Journal of Petrology Advance Access published online on January 7, 2009
Journal of Petrology, doi:10.1093/petrology/egn070
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Internal Differentiation of the Archean Continental Crust: Fluid-Controlled Partial Melting of Granulites and TTG–Amphibolite Associations in Central Finland
1Johannes Gutenberg University, Institute of Geosciences, C/O Stephen Foley, Becher-Weg 21, 55099 Mainz, Germany
2Johannes Gutenberg University, Institute of Geosciences, Becher-Weg 21, 55099 Mainz, Germany
3Geological Survey of Finland, Betonimiehenkuja 4, FI-02151 Espoo, Finland
4Georg August University of GöTtingen, Geoscience Centre, Goldschmidtstr. 3, 37077 GöTtingen, Germany
Received June 13, 2008; Revised typescript accepted November 21, 2008
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONFault bound blocks of granulite and enderbite occur within upper amphibolite-facies migmatitic tonalitic–trondhjemitic–granodioritic (TTG) gneisses of the Iisalmi block of Central Finland. These units record reworking and partial melting of different levels of the Archean crust during a major tectonothermal event at 2·6–2·7 Ga. Anhydrous mineral assemblages and tonalitic melts in the granulites formed as a result of hydrous phase breakdown melting reactions involving amphibole at peak metamorphic conditions of 8–11 kbar and 750–900°C. A nominally fluid-absent melting regime in the granulites is supported by the presence of carbonic fluid inclusions. The geochemical signature of light rare earth element (LREE)-depleted mafic granulites can be modelled by 10–30 wt % partial melting of an amphibolite source rock leaving a garnet-bearing residue. The degree of melting in intermediate granulites is inferred to be less than 10 wt % and was restricted by the availability of quartz. Pressure–temperature estimates for the TTG gneisses are significantly lower than for the granulites at 660–770°C and 5–6 kbar. Based on the P–T conditions, melting of the TTG gneisses is inferred to have occurred at the wet solidus in the presence of an H2O-rich fluid. A hydrous mineralogy, abundant aqueous fluid inclusions and the absence of carbonic inclusions in the gneisses are in accordance with a water-fluxed melting regime. Low REE contents and strong positive Eu anomalies in most leucosomes irrespective of the host rock composition suggest that the leucosomes are not melt compositions, but represent plagioclase–quartz assemblages that crystallized early from felsic melts. Furthermore, similar plagioclase compositions in leucosomes and adjacent mesosomes are not a ‘migmatite paradox’, as both record equilibration with the same melt phase percolating along grain boundaries.
KEY WORDS: Archean continental crust; fluid inclusion; granulite; migmatite; partial melting
|
INTRODUCTION |
|---|
The Archean continental crust predominantly consists of tonalite–trondhjemite–granodiorite (TTG) associations. These rocks have variable field aspects ranging from homogeneous gneisses to highly heterogeneous migmatites (Martin, 1994
) as a result of post-magmatic metamorphic overprinting. There is a general consensus that TTGs represent derivatives of partially melted basaltic crust (Drummond & Defant, 1990; Martin, 1993). However, it is still debated whether melting occurred in a subduction zone setting or at the base of thick basaltic crust (Bédard, 2006a, and references therein). Tonalitic to trondhjemitic melts have been generated in a variety of melting experiments on mafic source rocks (Johnston, 1986; Beard & Lofgren, 1991; Rapp et al., 1991; Rushmer, 1991; Sen & Dunn, 1994; Wolf & Wyllie, 1994; Patiño Douce & Beard, 1995; Rapp, 1995; Springer & Seck, 1997; Sisson et al., 2005). Experimental studies indicate that garnet, rutile and amphibole are required in the melt-depleted residue to induce the strong fractionation of the rare earth elements (REE), negative Nb–Ta anomalies and the low Nb/Ta ratios typical of Archean TTG (Martin, 1994; Foley et al., 2002; Rapp et al., 2003; Xiong, 2006). Lower crustal mafic granulites containing garnet and amphibole may therefore be suitable residues of TTG extraction alongside garnet-amphibolites and rutile-bearing eclogites (Rapp et al., 1991; Foley et al., 2002; Xiong, 2006).
Reworking and partial melting of pristine TTG crust is evident in many Archean crustal sequences. Recycling of continental crust as far back as the late Archean is recorded by zircon ages and isotope data (Nd and Hf isotopes) of rocks from the Superior Province in Canada and from southern China (Whalen et al., 2002; Bédard, 2006a; Zhang et al., 2006). The discovery of inherited zircon cores in addition to Nd isotopic constraints point to a variable contribution of older crustal sources in many TTG suites and greenstone belt volcanic rocks (Stern et al., 1994; Tomlinson et al., 2003, 2004; Whalen et al., 2004), thus complicating the picture of TTG genesis and of the evolution the continental crust. Recent studies by Hawkesworth & Kemp (2006) and Kemp et al. (2006) have shown that zircon crystallization ages can reflect crustal reworking rather than crustal growth. This suggests that global peaks in crustal growth at 2·7 Ga or 1·9 Ga (Condie, 1998) also involved reworking and remelting of older crust. Such processes are also recognized in the Archean rocks of Central Finland, where upper amphibolite- to granulite-facies metamorphism at 2·7–2·6 Ga reworked 3·1 Ga continental crust.
It is generally accepted that most of the TTG crust is primitive and only slightly modified by fractional crystallization (Bédard, 2006a). However, the strong migmatitic appearance of many TTG complexes indicates that recycling and remelting in most of the high-grade Archean terrains led to internal differentiation of the TTG complexes themselves and to an addition of felsic melt to the upper crust.
The importance of crustal melting for the production of largely granitic magmas is widely accepted. However, the fluid regime during crustal melting may be diverse. The present study shows that water-fluxed melting and melting triggered by hydrous phase breakdown may have operated coevally in central Finland but that different fluid regimes were restricted to different levels of the crust.
In this study we compare and contrast petrographical, geochemical and fluid-inclusion data to constrain the conditions of crustal melting in different lithologies. Modelling of melt and residue compositions allows us to draw conclusions on the extent of melt formation. We show that granulites from central Finland are anhydrous residues formed by incongruent hydrous phase breakdown reactions in amphibolitic source rocks. There is no mineralogical evidence that the upper amphibolite-facies TTG gneisses experienced hydrous phase breakdown reactions and we suggest that melting within these rocks could have occurred only if triggered by a hydrous fluid. Thus the availability of water is considered to be the controlling factor of melt formation reactions in different crustal levels.
|
GEOLOGICAL SETTING |
|---|
The Archean Iisalmi block lies at the SW margin of the Archean Karelian craton, bordering the craton against the Proterozoic Svecofennian Domain. The Iisalmi block consists predominantly of migmatitic TTG gneisses. Amphibolites occur as schollen, rafts, lenses and thick layers within the gneisses. The migmatitic TTG gneisses form the host rock for fault-bound blocks of intermediate and mafic granulites (Fig. 1). The region is subdivided into three major units according to the field appearance and mineralogy of the granulites. These units are from NW to SE the Iisalmi–Sukeva area, the Varpanen–Pällikäs area and the Jonsa area.
|
Zircon ion microprobe dating as well as Sm–Nd model ages of granulites from the Varpanen area indicate protolith ages of 3·1–3·2 Ga (Hölttä et al., 2000
; Mänttäri & Hölttä, 2002). Similarly, mesosomes of TTG gneisses from the western part of the region have an age of 3·2 Ga (Mänttäri et al., 1998). In contrast, zircon ages of granulites from the Jonsa area are 2·73–2·70 Ga and the Sm–Nd model age is 2·93 Ga (Hölttä et al., 2000). The age difference between the NW and the SE suggests terrane accretion during the late Archean associated with coeval metamorphism in both terranes between 2·70 and 2·63 Ga (Hölttä et al., 2000; Mänttäri & Hölttä, 2002). Enderbites intruded throughout the study area at 2·7–2·72 Ga. However, they were not the cause of granulite-facies metamorphism but themselves experienced metamorphism, which is recorded in metamorphic overgrowth on zircon in enderbites at 2·65–2·64 Ga (Mänttäri & Hölttä, 2002). Contemporaneous and compositionally similiar leucodiorites without orthopyroxene occur in the north of the region (Naimakangas type leucodiorites) and east of the Varpanen area (Rokanmäki tonalite) (Paavola, 1999, 2003).
Numerous dolerite dykes intruded into granulite blocks and host migmatites between 2·3 and 2·1 Ga (Toivola et al. 1991; Hölttä et al. 2000). Faults and shear zones in the study area developed during the 
1·9 Ga Svecofennian orogeny, which also caused uplift of lower crustal granulites and juxtaposition with middle crustal rocks. To the west of the study area Archean granulites experienced pervasive retrogression during this Proterozoic deformation, whereas retrogression in the study area was restricted to shear zones and fractures (Mänttäri & Hölttä, 2002).
|
PETROGRAPHY |
|---|
Granulites
Granulites of the Iisalmi block display a variety of modal mineralogies ranging from garnet–clinopyroxene–plagioclase granulites to orthopyroxene–clinopyroxene–plagioclase granulites lacking garnet (Appendix A; mineral abbrevations after Kretz, 1983
). The granulites have granoblastic textures with subhedral to euhedral grain shapes. Granoblastic garnet–clinopyroxene–plagioclase ± hornblende assemblages are typical for the mafic granulites from the Iisalmi–Sukeva area (Fig. 2a). In the mafic granulites orthopyroxene is present within symplectites around garnet or pseudomorphs after garnet (Fig. 2b).
|
The intermediate granulites of the Jonsa and Varpanen area show layering of lighter and darker bands. They characteristically contain granoblastic orthopyroxene and garnet is restricted to more mafic layers. Field appearance and compositional differences within the Jonsa and Varpanen granulites suggest a pre-metamorphic layering of broadly andesitic rocks probably representing original volcanic successions.
The modal content of amphibole in granulites is variable. Amphibole is rarely preserved in the mafic granulites from the Iisalmi–Sukeva area. It is, however, a typical constituent of the intermediate granulites from the Jonsa and Varpanen areas, where variations between amphibole-rich and amphibole-absent layers can be observed within a single outcrop. Amphibole in these rocks is usually overgrown by pyroxenes or occurs as inclusions in pyroxenes (Fig. 2c and d).
Porphyroblastic garnet in mafic granulites reaches grain sizes up to 2–3 cm and contains inclusions of mainly clinopyroxene and amphibole (Fig. 2a and b). Intergrowth of skeletal garnet and quartz is sometimes found in contact with leucosomes. Ilmenite and magnetite are abundant in all types of granulite (Fig. 3b and d). Biotite is a minor constituent of intermediate granulites from the Jonsa and Varpanen area but not of the mafic granulites from the Iisalmi–Sukeva area. Apatite is present as tiny euhedral crystals usually enclosed in plagioclase.
|
Late hydration of the granulites, indicated by the presence of low-grade amphiboles, epidote and chlorite, occurred along shear zones within and along the bordering faults of granulite blocks.
TTG gneisses and associated amphibolites
The structure of TTG gneisses varies from metatexites with small-scale leucocratic segregations (Fig. 3c) to schlieren migmatites with a higher degree of disaggregation (Fig. 3d). Mesosomes of the migmatitic TTG gneisses are predominantly tonalitic and consist of plagioclase + quartz ± potassic feldspar ± amphibole ± biotite. Mafic minerals make up 10–40% of the modal mineralogy but either biotite or amphibole can be absent (Fig. 2g). Potassic feldspar occurs in biotite-rich varieties.
The TTG gneisses commonly contain blocks, lenses or layers of amphibolite (Fig. 3e). Outcrop 038 in the NW part of the study area is an exception, as it is a larger block of amphibolitic gneiss compared with the small-scale lenses and layers of amphibolite (Fig. 1). Amphibole is the dominant mafic mineral in these amphibolite layers and lenses, generally accompanied by biotite. Clinopyroxene can be present but is usually altered into secondary green, Mg-rich amphibole (Fig. 2e). Plagioclase and quartz occupy interstices between granoblastic, subhedral to euhedral amphibole. Potassic feldspar occurs together with biotite (Fig. 2f). Amphibolite schollen and lenses are considered to represent disrupted pre-metamorphic mafic intrusions into the tonalitic basement.
Leucosomes
Granulites as well as the lower-grade TTG gneisses and amphibolites contain leucosomes, which are considered as the principal evidence for melt formation in all rock types. Leucosomes within mafic granulites from the Iisalmi–Sukeva area consist of thin veins and small leucocratic patches (Fig. 3a). Intermediate granulites from Varpanen and Jonsa areas are stromatic with leucosomes of several centimetres thickness (Fig. 3b). Additionally, diatexites with a high leucosome/mesosome ratio occur in the Jonsa and Varpanen areas. The amount of leucosome in the TTG gneisses is variable. Some outcrops display features of extensive melting and mobilization with host rock structures being nearly obliterated (Fig. 3c and d).
Leucosomes in all lithologies consist of large subhedral to euhedral grains of plagioclase with quartz occupying the interstices between plagioclase grains (Fig. 2h). Within granulites mafic phases are more abundant in small leucosome patches but are rare in thicker discordant leucosomes.
|
MINERAL CHEMISTRY AND METAMORPHIC CONDITIONS |
|---|
Representative mineral compositions determined by electron microprobe are given in Appendix B. Garnet in granulites has XMg [= Mg/(Mg + Fe)] of 0·26–0·36. Xgrs [= Ca/(Ca + Mg + Fe + Mn)] is normally 0·15–0·24, with one exception of a more calcic mafic granulite with Xgrs = 0·27. Amphiboles are ferropargasitic, pargasitic and edenitic in composition (after Leake, 1997
) with only minor within-sample variation. An exception is sample 036M1, where amphibole enclosed by garnet has higher XMg than larger amphibole grains lacking contact with garnet. Amphiboles from granulite patches in diatexite outcrops (080M, 102a, 114M) have higher K2O (1·5–1·8 wt %) than amphibole from outcrops with little leucosome (0·64–0·85 wt % K2O). Clinopyroxene is mainly augite and diopside with XMg between 0·6 and 0·8, and orthopyroxene has XMg of 0·5–0·7. Plagioclase composition is variable. In mafic granulites it has XAn 0·53–0·7 in agreement with the CaO-rich bulk composition. Plagioclase from two wide granulite bodies (076a, 112M) has XAn of 0·47–0·53 whereas plagioclase elsewhere in intermediate granulites has XAn of 0·26–0·33.
A detailed description of the metamorphic mineral assemblages and reactions in mafic and intermediate granulites and P–T estimates has been given by Hölttä & Paavola (2000). They obtained peak metamorphic conditions of 8–11 kbar and 800–950°C in granulites from garnet–clinopyroxene–plagioclase–quartz assemblages using the program TWQ 1.02 (Berman, 1991). Similar P–T conditions, although slightly lower temperatures of 750–850°C, were obtained using the new mineral compositions obtained as part of this study and the new TWQ version 2.32 (Berman, 2007). Pressure increases from the southeastern Jonsa block (8–9 kbar) to the Varpanen block and Iisalmi–Sukeva area (9–11 kbar), suggesting a deeper crustal origin for the granulites in the northwestern parts of the study area. Garnet decomposition to orthopyroxene and plagioclase in the granulites took place during decompression and cooling to 700°C and 7 kbar (Hölttä & Paavola, 2000). Symplectites after garnet and clinopyroxene are restricted to areas with late deformation and may have formed during Proterozoic tectonic movements.
TTG gneisses and enclosed amphibolites are free of garnet, which makes P–T estimates difficult. Amphibole–plagioclase thermometry following Holland & Blundy (1994) was applied in combination with the Al-in-hornblende barometer calibrated by Anderson & Smith (1995) to constrain P–T conditions for these rocks. TTG gneisses as well as their enclosed amphibolites usually contain more than one generation of amphibole. Prograde amphibole appears as brownish edenitic and pargasitic hornblende, and more rarely also as magnesiohornblende, following the nomenclature of Leake et al. (1997). Four samples from amphibolite lenses and two samples of amphibolitic mesosome in TTG gneisses were used for the calculations (Appendix C). XAn of plagioclase in these samples is 25–36. The samples yielded upper amphibolite-facies conditions with temperatures of 680–750°C and pressures of 4·5–5·3 kbar. The pressure differences between granulites and surrounding TTG gneisses strongly support a tectonic juxtaposition of the lower crustal granulites with the middle crustal TTG gneisses.
|
ANALYTICAL METHODS |
|---|
Bulk-rock samples were analysed for major and trace elements (Cr, Ni, Sc, V, Sr, Rb, Ba and Zr) by X-ray fluorescence (XRF) at the Institute for Geosciences, University of Mainz. The REE as well as Y, Hf, Nb, Ta, Th and U were analyzed by laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) and solution ICP-MS. Most samples with less than 55 wt % SiO2 were processed on an iridium-strip heater and the glass beads were measured by LA-ICP-MS. Further explanation of this method has been given by Fedorowich et al. (1993
) and Nehring et al. (2008). Samples with higher SiO2 contents were analysed following Li-metaborate fusion based on the method described by Mareels (2004). The solutions obtained by sample digestion after Li-metaborate fusion were analyzed using a VG Elemental Plasma Quad 3 at the University of Mainz, Institute for Geosciences. Thorium determination using the method of Mareels (2004) proved to be difficult, especially for the low concentrations in the leucosomes. Fluid inclusion studies were carried out using a Linkam THMS 600 heating–freezing stage mounted on an Olympus BH-2 microscope in combination with a TP 91 controller. The accuracy of the temperature measurements is generally about 1°C, and for CO2-melting 0·2°C.
|
FLUID INCLUSIONS |
|---|
Fluid inclusions were studied in mesosome–leucosome pairs from granulites and TTG gneisses. However, data were mainly derived from the leucosomes, because of their higher quartz content and more favourable properties for preserving fluid inclusions.
Fluid inclusions in granulites
All types of granulite are characterized by a high abundance of carbonic inclusions, which occur in quartz, and frequently also in garnet (mesosomes) or in plagioclase (leucosomes). The carbonic inclusions are situated along healed-fracture planes or within stretched and transposed clusters, none of which cross-cut crystal boundaries. They exhibit negative crystal shapes or are slightly rounded. Within garnet they are stretched and form large vermiform and only partly filled cavities. Melting of the CO2 inclusions occurs instantaneously close to the triple point of CO2 (Tm = –56·6°C), indicating that the fluid is pure CO2. The overwhelming majority of inclusions homogenize into the liquid phase. The homogenization temperatures (Th) range between –26°C and + 30°C and have a prominent mode between –5 and + 5°C (Fig. 4a). Secondary peaks can be recognized at –10°C and between + 20 and +23°C. Homogenization temperatures below –18°C were found only in one sample from the Varpanen area. Densities of CO2 inclusions corresponding to the range of Th are 
= 1·05–0·75 g/cm3.
|
Homogenization temperatures of inclusions in plagioclase agree with those in quartz within the same sample, whereas inclusions in garnet show higher Th. This may be due to a late recrystallization of quartz resulting in higher fluid densities in quartz compared with the rigid garnet porphyroblasts.
Granulites contain few aqueous inclusions, although these become more abundant near faults or close to the transition towards the TTG gneisses where hydration has occurred. Aqueous inclusions are irregularly formed and occur as isolated small clusters, within fractures or along grain boundaries, the latter indicating late formation of the quartz crystal boundaries. The aqueous inclusions are mostly monophase at room temperature. They usually melt in the temperature range –5 to –15°C (Fig. 4b) corresponding to salinities of 8–18 eq. wt % NaCl (Bodnar, 1993). Final melting temperatures below –30°C were observed in two samples; one of these is an amphibolite that most probably represents a retrogressed granulite.
Fluid inclusions in TTG gneisses and amphibolite
Irregular aqueous fluid inclusions are abundant in the migmatitic TTG gneisses, where they occur in settings similar to those in the granulites. Carbonic inclusions or mixed CO2–H2O inclusions are lacking in the migmatitic gneisses and amphibolites.
Final melting of ice within most of the aqueous inclusions is observed in the temperature range –5 to –15°C (Fig. 4c). According to freezing point depressions, this corresponds to less than 20 eq. wt % NaCl (Bodnar, 1993). Final melting temperatures below –30°C are rare. These brines rarely contain cubic crystals of salt. Formation of eutectic melt is frequently observed at temperatures below –50°C, clearly indicating the presence of CaCl2 in the aqueous fluid.
Aqueous inclusions suitable for measurements of homogenization temperatures are rare. One sample gave Th 240°C at low salinities of 8–14 eq. wt % NaCl whereas another gave Th 100°C at higher salinities of 18 eq. wt % NaCl.
|
GEOCHEMISTRY |
|---|
Granulites
Garnet–clinopyroxene granulites from the Iisalmi–Sukeva area are mafic with 48–50 wt % SiO2 (Table 1), whereas orthopyroxene–clinopyroxene granulites from the Jonsa area are intermediate in composition (52–60 wt % SiO2). Mafic granulites display fractionation trends as recognized in variation diagrams (Appendix D); a high modal abundance of clinopyroxene (35–50%) correlates with elevated CaO contents (8–16 wt %). Except for one location, mafic granulites have lower K2O (0·05–0·42 wt %) and Na2O contents (1·2–2·7 wt %) than intermediate granulites (K2O 0·26–1·25 wt %; Na2O 1·7–4·7 wt %) resulting in higher Ca/(Na + K) ratios. FeO and MgO contents are comparable in both types of granulite at the boundary between mafic and intermediate granulites, but they decrease with increasing SiO2 in the intermediate granulites.
|
The intermediate Jonsa granulites show a strong trend of increasing Al2O3 and decreasing TiO2 (Fig. 5) with increasing Mg-number [= Mg/(Mg + Fe2+)]. This correlation is not observed in the mafic granulites from the Iisalmi–Sukeva area, but a low- and a high-Al2O3 group may be distinguished in this rock type. Granulites from the Varpanen–Pällikäs area show compositional overlap with both groups. Trends in major element variation diagrams for both groups of granulites could indicate fractionation of clinopyroxene and An-rich plagioclase from a mafic precursor, although the proportion of the fractionating minerals most probably was different.
|
Granulites from the study area can be additionally subdivided according to their trace element characteristics (Hölttä, 1997
). Mafic granulites in the Iisalmi–Sukeva area as well as mafic interlayers in the Varpanen granulites have higher compatible element contents (Cr and Ni) but lower incompatible element abundances (Rb, Sr, Ba) than the intermediate granulites (Fig. 6, Table 2). Slight light REE (LREE) depletion is characteristic of the mafic granulites from the Iisalmi–Sukeva area whereas the Jonsa granulites are enriched in the middle REE (MREE) and LREE (Fig. 7). REE patterns of granulites from Pällikäs and Varpanen are flat or LREE enriched. Both types may occur in a single outcrop and flat REE patterns are restricted to more mafic, garnet-bearing layers. As shown by Hölttä (1997), a distinction between mafic and intermediate granulites can also be made according to their Ti/Zr ratios. Intermediate Jonsa granulites and intermediate samples from the Varpanen area have Ti/Zr = 40–61 whereas mafic granulites mainly show ratios of 67–180 (Fig. 6).
|
|
|
On multi-element plots intermediate granulites resemble basaltic andesites because of negative Nb–Ta, Ti and Zr anomalies. Mafic granulites lack such anomalies and broadly resemble tholeiitic basalts.
TTG gneisses
TTG gneisses are mainly tonalitic with SiO2 contents between 55 and 73 wt %. Their normalized REE patterns (Fig. 7) are strongly fractionated, as is typical for Archean TTG complexes (Martin, 1994). Two groups of TTG gneisses can be distinguished according to their major and trace element composition. The most striking difference can be observed in the slope of the REE pattern, so that the ratio of the chondrite-normalized values of La and Yb [(La/Yb)N] is 37–56 in group I gneisses and <20 in group II gneisses. Additionally, the two groups can be subdivided according to their Y contents, which are <8 ppm in group I and 11–32 ppm in group II. Low heavy REE (HREE) as well as low Y contents suggest a strong involvement of garnet in the petrogenesis of the group I gneisses. Sr/Y ratios are >55 in group I gneisses and mainly <52 in group II gneisses. Sr/Y ratios decrease with increasing Y contents but show no dependence on Sr contents, which are 250–400 ppm in group I and 500–860 ppm in group II gneisses.
Although there is compositional overlap, group I gneisses have higher SiO2 (68–73 wt %) and Na2O (> 4 wt %) but lower FeO (0·9–3·1 wt %), MgO (0·42–1·56 wt %) and CaO (1·7–4·0 wt %) contents than group II gneisses (SiO2 55–66 wt %, Na2O <4 wt %, FeO 4·5–6·3 wt %, MgO 2·07–4·3 wt %, CaO 3·9–4·7 wt %).
Amphibolitic gneisses and amphibolite lenses
Amphibolitic layers and lenses occurring within the TTG gneisses have SiO2 contents of 46–51 wt %, comparable with mafic granulites; however, their CaO and MnO contents are slightly lower and FeO and MgO are more variable (Appendix C). With the exception of the amphibolitic gneiss at outcrop 038, amphibolite lenses have high Cr, Ni, Sc and V contents, comparable with the mafic granulites, but in contrast to the depleted granulites they are enriched in incompatible elements (Fig. 6). Amphibolites have essentially flat chondrite-normalized REE patterns (Fig. 7).
Leucosomes
The SiO2 contents of leucosomes from granulites and TTG gneisses are in the range 57–80 wt %. In the normative feldspar classification diagram after Barker (1979), leucosomes from granulites are tonalitic whereas leucosomes from TTG gneisses are less calcic and cluster along the border between the trondhjemite and tonalite fields (Fig. 8). Few leucosomes in either rock type have K2O contents >1 wt %. Major element trends in leucosomes relate to the composition of the respective plagioclase in the leucosome (Fig. 9).
|
|
High Sr contents in leucosomes occur in host rocks with sodic plagioclase (intermediate granulites, TTG gneisses), whereas low Sr contents are observed in leucosomes from more calcic precursor rocks (mafic granulites, amphibolites) (Fig. 10). Rb contents >30 ppm are associated with Ba >500 ppm, reflecting the involvement of biotite. Zr contents in leucosomes exhibit a wide scatter between 18 and 340 ppm but are mostly below 150 ppm.
|
The REE patterns of leucosomes are very variable. Most leucosomes exhibit a strong positive Eu anomaly (Fig. 11), demonstrating the influence of plagioclase on their REE budget. Leucosomes enriched in REE or with rather flat REE patterns lack a Eu anomaly and only the most REE-enriched leucosomes show a slight negative Eu anomaly. Some leucosomes within granulites have high contents of HREE indicative of the presence of garnet as a peritectic or entrained phase.
|
|
GRANULITE AND MIGMATITE PETROGENESIS |
|---|
Fluid regime during partial melting of granulite- and amphibolite-facies rocks
Granulites
The lowest homogenization temperature of carbonic fluid inclusions found in only one diatexite sample is –24°C. This corresponds to an isochore that intersects the lower part of the P–T field for granulites yielding about 8 kbar at 800°C using the equation of state of Angus et al. (1976
) provided within the computer package FLUIDS by Bakker (2003). These inclusions may represent early inclusions trapped close to peak metamorphic conditions or during the early phases of decompression and cooling (Fig. 12). Although such early CO2 inclusions are rare, their presence confirms the experiments of Vityk & Bodnar (1995), which showed that a small percentage of inclusions may survive the discrepancy between inclusion and ambient pressure during uplift. Isochores corresponding to CO2 inclusions with lower densities clearly lie below the P–T field of granulites (Fig. 12). Density resetting of CO2 inclusions could have been triggered by extensional tectonics, which also allowed the intrusion of dolerite dykes during the Paleoproterozoic (2·3–2·1 Ga). Furthermore, disturbance of inclusion densities during the 1·9 Ga Svecofennian overprint at 6–7 kbar and 600–700°C (Hölttä & Paavola, 2000; Mänttäri & Hölttä, 2002) and adjustment of fluid-inclusion pressures to lithostatic pressure during uplift could have caused the wide range in homogenization temperatures.
|
The overall abundance of carbonic inclusions compared with the limited number of aqueous inclusions in granulites most probably results from the low wetting ability of CO2, which promotes preferential capture of the carbonic fluid (Watson & Brenan, 1987
; Holness, 1993). This also precludes infiltration of CO2-rich fluids from igneous enderbites as the principal mechanism of dehydration. Igneous enderbites that contain abundant carbonic fluid inclusions (Poutiainen, 1992) occur throughout the area. They intruded during the early phases of metamorphism and experienced a granulite-facies overprint themselves (Hölttä & Paavola, 2000; Mänttäri & Hölttä, 2002). Although dehydration induced by infiltration of a carbonic fluid from enderbites into their wall-rocks has frequently been observed (Knudsen & Lidwin, 1996; Van den Kerkhof & Grantham, 1999; Harlov et al., 2006), the low wetting ability of carbonic fluids requires intense fracturing or shearing to allow fluid transport over larges distances from the source. The widespread and pervasive occurrence of granulites in the study area argues against granulite formation along channelized fluid pathways.
A limited number of aqueous inclusions with >20 eq. wt % NaCl is present in most of the granulite samples. The occurrence of such saline aqueous inclusions (brines) together with carbonic inclusions and the absence of mixed gaseous–aqueous inclusions in granulites supports experimental findings of immiscibility between brines and carbonic fluids under lower crustal conditions (Bowers & Helgeson, 1983; Johnson, 1991; Shmulovich & Graham, 2004). Both fluid types may evolve by unmixing from a single fluid phase during progressive metamorphism. H2O activities in brines are low at pressures of lower crustal metamorphism, allowing the coexistence with the anhydrous mineralogy of the granulites (Aranovich & Newton, 1996; Newton et al., 1998; Newton & Manning, 2000).
Nevertheless, the majority of aqueous inclusions are less saline. These inclusions may have developed by expulsion of aqueous fluid upon crystallization of the leucosomes from a hydrous melt. This is also consistent with the presence of hydrous minerals within and along leucosome boundaries, which were produced by retrograde reactions of the anhydrous mineral assemblage with the hydrous melt. Hydration reactions along or within melt channels may have the potential to increase the salinity of the coexisting fluid phase as deduced from the works of Svensen et al. (1999), Markl & Bucher (1998) or Trommsdorff et al. (1985); this could explain the rather wide spread in fluid salinities observed in the granulite samples.
TTG gneisses
A major reduction in water activity across the amphibolite- to granulite-facies transition has been described from many localities (e.g. Touret & Olsen, 1985; Andersen et al., 1997; Perchuk et al., 2000). Aqueous fluid inclusions with variable salinities as observed in the Finnish TTG gneisses seem to be typical for rocks equilibrated under amphibolite-facies conditions in a variety of geological settings (Andersen et al., 1991; Klemd et al., 1995; Carson et al., 2002; Yardley & Graham, 2002).
The lack of carbonic inclusions in the gneisses and the presence of aqueous inclusions with variable salinities point to fluid-present conditions during peak metamorphism. Furthermore, apart from the formation of clinopyroxene along amphibole rims there is no mineralogical evidence that these rocks ever experienced metamorphic conditions sufficient to promote hydrous phase breakdown melting. However, anatexis is evidenced by the strong migmatization. It is suggested that a H2O-rich fluid aided melt formation in rocks that could otherwise not have melted at the metamorphic conditions under consideration. The mole fraction of NaCl in inclusions with the lowest melting temperatures of –35°C is XNaCl = 0·16 resulting in aH2O = 0·72 at 10 kbar, which is sufficient for fluid-induced melt formation in the granitic system at temperatures close to 700°C (Aranovich & Newton, 1996). If these brines represent the peak metamorphic fluid, they probably evolved by scavenging of water in the melt phase (Yardley & Graham, 2002).
Devolatilization of hydrous minerals contained in the TTG gneisses along the prograde path may have provided small quantities of water sufficient to produce the low melt fractions observed in the metatexites. The transition to melt-rich schlieren migmatites in single outcrops most probably was a result of influx of an external fluid. White et al. (2005) considered fluid release by subsolidus dehydration reactions in metapelites and metapsammites as a source for fluid-enhanced melt production in adjacent gneisses. Metasediments occur in the Jonsa area but are absent from the rest of the region and are therefore not likely sources for hydrous fluids. Nevertheless, the exposed granulites continue to deeper levels, as suggested by geophysical data (Korsman et al., 1999), so that dehydration reactions of hydrous mafic rocks could provide a possible source of hydrous fluids. Additionally, hydrous phase breakdown melting at deeper crustal levels of mafic rocks and crystallization of these melts could have released the hydrous fluids required for melting of the overlying TTG gneisses. The migration ability of the aqueous fluid would be enhanced compared with carbonic fluids because of the better wetting ability of the aqueous fluid.
Aqueous inclusions with homogenization temperatures of 240°C could have been trapped anywhere along the uplift path following an isochore that crosses through the P–T conditions of all major tectonic events (Fig. 12). Therefore salinities of 8–14 eq. wt % NaCl could closely correspond to the peak metamorphic fluid composition. More saline, high-density aqueous inclusions with Th 100°C most probably developed by decrepitation processes and quartz recovery during uplift and finally equilibrated under near-surface conditions (Touret & Dietvorst, 1983; Van den Kerkhof et al., 2004). Simultaneous water uptake in the formation of low-grade hydrous minerals could explain high salinities in the residual fluids.
Melt formation in granulites
The common depletion of incompatible elements in large Archean granulite terranes (Lambert & Heier, 1968; Pride & Mücke, 1980; Vielzeuf et al., 1990; Guernina & Sawyer, 2003; Hansen & Harlov, 2007) suggests that granulites are residues of melt extraction. The amount of melt extracted from granulite terranes strongly depends on the amount of hydrous phases present at the onset of melting and on the melt-forming reaction, which is mainly controlled by the P–T conditions. Hydrous phase breakdown melting reactions of amphibole have been examined in a variety of experimental studies (Beard & Lofgren, 1991; Rapp et al., 1991; Rushmer, 1991; Sen & Dunn, 1994; Wolf & Wyllie, 1994; Patiño Douce & Beard, 1995; Rapp, 1995; Springer & Seck, 1997; Sisson et al., 2005). Therefore the P–T dependence of melt-forming reactions in mafic and intermediate source rocks is well known. This allows us to model the observed mineral modes and compositions of granulites as residues from partial melting of amphibole-bearing source rocks.
Mafic granulites
The modal amphibole content of the mafic granulites is <10%. The relatively dry character of the rocks can be accounted for by two scenarios.
- The igneous precursor of the mafic granulites intruded into the middle to lower crust and never experienced hydration so that the rock was dry at the onset of metamorphism. In this case, the mineral assemblage of the mafic granulites represents igneous minerals recrystallized to granulite-facies conditions and the trace element composition of mafic granulites is a feature inherited from a source rock with a composition comparable with normal mid-ocean ridge basalt (N-MORB).
- The rocks were moderately hydrous at the onset of metamorphism before amphibole was nearly consumed during the formation of clinopyroxene, garnet and melt. In this case, the trace element characteristics of the mafic granulites can be modelled as residues of melting of a mafic precursor.
We favour the second option for the petrogenesis of the mafic granulites from the Iisalmi–Sukeva area. The dry solidus of mafic rocks is located between 800 and 900°C at 5–10 kbar (Beard & Lofgren, 1991; Rushmer, 1991; Sisson et al., 2005). Temperatures in excess of 800°C and pressures of 9–11 kbar were determined for these granulites and allowed partial melting of mafic rocks in agreement with the following general hydrous phase breakdown melting reaction (Wolf & Wyllie, 1994; Hartel & Pattison, 1996):
|
| (1) |
Hartel & Pattison (1996) predicted that garnet is produced in larger amounts than clinopyroxene and that titanite will be an additional crystallizing phase. Titanite is rarely observed and clinopyroxene is always more abundant than garnet in the mafic granulites, as deduced from field appearance, point counting and calculation of modal abundances. This may be attributed to subsolidus clinopyroxene formation on the prograde path before the rocks entered the stability field of garnet.
Patches and veins of leucosome as well as low Rb, Ba and K contents within the mafic granulites are indicative of the loss of a melt fraction. The mafic granulites are depleted in LREE and have fractionated Th/U ratios that may be an additional feature produced during partial melting. Similar processes have been described by Garrido et al. (2006) for the formation of garnet granulites in the Kohistan Arc Complex by hydrous phase breakdown melting of amphibole-bearing arc plutonic rocks.
Amphibolite lenses and layers abundantly present in the TTG gneisses resemble mafic granulites in terms of major elements and compatible trace elements. These amphibolites are the only mafic rocks of suitable age and composition in the study area that could represent lower grade analogues of mafic granulites. The apparent depletion of the mafic granulites in incompatible elements compared with the enrichment of these elements in the amphibolites could then be produced during partial melting. To investigate this we applied a batch melting model using observed modal compositions of mafic granulites as likely residues from melting of an average amphibolite composition. Partition coefficients experimentally derived under conditions appropriate for crustal melting are limited. Klein et al. (2000) reported DCpx and DGrt in equilibrium with tonalitic liquids at 15 kbar and 1050°C and 900°C, respectively, which are consistent with the results of Barth et al. (2002). Amphibole partition coefficients were taken from Klein et al. (1997), as they were obtained in equilibrium with a tonalitic liquid at 900°C and 10 kbar. However, more recent research (Tiepolo et al., 2000) showed that amphibole fractionates Nb from Ta with AmphDNb > AmphDTa. We therefore used AmphDNb 0·44 and AmphDTa 0·26, which are suitable for the observed amphibole compositions (TiTot 0·2 a.p.f.u., Mg-number 60). Plagioclase partition coefficients for XAn 50 were calculated using the equations of Bédard (2006b).
About 10% partial melting of an average amphibolite lens leaving 10% garnet, 35% plagioclase, 45% clinopyroxene and 10% amphibole in the residue is sufficient to explain the trace element pattern of most mafic granulites. Some strongly fractionated mafic granulites require up to 30% partial melting, higher amounts of residual garnet (30%) and lower amounts of residual plagioclase (20%) (Fig. 13a). Incidentally, these strongly fractionated granulites record pressures of 11 kbar, which points to an increased incorporation of plagioclase in the melt at pressures >20 kbar as observed, for instance, by Sen & Dunn (1994).
|
In model residues, low Ba contents similar to those of the mafic granulites are obtained only if Ba in the source rock is less than 150 ppm. This requires omission of two biotite-rich amphibolite lenses with Ba contents of 454 ppm and 894 ppm from the calculation of an average amphibolite composition. Ba contents <20 ppm observed in mafic granulites with strongly fractionated REE patterns are not reproduced by our model (Fig. 13b). They require less Ba in the source as well as total consumption of amphibole and lower modal plagioclase in the residue. Likewise, low Sr contents in strongly fractionated mafic granulites are reproduced only by low source rock Sr contents <100 ppm instead of 140 ppm of an average Iisalmi amphibolite. Slightly fractionated mafic granulites exhibit a stronger positive Sr anomaly than the strongly fractionated granulites. This Sr anomaly in the granulites could be caused by the storage of Sr in residual plagioclase. In this case, the smaller positive Sr anomaly in strongly fractionated granulites points to an increased incorporation of plagioclase in the melt phase at higher pressures.
Adjustment of Nb and Ta values requires a phase with high DNb and DTa. Ilmenite is a common accessory phase in the granulites from the study area whereas rutile is not observed. LA-ICP-MS analyses of ilmenites show high concentrations of Nb, Ta, Sc and V, and including ilmenite in the melting model yields Nb and Ta contents in good agreement with the mafic granulites. Additionally, ilmenite will retain Ti during melting.
The residual garnet in the mafic granulites is expected to cause HREE depletion in the corresponding melts. This implies that hydrous phase breakdown melting of amphibolitic source rocks under lower crustal conditions leaving a garnet-bearing mafic granulite residue produces strongly fractionated tonalitic melts similar to typical Archean TTGs (Fig. 13c). Different amounts of garnet in the residue will cause variable HREE depletion similar to observed HREE variations in the TTG gneisses. However, Sr and Eu contents in the melt are strongly controlled by the amount and composition of residual plagioclase. High Sr contents (>250 ppm) are typical for Archean TTG (Martin, 1997; Martin et al., 2005); the lack of negative Sr and Eu anomalies in melts can arise only if plagioclase abundance in the source progressively decreases at pressures above 10 kbar, leading to plagioclase-free eclogite melting as originally suggested for TTG genesis by Drummond & Defant (1990).
Intermediate granulites
Compared with mafic granulites, in which prograde amphibole is rare, intermediate granulites provide abundant evidence for amphibole breakdown such as overgrowth of pyroxenes on amphibole. It is suggested that prograde clinopyroxene and orthopyroxene were produced by a reaction corresponding to reaction (2) as found by Patiño Douce & Beard (1995):
|
| (2) |
This reaction is consistent with lower pressures of 8–9 kbar determined for the intermediate granulites from the Jonsa block in the SE of the study area compared with mafic granulites from the NW part of the study area, which equilibrated at 9–11 kbar. A similar clinopyroxene/orthopyroxene ratio to that observed by Patiño Douce & Beard (1995) is found in the intermediate granulites (Table 3). Metamorphism and melting of a layered suite of rocks such as the granulites from the Jonsa and Varpanen area produces inhomogeneous residues whose mineralogy depends on source rock composition on the one hand and degree of amphibole breakdown on the other. Accordingly, intermediate granulites exhibit significant variations in modal mineralogy, with plagioclase constituting 30–50%, amphibole 10–40% and clinopyroxene 10–30% of the modal abundance. The high proportion of plagioclase in intermediate granulites indicates that it was a stable phase during partial melting and most probably also formed as a peritectic product as predicted from reaction (2). The garnet-producing reaction (1) is restricted to Fe-rich layers in intermediate granulites but within these layers garnet can make up more than 20%.
|
By assuming that all clinopyroxene in the intermediate granulites was produced by the breakdown of amphibole in the presence of quartz [reaction (2)] it is possible to calculate the amount of melt that formed as a result of reaction (2). The melt proportion was small, ranging between 5 and 10% for most of the samples (Table 3). Because amphibole is still present in most intermediate granulites the rocks are fertile with respect to melt formation. Free quartz is absent from the intermediate granulites, so the consumption of quartz in the melt-forming reaction probably controlled the reaction progress.
The low degree of melting in the intermediate granulites should not change the REE contents significantly so that the observed REE contents will closely approach source compositions. Sub-parallel REE patterns for the intermediate granulites additionally indicate that HREE-retaining garnet did not play a significant role in fractionation during partial melting, except for some HREE-enriched layers. The consumption of amphibole during partial melting mainly liberates large ion lithophile elements (LILE) as well as Nb and Ta. However, the concentration of Nb and Ta in the residue is buffered by the formation of ilmenite. Accessory biotite and residual amphibole buffer Ba contents. Sr contents are generally high in the intermediate granulites but positive Sr anomalies on multielement plots are rare. Samples lacking Sr anomalies or having small negative Sr anomalies occur more frequently. Variable concentrations of Sr must be inherited from the precursor rocks, as Sr remains in the residue because of the large modal abundance and stability of plagioclase.
Melts derived from intermediate granulites mainly inherit the trace element pattern of their sources but will be more LILE- and LREE-rich. The HREE will not be fractionated because of the absence of garnet. The shape of the MREE pattern is influenced by the amount of amphibole in the residue. Negative Eu and Sr anomalies may be characteristic for melts derived from intermediate granulites because of the high amounts of restitic plagioclase. Altogether, this implies that the trace element patterns of partial melts of intermediate granulites formed at pressures less than 10 kbar do not resemble typical Archean TTG although, in terms of major elements, they are tonalitic.
Melt formation in TTG gneisses
The mesosomes of migmatitic TTG gneisses are hydrous comprising amphibole and biotite as mafic phases. Additionally, clinopyroxene can be present especially in amphibolite lenses and layers. The lack of anhydrous mineral assemblages indicates that the leucosomes in the TTG gneisses were not formed as a result of hydrous phase breakdown melting but that melting at low temperatures was probably fluxed by a hydrous fluid. The stability of plagioclase is greatly lowered in the presence of water. Therefore melting in hydrous quartzofeldspathic compositions starts at temperatures considerably lower than required for hydrous phase breakdown melting and the melt productivity is greatly enhanced in the H2O-present environment (Beard & Lofgren, 1991).
Water-saturated experiments on natural granitoid compositions were conducted at 15 kbar by Johnston & Wyllie (1988) and Carroll & Wyllie (1990), and at 20 kbar by Acosta-Vigil et al. (2006). The melting interval of the trondhjemitic composition used by Johnston & Wyllie (1988) lies between 610 an 745°C. Carroll & Wyllie (1990) detected the solidus for their tonalite below 700°C at water contents of 5 wt %. Acosta-Vigil et al. (2006) observed an interconnected melt phase already at 690°C and low degrees of melting. According to Johannes (1984) the beginning of melting in the granite system qz–kfs–ab–an–H2O can be observed between 690°C at 2 kbar and 630°C at 17 kbar.
The transition from stromatic migmatites to strongly disaggregated schlieren migmatites observed in some outcrops of TTG gneisses (Fig. 3d), resembles the metatexite to diatexite transitions described by White et al. (2005). The amount of melt required to produce diatexites is estimated to be of the order of 20–40 vol. % (Greenfield et al., 1996; Sawyer, 1998). Becausee most migmatitic gneisses preserve pre-migmatitic structures and classify as metatexites, the amount of melt in the system most probably was less than 25 vol. %. Redistribution of melt could explain the enrichment of melt and formation of diatexites in some migmatite outcrops. Nevertheless, the dependence of Ca/Na ratios (see above) and Sr and Rb contents in leucosomes on mesosome composition, as indicated in Fig. 10, argues against a large degree of melt injection and melt mixing.
Leucosome plagioclase compositions are slightly more sodic in some outcrops but virtually the same as mesosome plagioclase composition in other outcrops (Fig. 14). More sodic plagioclase is expected in the melt phase compared with the residual mesosome, but equal plagioclase compositions in leucosomes and mesosomes have often been described from migmatites (Gupta & Johannes, 1982; Whitney & Irving, 1994). This contradiction to experimental data may be explained by percolation of melt along grain boundaries leading to re-equilibration of mesosome and plagioclase leucosome with the same melt phase (Marchildon & Brown, 2001).
|
Larger intrusive bodies that could represent melts segregated from the migmatitic gneisses are lacking at the present outcrop scale. Small leucodiorite intrusions (Naimakangas leucodiorites) from the far NW of the field area would be likely candidates supported by conventional U–Pb age data on zircons that give 2706 ± 3 Ma (Paavola, 2003
). However, neodymium isotopes yield TDM model ages of 2800 Ma (H. Huhma, personal communication), thus showing very limited contribution of older crustal material to the sources of the Naimakangas leucodiorites.
Origin of leucosomes in granulites and amphibolite-facies rocks
The origin of leucosomes has long been debated. In most circumstances, leucosomes are related to the anatexis of their host rocks, but they may also represent minimum melts, equilibrium melts, disequilibrium melts, fractionated melts or cumulates. Comparison of the major element composition of leucosomes from different rock types with melt compositions generated by Johannes (1989) in the tonalitic system Qtz–(Ab + Or)–An reveals that leucosomes from the study area are poorer in Qtz than tonalitic melts at 2 kbar (Fig. 15). Lower quartz contents than in the 2 kbar experiments can be produced by higher pressures that allow plagioclase to enter the melt in a higher proportion than in the experiments. Leucosome compositions from the TTG gneisses overlap with melt compositions derived in melting experiments on tonalitic compositions under variable H2O activities, temperatures and pressures. All of these experiments yielded compositions displaced towards the (Ab + Or)–An join of the triangle (Fig. 15). The position of most granulite leucosomes is removed towards the An–Qtz join of the triangle compared with the leucosomes from the TTG gneisses. This clearly shows the influence of the more calcic mafic source rocks.
|
There is little difference in the trace element pattern of leucosomes from different rock types despite the variable trace element budget of the host rocks. This suggests that leucosome trace element composition results from a process that might equally take place in leucocratic partial melts from different source rocks. For instance, the overall positive Eu- and Sr anomalies in leucosomes may indicate that plagioclase fractionation was important in their petrogenesis because of the high partition coefficients of plagioclase for Sr and Eu (Blundy & Wood, 1991
; Wilke & Behrens, 1999; Bédard, 2006b). Leucosomes may therefore consist of plagioclase and quartz crystallized from a percolating and finally extracted melt phase (Marchildon & Brown, 2001; Sawyer, 2001; Solar & Brown, 2001). This is also supported by the coarse grain size and the euhedral shape of plagioclase crystals as already pointed out by Brown (2001). Furthermore, leucosomes are richer in CaO (Fig. 16) but poorer in K2O at high SiO2 contents than the vast majority of Archean TTG, pointing to the effect of a pronounced input of fractionated plagioclase on leucosome composition.
|
Several parameters can influence leucosome composition if the fractional crystallization model applies:
- composition of the parent melt;
- composition of the fractionating plagioclase and changing partition coefficients with changing XAn;
- degree of fractional crystallization;
- amount of fractionating plagioclase and other potential minerals such as quartz or K-feldspar;
- entrapment of evolved liquids and entrainment of host rock minerals;
- fractionation from already fractionated liquids.
To test the fractional crystallization model we determined the modal abundance of plagioclase in migmatite and granulite leucosomes by least-square regressions using appropriate plagioclase compositions. Plagioclase abundances were found to range between 40 and 60 modal% and a mean value of 50 modal% was used in modelling. XAn in the leucosomes ranges between 20 and 40. We used partition coefficients (Di) in the calculations appropriate for XAn 30 after Bédard (2006b). The most arguable variable in the fractional crystallization model for leucosomes is the composition of the parent melt. Two out of 32 migmatite leucosomes analyzed have a smooth REE pattern and a major and trace element composition closely resembling the Naimakangas leucodiorites as well as mesosomes from the TTG gneisses (Table 4). Therefore the two leucosomes as well as an average Naimakangas leucodiorite and two groups of migmatite mesosomes were used as potential parent melts.
|
Migmatite leucosomes with the lowest REE contents probably represent the most pristine cumulates, and the degree of fractional crystallization to produce such leucosomes from a leucocratic melt with an REE content typical for the TTG gneisses is
10% (Fig. 17). HREE in the pristine leucosomes are slightly higher than could be explained by fractional crystallization of plagioclase alone. Zircon has high DHREE and its crystallization as an accessory phase raises the HREE contents of the leucosomes. Alternatively, the entrainment of zircon from the host rock can produce the same effect. Higher REE contents and less pronounced positive Eu anomalies in many other leucosomes from TTG gneisses suggest that they formed by fractionation from an already fractionated liquid or that they contain trapped fractionated melts as suggested by Sawyer (1987). Both processes may operate equally.
|
Leucosome 036–04 L2 was used as the parent melt for modelling the granulite leucosomes because of its good agreement with the 10% model melt from the partial melting model (Fig. 13c). As observed for the leucosomes within the TTG gneisses the whole spread of leucosome REE contents in granulites can be explained only by fractionation from evolved liquids or by entrainment of residual melts and if accessory zircon accounts for the typical trough-shaped pattern of the HREE.
Sr contents of leucosomes are extremely sensitive to the fractional crystallization model, as Sr will be effectively removed from the liquid as fractionation progresses. The Sr contents of the respective parent melt for each leucosome is source rock dependent (Fig. 10) and will strongly influence the Sr content in the cumulate. Sr contents are usually below 400 ppm in granulite leucosomes. About 30–50% fractional crystallization is required to account for the observed Sr contents in granulite leucosomes using an initial Sr content of 300 ppm in the parent melt and D
= 6·5 for XAn 30.
The formation of leucosomes by crystallization of plagioclase and quartz from a melt phase involves generation of fractionated residual liquids with negative Eu and Sr anomalies. Very few leucosomes with these geochemical characteristics have been found in the Iisalmi block. The lack of residual liquids among the sampled lithologies could be due to incomplete sampling but may also indicate efficient melt loss from the system.
|
CONCLUSIONS |
|---|
We present an example of internal differentiation of Archean continental crust by coeval partial melting of middle and lower crustal lithologies. Melt-forming reactions in the lithologies of different levels of the crust strongly depend on the activity of H2O in the peak metamorphic fluid. Whereas lower crustal granulites represent the anhydrous residues of hydrous phase breakdown melting of amphibole, melting within middle crustal TTG gneisses and amphibolites requires the presence of a hydrous fluid to allow melting at the wet solidus.
About 10–30% tonalitic melt was produced by melting of an amphibolitic precursor leaving mafic garnet–clinopyroxene granulites as residues. Intermediate clinopyroxene–orthopyroxene granulites record 10% of melting according to the clinopyroxene/amphibole ratio. TTG gneisses show variable degrees of migmatization. The obliteration of pre-migmatitic structures by disaggregation and the formation of schlieren migmatite indicates that melt production within the gneisses may have been as high as 25 vol. %.
Melting of the TTG gneisses yielded leucosomes that are rich in Na2O and have an affinity towards trondhjemite, whereas leucosomes within granulites are CaO-rich and classify as tonalites. This indicates that remelting of a Na2O-rich precursor under conditions that allow plagioclase taking part in the melt-forming reactions is a viable mechanism to produce trondhjemitic melts. In the present study plagioclase stability most probably was lowered because of the high activity of H2O in the migmatitic gneisses.
Leucosome trace element patterns with strong positive Eu anomalies reflect the importance of plagioclase in leucosome petrogenesis. These patterns can be reproduced by fractional crystallization of plagioclase, quartz and accessory zircon from a leucocratic melt including the storage of small amounts of fractionated liquid. Efficient melt extraction from migmatite and granulite terrains is suggested because of the residual character of the leucosomes and their host rocks.
|
APPENDIX A: OUTCROP LOCALITIES AND SAMPLE MINERALOGY (TWO COORDINATES ARE GIVEN FOR LARGER OUTCROPS) |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
APPENDIX B: REPRESENTATIVE MINERAL COMPOSITIONS |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
*Fe2O3 in garnet calculated from the charge balance of the analyses.
|
APPENDIX C: REPRESENTATIVE AMPHIBOLE AND PLAGIOCLASE COMPOSITIONS USED FOR AMPHIBOLE–PLAGIOCLASE THERMOBAROMETRY ON AMPHIBOLITES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
APPENDIX D: MAJOR ELEMENT VARIATIONS VS. SIO2 FOR MAIN LITHOLOGIES FROM THE IISALMI BLOCK |
|---|
|
ACKNOWLEDGEMENTS |
|---|
The support of the ICP-MS group of the Max-Planck-Institute for Chemistry (subdivision Geochemistry) with the whole-rock trace element analyses is gratefully acknowledged. We also want to thank the Geological Survey of Finland for its support during the field work. The manuscript benefited from constructive reviews by Michael Brown, Trevor Green and Daniel Harlov. This work was funded by the German Research Foundation and the state of Rheinland–Pfalz within the framework of the graduate school dedicated to research on the ‘Composition and Evolution of Crust and Mantle’.
*Corresponding author. Telephone: +49-471-4831-1927. Fax: +49-471-4831-1929. E-mail: Franziska.Nehring{at}awi.de
|
REFERENCES |
|---|
Acosta-Vigil A., London D., Morgan G. B. VI. Experiments on the kinetics of partial melting of a leucogranite at 200 MPa H2O and 690–800°C: Compositional variability of melts during the onset of H2O saturated crustal anatexis. Contributions to Mineralogy and Petrology (2006) 151:539–557.[CrossRef][Web of Science]
Andersen T., Austrheim H., Burke E. A. J. Fluid-induced retrogression of granulites in the Bergen Arcs, Caledonides of W. Norway: Fluid inclusion evidence from amphibolite-facies shear zones. Lithos (1991) 27:29–42.[CrossRef][Web of Science]
Andersen T., Whitehouse M. J., Burke E. A. J. Fluid inclusions in Scourian granulites from the Lewisian complex of NW Scotland: Evidence for CO2-rich fluid in Late Archaean high-grade metamorphism. Lithos (1997) 40(2–4):93–104.[CrossRef][Web of Science]
Anderson J. L., Smith D. R. The effects of temperature and fO2 on the Al-in-hornblende barometer. American Mineralogist (1995) 80:549–559.[Abstract]
Angus S, Armstrong B, de Reuk KM, Altunin VV, Gadetskii OG, Chapela GA, Rowlinson JS. International Thermodynamic Tables of the Fluid State, 3, Carbon Dioxide (1976) Oxford: Pergamon Press.
Aranovich L. Y., Newton R. C. H2O activity in concentrated NaCl solutions at high pressures and temperatures measured by the brucite–periclase equilibrium. Contributions to Mineralogy and Petrology (1996) 125:200–212.[CrossRef][Web of Science]
Bakker R. J. Package FLUIDS 1. Computer programs. Chemical Geology (2003) 194:3–23.[CrossRef][Web of Science]
Barker F. Trondhjemite: Definition, environment and hypotheses of origin. In: Trondhjemites, Dacites and Related Rocks—Barker F., ed. (1979) Amsterdam: Elsevier. 1–12.
Barth M. G., Foley S. F., Horn I. Partial melting in Archean subduction zones: Constraints from experimentally determined trace element partition coefficients between eclogitic minerals and tonalitic melts under upper mantle condiditons. Precambrian Research (2002) 113:323–340.[CrossRef][Web of Science]
Beard J. S., Lofgren G. E. Dehydration melting and water-saturated melting of basaltic and andesitic greenstones and amphibolites at 1, 3 and 6–9 kb. Journal of Petrology (1991) 32:365–401.
Bédard J. H. A catalytic delamination-driven model for coupled genesis of Archaean crust and sub-continental lithospheric mantle. Geochimica et Cosmochimica Acta (2006a) 70:1188–1214.[CrossRef]
Bédard J. H. Trace element partitioning in plagioclase feldspar. Geochimica et Cosmochimica Acta (2006b) 70:3717–3742.[CrossRef][Web of Science]
Berman R. G. Thermobarometry using multi-equilibrium calculations: a new technique, with petrological applications. Canadian Mineralogist (1991) 29:833–855.[Web of Science]
Berman RG. winTWQ (version 2.3): A software package for performing internally-consistent thermobarometric calculations. Geological Survey of Canada, Open File (2007) 5462 (revised).
Blundy J., Wood B. Crystal-chemical controls on the partitioning of Sr and Ba between plagioclase feldspar, silicate melts and hydrothermal solutions. Geochimica et Cosmochimica Acta (1991) 55:193–209.[CrossRef][Web of Science]
Bodnar R. J. Revised equation and table for determining the freezing point depression of H2O–NaCl solutions. Geochimica et Cosmochimica Acta (1993) 57:683–684.[CrossRef][Web of Science]
Bowers T. S., Helgeson H. C. Calculation of the thermodynamic and geochemical consequences of nonideal mixing in the system H2O–CO2–NaCl on phase relations in geologic systems: Metamorphic equilibria at high pressures and temperatures. American Mineralogist (1983) 68:1059–1075.[Abstract]
Brown M. Orogeny, migmatites and leucogranites: A review. Proceedings of the Indian Academy of Sciences—Earth and Planetary Sciences (2001) 110(4):313–336.
Carroll M. R., Wyllie P. J. The system tonalite–H2O at 15 kbar and the genesis of calc-alkaline magmas. American Mineralogist (1990) 75:345–357.[Abstract]
Carson C. J., Ague J. J., Grove M., Coath C. D., Harrison T. M. U–Pb isotopic behaviour of zircon during upper-amphibolite facies fluid infiltration in the Napier Complex, east Antarctica. Earth and Planetary Science Letters (2002) 199:287–310.[CrossRef][Web of Science]
Condie K. C. Episodic continental growth and supercontinents: A mantle avalanche connection? Earth and Planetary Science Letters (1998) 163(1–4):97–108.[CrossRef][Web of Science]
Condie K. C. TTGs and adakites: Are they both slab melts? Lithos (2005) 80:33–44.[CrossRef][Web of Science]
Drummond M. S., Defant M. J. A model for trondhjemite–tonalite–dacite genesis and crustal growth via slab melting—Archean to modern comparisons. Journal of Geophysical Research—Solid Earth and Planets (1990) 95(B13):21503–21521.[CrossRef]
Fedorowich J. S., Richards J. P., Jain J. C., Kerrich R., Fan J. A rapid method for REE and trace-element analysis using laser sampling ICP-MS on direct fusion whole-rock glasses. Chemical Geology (1993) 106:229–249.[CrossRef][Web of Science]
Foley S., Tiepolo M., Vannucci R. Growth of early continental crust controlled by melting of amphibolite in subduction zones. Nature (2002) 417(6891):837–840.[CrossRef]
Garrido C. J., Bodinier J.-L., Burg J.-P., Zeilinger G., Hussain S. S., Dawood H., Chaudhry S., Gervilla F. Petrogenesis of mafic garnet granulite in the lower crust of the Kohistan Paleo-arc Complex (Northern Pakistan): Implications for intra-crustal differentiation of island arcs and generation of continental crust. Journal of Petrology (2006) 47(10):1873–1914.
Greenfield J. E., Clarke G. L., Bland M., Clark D. J. In-situ migmatite and hybrid diatexite at Mt. Stafford, Central Australia. Journal of Metamorphic Geology (1996) 14(4):413–426.[CrossRef][Web of Science]
Guernina S., Sawyer E. W. Large-scale melt-depletion in granulite terranes: An example from the Archean Ashuanipi Subprovince of Quebec. Journal of Metamorphic Geology (2003) 21(2):181–201.[CrossRef][Web of Science]
Gupta L. N., Johannes W. Petrogenesis of a stromatic migmatite (Nelaug, southern Norway). Journal of Petrology (1982) 23(4):548–567.
Hansen E. C., Harlov D. E. Whole-rock, phosphate, and silicate compositional trends across an amphibolite- to granulite-facies transition, Tamil Nadu, India. Journal of Petrology (2007) 48:1641–1680.
Hansen E. C., Stuk M. Orthopyroxene-bearing, mafic migmatites at Cone Peak, California: Evidence for the formation of migmatitic granulites by anatexis in an open system. Journal of Metamorphic Geology (1993) 11:291–307.[Web of Science]
Harlov D. E., Johansson L., Van den Kerkhof A., Förster H.-J. The role of advective fluid flow and diffusion during localized, solid-state dehydration: Söndrum Stenhuggeriet, Halmstad, SW Sweden. Journal of Petrology (2006) 47:3–33.
Hartel T. H. D., Pattison D. R. M. Genesis of the Kapuskasing (Ontario) migmatitic mafic granulites by dehydration melting of amphibolite: The importance of quartz to reaction progress. Journal of Metamorphic Geology (1996) 14:591–611.[CrossRef][Web of Science]
Hawkesworth C. J., Kemp A. I. S. Using hafnium and oxygen isotopes in zircons to unravel the record of crustal evolution. Chemical Geology (2006) 226:144–162.[CrossRef][Web of Science]
Holland T. J. H., Blundy J. Non-ideal interactions in calcic amphiboles and their bearing on amphibole–plagioclase thermometry. Contributions to Mineralogy and Petrology (1994) 116:433–447.[CrossRef][Web of Science]
Holness M. B. Temperature and pressure dependence of quartz–aqueous fluid dihedral angles: The control of adsorbed H2O on the permeability of quartzites. Earth and Planetary Science Letters (1993) 117:363–377.[CrossRef][Web of Science]
Hölttä P. Geochemical characteristics of granulite facies rocks in the Archean Varpaisjärvi area, Central Fennoscandian Shield. Lithos (1997) 40(1):31–53.[CrossRef][Web of Science]
Hölttä P., Paavola J. P–T–t development of Archaean granulites in Varpaisjärvi, Central Finland. I. Effects of multiple metamorphism on the reaction history of mafic rocks. Lithos (2000) 50:97–120.[CrossRef][Web of Science]
Hölttä P., Huhma H., Mänttäri I., Paavola J. P–T–t development of Archaean granulites in Varpaisjärvi, Central Finland. II. Dating of high-grade metamorphism with the U–Pb and Sm–Nd methods. Lithos (2000) 50:121–136.[CrossRef][Web of Science]
Johannes W. Beginning of melting in the granite system Qz–Or–Ab–An–H2O. Contributions to Mineralogy and Petrology (1984) 86:264–273.[CrossRef][Web of Science]
Johannes W. Melting of plagioclase–quartz-assemblages at 2 kbar water pressure. Contributions to Mineralogy and Petrology (1989) 102:270–276.
Johnson E. L. Experimentally determined limits for H2O–CO2–NaCl immiscibility in granulites. Geology (1991) 19:925–928.
Johnston A. D. Anhydrous P–T phase relations of near-primary high-alumina basalt: Implications for the origin of island arcs and tonalite–trondhjemite series rocks. Contributions to Mineralogy and Petrology (1986) 92:368–382.[CrossRef][Web of Science]
Johnston A. D., Wyllie P. J. Constraints on the origin of Archean trondhjemites based on phase relations of the Nuk Gneiss with H2O at 15 kbar. Contributions to Mineralogy and Petrology (1988) 100:35–46.[CrossRef][Web of Science]
Kemp A. I. S., Hawkesworth C. J., Paterson B. A., Kinny P. D. Episodic growth of the Gondwana supercontinent from hafnium and oxygen isotopes in zircon. Nature (2006) 439:580–582.[CrossRef][Medline]
Klein M., Stosch H.-G., Seck H. A. Partitioning of high field-strength and rare-earth elements between amphibole and quartz-dioritic to tonalitic melts: An experimental study. Chemical Geology (1997) 138:257–271.[CrossRef][Web of Science]
Klein M., Stosch H. G., Seck H. A., Shimizu N. Experimental partitioning of high field strength and rare earth elements between clinopyroxene and garnet in andesitic to tonalitic systems. Geochimica et Cosmochimica Acta (2000) 64(1):99–115.[CrossRef][Web of Science]
Klemd R., Bröcker M., Schramm J. Characterisation of amphibolite-facies fluids of Variscan eclogites from the Orlica–Snieznik dome (Sudetes, SW Poland). Chemical Geology (1995) 119:101–113.[CrossRef][Web of Science]
Knudsen T. L., Lidwin A. Magmatic CO2, brine and nitrogen inclusions in Sveconorwegian enderbitic dehydration veins and a gabbro from the Bamble sector, southern Norway. European Journal of Mineralogy (1996) 8(5):1041–1063.
Korsman K., Korja T., Pajunen M., Virransalo P. & GGT/SVEKA Working Group. The GGT/SVEKA transect: Structure and evolution of the continental crust in the Paleoproterozoic Svecofennian orogen in Finland. International Geology Review (1999) 41(4):287–333.[Web of Science]
Kretz R. Symbols of rock-forming minerals. American Mineralogist (1983) 68:277–279.[Abstract]
Lambert I. B., Heier K. S. Geochemical investigations of deep-seated rocks in the Australian shield. Lithos (1968) 1:30–53.
Leake B. E., Woolley A. R., Charles E. S., Birch W. D., Gilbert M. C., Grice J. D., Hawthorn 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. Nomenclature of amphiboles: report of the subcommittee on amphiboles of the international mineralogical association, commission on new minerals and mineral names. The Canadian Mineralogist (1997) 35:219–246.
Mänttäri I., Hölttä P. U–Pb dating of zircons and monazites from Archean granulites in Varpaisjärvi, Central Finland: Evidence for multiple metamorphism and Neoarchean terrane accretion. Precambrian Research (2002) 118:101–131.[CrossRef][Web of Science]
Mänttäri I, Sergeev SA, Paavola J, Pakkanen L, Vestin J. From conventional to ionprobe U–Pb dating of inhomogeneous zircon population: 3·2 Ga basement gneiss, Lapinlahti, Central Finland. 23rd Nordiske Vintermøde (1998) 197. Abstract Volume, Århus, 1998.
Marchildon N., Brown M. Melt segregation in late syn-tectonic anatectic migmatites: An example from the Onawa contact aureole, Maine, USA. Physics and Chemistry of the Earth Part a—Solid Earth and Geodesy (2001) 26:225–229.[CrossRef]
Mareels J. ICP-MS analysis, geochemistry and petrogenesis of granites from the Variscan northern Vosges (France). PhD thesis (2004) Katholieke Universiteit Leuven.
Markl G., Bucher K. Composition of fluids in the lower crust inferred from metamorphic salt in lower crustal rocks. Nature (1998) 391:781–783.[CrossRef]
Martin H. The mechanisms of petrogenesis of the Archean continental-crust—comparison with modern processes. Lithos (1993) 30(3–4):373–388.[CrossRef][Web of Science]
Martin H. The Archean Grey Gneisses and the genesis of continental crust. In:. Archean Crustal Evolution. Developments in Precambrian Geology—Condie KC, ed. (1994) 11:205–247.
Martin H, Smithies RH, Rapp R, Moyen J.-F, Champion D. An overview of adakite, tonalite–trondhjemite–granodiorite (TTG), and sanukitoid: Relationships and some implications for crustal evolution. (2005) Lithos 79, 1–24.
Nehring F., Jacob D. E., Barth M. G., Foley S. F. Laser-ablation ICP-MS analysis of siliceous rock glasses fused on an iridium strip heater using MgO dilution. Microchimica Acta (2008) 160:153–163.[CrossRef][Web of Science]
Newton R. C., Manning C. E. Quartz solubility in H2O–NaCl and H2O–CO2 solutions at deep crust–upper mantle pressures and temperatures: 2–15 kbar and 500–900°C. Geochimica et Cosmochimica Acta (2000) 64:2993–3005.[CrossRef][Web of Science]
Newton R. C., Aranovich L. Y., Hansen E. C., Vandenheuvel B. A. Hypersaline fluids in Precambrian deep-crustal metamorphism. Precambrian Research (1998) 91:41–63.[CrossRef][Web of Science]
Paavola J. Pre-Quaternary rocks of the Rautavaara map-sheet area (sheet 3343). In:. In: Explanation to the maps of Pre-Quaternary rocks (1999) Espoo: Geological Survey of Finland.
Paavola J. Pre-Quaternary rocks of the Vieremä map-sheet area (sheet 3342). In:. In: Explanation to the maps of Pre-Quaternary rocks (2003) Espoo: Geological Survey of Finland.
Patiño Douce A. E., Beard J. S. Dehydration-melting of biotie gneiss and quartz amphibolite from 3 to 15 kbar. Journal of Petrology (1995) 36(3):707–738.
Perchuk L. L., Safonov O. G., Gerya T. V., Fu B., Harlov D. E. Mobility of components in metasomatic transformation and partial melting of gneisses: An example from Sri Lanka. Contributions to Mineralogy and Petrology (2000) 140:212–232.[CrossRef][Web of Science]
Poutiainen M. Fluid inclusions in rocks of the Archean high-grade Varpaisjärvi–Lapinlahti area, Central Finland. Bulletin of the Geological Society of Finland (1992) 64:3–12.
Pride C., Mücke G. K. Rare earth element geochemistry of the Scourian Complex N.W. Scotland—Evidence for the granite–granulite link. Contributions to Mineralogy and Petrology (1980) 73:403–412.[CrossRef][Web of Science]
Rapp R. P. Amphibole-out phase boundary in partially melted metabasalt, its control over liquid fraction and composition, and source permeability. Journal of Geophysical Research—Solid Earth (1995) 100:15601–15610.[CrossRef]
Rapp R. P., Watson E. B., Miller C. F. Partial melting of amphibolite/eclogite and the origin of Archean trondhjemites and tonalites. Precambrian Research (1991) 51:1–25.[CrossRef][Web of Science]
Rapp R. P., Shimizu N., Norman M. D. Growth of early continental crust by partial melting of eclogite. Nature (2003) 425:605–609.[CrossRef][Medline]
Rushmer T. Partial melting of two amphibolites: Contrasting experimental results under fluid-absent conditions. Contributions to Mineralogy and Petrology (1991) 107:41–59.[CrossRef][Web of Science]
Sawyer E. W. The role of partial melting and fractional crystallization in determining discordant migmatite leucosome compositions. Journal of Petrology (1987) 28:445–473.
Sawyer E. W. Disequilibrium melting and rate of melt–residuum separation during migmatization of mafic rocks from the Grenville front, Quebec. Journal of Petrology (1991) 32:701–738.
Sawyer E. W. Formation and evolution of granite magmas during crustal reworking: The significance of diatexites. Journal of Petrology (1998) 39(6):1147–1167.[CrossRef][Web of Science]
Sawyer E. W. Melt segregation in the continental crust: Distribution and movement of melt in anatectic rocks. Journal of Metamorphic Geology (2001) 19(3):291–309.[CrossRef][Web of Science]
Sen S. K., Dunn T. Dehydration melting of a basaltic composition amphibolite at 1·5 and 2 GPa: Implications for the origin of adakites. Contributions to Mineralogy and Petrology (1994) 117:394–409.[CrossRef][Web of Science]
Shmulovich K. I., Graham C. M. An experimental study of phase equilibria in the systems H2O–CO2–CaCl2 and H2O–CO2–NaCl at high pressures and temperatures (500–800°C, 0·5–0·9 GPa): Geological and geophysical applications. Contributions to Mineralogy and Petrology (2004) 146:450–462.[CrossRef][Web of Science]
Singh J., Johannes W. Dehydration melting of tonalites. Part I. Beginning of melting. Contributions to Mineralogy and Petrology (1996) 125:16–25.[CrossRef][Web of Science]
Sisson T. W., Ratajeski K., Hankins W. B., Glazner A. F. Voluminous granitic magmas from common basaltic sources. Contributions to Mineralogy and Petrology (2005) 148:635–661.[CrossRef][Web of Science]
Solar G. S., Brown M. Petrogenesis of migmatites in Maine, USA: Possible source of peraluminous leucogranite in plutons? Journal of Petrology (2001) 42(4):789–823.
Springer W., Seck H. A. Partial fusion of basic granulites at 5 to 15 kb: Implications for the origin of TTG magmas. Contributions to Mineralogy and Petrology (1997) 127:30–45.[CrossRef][Web of Science]
Stern R. A., Percival J. A., Mortensen J. K. Geochemical evolution of the Minto block: A 2·7 Ga continental magmatic arc built on the Superior proto-craton. Precambrian Research (1994) 65:115–153.[CrossRef][Web of Science]
Svensen H., Jamtveit B., Yardley B., Engvik A. K., Austrheim H., Broman C. Lead and bromine enrichment in eclogite-facies fluids: Extreme fractionation during lower-crustal hydration. Geology (1999) 27:467–470.
Tiepolo M., Vannucci R., Oberti R., Foley S., Bottazzi P., Zanetti A. Nb and Ta incorporation and fractionation in titanian pargasite and kaersutite: Crystal-chemical constraints and implications for natural systems. Earth and Planetary Science Letters (2000) 176:185–201.[CrossRef][Web of Science]
Toivola V., Huhma H., Vaasjoki M. The diabase dykes in the Sonkajärvi–Varpaisjärvi Area, Central Finland. Geological Survey of Finland Special Paper (1991) 12:59–61.
Tomlinson K. Y., Davis D. W., Stone D., Hart T. R. U–Pb age and Nd isotopic evidence for Archean terrane development and crustal recycling in the south–central Wabigoon subprovince, Canada. Contributions to Mineralogy and Petrology (2003) 144:684–702.
Tomlinson K. Y., Stott G. M., Percival J. A., Stone D. Basement terrane correlations and crustal recycling in the western Superior Province: Nd isotopic character of granitoid and felsic volcancic rocks in the Wabigoon subprovince, N. Ontario, Canada. Precambrian Research (2004) 132:245–274.[CrossRef][Web of Science]
Touret J. L. R., Dietvorst P. Fluid inclusions in high-grade anatectic metamorphites. Journal of the Geological Society, London (1983) 140:635–649.
Touret J. L. R., Olsen S. N. Fluid inclusions in migmatites. In: Migmatites—Ashworth J. R., ed. (1985) Glasgow: Blackie. 265–286.
Trommsdorff V., Skippen G., Ulmer P. Halite and sylvite as solid inclusions in high-grade metamorphic rocks. Contributions to Mineralogy and Petrology (1985) 89:24–29.[CrossRef][Web of Science]
Van den Kerkhof A. M., Grantham G. H. Metamorphic charnockite in contact aureoles around intrusive enderbite from Natal, South Africa. Contributions to Mineralogy and Petrology (1999) 137(1–2):115–132.[CrossRef][Web of Science]
Van den Kerkhof A., Kronz A., Simon K., Scherer T. Fluid-controlled quartz recovery in granulite as revealed by cathodoluminescence and trace element analysis (Bamble sector, Norway). Contributions to Mineralogy and Petrology (2004) 146:637–652.[CrossRef][Web of Science]
Vielzeuf D., Clemens J. D., Pin C., Moinet E. Granites, granulites and crustal differentiation. In: Granulites and Crustal Evolution—Vielzeuf D., Vidal P., eds. (1990) Dordrecht: Kluwer Academic. 59–85.
Vityk M. O., Bodnar R. J. Do fluid inclusions in high grade metamorphic terranes preserve peak metamorphic density during retrograde decompression? American Mineralogist (1995) 90:641–644.
Watson E. B., Brenan J. M. Fluids in the Lithosphere. 1. Experimentally determined wetting characteristics of CO2–H2O fluids and their implications for fluid transport, host rock physical properties and fluid inclusion formation. Earth and Planetary Science Letters (1987) 85:497–515.[CrossRef][Web of Science]
Whalen J. B., Percival J. A., McNicoll V. J., Longstaffe F. J. A mainly crustal origin for tonalitic granitoid rocks, Superior Province, Canada: Implications for late Archean tectonomagmatic processes. Journal of Petrology (2002) 43(8):1551–1570.
Whalen J. B., Percival J. A., McNicoll V. J., Longstaffe F. J. Geochemical and isotopic (Nd–O) evidence bearing on the origin of late- to post-orogenic high-K granitoid rocks in the Western Superior Province: Implications for late Archean tectonomagmatic processes. Precambrian Research (2004) 132:303–326.[CrossRef][Web of Science]
White R. W., Pomroy N. E., Powell R. An in-situ metatexite–diatexite transition in upper amphibolite facies rocks from Broken Hill, Australia. Journal of Metamorphic Geology (2005) 23:579–602.[CrossRef][Web of Science]
Whitney D. L., Irving A. J. Origin of K-poor leucosomes in a metasedimentary migmatite complex by ultrametamorphism, syn-metamorphic magmatism and subsolidus processes. Lithos (1994) 32:173–192.[CrossRef][Web of Science]
Wilke M., Behrens H. The dependence of the partitioning of iron and europium between plagioclase and hydrous tonalitic melt on oxygen fugacity. Contributions to Mineralogy and Petrology (1999) 137(1–2):102–114.[CrossRef][Web of Science]
Williams M. L., Hammer S., Kopf C., Darrach M. Syntectonic generation and segregation of tonalitic melts from amphibolite dikes in the lower crust, Striding–Athabasca mylonite zone, northern Saskatchewan. Journal of Geophysical Research—Solid Earth (1995) 100:15717–15734.[CrossRef]
Wolf M. B., Wyllie P. J. Dehydration-melting of amphibolite at 10 kbar—the effects of temperature and time. Contributions to Mineralogy and Petrology (1994) 115(4):369–383.[CrossRef][Web of Science]
Xiong X.-L. Trace element evidence for growth of early continental crust by melting of rutile-bearing hydrous eclogite. Geology (2006) 34(11):945–948.
Yardley B. W. D., Graham J. T. The origins of salinity in metamorphic fluids. Geofluids (2002) 2:249–256.[CrossRef][Web of Science]
Zhang S.-B., Zheng Y.-F., Wu Y.-B., Zhao Z.-F., Gao S., Wu F.-Y. Zircon U–Pb age and Hf isotope evidence for 3·8 Ga crustal remnant and episodic reworking of Archean crust in South China. Earth and Planetary Science Letters (2006) 252:56–71.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||