TIMS 1027 9ROG218b 314258 6532437 Porpyritic granite SIMS 1028 20ROG033045 371238 6472562 Porpyritic granite LA-ICP-MS 1030 10ROG033044 363750 6478250 Porpyritic granite LA-ICP-MS 1032 10073176 382099 6487939 Porpyritic bt granite LA-ICP-MS 1033 15073177 406888 6481794 Porpyritic bt granite LA-ICP-MS 1033 7NC080812 384314 6514426 Porpyritic granite LA-ICP-MS 1033 7ROG033129 376265 6487857 Porpyritic granite LA-ICP-MS 1033 14MM02247b 338597 6572694 Porphyritic bt(hbl) granite SIMS 1035 9ROG275 318862 6531428 Garnet granite ID-TIMS 1035 6073184 406995 6477340 Porpyritic bt granite LA-ICP-MS 1036 9ROG033124 371003 6507925 Porpyritic granite LA-ICP-MS 1038 8MM36470 335134 6553657 Porpyritic granite LA-ICP-MSA 1040 10SA3-01 357948 6615161 Porpyritic granite CA-LA-ICP-MSNP 1040 10MM057845 305958 6590515 Porpyritic granite LA-ICP-MSS,E 1042 11ROG080b 371806 6546058 Porpyritic granite, foliated SIMS 1043 7MM026306b 340075 6579722 Porpyritic granite SIMS 1047 6ROG076428 383998 6458992 Porpyritic granite, foliated LA-ICP-MS 1047 9ROG076442 393173 6471857 Sparesly prophyritic granite LA-ICP-MS 1047 7SA7-58 359965 6596954 Porpyritic granite (Stropa granite) ID-TIMS 1049 5MM026297b 361312 6548784 Leucogranite SIMS 1048 16NC084354 392759 6504616 Biotite granite LA-ICP-MS 1050 37ROG032999 355935 6475987 Porpyritic granite LA-ICP-MS 1050 9ROG033000 360535 6480446 Porpyritic granite LA-ICP-MS 1050 12ROG033126 369013 6507064 Coarse grained granite LA-ICP-MS 1051 14073199 396456 6547682 Porpyritic bt granite LA-ICP-MS 1051 14ROG032997 356811 6471183 Porpyritic granite LA-ICP-MS 1052 9ROG076439 404642 6461267 Porpyritic granite LA-ICP-MS 1061 9SA8-69 370560 6607698 Porpyritic granite CA-LA-ICP-MSNP 1066 10

A and NP denote samples analyzed by the Attom and Nu Plasma ICP-MS, respectively.R,E and S,E denote samples analyzed by a Finnigan ELEMENT1 ICP-MS, using a 266 and 193 nm laser, respectively.All other samples were analyzed by an ELEMENT XR at the NGU, using a 193 nm laser.

a UTM WGS84 coordinates in zone 32N.b Previously presented in Slagstad et al. (2013a).

6

(m(lPNh(Ct(ca

6

Tt

. Geochronology

Samples of 38 granitoids have been dated to assess the spatialto identify from older granites) and temporal range of SMB mag-

atism. Data from 28 of these samples are previously unpublishedsee Table 1). Several different techniques have been used to ana-yze these samples, including Laser-Ablation Inductively Coupledlasma Mass Spectrometry (LA-ICP-MS) at the Geological Survey oforway, and at the NERC Isotope Geosciences Laboratory, Notting-am (see Roberts et al., 2011), Secondary Ion Mass SpectrometrySIMS) at NORDSIM, Stockholm (see Slagstad et al., 2013a) andhemical Abrasion Isotope Dilution Thermal Ionisation Mass Spec-rometry (CA-ID-TIMS) at the NERC Isotope Geosciences Laboratorysee Roberts et al., 2013). A detailed overview of the methods usedan be found in Electronic Supplement 1, whereas the U–Pb datare presented in Electronic Supplement 3.

.1. SMB granites

The ages of SMB granitoids are compiled in Table 1 and Fig. 12.he ages range from ca. 1060 to 1020 Ma, with what appears to bewo clusters at around 1050 and 1030 Ma. As yet, it is not known

whether these two clusters represent two main periods of mag-matic activity, or if they are artifacts of sampling and/or varyinganalytical precision.

The uncertainties of calculated ages vary greatly, from ±61 Ma(2�) for one of the samples analyzed by SIMS, to ±1 Ma for a sampleanalyzed using ID-TIMS. Most calculated uncertainties, however,are between 6 and 15 Ma. Reported ages are either concordia ages orweighted averages of 207Pb/206Pb ages. Most samples display con-cordant analyses with minor lead loss. Some samples show reversediscordance in Tera-Wasserburg plots and scatter in the 238U/206Pbratios. However, the very minor scatter in the 207Pb/206Pb data sug-gests that the discordance is due to instrumental- or lased-inducedfractionation rather than a tectonothermal event.

Discordia ages are reported from those samples that do showclear signs of non-zero age lead loss. No robust age can be extractedfrom the lower intercepts, but they commonly coincide with thetime of Mid-Paleozoic Caledonian orogenesis. Interestingly though,few granite samples show any evidence for a spread of analysesalong concordia, which may be expected if the high-grade meta-morphic event that followed at ca. 1000 Ma (e.g. Drüppel et al.,
2013; Möller et al., 2002, 2003), was regional and penetrative (seeKorhonen et al., 2013; O’Brien and Miller, 2014).


Fig. 12. Cathodoluminescence images of a few, selected zircons from some of the dated SMB granites, with arrows indicating laser traces. Numbers in parentheses refer toindividual analyses (see Electronic Supplement 3 for details), along with the 207Pb/206Pb for that analysis. Most zircons from the SMB granites show features such as oscillatoryzoning, that are interpreted to be magmatic. Rims or overgrowths that can be interpreted to reflect metamorphism are generally lacking. The figure also shows a combinedprobability plot and stock diagram showing the age distribution of SMB granites. Each bar in the stock diagram is scaled to represent the 1� and 2� uncertainty.

D

6

fsu

6

op

ata from Slagstad et al. (2013a) and this study.

.2. Garnet granite

Zircon grains from the garnet granite are commonly very dark,airly rich in inclusions and possibly metamict. Zircon from twoamples of the garnet granite were chemically abraded and datedsing ID-TIMS at the NIGL.

.2.1. ROG275Five zircon grains were analyzed from this sample, out of which

ne was clearly inherited. The remaining four grains yielded aoorly constrained upper intercept age of 1060 ± 70 Ma. However,

the youngest grain appears to have been affected by late Sveconor-wegian lead loss. Regression of the three remaining analyses yieldsan age of 1035 ± 6 Ma (Fig. 13a), which is interpreted as the mag-matic age of this rock.

6.2.2. MM02277Four zircon grains were analyzed from this sample, once again

yielding a relatively poorly constrained upper intercept age of1024 ± 23 Ma. With this sample, it also appears that the youngestand most discordant grain is not completely in line with the others.Excluding this discordant grain from the regression yields an age


1032

1028

1024

1020

1016

0.1705

0.17 15

0.1725

0.17 35

0.1745

1.715 1.725 1.735 1.745 1.755 1.765 1.775207Pb/235U

206 Pb/238 U

Intercepts at374±290 & 1034.9±5.5 [±9.1] Ma

MSWD = 0.0044

a)

1040

1020

1000

980

0.158

0.162

0.166

0.170

0.174

0.178

1.58 1.62 1.66 1.70 1.74 1.78207Pb/235U

206 Pb/238 U

Intercepts at597±33 & 1027.4±4.1 [±11] Ma

MSWD = 0.31

b)

MM02277

ROG275

CA-ID-TIMS Garnet Granite

F 02277

opt

6

6K

(fmhbeo-‘tct(fCm

2

pdstswestgttt

ig. 13. Concordia plots of zircon U–Pb data from samples (a) ROG275 and (b) MM

f 1027 ± 4 Ma (Fig. 13b). This age is within uncertainty of the ageresented for ROG275 and is interpreted as the magmatic age ofhis rock.

.3. Xenoliths and host rocks to the SMB

.3.1. Large migmatitic screen within the SE part of the SMB, nearonsmo

We separated zircon from a migmatite sample near KonsmoROG076438, Fig. 1a, b for location) in order to obtain an ageor the protolith and, more importantly, determine the age of

igmatization. The zircons from the migmatite sample are quiteeterogeneous, ranging from elongate to equidimensional andetween 100 and 200 �m. Internal textures are also highly het-rogeneous, and include (Fig. 7): (a) CL-dark grains, in some casesscillatory zoned; (b) CL-dark cores with thick, CL-intermediate tobright rims that display sector and fir-tree zoning, or more chaoticflow’ textures and that in many cases truncate oscillatory zoning inhe cores; and (c) CL-bright grains without visible cores that displayhaotic flow textures. Corfu et al. (2003) attributed similar zirconextures to reflect high-grade metamorphism, and Pidgeon et al.2000) suggested that zircon with this type of zoning crystallizedrom melt segregations in felsic gneisses. We therefore interpret theL-intermediate to -bright rims and grains to have formed duringigmatization.Five analyses of CL-dark, oscillatory-zoned grains or cores yield

07Pb/206Pb ages between 1393 and 1552 Ma (see Electronic Sup-lement 4 for data and sample coordinates). The younger age isiscordant, and excluding this analysis the remaining four analy-es yield a concordia age of 1522 ± 21 Ma. This age is interpreted ashe age of the igneous protolith of the migmatite. Nineteen analy-es of rims and CL-bright grains yield nearly concordant data with aeighted 207Pb/206Pb age of 1026 ± 14 Ma (Fig. 7). One analysis was

xcluded by Isoplot. This age is similar within error to that of theurrounding SMB granites, and probably reflects the age of migma-ization. It is possible, however, that resetting of the zircons during
ranite emplacement means that the age does not correspond tohe age of formation of the metamorphic fabric, which is consis-ently cut by the granites on a regional scale. More work is neededo elucidate the significance of these analyses.
, both from the garnet granite. Arrows in (b) indicate the position of error ellipses.

6.3.2. Garnet-bearing, migmatitic paragneiss in OltedalIn an attempt to date the age of high-grade metamorphism

of the host rock to the garnet granite, we separated zircon froma sample (ROG643A) of garnet-bearing, migmatitic paragneiss atOltedal (Fig. 1a, c for location). The separated zircons are mainlyequidimensional to elongate with rounded exteriors. Most grainsare ca. 100 �m or slightly smaller. All the grains have a CL-dark to-intermediate, diffuse, irregular zoning, either as single grains ormantling typically small, CL-bright cores, some of which are oscil-latory zoned (Fig. 14). We interpret the cores to represent detritalzircons, and the CL-darker grains and rims to have grown duringhigh-grade metamorphism of the paragneiss. The cores were toosmall to be analyzed by LA-ICP-MS, so most analyses are from thedarker grains/rims. Although we attempted to avoid mixed anal-yses, some of the analyses may represent a mix of core and rim,indicated in Fig. 14.

We analyzed 13 grains in all (see Electronic Supplement 5 fordata and sample coordinates). Additional analyses were hamperedby small grain size and problems avoiding mixed analyses. Of the13 analyses, three analyses represent mixed core/rim analyses andhave been excluded from the age calculation. The mixed analysescome from grains like those shown in Fig. 14, which have small,irregular, CL-bright cores overgrown by metamorphic zircon. Dueto the rather large spot size of the laser, we did not manage tocompletely avoid hitting core despite aiming for overgrowth. Theremaining ten analyses yield a weighted average 207Pb/206Pb ageof 1039 ± 27 Ma (Fig. 14). This age probably reflects the age of high-grade metamorphism of the Oltedal paragneiss, but similarly to theKonsmo migmatite it is possible that the zircons have been affectedby intrusion of the nearby granite.

The age of high-grade metamorphism in Oltedal overlapswith the age of migmatization at Konsmo (1026 ± 14 Ma) anda metapelite in Hunnedalen (1035 ± 9 Ma, Fig. 1a for location,Tomkins et al., 2005). This age is also similar to that suggested byBingen et al. (2008a) for the onset of Sveconorwegian metamor-phism in this region.

7. Depth of crystallization of SMB granites

Mineral composition data, acquired by microprobe, wereobtained from different parts of the SMB (black triangles in Fig. 1a)


Table 2(a) Summarized mineral chemistry and mineral assemblages for all the samples used for hornblende-plagioclase thermobarometry. (b) Pressure and temperature estimatesfor the SMB granites associated with their respective standard deviation. all: allanite, ap: apatite, bt: biotite, hbl: hornblende, ksp: K-feldspar, ilm: ilmenite, magn: magnetite,pl: plagioclase, qz: quartz, tit: titanite/sphene, zr: zircon.

Sample Hbl chemistry Pl chemistry Mineral assemblage

av Mg# sdt dev Mg# range Fetot/(Fetot + Mg) Fe3+/(Fe3+/Fe2+) An range

(a)MM26252 0.63 0.01 0.61–0.65 0.42–0.44 0.18–0.26 24–26 pl, ksp, qz, hbl, bt, tit, magn, ap, zrMM26270 0.63 0.01 0.61–0.64 0.42–0.45 0.19–0.28 23–28 pl, ksp, qz, hbl, bt, tit, magn, ap, zrMM26269 0.66 0.01 0.64–0.68 0.41–0.43 0.22–0.33 26–29 pl, Ksp, qz, hbl, bt, tit, ap, zr, oxideVAG98888 0.63 0.02 0.61–0.68 0.44–0.46 0.23–0.38 26–27 pl, ksp, qz, cpx, hbl, bt, tit, ap, zr, all, ilm, magnVAG4 0.57 0.02 0.53–0.59 0.50–0.55 0.24–0.34 24–25 pl, ksp, qz, hbl, bt, tit, ap, zr, oxideVAG98889 0.64 0.01 0.61–0.66 0.43–0.46 0.24–0.36 26–28 pl, ksp, qz, hbl, bt, ap, zr, oxideVAG98767 0.41 0.01 0.39–0.42 0.63–0.65 0.15–0.20 19–21 pl, ksp, qz, hbl, bt, tit, magn, ilm, ap, all, zr

Sample Pa sdt dev Tb sdt dev

(b)MM26252 4.01 0.10 718.15 7.25MM26270 3.78 0.14 727.85 2.73MM26269 3.87 0.26 741.88 18.64VAG98888 3.92 0.27 752.44 11.51VAG4 4.80 0.51 743.97 11.18VAG98889 4.06 0.03 744.86 4.20VAG98767 na na na na

a Anderson and Smith (1995).

idpi

Ffl

b Holland and Blundy (1994).

n order to determine the pressure and temperature (P–T) con-
itions of the emplacement of the belt. Microprobe data onlagioclase and hornblende were acquired on a CAMECA SX100
nstrument at the Microsonde Ouest, in Brest, France. The operating

ig. 14. CL images of dated zircons, and Tera-Wasserburg plot of the U–Pb datarom a migmatitic grt-bearing paragneiss near the garnet granite (see Fig. 1a, c forocation).

conditions were 15 kV and 20 nA with a focused beam of 1 �m. Con-centrations below 0.3% are considered qualitative. The full datasetis presented in Electronic Supplement 6.

In the following section, we calculated the P–T conditions ofcrystallization of the SMB using plagioclase-hornblende geother-mobarometry (Holland and Blundy, 1994; Anderson and Smith,1995). Microprobe data used in the following section can be foundin the Electronic Supplement 6. A summary of P–T estimates arepresented in Table 2.

The structural formulae of amphiboles were normalized to 13cations (Hawthorne and Oberti, 2007; Leake et al., 1997). Amphi-boles from the SMB granitoids range from edenite to hastingsitewith the majority of them being magnesiohastingsite. For simplic-ity, we will refer to them as hornblende in the rest of the paper.The Mg# in hornblende varies between 0.53 and 0.68, whereasAlIV varies between 1.39 and 1.83 pfu. Anorthite content in plagio-clase varies between An 23 and 30. Single samples usually displaya variation of 2–3 An units.

One sample, VAG98767, differs from the others. Hornblende inthis sample has a lower Mg# (0.39–0.42) for similar AlIV content(1.61–1.70) and plagioclase lower An content (19–21).

Pressures were estimated using the empirical formula pre-sented in Anderson and Smith (1995), whereas temperatures derivefrom Holland and Blundy (1994). All the rocks included in these cal-culations contain the required assemblage quartz, two feldspars,biotite, hornblende, Fe–Ti oxides, and titanite, except sampleVAG98889, which lacks magmatic titanite. The effect of the latteron the Al-in-Ca-amphibole barometry is not well understood andin several cases the calculated pressures in rocks lacking magmatictitanite have given consistent results with assemblages contain-ing titanite (Anderson and Smith, 1995). We therefore includethe pressure obtained for that particular sample as well. Horn-blende crystals display Fetot/(Fetot + Mg) between 0.42 and 0.59,which indicates that the amphibole grew in high fO2 conditions.The Fe3+/(Fe2+ + Fe3+) ratios vary between 0.18 and 0.38, which
for some hornblendes is below the value suitable for applying thebarometer (0.20–0.22) (Anderson and Smith, 1995). Therefore, weuse only amphibole rims with an appropriate ratio for our calcu-lations. Sample VAG98767 does not meet the required criteria for

an Res

galn

gm

N

Si

Fara5aa

settal

8

8

aetBebi2RtsrNgtdo

8

(gepTwa(w

N. Coint et al. / Precambri

eobarometry as its Fetot/(Fetot + Mg) ranges between 0.63 and 0.65nd its Fe3+/(Fe2+ + Fe3+) ratio is lower than 0.2. In addition, theimited number of hornblende crystals (2) in the thin section, didot allow to pursue the calculations.

Temperature was estimated using the hornblende-plagioclaseeothermometer (Holland and Blundy, 1994) based on the ther-odynamic formulation of the following reaction:

aCa2Mg5Si4(AlSi3)O22(OH)2edenite

+ 4SiO2quartz

= Ca2Mg5Si8O22(OH)2tremolite

+ NaAlSi3O8albite

Errors associated with the method are ±40 ◦C. In the case of theMB, all rocks are clearly saturated in silica, as quartz is abundantn all rocks (Fig. 3).

The results from these calculations are presented in Table 2 andig. 1a. Pressures vary from 3.8 ± 0.6 to 4.8 ± 0.5 kbar. Temperaturesre consistent throughout the belt between 720 and 750 ◦C. Theseesults are consistent with the fact that epidote is not present as

magmatic phase in the rocks, limiting the pressure to less than kbar (Schmidt and Poli, 2004; Schmidt and Thompson, 1996; Zennd Hammarstrom, 1984). Temperatures of 720–750 ◦C (±40 ◦C)re reasonable for granitic systems.

Calculated pressures, although all identical within error, arelightly higher in the eastern part of the belt, suggesting that theastern part of the belt might represent a slightly deeper section ofhe batholith, however, more data need to be collected to confirmhe pattern. Overall, the pressure variations over tens of kilometersre small, and it is likely that we are looking at the same crustalevel throughout the belt.

. Discussion

.1. Relationships between metamorphism and SMB magmatism

The metamorphic evolution of SW Norway has generally beenscribed to a ca. 1035–970 Ma (Bingen et al., 2008a) regionalvent, followed by a younger, contact-metamorphic event relatedo emplacement of the Rogaland Igneous Complex and thejerkreim–Sokndal Layered Intrusion at ca. 930 Ma (see Bingent al., 2008a,b and references therein). This picture, however, hasecome significantly more complicated recently with the proposed

dentification of a UHT event at ca. 1006 ± 4 Ma (Drüppel et al.,013), and indications that formation and emplacement of theogaland Igneous Complex was much longer lived than previouslyhought (Bybee et al., 2014; Sauer et al., 2013). It is outside thecope of this paper, and beyond our understanding given the cur-ent dataset, to discuss the entire metamorphic evolution of SWorway in detail. We therefore limit the discussion to constraintsiven by the SMB granites, and rather than presenting a model forhe Sveconorwegian metamorphic evolution of SW Norway, theiscussion highlights some of the problems in our understandingf this evolution.

.1.1. Did regional metamorphism at 1030–970 Ma take place?Based on new mapping, field observations and petrography

Figs. 1, 8, 10, and 11), it appears that many of the rocks around thearnet granite, xenoliths within the latter, and rocks on the west-rn side of the SMB, underwent granulite-facies metamorphismrior to emplacement of the SMB rocks (i.e., prior to ca. 1030 Ma).he timing of this/these high-grade metamorphic event(s) is not
ell constrained. The metamorphic ages obtained on the host rocks
round the garnet granite (Fig. 14) and on the Konsmo migmatiteFig. 7) indicate that one of these events might have been coevalith the emplacement of the SMB, however, this remains to be

earch 265 (2015) 57–77 73

better constrained. It is unclear whether these ages are related tothe emplacement of the SMB or relate to a tectono-metamorphicevent that could have triggered the genesis of these granites. In anycase, it is likely that previous attempts to delineate the extent ofUHT metamorphism around the anorthosite by mapping mineraloccurrences in the rocks resulted in mixing these various events(see section on opx-in isograd below).

SW Norway is said to have undergone regional high-grade meta-morphism between 1030 and 970 Ma (Bingen et al., 2008a), and theexposed crust in that region to have been ductile after 970 Ma dur-ing gravitational collapse of the orogen (Bingen et al., 2006). Theseinterpretations stem largely from geochronological work in the SWpart of the SMB (e.g., Falkum, 1998), which has undergone duc-tile deformation at high metamorphic grades. These observations,however, do not apply to most of the SMB, where ductile defor-mation is limited to rather narrow, xenolith-rich zones. There is,therefore, no evidence for orogen-scale ductile flow after emplace-ment of the SMB.

As presented in this paper, there is growing evidence thatthe region is polycyclic, possibly involving pre-Sveconorwegianas well as multiple Sveconorwegian events of different character.In the northern part of the SMB, east of Sand (Fig. 1a), an unde-formed SMB granite dated at 1047 ± 1 Ma cross-cuts the fabric inamphibolites-facies 1.5 Ga metavolcanic rocks (Fig. 5f), providing aminimum age for this metamorphic event. It is, however, unclearwhether this is an early Sveconorwegian event or a much olderevent unrelated to the Sveconorwegian orogeny. Farther north,in the Hardangervidda region, a supracrusal sequence only meta-morphosed to upper greenschist facies (Sæsvatn–Valldal; Sigmond,1978), implies that any amphibolites-facies or higher-grade meta-morphism during the Sveconorwegain orogeny was not regionallypervasive this far north.

Farther south, in the xenolith-rich zones (Fig. 1b), granitesdated at ca. 1030 Ma intrude previously deformed orthogneisses.Although field relationships clearly show that deformation tookplace prior to intrusion of the host granites (Fig. 5e), geochronol-ogy on one such migmatitic orthogneiss at Konsmo suggests thatmetamorphism took place at 1026 ± 14 Ma, i.e., just before or dur-ing granite emplacement. Similarly, geochronological data from agrt-bearing paragneiss in Oltedal (Fig. 1c) yield an age of meta-morphism of 1039 ± 27 Ma, i.e., overlapping with the age of thenearby cross-cutting garnet granite dated at between 1035 ± 6 and1027 ± 4 Ma.

The geographical extent and tectonic significance of these1026 ± 14 to 1039 ± 27 Ma syn-SMB (metamorphic) ages are as yetunclear. Tomkins et al. (2005) obtained a similar age of 1035 ± 9 Mafrom a metapelite between the garnet granite and the main SMB,interpreted to reflect incipient migmatization of the metapelite.This result shows that the effects of this metamorphic eventextends to areas that are not in direct contact with the SMBgranites.

Large parts of the SMB are older than the inferredca.1030–1040 Ma metamorphism, but do not record any evidenceof deformation and metamorphic overprinting. For example,ca. 1.5 km east of the migmatite at Konsmo, an undeformedporphyritic bt granite has been dated at 1061 ± 9 Ma, and else-where in the belt numerous granites dated at between 1040and 1050 Ma show no sign of metamorphic overprinting. Theseobservations suggest that deformation and metamorphism waslocalized, rather than reflecting a general thickening of the crust.We therefore speculate that much of the Sveconorwegian-agemetamorphism observed in SW Norway results from anoma-lously high heat flow related to emplacement of granitic,anorthositic, and mafic magmatism, concomitant with volu-
minous basaltic underplating starting at ca. 1040 Ma (Bybee et al.,2014).

7 ian Re

8

Serurad(AptifF(utiawirp

aFs(SMton1pat

clptmmSppa

ftuiedd

8

SWgrB


.1.2. Does the Sveconorwegian ‘opx-in isograd’ exist?An important boundary in the geological literature from

W Norway is the N–S-trending opx-in isograd of Hermanst al. (1975). This isograd is generally interpreted to separateocks that underwent Sveconorwegian, regional granulite- andpper-amphibolite-facies metamorphism to the west and east,espectively (e.g., Bingen and van Breemen, 1998b). It also figuress an important boundary in models trying to explain relativelyry AMCG magmatism to the west, and more hydrous HBG-graniteHornblende Biotite Granite) magmatism to the east (Vanderuwera et al., 2003, 2011, 2014). As shown in Fig. 1a, the previouslyroposed opx-in isograd locally follows the western boundary ofhe SMB quite closely. In an attempt to better understand the opx-n isograd, we mapped the Gyadalen transect (Fig. 9) westwardsrom Tonstad (which is within the SMB and east of the isograd,ig. 1a), and found that granulite-facies migmatitic orthogneissesFigs. 10a and 11a), are intruded by and found as xenoliths withinndeformed to weakly deformed and unmetamorphosed graniteshat most likely form part of the SMB (Fig. 9) although the gran-te remains to be dated. This suggests that the SMB granites in thisrea intruded rocks that had undergone earlier metamorphism, butere unaffected by later high-grade metamorphism. The opx-in

sograd is therefore not a metamorphic isograd in this area, butather the intrusive contact between the SMB and older, metamor-hic rocks.

On the same transect, two opx-bearing gneisses located 3 kmway from the Bjerkreim Sokndal intrusion (see white star inig. 9 for the location of the samples on the Gyadalen tran-ect), were dated at 1034 ± 7 Ma and 1052 ± 5 Ma, respectivelyMöller et al., 2002), indicating that rocks with similar age as theMB rocks underwent granulite-facies metamorphism in that area.etapelites east of these orthogneisses display symplectitic tex-

ures and present mineral assemblages that appear characteristicf UHT conditions with opx, green Zn-free spinel, ilmenite, mag-etite, garnet, orthopyroxene, quartz and plagioclase (e.g. Spear,995; Harley, 1998). In this particular case, the opx-isograd encom-asses two different granulite-facies events that cannot be pickedpart without further detailed work (Fig. 9) and was placed closeo the contact with the unmetamorphosed SMB.

Farther south, in the vicinity of Moi, the SMB granites, relativelylose to the Rogaland Igneous Complex and Bjerkreim–Sokndalayered intrusion, are also overprinted by high-grade metamor-hism; thus it is possible that there exists an opx-in isograd relatedo contact metamorphism around these complexes. This contact

etamorphism must overprint a pre-/syn-SMB high-grade meta-orphism, as suggested by Tobi et al. (1985) and discussed above.

orting these events from each other will require detailed map-ing to reveal cross-cutting relationships, and detailed, in situetrochronology where ages can be linked to particular texturesnd reactions (e.g. Regis et al., 2014; Mottram et al., 2014).

The opx-in isograd is therefore a mix between several granulite-acies events and locally, such as in the eastern part of the Gyadalenransect (Fig. 9), corresponds to the intrusive contact between thenmetamorphosed SMB and the high-grade gneisses. Further work

s needed to characterize the extent of the various high-gradevents in the area but we argue that the opx-isograd, as currentlyefined, has little or geologic significance, and should be aban-oned.

.1.3. Implications for UHT metamorphism in SW NorwayUHT metamorphic assemblages are well preserved in rocks

W of the SMB (e.g., Drüppel et al., 2013; Hermans et al., 1975;
estphal et al., 2003 and Fig. 9). Although most studies have sug-
ested that the UHT metamorphism represents contact heatingelated to emplacement of the Rogaland Igneous Complex andjerkreim–Sokndal layered intrusion, the most recent study by

search 265 (2015) 57–77

Drüppel et al. (2013) dated the UHT metamorphism to between1010 and 1005 Ma, i.e., much earlier than crystallization of theAMCG suite. We note, however, that earlier geochronologicalinvestigations from the UHT area have shown zircon systemat-ics to be highly complex (Möller et al., 2003; own unpublisheddata), suggesting that care must be taken when interpreting suchdata. Based on the new age data, that matched, or at least fellwithin, the agreed-upon age of Sveconorwegian regional meta-morphism (1030–970 Ma), Drüppel et al. (2013) argued that theUHT metamorphism must be a regional metamorphism related tocontinent–continent collision and crustal thickening. Slagstad et al.(2013b) discussed why the UHT metamorphism was unlikely toresult solely from crustal thickening, and as shown here there is noevidence that such a metamorphic event affected SMB rocks fartherthan ca. 15–20 km to the north and east of the Rogaland IgneousComplex and Bjerkreim–Sokndal layered intrusion. It is highlyunlikely that a biotite-bearing granite would go unscathed throughtemperatures approaching and possibly exceeding 1000 ◦C, whichtherefore limits the geographical extent of this UHT metamorphicevent significantly.

We therefore agree with most earlier workers that theobserved UHT metamorphism is probably related to the thermaleffects of emplacement of the Rogaland Igneous Complex andBjerkreim–Sokndal layered intrusion. That said, it is possible, andmaybe even likely, that the UHT event is much more complex thanthat outlined by e.g., Westphal et al. (2003). Although interpretedin such a way as to fit the current view of Sveconorwegian meta-morphism, the zircon data from Möller et al. (2003) show a complexage distribution with a large spread of ‘post-M1’ (i.e., the purportedSveconorwegian regional metamorphic event at ca. 1000 Ma) ages.Our own unpublished zircon and monazite data from the UHT areamimics the data from Möller et al. (2003), and combined with dataindicating long-lived (possibly episodic) anorthositic magmatism(Bybee et al., 2014; Sauer et al., 2013), it is possible to envisagemultiple or long-lived contact metamorphic events, resulting incomplex dissolution and regrowth of zircon.

8.2. Emplacement history and tectonic setting of the SMB

Emplacement of the SMB started at ca. 1070–1060 Ma andceased at ca. 1020 Ma, i.e., representing 40–50 million years ofapparently continuous granitic magmatism. A growing geochrono-logical database suggests that the intensity in magmatic activitymay have varied, with peaks at ca. 1050 and 1035 Ma, however,more detailed mapping within the SMB is needed to determinetime-integrated magma volumes. To this end, we have started toundertake detailed mapping of the xenolith-rich zones (Fig. 1b) thatwe think may represent contact zones between different intrusivebodies. If this is correct, the map pattern suggests that much of theSMB is composed of N–S-oriented, gently folded, sheet-like intru-sions. An exception to this pattern is the garnet granite, whichhas a nearly circular geometry. These different geometries mayhave been governed by pre-existing structures or local variations instress fields at the time of emplacement, and can serve as markersfor linking deformation and magmatism.

Estimates of crystallization pressures yield a rather narrowrange of pressures between 4 and 5 kbar, corresponding to mid-crustal depths of 12–15 km assuming an average density of 3 g/cm3.The similar pressures recorded in northern and south-central partsof the SMB suggest minor post-emplacement tilting of the belt. Thisimplies that the trend of granulite-facies to amphibolites-faciesmetamorphism that is recorded in pre-SMB rocks, from southwest
to the north and east, respectively, is not due to any syn- or post-Sveconorwegian tilting or variable exhumation.
Determining the tectonic affinity of the SMB granites throughpetrological means is inherently difficult due to their strongly

an Res

eafgiaSrla

sAdbDth1olecthlseSBtpG(t

cclcrdttmt1t

8

fcevltnmirtetg


volved compositions (Slagstad et al., 2013a and unpublished data),nd crustal isotopic signatures (Bingen et al., 1993). It is, there-ore, possible that the arc-like geochemical signature of the SMBranites (Bingen and Van Breemen, 1998a; Slagstad et al., 2013a)s solely an inherited source characteristic, e.g., from 1.5 Ga calc-lkaline metavolcanic and -granitoid rocks that underlie much of

Norway (Bingen et al., 2005; Roberts et al., 2013), and does noteflect the tectonic setting in which the granites formed. Severalines of evidence, listed below, nevertheless argue for formation inn Andean- or Cordilleran-type, accretionary setting.

The emplacement of large, linear belts of granite batholiths overeveral tens of millions of years, such as the SMB, is characteristic ofndean-type accretionary orogens (e.g., Weaver et al., 1990). Theuration and apparent peaks and lulls in magmatic activity resem-le that observed for Cordilleran systems (DeCelles et al., 2009;ucea and Barton, 2007). Recently, Bybee et al. (2014) showed

hat high-alumina opx megacrysts in the Rogaland anorthositesave rather juvenile isotopic compositions and probably grew at041 ± 17 Ma, i.e., during SMB emplacement. These high-aluminapx megacrysts probably represent cumulates from the crystal-izing basaltic magma that may have ponded at the MOHO andventually given rise to the anorthosites (and SMB through lower-rustal melting). The age of the basaltic underplate corresponds tohe age of the crystallization of the SMB, which provides a potentialeat source for generating such large volumes of granite. A long-

ived Andean-type margin appears to be the most likely setting foruch a long-lived magmatic system (Slagstad et al., 2013a; Bybeet al., 2014). The UHT region in SW Norway is surrounded by theMB to the north and east, and the Rogaland Igneous Complex andjerkreim–Sokndal layered intrusion to the southwest; therefore,he extreme metamorphic conditions that led to UHT metamor-hism may well have been related to heat from this magmatism.enerating UHT (900–1000 C) conditions at mid-crustal conditions

7.5–5.5 kbar, Drüppel et al., 2013) in such a small area, by crustalhickening is very unlikely.

Alternative mechanisms for the formation of the SMB includeontinental collision and post-orogenic extension. In both theseases, the granites likely would have formed from partial melting ofower to middle crustal sources, both permissible from geochemi-al and isotopic data. During collision, the magmatism would haveesulted from crustal thickening and thermal relaxation, whereasecompression melting could be envisaged in a post-orogenic set-ing. However, both models require a phase of orogen-wide crustalhickening that preceded the onset of magmatism by at least 10–20

illion years, and no such available data suggest that any crustalhickening took place prior to the earliest SMB magmatism at ca.070–1060 Ma. Thus, currently an Andean- or Cordilleran-type set-ing remains the preferred choice of tectonic setting.

.3. Implications for the Sveconorwegian Province

Our suggestion that the Sveconorwegian Province may haveormed in an accretionary orogen rather than an orogen related toontinent–continent collision has led to significant debate (Möllert al., 2013; Slagstad et al., 2013b). The debate stems from obser-ations in the eastern part of the province in Sweden that showarge-scale contractional deformation and evidence of continen-al subduction (see Möller et al., 2013 for a discussion). We doot disagree with the conclusions that the structures here reflectajor contractional deformation, however, we find no evidence of

ndenter tectonism having affected the crust of SW Norway, such asegionally penetrative deformation and metamorphism. The con-
ractional deformation in the eastern part of the province, andxhibited in the central area around the Bamble and Kongsbergerranes (see Bingen et al., 2008a), may instead relate to amal-amation of the different domains or blocks that comprise the
earch 265 (2015) 57–77 75

Sveconorwegian Province, without requiring collision of an exoticcontinent. This broad polycyclic orogenic evolution, from 1140 to930 Ma (Bingen et al., 2008a), is referred to by Roberts and Slagstad(2015) as a period of terminal orogenesis. In this case, it marksthe end of a long-lived (∼1.86 to <1.1 Ga) accretionary orogen, ina major phase of contractional deformation that may be related toplate-scale movements and stresses. Observed accretionary eventsceased after ca. 0.9 Ga, and in the case of an accretionary ratherthan continent–continent collision-related orogen, the convergentmargin likely retreated or jumped outboard, eventually leaving SWNorway to lie inboard of a passive margin.

9. Conclusions

A major conclusion of this work is that magmatic activity wasa much more voluminous and intergral part of the orogenic evo-lution than hitherto known (cf., Bingen et al., 2008a), and that themetamorphic evolution in SW Norway includes events that pre-date and/or are synchronous with emplacement of the SMB, as wellas events post-dating the belt. Cross-cutting relationships alongthe length of the belt show that the SMB was emplaced into aregionally deformed and metamorphosed crust. In contrast, eventspost-dating the SMB are geographically much more restricted asmost of the belt is unaffected by these events.

The observed voluminous magmatism, of metamorphism (UHT)and lack of widespread penetrative deformation, which one wouldexpect beneath a long-lived orogenic plateau, is best interpreted toreflect accretionary processes along an active continental margin.

Our new observations suggest that the ‘opx-in isograd’ in SWNoray may partly be an intrusive contact between the SMB andpreviously metamorphosed rocks. In the SW part of the SMB,the granites have been overprinted by geographically restricted,high-grade metamorphism, interpreted to reflect contact meta-morphism during emplacement of the Rogaland Igneous Complex.Thus the ‘opx-in isograd’ may reflect different types of contactswith different ages along its length, and as such does not representa metamorphic isograd.

Acknowledgements

We would like to thank Benjamin Berge, Bengt Johansen andBrit Inger Vongraven for preparing the samples, Øyvind Skår andMartin Whitehouse for their help acquiring geochronological data,and Jessica Langlade for the microprobe analyses. We would liketo thank Charles Jourdan and Andreas Sigersvold for guiding us inthe Knaben area, Giulio Viola, Randall Parrish for the editing, Fed-erico Farina and an anonymous reviewer for comments that helpedimprove the manuscript. We also thank the Norwegian GeologicalSurvey (NGU) for funding the research. Discussions with Iain Hen-derson, Axel Müller, Martin Stormoen and Stine Bang are greatlyappreciated.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, inthe online version, at http://dx.doi.org/10.1016/j.precamres.2015.05.002

References

Anderson, L.J., Smith, D.R., 1995. The effect of temperature and fO2 on the Al-in-
hornblende barometer. Am. Mineral. 80, 549–559.
Bartley, J.M., Coleman, D.S., Glazner, A.F., 2006. Incremental pluton emplacement bymagmatic crack-seal. Trans. R. Soc. Edinb.: Earth Sci. 97, 383–396.

Bateman, P.C., 1983. A summary of the critical relations in the central part of theSierra Nevada Batholith, California, U.S.A. Geol. Soc. Am. Mem. 159, 241–254.

http://dx.doi.org/10.1016/j.precamres.2015.05.002











http://refhub.elsevier.com/S0301-9268(15)00142-4/sbref0005



























































7 ian Re

B

B

B

B

B

B

B

B

B

B

C

D

D

D

F

F

F

F

G

G

G

H

H

H

H

J

K

K


ingen, B., Demaiffe, D., Hertogen, J., Weis, D., Michot, J., 1993. K-rich calc-alkalineaugen gneisses of Grenvillian age in SW Norway: mingling of mantle-derivedand crustal components. J. Geol. 101, 763–778.

ingen, B., Mansfeld, J., Sigmond, E.M.O., Stein, H., 2002. Baltica-Laurentia link duringthe Mesoproterozoic: 1.27 Ga development of continental basins in the Sve-conorwegian Orogen, southern Norway. Can. J. Earth Sci. 39, 1425–1440.

ingen, B., Davis, W.J., Hamilton, M.A., Engvik, A.K., Stein, H.J., Skår, Ø., Nordgulen,Ø., 2008a. Geochronology of high-grade metamorphism in the Sveconorwegianbelt, S. Norway: U–Pb, Th–Pb and Re–Os data. Nor. J. Geol. 88, 13–42.

ingen, B., Nordgulen, O., Viola, G., 2008b. A four-phase model for the Sveconorwe-gian orogeny, SW Scandinavia. Nor. J. Geol. 88, 43–72.

ingen, B., Skar, O., Marker, M., Sigmond, E.M.O., Nordgulen, O., Ragnhildstveit, J.,Mansfeld, J., Tucker, R.D., Liegeois, J.P., 2005. Timing of continental building inthe Sveconorwegian orogen, SW Scandinavia. Nor. J. Geol. 85, 87–116.

ingen, B., Stein, H.J., Bogaerts, M., Bolle, O., Mansfeld, J., 2006. Molybdenite re-os dating constrains gravitational collapse of the Sveconorwegian orogen, SWScandinavia. Lithos 87, 328–346.

ingen, B., Van Breemen, O., 1998a. Tectonic regimes and terrane boundaries inthe high-grade Sveconorwegian belt of SW Norway, inferred from U–Pb zircongeochronology and geochemical signature of augen gneiss suites. J. Geol. Soc.Lond. 155, 143–154.

ingen, B., van Breemen, O., 1998b. U–Pb monazite ages in amphibolite- togranulite-facies orthogneiss reflect hydrous mineral breakdown reactions:Sveconorwegian Province of SW Norway. Contrib. Mineral. Petrol. 132,336–353.

rewer, T.S., Ahall, K.I., Menuge, J.F., Storey, C.D., Parrish, R.R., 2004. Mesoproterozoicbimodal volcanism in SW Norway, evidence for recurring pre-Sveconorwegiancontinental margin tectonism. Precambrian Res. 134, 249–273.

ybee, G.M., Ashwal, L.D., Shirey, S.B., Horan, M., Mock, T., Andersen, T.B., 2014.Pyroxene megacrysts in Proterozoic anorthosites: implications for tectonic set-ting, magma source and magmatic processes at the Moho. Earth Planet. Sci. Lett.389, 74–85.

orfu, F., Hanchar, J.M., Hoskin, P.W.O., Kinny, P.D., 2003. Atlas of igneous zircon.Rev. Mineral. Geochem. 53, 470–500.

eCelles, P.G., Ducea, M.N., Kapp, P., Zandt, G., 2009. Cyclicity in Cordilleran orogenicsystems. Nat. Geosci. 2, 251–257.

rüppel, K., Elsasser, L., Brandt, S., Gerdes, A., 2013. Sveconorwegian mid-crustalultrahigh-temperature metamorphism in Rogaland, Norway: U–Pb LA-ICP-MSgeochronology and pseudosections of sapphirine granulites and associatedparagneisses. J. Petrol. 54, 305–350.

ucea, M.N., Barton, M.D., 2007. Igniting flare-up events in Cordilleran arcs. Geology35, 1047–1050.

alkum, T., 1982. Geologisk kart over Norge, berggrunnskart MANDAL, 1:250,000.Nor. Geol. Unders.

alkum, T., 1998. The Sveconorwegian magmatic and tectonometamorphic evolu-tion of the high grade Proterozoic Flekkefjord complex, south Norway. Nor. Geol.Unders. Bull. 434, 5–33.

alkum, T., Petersen, J.S., 1980. The Sveconorwegian orogenic belt, a case of late-proterozoic plate-collision. Geol. Rundsch. 69, 622–647.

alkum, T., 1985. Geotectonic evolution of southern Scandinavia in light of a late-Proterozoic plate-collision. In: Tobi, A.C., Touret, J. (Eds.), The Deep ProterozoicCrust in the North Atlantic Provinces. Nato Advanced Study Institute Series C158.D. Reidel Publishing Company, pp. 309–322.

ower, C., 1985. Correlations between the Grenville Province and Sveconorwegianorogenic belt – implications for Proterozoic evolution of the southern mar-gin of the Canadian and Baltic Shields. In: Tobi, A.C., Touret, J.L.R. (Eds.), TheDeep Proterozoic Crust in the North Atlantic Provinces. Reidel, Dordrecht, pp.247–257.

ower, C.F., Kamo, S.L., Kwok, K., Krogh, T.E., 2008. Proterozoic southward accre-tion and Grenvillian orogenesis in the interior Grenville Province in easternLabrador: evidence from U–Pb geochronological investigations. PrecambrianRes. 165 (1/2), 61–95.

ower, C.F., Ryan, A.B., Rivers, T., 1990. Mid-Proterozoic Laurentia-Baltica: anoverview of its geological evolution and a summary of the contributions madeby this volume. In: Gower, C.F., Rivers, T., Ryan, A.B. (Eds.), Mid-ProterozoicLaurentia-Baltica. Geological Association of Canada, pp. 1–20 (special paper).

arley, S., 1998. On the occurrence and characterization of ultrahigh-temperaturecrustal metamorphism. In: Treoloar, P.J., O’Brien, P.J. (Eds.), What Drives Meta-morphism and Metamorphic Reactions?, vol. 138. Geological Society, London,pp. 81–107 (special publications).

awthorne, F.C., Oberti, R., 2007. Classification of the amphiboles. Rev. Mineral.Geochem. 67, 55–88.

ermans, G.A.E.M., Tobi, A.C., Poorter, R.P.E., Maijer, C., 1975. The high-grademetamorphic Precambrian of the Sirdal-Ørsdal area, Rogaland/Vest-Agder,south-west Norway. Nor. Geol. Unders. 318, 51–74.

olland, T., Blundy, J., 1994. Non-ideal interactions in calcic amphiboles and theirbearing on amphibole-plagioclase thermometry. Contrib. Mineral. Petrol. 116,433–447.

ohansson, Å., 2009. Baltica, Amazonia and the SAMBA connection –1000 millionyears of neighbourhood during the Proterozoic? Precambrian Res. 175, 221–234.

arlstrom, K.E., Ahall, K.I., Harlan, S.S., Williams, M.L., McLelland, J., Geissman,
J.W., 2001. Long-lived (1.8–1.0 Ga) convergent orogen in southern Laurentia,its extensions to Australia and Baltica, and implications for refining Rodinia.Precambrian Res. 111, 5–30.
orhonen, F.J., Clark, C., Brown, M., Bhattacharya, S., Taylor, R., 2013. How long-livedis ultrahigh temperature (UHT) metamorphism? Constraints from zircon and

search 265 (2015) 57–77

monazite geochronology in the Eastern Ghats orogenic belt, India. PrecambrianRes. 234, 322–350.

Leake, B.E., Woolley, A.R., Arps, C.E.S., Birch, W.D., Gilbert, M.C., Grice, J.D.,Hawthorne, F.C., Kato, A., Kisch, H.J., Krivovichev, V.G., Linthout, K., Laird, J.,Mandarino, J.A., Maresch, W.V., Nickel, E.H., Rock, N.M.S., Schumacher, J.C.,Smith, D.C., Stephenson, N.C.N., Ungaretti, L., Whittaker, E.J.W., Youzhi, G., 1997.Nomenclature of amphiboles: report of the subcommittee on amphiboles ofthe International Mineralogical Association Commission on new minerals andmineral names. Can. Mineral. 35, 219–246.

Le Maitre, R.W., Bateman, P., Dudek, A., Keller, J., Lameyre, J., Le Bas, M.J., Sabine,P.A., Schmid, R., Sorensen, H., Streckeisen, A., Woolley, A.R., Zanettin, B., 1989. AClassification of Igneous Rocks and Glossary of Terms: Recommendations of theInternational Union of Geological Sciences Subcommission on the Systematicsof Igneous Rocks. Blackwell Scientific Publications, Oxford, UK.

Li, Z.X., Bogdanova, S.V., Collins, A.S., Davidson, A., De Waele, B., Ernst, R.E., Fitzsi-mons, I.C.W., Fuck, R.A., Gladkochub, D.P., Jacobs, J., Karlstrom, K.E., Lu, S.,Natapov, L.M., Pease, V., Pisarevsky, S.A., Thrane, K., Vernikovsky, V., 2008.Assembly, configuration, and break-up history of Rodinia: a synthesis. Precam-brian Res. 160, 179–210.

Möller, A., O’Brien, P.J., Kennedy, A., Kroner, A., 2002. Polyphase zircon in ultrahigh-temperature granulites (Rogaland, SW Norway): constraints for Pb diffusion inzircon. J. Metamorph. Geol. 20, 727–740.

Möller, A., O’Brien, P.J., Kennedy, A., Kröner, A., 2003. Linking Growth Episodes ofZircon and Metamorphic Textures to Zircon Chemistry: An Example from theUltrahigh-Temperature Granulites of Rogaland (SW Norway), vol. 220. Geolog-ical Society, London, pp. 65–81 (special publications).

Möller, C., Bingen, B., Andersson, J., Stephens, M.B., Viola, G., Schersten, A., 2013. Anon-collisional, accretionary Sveconorwegian orogen – comment. Terra Nova25, 165–168.

Mottram, C.M., Warren, C.J., Regis, D., Roberts, N.M.W., Harris, N.B.W., Argles, T.W.,Parrish, R.R., 2014. Developing an inverted Barrovian sequence; insights frommonazite petrochronology. Earth Planet. Sci. Lett. 403, 418–431.

O’Brien, T., Miller, E., 2014. Continuous zircon growth during long-lived granulitefacies metamorphism: a microtextural, U–Pb, Lu–Hf and trace element study ofCaledonian rocks from the Arctic. Contrib. Mineral. Petrol. 168, 1–19.

Park, R.G., 1992. Plate kinematic history of Baltica during the Middle to Late Protero-zoic: a model. Geology 20, 725–728.

Paterson, S.R., Memeti, V., Pignotta, G., Erdmann, S., Zák, J., Chambers, J., Ianno, A.,2012. Formation and transfer of stoped blocks into magma chambers: the high-temperature interplay between focused porous flow, cracking, channel flow,host-rock anisotropy, and regional deformation. Geosphere 8 (2), 443–469.

Paterson, S.R., Zák, J., Janousek, V., 2008. Growth of complex sheeted zones duringrecycling of older magmatic units into younger: Sawmill Canyon area, Tuolumnebatholith, Sierra Nevada, California. J. Volcanol. Geotherm. Res. 177, 457–484.

Pidgeon, R.T., Nemchin, A.A., Kinny, P.D.,2000. Fir-tree and nebulously zoned zirconsfrom granulite facies rocks: evidence for growth and interaction with metamor-phic fluids. In: Goldschmidt 2000. Cambridge Publications, Oxford, UK.

Pisarevsky, S.A., Bylund, G., 2006. Palaeomagnetism of 935 Ma mafic dykes in south-ern Sweden and implications for the Sveconorwegian Loop. Geophys. J. Int. 166,1095–1104.

Regis, D., Warren, C.J., Young, D., Roberts, N.M.W., 2014. Tectono-metamorphicevolution of the Jomolhari massif: variations in timing of syn-collisional meta-morphism across western Bhutan. Lithos 190, 449–466.

Roberts, N.M.W., Parrish, R.R., Horstwood, M.S.A., Brewer, T.S., 2011. The1.23 Ga Fjellhovdane rhyolite Grossae-Totak; a new age within the Telemarksupracrustals, southern Norway. Nor. J. Geol. 91, 239–246.

Roberts, N.M.W., Slagstad, T., 2015. Continental growth and reworking on the edgeof the Columbia and Rodinia supercontinents; 1.86–0.9 Ga accretionary orogenyin SW Fennoscandia. Int, Geol. Rev., 1–25, http://dx.doi.org/10.1080/00206814.2014.958579

Roberts, N.M.W., Slagstad, T., Parrish, R.R., Norry, M.J., Marker, M., Horstwood, M.S.A.,2013. Sedimentary recycling in arc magmas: geochemical and U–Pb–Hf–O con-straints on the Mesoproterozoic Suldal Arc, SW Norway. Contrib. Mineral. Petrol.165, 507–523.

Romer, R.L., 1996. Contiguous Laurentia and Baltica before the Grenvillian-Sveconorwegian orogeny? Terra Nova 8, 173–181.

Sauer, S., Slagstad, T., Andersen, T., Kirkland, C., 2013. Zircon Lu–Hf isotopes inhigh-alumina orthopyroxene megacrysts from the Neoproterozoic RogalandAnorthosite Province, SW Norway: a window into the Sveconorwegian lowercrust. Geophys. Res. Abstr. 15, EGU2013-13958.

Sawyer, E.W., 2008. Atlas of Migmatites. Canadian Mineralogist Special Publication9. NRC Research Press and Mineralogical Association of Canada, pp. 387.

Schmidt, M.W., Poli, S., 2004. Magmatic epidote. Rev. Mineral. Geochem. 56,399–430.

Schmidt, M.W., Thompson, A.B., 1996. Epidote in calc-alkaline magmas: an exper-imental study of stability, phase relationships, and the role of epidote inmagmatic evolution. Am. Mineral. 81, 462–474.

Sigmond, E.M.O., 1978. Beskrivelse til det berggrunnsgeologiske kartbladet Sauda1:250000. NGU Bull. 341, 1–94.

Slagstad, T., Roberts, N.M.W., Marker, M., Rohr, T.S., Schiellerup, H., 2013a. Anon-collisional, accretionary Sveconorwegian orogen – reply. Terra Nova 25,
169–171.
Slagstad, T., Roberts, N.M.W., Marker, M., Røhr, T.S., Schiellerup, H., 2013b. A non-collisional, accretionary Sveconorwegian orogen. Terra Nova 25, 30–37.

Spear, F.S., 1995. Metamorphic Phase Equilibria and Pressure–Temperature–TimePaths. Mineralogical Society of America Monograph, pp. 799.





















































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































































dx.doi.org/10.1080/00206814.2014.958579

dx.doi.org/10.1080/00206814.2014.958579

dx.doi.org/10.1080/00206814.2014.958579

dx.doi.org/10.1080/00206814.2014.958579

dx.doi.org/10.1080/00206814.2014.958579

dx.doi.org/10.1080/00206814.2014.958579

dx.doi.org/10.1080/00206814.2014.958579

dx.doi.org/10.1080/00206814.2014.958579

dx.doi.org/10.1080/00206814.2014.958579























































































































































































an Res

S

T

T

T

T

V

V


tephens, M.B., Wahlgren, C.-H., Weijermars, R., Cruden, A.R., 1996. Left-lateraltranspressive deformation and its tectonic implications, Sveconorwegian oro-gen, Baltic Shield, southwestern Sweden. Precambrian Res. 79, 261–279.

obi, A.C., Hermans, G.A.E.M., Maijer, C., Jansen, J.B.H., 1985. Metamorphic zoning inthe high-grade Proterozoic of Rogaland-Vest Agder, SW Norway. In: Tobi, A.C.,Touret, J.L.R. (Eds.), The Deep Proterozoic Crust in the North Atlantic Provinces.Reidel, Dordrecht, pp. 477–497.

omkins, H.S., Williams, I.S., Ellis, D.J., 2005. In situ U–Pb dating of zircon formed fromretrograde garnet breakdown during decompression in Rogaland, SW Norway.J. Metamorph. Geol. 23, 201–215.

orske, T., 1976. Metal provinces of a cordilleran-type orogen in the Precambrian ofSouth Norway. Geol. Assoc. Can. Spec. Pap. 14, 597–613.

orske, T., 1977. The South Norway Precambrian region – a Proterozoic coredilleran-type orogenic segment. Nor. Geol. Tidsskr. 57, 97–120.

ander Auwera, J., Bogaerts, M., Liégeois, J.-P., Demaiffe, D., Wilmart, E., Bolle, O.,Duchesne, J.C., 2003. Derivation of the 1.0–0.9 Ga ferro-potassic A-type gran-itoids of southern Norway by extreme differentiation from basic magmas.
Precambrian Res. 124, 107–148.
ander Auwera, J., Bolle, O., Bingen, B., Liégeois, J.-P., Bogaerts, M., Duchesne, J.-C., DeWaele, B., Longhi, J., 2011. Sveconorwegian massif-type anorthosites and relatedgranitoids result from post-collisional melting of a continental arc root. EarthSci. Rev. 107, 375–397.

earch 265 (2015) 57–77 77

Vander Auwera, J., Bolle, O., Dupont, A., Pin, C., Paquette, J.-L., Charlier, B., Duch-esne, J.C., Mattielli, N., Bogaerts, M., 2014. Source-derived heterogeneities inthe composite (charnockite-granite) ferroan Farsund intrusion (SW Norway).Precambrian Res. 251, 141–163.

Vernon, R.H., Paterson, S.R., 2006. Mesoscopic structures resulting from crystal accu-mulation and melt movement in granites. Trans. R. Soc. Edinb.: Earth Sci. 97,369–381.

Viola, G., Henderson, I.H.C., Bingen, B., Hendriks, B.W.H., 2011. TheGrenvillian-Sveconorwegian orogeny in Fennoscandia: back-thrustingand extensional shearing along the Mylonite Zone. Precambrian Res. 189,368–388.

Weaver, S.G., Bruce, R., Nelson, E.P., Brueckner, H.K., LeHuray, A.P., 1990. ThePatagonian batholith at 48◦S, Chile; geochemical and isotopic variations.In: Kay, S.M., Rapela, C.W. (Eds.), Plutonism from Antarctica to Alaska. ,pp. 33–50.

Westphal, M., Schumacher, J.C., Boschert, S., 2003. High-temperature metamor-phism and the role of magmatic heat sources at the Rogaland anorthosite
complex in Southwestern Norway. J. Petrol. 44, 1145–1162.
Whitney, D.L., Evans, B.W., 2010. Abbreviations for names of rock-forming minerals.Am. Mineral. 95, 185–187.

Zen, E.A., Hammarstrom, J., 1984. Magmatic epidote and its petrologic significance.Geology 12, 515–518.












































































































































































































































































































The Late Mesoproterozoic Sirdal Magmatic Belt, SW Norway: Relationships between magmatism and metamorphism and implications for Sveconorwegian orogenesis

Documents