J. metamorphic Geol., 2011

doi:10.1111/j.1525-1314.2011.00951.x

Eclogitization of the Monviso ophiolite (W. Alps) and implications on subduction dynamics S. ANGIBOUST,1 R. LANGDON,2 P. AGARD,1 D. WATERS2 AND C. CHOPIN3 1 ISTEP, Universite´ Paris 6-UPMC, UMR CNRS 7193, F-75005 Paris, France ([email protected]) 2 Department of Earth sciences, Oxford University, Parks road, Oxford OX1 3PR, UK 3 Laboratoire de Ge´ologie, Ecole Normale Supe´rieure, UMR CNRS 8538, F-75005, Paris, France

ABSTRACT

To constrain deep (40–100 km) subduction dynamics, extensive P–T data are provided on the eclogitic Monviso ophiolite derived from the subducted Liguro-Piemontese oceanic lithosphere (which was exhumed, together with associated continental units, before the Alpine collision). The Monviso ophiolite has so far been interpreted either as a fossilized subduction channel, with tectonic blocks detached from the slab at different depths and gathered in a weak serpentinized matrix, or as a more or less continuous portion of oceanic lithosphere. To evaluate potential heterogeneities within and between the various subunits, extensive sampling was undertaken on metasedimentary rocks and Fe–Ti metagabbros. The results indicate that the Monviso ophiolite comprises two main coherent tectonic subunits (the Monviso and Lago Superiore Units) detached during subduction at different depths and later juxtaposed at epidote–blueschist facies during exhumation along the subduction interface. Raman spectroscopy of carbonaceous material suggests (i) a difference in peak temperature of 50 C between these two subunits and (ii) a good temperature homogeneity within each subunit. Pseudosections and average P–T estimates using THERMOCALC in the Lago Superiore Unit suggest for the first time homogeneous HP to UHP conditions (550 C, 26–27 kbar). Parageneses, peak conditions and tectonic setting are very similar to those of the Zermatt-Saas ophiolite, 200 km northwards, thus suggesting a common detachment mechanism for the whole Western Alpine belt. Key words: eclogites; monviso ophiolite; subduction channel; thermobarometry;

INTRODUCTION

Understanding deep subduction zones processes (such as mechanical coupling, fluid flow, phase relations and reactions) is critical for geodynamics, geochemical budgets as well as risk assessment. High-pressure rocks returned from former subduction zones and commonly exposed within accretionary wedges (e.g. Cloos, 1982; Platt, 1986; Agard et al., 2001), eclogite facies ophiolitic nappes (e.g. Ernst, 1999; Kurz et al., 1999; Davis & Whitney, 2008) or below obducted ophiolites (e.g. Searle et al., 1994) provide invaluable information on present-day processes taking place between 40 and 100 km at the subduction interface. Great uncertainties remain, however, as to how eclogite facies rocks migrate along the interface and what controls their exhumation, either as large coherent slab portions or tectonic blocks in a me´lange (Fig. 1c,d). This problem is particularly salient in the W. Alps. The Zermatt-Saas, Monviso, Voltri and Corsican Liguro-Piemontese ophiolites (Fig. 1a), which represent well-preserved portions of the oceanic lithosphere detached from the same Tethyan slab under eclogite facies conditions during the middle Eocene (e.g. Balle`vre & Merle, 1993; Schwartz et al., 2000; Agard et al., 2001; Rubatto & Hermann, 2003) were  2011 Blackwell Publishing Ltd

THERMOCALC.

interpreted in very different ways (Fig. 1b–d). The Monviso ophiolite was interpreted either as a fossilized subduction channel, where tectonic blocks detached from the slab at different depths were chaotically juxtaposed in a weak serpentinized matrix (Schwartz et al., 2000, 2001; Guillot et al., 2004), or as an almost continuous portion of oceanic lithosphere (Lombardo et al., 1978; Castelli & Lombardo, 2007). Results suggesting that Monviso and Voltri may represent serpentinite me´langes (Blake et al., 1995; Guillot et al., 2004; Federico et al., 2007; see also Bousquet, 2008 for discussion on tectonic mixing) also contrast with recent conclusions on the Zermatt-Saas ophiolite, demonstrating the existence of a coherent, 70 km long piece of oceanic lithosphere detached at similar depths (80 km; Angiboust et al., 2009; Fig. 1c). Recent investigations on Zermatt-Saas and Corsican ophiolites also demonstrated the preservation of inherited ocean–continent-transitions, suggesting only restricted or localized tectonic mixing during exhumation and nappe-stacking (Angiboust & Agard, 2010; Beltrando et al., 2010; Vitale-Brovarone et al., 2011a). To clarify tectonic relationships between eclogite facies oceanic terranes in the Western Alps and shed light on subduction dynamics, we thus reinvestigated the tectonic and petrological setting of the Monviso 1

2 S. ANGIBOUST ET AL.

(b)

(c)

(a)

(d)

Fig. 1. (a) Spatial distribution of sedimentary (light blue) and mafic ⁄ ultramafic rocks (dark blue) in Western Alps. Zermatt-Saas and Monviso ophiolite belong to this domain, referred as Liguro-Piemontese realm. (b) Compilation of recent (since 1998) P–T estimates on rocks from the Liguro-Piemontese unit. 1. Bucher & Grapes, 2009; 2,4. Angiboust et al., 2009; 3a (Lago Di Cignana). Groppo et al., 2009; 3b. Reinecke, 1998; 5. Martin et al., 2008; 6. Bousquet, 2008; 7–10. Agard et al., 2001; 11. Messiga et al., 1999; 12. Schwartz et al., 2000; 13. Groppo & Castelli, 2010; 14. Vignaroli et al., 2005 15. Ravna et al., 2010; 16. Vitale-Brovarone et al., 2011b. (c) Sketch showing a possible mechanism for exhumation of large volumes of oceanic crust (modified after Angiboust et al., 2009). (d) Sketch illustrating exhumation processes within a serpentinized me´lange such as Voltri (e.g. Federico et al., 2007).

ophiolite. The approach is similar to that of Angiboust et al. (2009), using Raman spectroscopy of carbonaceous material (RSCM), conventional thermobarometry and pseudosection modelling to evaluate the homogeneity of peak conditions across the ophiolite. Results and implications are discussed in the light of previous work and compared with other Alpine ophiolites.

GEOLOGICAL SETTING OF THE MONVISO OPHIOLITE The Monviso ophiolite within the Alpine belt

The studied area belongs to the Liguro-Piemontese domain, which stretches along the W. Alps over 350 km, from Voltri (Italy) to Zermatt-Saas (Switzerland;  2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 3

Fig. 1a). The Liguro-Piemontese domain (comprising the Schistes Lustre´s and the Monviso ophiolite) formed by subduction and exhumation of the slow spreading Tethyan ocean below the Apulian margin during the Late Cretaceous (Lagabrielle & Cannat, 1990; Lagabrielle & Lemoine, 1997). The Schistes Lustre´s complex (Fig. 1) represents a fossil accretionary wedge, with relatively scarce blocks of mafic and ultramafic origin (including blueschist facies metagabbros) in a dominantly sedimentary-rich matrix (Lemoine et al., 1987; Deville et al., 1992; Schwartz et al., 2000) and shows a general increase in metamorphic conditions from West to East (e.g. Goffe´ & Chopin, 1986; Agard et al., 2001; Oberha¨nsli & Goffe´, 2004). By contrast, the structurally lower Monviso ophiolite abounds in ultramafic rocks equilibrated under eclogite facies conditions during the early Eocene, between 50 and 40 Ma (e.g. Ducheˆne et al., 1997; Rubatto & Hermann, 2003). The eclogitic Monviso ophiolite is separated from the blueschist facies Schistes Lustre´s accretionary wedge by a major westward-dipping ductile normal fault thought to be responsible for the metamorphic gap between both units (Balle`vre et al., 1990; Philippot, 1990). To the east, the Monviso ophiolite is in turn separated from the underlying continental eclogitic unit of Dora Maira by another ductile normal fault (Blake & Jayko, 1990). Subduction of the oceanic Liguro-Piemontese domain during the Eocene was followed by continental subduction of the thinned leading edge of the European margin. Parts of this continental margin detached from the downgoing slab at variable depths, ranging from 80 km at 43–37 Ma for Gran Paradiso or Monte Rosa massifs (e.g. Meffan-Main et al., 2004; Lapen et al., 2007; Gabudianu Radulescu et al., 2009) to 120 km at 39–33 Ma for the coesite-bearing DoraMaira massif (e.g. Chopin, 1984, 2003; Schertl et al., 1991; Tilton et al., 1991; Gebauer et al., 1997; Castelli et al., 2007). Structure of the Monviso ophiolite complex

The Monviso ophiolitic domain extends over 30 km along the French–Italian border (Fig. 2a,b). Internally, this well-preserved portion of the Tethyan ocean floor was described as a series of six tectonic slices comprising calcschists, pillow lavas, banded metabasalts, diabases, metagabbros and serpentinites separated by several west-dipping shear zones (Lombardo et al., 1978; Fig. 2b). Extensive structural and petrological field observations (confirmed by thermobarometric data, as shown below) allow simplification of the overall structure of the Monviso ophiolite into two major tectonic units: the ÔMonviso UnitÕ to the west and the ÔLago Superiore UnitÕ to the east (Fig. 2). The two westernmost slices of the ÔclassicalÕ cross-section (namely the Vallanta and Costa Ticino units in Lombardo et al., 1978; Fig. 2b) correspond to the ÔMonviso UnitÕ in this study. The  2011 Blackwell Publishing Ltd

Vallanta unit, which is a 300 m thick metabasaltic sequence with relicts of eclogite facies metamorphism, is separated from the underlying Costa Ticino unit by a tectonic contact. This latter unit corresponds to a remarkably overturned 1100 m thick sequence with, from top to bottom, Mg–Al gabbros overlying pillow basalts, banded metabasites and thin layers of metasedimentary rocks interbedded with banded metabasites (20–30 m). The Monviso Unit is in turn separated from the underlying sequence by a major tectonic contact underlined by many mylonitized, 1–50 m thick serpentinite slivers, in which some Fe–Ti eclogite blocks are dispersed (Fig. 2a). Other units (Passo Gallarino complex, Viso Mozzo unit, Lago Superiore metagabbros and the basal serpentinite; Fig. 2b) are named Lago Superiore Unit in this study. The Lago Superiore Unit corresponds to a dismembered portion of the Tethyan ocean floor which preserved a normal polarity and shows, from top to bottom: (i) calcschists and basalts in places intruded by Fe–Ti gabbros, (ii) a variably thick metagabbroic sequence cut by Fe–Ti gabbro dykes and sills, (iii) a thick serpentinite sole (400 m) made of fully serpentinized antigorite schists (Auzende et al., 2006) with rounded lenses of Fe–Ti and Mg–Al metagabbroic bodies and occasional pods of metaplagiogranite intrusions (Castelli et al., 2002; Castelli & Lombardo, 2007). Previous P–T estimates

Lombardo et al. (1978) reported that Ôall the units of the Monviso ophiolite show a similar metamorphic historyÕ, yet more recent work reported variable peak conditions across the Monviso Unit (Schwartz et al., 2000: 580 C ⁄ 20 kbar for the Lago Superiore area and 450 C ⁄ 12 kbar for the Passo Gallarino unit, with average P–T THERMOCALC estimates). These latter results are in line with Blake et al. (1995) results showing heterogeneous peak conditions (400–650 C ⁄ 12–20 kbar) using garnet-clinopyroxene ± phengite geothermobarometers. Such estimates contrast, however, with those from Messiga et al. (1999) on a magnesian gabbro sample from the lowermost part of the Lago Superiore gabbro (620 C ⁄ 24 kbar). Garnet-clinopyroxene geothermometry from a Fe–Ti metagabbro body from the basal serpentinite unit yielded slightly lower peak conditions (545 C, 20 kbar; Castelli et al., 2002). Recently, Groppo & Castelli (2010) obtained peak conditions of 550 C and 25 kbar on a lawsonite-eclogite boulder from the serpentinite sole south of the studied area (Fig. 2; sample OF2727). These differences among previous estimates of peak metamorphic conditions in part stem from the comparison of results from methods with (i) large (± 50–70 C) uncertainties, (ii) poorly constrained Fe3+ clinopyroxene content, and (iii) with a high sensitivity to the bulk-rock composition (for further discussion on garnet-clinopyroxene geothermometry: Carswell & Zhang, 1999; Ravna & Paquin, 2004).

4 S. ANGIBOUST ET AL.

(a)

(b)

Fig. 2. (a) Geological map based on our observations together with Lombardo et al. (1978) original map. RSCM sampling sites (white boxes) and studied mafic samples (black boxes) are shown. (b) Cross-section across the Monviso ophiolite showing the main tectonic units and tectonic relationships between them (modified after Lombardo et al., 1978 and Schwartz et al., 2000). Abbreviations for tectonic subunits are: VU (Vallanta Unit), CT (Costa Ticino unit), VM-LS (Viso Mozzo, Lago Superiore units), BSU (Basal Serpentinite Unit) and DM (Dora Maira).

 2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 5

By contrast, the retrograde exhumation history is well-constrained and characterized by a pervasive epidote blueschist facies recrystallization (450 C, 10–12 kbar) followed by a localized greenschist facies overprint (e.g. Lombardo et al., 1978; Blake et al., 1995). Schwartz et al. (2000) reported 390–450 C and 6–9 kbar for the blueschist- to greenschist transitional assemblage, which is widespread in the Monviso ophiolite. SAMPLING STRATEGY AND ANALYTICAL TECHNIQUES Sampling strategy

To determine if the Monviso ophiolite is a Ôme´langeÕ, where heterogeneous P–T conditions are expected, or a large coherent tectonic slice, independent methods applicable to the entire area were used: maximum temperatures were determined on metapelites using Raman Spectroscopy of Carbonaceous Matter (RSCM; Beyssac et al., 2002, 2003) and average P–T calculations and pseudosection modelling were performed on both metabasites and metapelites using THERMOCALC (Holland & Powell, 1990, 1998). To gather a sufficiently representative data set from the different localities mentioned above, 50 metasedimentary rocks and 50 metabasites samples were collected across the whole ophiolite. Petrology and chemistry of Monviso metasedimentary rocks have been investigated and their P–T history has been investigated for the first time. Analytical techniques RSCM

Raman spectroscopy of carbonaceous material (RSCM) thermometry is based on the quantitative study of the degree of graphitization of carbonaceous material, which is irreversible and depends on the maximum temperature reached during metamorphism (Beyssac et al., 2002). This method is thus convenient to determine peak conditions even in strongly retrogressed metasedimentary rocks, as is the case for most rocks in the Schistes Lustre´s (e.g. Gabalda et al., 2009) and Monviso units. Because of uncertainties on the petrological data used for the calibration, absolute temperature can be determined in the range 350– 600 C with an accuracy of ±50 C (Beyssac et al., 2002; see also Aoya et al., 2010 for further discussion on the method). However, relative uncertainties on temperature are much smaller, probably 10–15 C (Beyssac et al., 2004). A good consistency between RSCM data and THERMOCALC average P–T estimates has been recently shown on the Zermatt-Saas ophiolitic unit (Angiboust et al., 2009). Raman spectra were obtained using the Renishaw InVia micro3spectrometer in the Geology Laboratory at the Ecole Normale  2011 Blackwell Publishing Ltd

Supe´rieure in Paris following the approach described in Beyssac et al. (2002). Between 12 and 18 spectra were recorded for each sample to smooth out the within-sample structural heterogeneity. Our data set consists of 40 CM-bearing samples collected in many calcschists and metapelite lenses interleaved with metabasalts, metagabbros and serpentinites. Mineral analysis

Most analyses were performed with a Cameca SX100 electron microprobe (Camparis, Univ. Paris 6; a Cameca SX50 was also used for additional conventional analyses). Classical analytical conditions were adopted for spot analyses [15 kV, 10 nA, wavelengthdispersive spectroscopy (WDS) mode], using Fe2O3 (Fe), MnTiO3 (Mn, Ti), diopside (Mg, Si), CaF2 (F), orthoclase (Al, K), anorthite (Ca) and albite (Na) as standards. Complementary garnet and omphacite chemical analyses were performed using a JEOL JSM840A scanning electron microscope in the Department of Earth Sciences at the University of Oxford with an Oxford Instruments Isis 300 energy-dispersive analytical system. Accelerating voltage was 20 kV, with a beam current of 6 nA, and a live counting time of 100 s. It was calibrated with a range of natural and synthetic standards and a ZAF correction procedure was used. Ferric iron contents for clinopyroxene, chloritoid and garnet were estimated using the stoichiometric criteria described in Droop (1987) except for blue amphibole for which structural formulae have been recalculated according to the procedure outlined by J. Schumacher (in Leake et al., 1997). Mineral abbreviations used in this work are after Whitney & Evans (2010). Chemical maps were obtained at the Geology Laboratory of the Ecole Normale Supe´rieure (Paris) using a Zeiss Sigma field-emission-gun SEM with an X-max Oxford detector (50 mm2), an acceleration voltage of 20 kV and a count time of 240 min. Whole-rock chemical analysis

Whole-rock chemical analyses were undertaken at the SARM (CRPG Nancy) from 10 cm3 of each representative rock sample (VI17 & C32). The major elements were analysed by inductively coupled plasmaoptical emission (ICP-OES) spectroscopy after fusion with LiBO3 and dissolution in HNO3. More details on the method are given in Carignan et al. (2001). H2O content was determined by loss on ignition. PETROGRAPHY OF METASEDIMENTARY ROCKS Metasedimentary rocks from the Monviso Unit

In the Monviso Unit, calcschists and metapelites are mainly located at the eastern foot of the Monviso cliff (Fig. 2a), interleaved with a continuous quartzite layer which lies stratigraphically over foliated metabasalts.

6 S. ANGIBOUST ET AL.

The metasedimentary layers from the Monviso Unit frequently show chloritoid either as porphyroblasts in the main schistosity or as post-kinematic radiating needles growing over the main foliation. Chloritoid shows relatively low values of XMg (between 0.06 and 0.14). Between 1 and 5 wt% of the iron in chloritoid is ferric. Phengite and chlorite are abundant in the main quartz–calcite bearing foliation. Garnet is very rare (< 1 vol.%), generally small (150–200 lm), poorly zoned (Alm79Grs12Sps6Py3, on average) and never observed associated with chloritoid in metapelites from the Monviso Unit (Table 1). In quartzitic layers, a different paragenesis can be locally observed, made of blue amphibole (mostly crossite), phengite (Si4+ content up to 3.75 p.f.u.), Mn-rich garnet ± epidote ± jadeite-aegirine clinopyroxene ± hematite and occasionally chlorite (Lombardo et al., 1978; Kienast, 1983; this study). This quartzitic layer can be traced almost continuously, for more than 10 km, from Colle di Luca to the south to Monte Granero to the north (Fig. 2a).

Metasedimentary rocks from the Lago Superiore Unit

Calcschist and metapelitic lenses are mainly found in serpentinite as N–S trending bodies ranging in size from metre to hundred metre-sized boudins wrapped by the fabric of the host unit. Metasedimentary layers in the Lago Superiore Unit are strongly foliated and retrogressed calcschists with few HP minerals and have been little studied so far. Their mineralogy comprises in order of decreasing abundance Qz + Chl + Ph ± Cal ± Cld ± Grt (Table 1). Occasionally, smaller amounts of paragonite, zoisite, rutile, tourmaline and apatite have been observed in the main schistosity. In some samples (e.g. C18), some 5 mm to 1 cm long patchy aggregates of phengite, paragonite and locally epidote can be interpreted as possible pseudomorphs after lawsonite, in line with the early interpretation of Lombardo et al. (1978). Similar micaceous aggregates have not been observed in the Monviso Unit. The HP paragenesis is best preserved in sample C32. This quartz–chlorite micaschist is characterized by the

Table 1. Compilation table gathering petrological and thermometric estimates on calcschists from the Monviso ophiolite. Pseudomorphs after garnet or chloritoid are indicated by brackets. Samples are located in Fig. 2. XMg of chloritoid is given for some samples. XMg of HP-stage chloritoid relicts, with significantly high values, are italicized. Number of RSCM spectra (n), R2 ratio, average temperature (T) and standard deviation (SD) are given for all the samples. See text and Beyssac et al. (2002) for details. Unit

Paragenesis Sample

LS LS LS LS LS MV LS LS LS LS LS MV LS LS ? LS LS LS LS LS LS LS LS LS LS LS LS LS LS LS LS LS LS LS LS LS MV LS MV LS

C1 C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 Cl2 Cl3 Cl4 C15 C16 C17 Cl8 Cl9 C20 C21 C22 C23 C24 C25 C26 C27 C28 C29 C30 C31 C32 C33 C34 C35 C36 C37 C38 C39 C40

Grt

Cld

Ph

Chl

Cal

x

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

x

(x) (x)

x x x x (x) x x (x)

x x x x (x)

(x)

(x)

x x

EMPA

x x x x x x x (x) x x x

(x) x x x

x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x x

Ps.Lws

XMg of Cld

0.06–0.1 x x x 0.08–0.11

0.07–0.1 0.07–0.08

x x

0.06–0.1 0.06–0.07 0.07–0.09

x 0.06–0.12

0.08–0.1 ⁄ 0.2–0.24 x 0.16–0.2 x x 0.07–0.1

RSCM results Sample

n

R2

T

SD

Cl C2 C3 C4 C5 C6 C7 C8 C9 C10 C11 C12 C13 C14 C15 Cl6 C17 C18 C19 C20 C21 C22 C23 C24 C25 C26 C27 C28 C29 C30 C31 C32 C33 C34 C35 C36 C37 C38 C39 C40

13 13 12 14 13 13 12 12 14 12 12 17 12 13 12 14 12 12 12 18 12 18 17 12 15 12 12 13 12 15 15 13 12 12 12 12 17 14 15 12

0.21 0.25 0.27 0.22 0.21 0.36 0.19 0.24 0.22 0.23 0.29 0.35 0.22 0.23 0.28 0.27 0.24 0.17 0.16 0.23 0.21 0.24 0.26 0.26 0.19 0.19 0.22 0.23 0.24 0.26 0.26 0.26 0.24 0.25 0.21 0.23 0.34 0.24 0.33 0.26

549 530 520 544 548 480 556 536 544 541 514 484 544 531 515 520 532 565 571 538 546 535 524 526 555 555 541 531 532 524 525 524 534 531 548 540 489 534 494 523

20 13 16 18 16 21 11 17 22 19 21 14 29 19 15 17 17 15 13 17 17 30 28 27 18 23 14 12 20 20 17 13 27 24 12 13 17 17 24 23

 2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 7

presence of 200–300 lm wide garnet and chloritoid porphyroblasts growing in the main schistosity. Garnet cores (XMn = 0.13–0.14; XFe = 0.7–0.72) are enriched in Mn and depleted in Fe with respect to mantle and rim (XMn = 0.06–0.1; XFe = 0.76–0.81). As garnet outer rims are always retrogressed to chlorite, the chemical composition corresponding to the last garnet generation may have been erased. No significant zoning has been noticed for Ca (XCa = 0.07) and Mg (XMg = 0.08). Garnet is often retrogressed and replaced by rounded aggregates of chlorite wrapped by the schistosity. Interestingly, chloritoid shows two clearly distinct chemical compositions in this sample. Relicts of peak chloritoid (XMg = 0.21–0.24) are occasionally preserved in the middle of chlorite and phengite aggregates (Fig. 3a). Many smaller late patchy chloritoid crystals characterized by a significantly lower XMg (0.08–0.11) have grown either over the main schistosity or as thin rims around peak chloritoid (Fig. 3a). Chlorite is abundant along the main schistosity and texturally secondary, growing at the expense of garnet and chloritoid, and shows almost constant composition with XMg ranging from 0.42 to 0.44. The white mica is mainly phengitic in composition and shows variable silica content. The latter ranges from 3.06 to 3.2 p.f.u. in late shear bands and on the margin of crystals within the schistosity; it increases continuously towards the middle of the largest phengite crystals, where it locally reaches 3.55 to 3.65 p.f.u. Rare paragonite crystals are intergrown with phengite along the main schistosity. Calcite is scarce in this sample (< 1 vol.%). PETROGRAPHY OF METABASALTIC ROCKS Metabasalts from the Monviso Unit

Metabasalts occur as dykes cross-cutting the main gabbroic series or as pillowed basalts structurally below the overturned Monviso Unit gabbro. Pillow basalts, in which the pillows range in length from 10 to 50 cm, are located mainly in the middle-eastern part of the Monviso Unit, below the gabbroic complex. Glaucophane and tremolite, epidote, chlorite and albite typically replace the augite, plagioclase and olivine of the magmatic assemblage. The garnet-omphacite ± glaucophane ± epidote paragenesis (Lombardo et al., 1978; Blake et al., 1995) may exceptionally be preserved in areas with suited bulk-rock composition and protected from retrogression. Metabasalts from the Lago Superiore Unit

Metabasalts, which locally display brecciated and pillowed facies, were later stretched and flattened during blueschist- and greenschist facies retrogression forming the series of Rocce Alte, Viso Mozzo and Balze di Cesare, where clear stratigraphic contacts with calcschists lenses are exposed. In preserved portions of  2011 Blackwell Publishing Ltd

the unit, garnet is found dispersed in the clinopyroxene-rich matrix, underlined by a crude rutile foliation (Blake et al., 1995). PETROGRAPHY OF METAGABBROS Metagabbros from the Monviso Unit

These gabbros generally display a well-developed coarse-grained ophitic texture, with rapid grain-size variations from fine-grained layered gabbro to pegmatoid varieties (Lombardo et al., 1978; Lombardo & Pognante, 1982). Bulk-rock analyses reported in Lombardo et al. (1978) show that these metagabbros bear many chemical and textural similarities with the Lago Superiore metagabbroic unit. Igneous minerals (plagioclase, Opx, Cpx and olivine) are sometimes preserved as relicts. Thin hornblende rims occasionally occur around clinopyroxene. Garnet has not been reported so far in these Mg–Al metagabbros. Fe-rich gabbros or Fe–Ti intrusives, which have smaller grain size and ⁄ or a more appropriate composition, locally developed a very fine-grained chloritoid-bearing eclogitic paragenesis, as mentioned by Lombardo et al. (1978). Metagabbros from the Lago Superiore Unit Mg–Al metagabbros

The Mg–Al metagabbros are mainly found as 100– 1000 m long boudins on the western side of the Lago Superiore Unit, embedded within serpentinite. Mg–Al metagabbros, which locally show compositional layering, were originally identified as cumulates (Monviso, 1980; Lombardo & Pognante, 1982). Interestingly, the Lago Superiore Unit Mg–Al gabbro is much more pervasively eclogitized than the Mg–Al gabbro from the Monviso Unit, where relicts of igneous minerals are widespread. Gabbro boudins are strongly altered under greenschist facies on their margins and show abundant evidence of rodingitization. Away from highly deformed areas, the typical assemblage is composed of centimetre-sized omphacite (Di62Jd33Acm5) porphyroblasts (after magmatic augite) in a fine-grained matrix of epidote, Ca–Na amphibole and chlorite (XMg = [0.8–0.9]). Talc (XMg = 0.94), associated with glaucophane, are often included within the omphacite cores but also within omphacite fractures. Note that garnet is rare or absent from most typical low-strained Mg–Al gabbros. Retrogressed magnesian gabbros are constituted of albite, epidote, actinolite and chlorite. In mylonitized areas (Vi01), a new generation of omphacite (Di50Jd42Acm8) together with garnet (Alm41Grs31Prp27Sps1), glaucophane, phengite (Si = 3.55–3.57 p.f.u.) and quartz crystallized along the main foliation. Finely brecciated omphacitic porphyroblasts are preserved as small (100–500 lm) fragments

8 S. ANGIBOUST ET AL.

Ph

Vi56

C32

Si=3.6

Grt Omp Ib Rt

Cld I XMg= 0.21-0.24

Omp Ia Omp II Grt

Cld II XMg= 0.10

100 μm

(a)

(b)

200 μm

Vi56

1 mm

Omp II

Vi42

Omp II

Grt - Omp II inclusions

Ep+Pg

Omp II

Grt Omp II

Ep

Grt

Pg

(c)

1 cm

(d)

Vi42

Omp II

Vi52d

Grt Ep+Pg Omp I Omp II

Omp III

Omp II

50 μm

(e)

x (Acm)

0.20

I

0.15

Bright core

III

0.1

0 0.3

Outer rim

II

0.05 0.35

0.40

Grt

Dark rim

0.45

(f)

5 mm

x (Jd)

 2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 9

wrapped by the main eclogitic schistosity. Rare outcrops of eclogite facies, olivine-bearing Mg–Al metagabbro (meta-troctolite: sample Vi10) are found in a structurally lower position towards the base of the main Mg–Al-gabbro cliff above Lago Fiorenza (Philippot & Kienast, 1989). Such lithologies are also found as rare boulders east of Viso Mozzo cliff (Fig. 2). Metatroctolitic layers are characterized by the presence of coronitic garnet and Mg-chloritoid (XMg = 0.8) crystallized around omphacite and pseudomorphed plagioclase (see Kienast & Messiga, 1987; and Messiga et al., 1999, for a detailed description). Fe–Ti metagabbros

Fe–Ti metagabbros are found (i) as sills and dykes cross-cutting the Mg–Al gabbro body (Lago Superiore area) and the metavolcanic sequence (Passo Gallarino area), (ii) as dismembered bodies, 10 to 50 m in length lenses embedded by serpentinite within the main shear zones (e.g. Punta Forcion), (iii) as rare, variably sized rounded blocks in the thick serpentinite sole at the bottom of the unit. Fe–Ti metagabbro characteristic assemblages observed in the field are Grt ± Omp + Rt ± Ph ± Lws ± Gln. Retrogression often led to the development of albite-chloriteglaucophane veins and actinolite rims around omphacite. Parageneses and phase proportions observed within studied samples are given in Table 2. Fe–Ti metagabbros show variable eclogite facies metamorphic fabrics. Two contrasting end-member microstructures (and transitional cases) have been recognized, in agreement with earlier observations (Philippot & van Roermund, 1992): coarse-grained low-strained eclogites and highly strained mylonitic eclogites. Deformation within these eclogites led to the formation of several generations of omphacite (Philippot & Kienast, 1989). Three different omphacite generations with various aegirine contents occupying distinct micro-structural positions have been identified in this work. Rutile aggregates are a reliable criterion to assess the degree of mylonitization at the thin section-scale, ranging from pseudomorphing clusters

(after the igneous precursor Ti-bearing phase) to highly stretched stringers in the fabric of strongly deformed samples. We focus below on the textural and chemical description of peak assemblages observed in these Fe–Ti gabbros. DEFORMATION TEXTURES OF Fe–Ti METAGABBROS FROM THE LAGO SUPERIORE UNIT Low-strained Fe–Ti metagabbros

Coarse-grained eclogitized Fe–Ti metagabbros are found as sills or dykes within the mafic sequence (Vi14). Fe–Ti metagabbros also occur more rarely as rounded blocks in the serpentinite sole (Vi56) and occasionally in sheltered areas within more deformed areas. Sample Vi56 comes from a series of several 50–100 m wide rounded blocks embedded in the serpentinite sole. In this sample, the igneous assemblage of augite, plagioclase and Fe–Ti oxides is pseudomorphed by megacrysts (0.3–2 mm) of omphacite Ia, a fine-grained omphacite Ib (10–100 lm), and by rutile (Fig. 3b), respectively. Omphacite I generally exhibits decreasing Fe3+ from dark green cores to thin light green omphacite II rims (Di60Jd15Acm25 to Di53Jd37Acm10). Boundaries between omphacite microdomains are filled by a coronitic subidioblastic garnet (Fig. 3b) containing rare omphacite II and rutile inclusions. Garnet is weakly zoned with an average mantle composition of Alm74Grs16Py10 and an average rim composition of Alm71Grs14Py14Sps01. Phengite is extremely rare in these lithologies and has average Si content of 3.66 p.f.u. Quartz is also very small and rare in the matrix. Talc (XMg = 0.85) is found as inclusions within omphacite cores and mantles (i.e. Omp I and Omp II compositions). Lozenge-shaped 0.5–1 mm long aggregates of epidote (Ps20) and paragonite are fairly rare in this sample and classically interpreted as lawsonite pseudomorphs (Fig. 3c). Although these pseudomorphs are locally frequent, no fresh lawsonite has been observed so far. However, its exceptional preservation shielded within garnet and omphacite has

Fig. 3. (a) Photomicrograph of a zoned chloritoid aggregate in sample C32 (crossed polars). (b) Photomicrograph in plane polarized light of a typical low-strain Fe–Ti metagabbro (Vi56) with coronitic growth of garnet between omphacite Ia microdomain (after augite) and omphacite Ib (after plagioclase). (c) Back-scattered SEM image showing a euhedral lawsonite pseudomorph in an unretrogressed eclogite facies low-strain sample (Vi56). (d) Close-up outcrop view of sample Vi42 showing relatively large garnet porphyroblasts and small lawsonite crystals (now pseudomorphed by epidote and paragonite) in an omphacite-rich matrix. (e) BSE SEM image illustrating the zoning pattern occasionally preserved in Vi42 matrix. A plot of clinopyroxene compositions shows that omphacite I cores roughly have similar compositions to rim compositions (Omp III). Omphacite II, depleted in ferric iron, is growing over omphacite I cores. (f) Polished-sample view of sample Vi52d showing lawsonite pseudomorphs dispersed in the omphacite-rich domain of a banded mylonitized eclogite. (g) FEG-SEM quantified Ca map of mylonite Vi17 showing embayments in garnet rimmed by a Ca-poor composition (lower square). Note (upper square) that garnet III composition is healing garnet blasts. A compositional zoning profile showing the three different garnet generations is also shown (proportions of Grt end-members normalized to 1). (h) BSE-image of omphacite zoning showing cores depleted in ferric iron (Omp II) and sharp rims and fracture edges enriched in ferric iron (Omp III; sample Vi17). (i) FEG-SEM Ca map (counts) showing an aggregate of epidote and paragonite (lawsonite pseudomorph) growing in an eclogite facies mylonite (Vi52d). (j) BSE-image of a small lawsonite pseudomorph inclusion within a garnet core from a mylonite eclogite from Lower Shear Zone (Vi21). (k) SE image of sample Vi05 showing radial cracks around quartz inclusion in garnet (Lower Shear Zone). (l) Thin-section view of a typical mylonite eclogite foliation cross-cut by a vein mainly filled by epidote and paragonite, possibly after lawsonite (sample Vi21, Lower Shear Zone).  2011 Blackwell Publishing Ltd

10 S. ANGIBOUST ET AL.

13

Vi17 Rt

Vi17

Grt

n

CaO weight %

10

Omp II

=

Di43 Jd42 Acm15

20

5

Omp II

Omp III

Di45 Jd26Acm29

500 μm

(g) 0.6

Grt I Alm

II

III

0.4

Grs

0.2

Sps

0

Prp

0

II

III

Fractured garnet

100 μm

(h)

Vi52d

Vi21

Grt Omp II Pg

Grt

Omp II

Pg

Omp II

Ep

Ep Omp II Grt

300 μm

(i)

Omp

Vi05

Omp

Radial cracks

Qtz

100 μm

(j)

Ep + Pg (Lws)

Vi21 Mylonite foliation

Omp

Grt (k)

100 μm

(l)

1 cm

Fig. 3. (Continued)  2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 11

Table 2. Summary of eclogite facies assemblages and relative abundances observed in Monviso ophiolite metagabbros. Sample Fe–Ti gabbro VI05 VI14 VI17 VI21 VI42 VI52d VI56 Mg–Al gabbro VI01 (mylonite) VI10 (metatroctolite)

Grt

Cpx

Ps Lws

Ph

++ ++ ++ ++ + ++ +

+++ +++ +++ ++ +++ ++ +++

+

+ ) ) ) +

++ ++

++ ++

) + + + + +

Gln

+ ) )

)

)

) )

)

Rt

Qz

+ + + + + + +

) ) ) ) ) ) )

+ )

)

Tlc

Pg

Cld

Ep

Chl

) )

?

) ) ++

) )

+

+ ?

+++: over 70 vol.%; ++: 25 to 70 vol.%; +: 1 to 25 vol.%; ): less than 1 vol.%.

been recently demonstrated by Groppo & Castelli (2010) in a low strained Fe–Ti metagabbro sample from the serpentinite sole, south of the studied area. Fe–Ti metagabbros constituting the top of the mafic sequence of Lago Superiore Unit (Viso Mozzo, Passo Gallarino: sample Vi14) are characterized by the abundance of glaucophane both in the matrix and as inclusion within garnet and omphacite, suggesting glaucophane stability under peak conditions. As already noticed by Schwartz et al. (2000), garnet and omphacite are respectively characterized by high almandine (65–75 mol.%) and high aegirine contents (20–30 mol.%). Omphacite II, showing the lowest ferric iron contents (Di55Jd25Acm20), occurs around omphacite I porphyroblasts (Di53Jd20Acm27) or as needles in strained domains. Phengite is frequently found included within rutile and garnet. Highest Si content of phengite ranges between 3.6 and 3.7 p.f.u. (Fig. 4). Small pseudomorphs after lawsonite are rare and scattered along the foliation. In strongly oxidized rock types, epidote occurs randomly dispersed along the foliation in textural equilibrium with omphacite and is apparently stable under peak conditions instead of lawsonite. Balle`vre et al. (2003), Warren & Waters (2006) and Rebay et al. (2010) have shown that epidote 0.9

Peak Ph

0.8

Fe2+ + Mg

0.7

Ph in Lws pseudomorphs

0.6 Samples VI05 VI14 VI17 VI21 VI42 VI56

0.5 0.4 0.3 0.2 3.2

3.3

3.4 3.5 3.6 Si4+ content of phengite

3.7

3.8

Fig. 4. Chemical composition of phengite in Fe–Ti metagabbro. Note that retrograde phengite growing during breakdown of lawsonite pseudomorph has much lower Si4+ content than peak phengite.  2011 Blackwell Publishing Ltd

may occasionally be stable in the lawsonite domain for relatively high oxygen fugacities. Note that epidote within pseudomorphs after lawsonite generally has lower pistachite content (Ps10–Ps20) than late blueschist facies epidote growing as patches on the matrix foliation (Ps30–Ps40) Intermediately strained Fe–Ti metagabbros

Intermediately strained Fe–Ti metagabbros occur either on the outer rim of blocks in the serpentinite sole or within Fe–Ti metagabbros within the main mafic sequence. Sample Vi42 represents an intermediate sample because the original magmatic fabric is not completely blurred by the mylonitic foliation. Subidioblastic garnet (100–900 lm) is randomly scattered throughout the sample. Locally, garnet reaches several millimetres in width (Fig. 3d). Mimetically replaced augite with low-strain-type deformation features is rare (< 10 vol.%), whereas microcrystalline omphacite Ib after plagioclase is absent. Dynamic recrystallization led to the formation of a planar fabric orientation marked by omphacite II. Intracrystalline strain is recorded by undulose extinction within the rare omphacite I porphyroblasts. Interestingly, an early generation of aegirine-rich omphacite I is preserved in some omphacite cores (Fig. 3e). A dark omphacite II rim surrounding the core shows a strong depletion in aegirine. The outer crystal rim is again richer in aegirine component (omphacite III). Large (0.5–3 mm) phengite laths characterized by extremely high Si content (3.6–3.7 p.f.u.) occur aligned within a crude eclogitic foliation and contain inclusions of fine-grained omphacite and garnet. Flattened aggregates of former lawsonite crystals (now pseudomorphed by epidote and paragonite) are texturally in equilibrium with phengite laths, rutile, garnet, omphacite and quartz. Some rare (< 0.5 vol.%) and small (20–40 lm) glaucophane needles seem also in textural equilibrium with the peak assemblage. Mylonitic Fe–Ti metagabbros

Highly strained samples (Vi05, Vi17, Vi21 and Vi52d) are mostly disseminated within the main shear zones

12 S. ANGIBOUST ET AL.

strained eclogite blocks scattered along the Lower Shear Zone (Fig. 2) in a serpentinized matrix. In these samples (sample Vi05), garnet porphyroblasts (100– 500 lm) coexist with small (5–20 lm) acicular omphacite crystals together with phengite, Mg-chlorite, quartz and lawsonite pseudomorphs (see also Lombardo et al., 1978; Kienast, 1983; Blake et al., 1995; Cliff et al., 1998). THERMOBAROMETRY

Three thermobarometric methods were applied to a set of over 60 samples to obtain widespread P–T constraints in the different units and on different lithologies. Average P–T calculations using THERMOCALC have been performed on eight samples (seven Fe–Ti gabbros and one quartzite). Two sets of pseudosections using THERMOCALC have been plotted for (i) a mylonitic eclogite facies Fe–Ti gabbro and (ii) a metapelitic sample from the Lago Superiore Unit. These results are then compared with 40 RSCM maximum temperatures from both Monviso and Lago Superiore Units. RSCM

All RSCM temperatures obtained are located on the geological map in Fig. 2. Detailed calculation (R2, number of spectra, temperature and standard deviation) is presented in Table 1. Maximum temperatures have been estimated on 40 samples. Other samples either did not contain organic matter (mostly quartzites) or organic matter was not preserved, possibly because of oxidizing conditions. At a first glance, two sets of temperatures can be identified from Fig. 5. Samples from the sedimentary layers at the bottom of the Monviso Unit show average temperatures ranging from 480 to 500 C. Metasedimentary lenses interleaved with metabasalts,

W Upper shear zone

Lower shear zone

E

600

550 Serpentinite sole

Max temp. (°C)

identified in the field (Fig. 2a). These eclogite facies mylonites have their relict igneous fabric completely obliterated by 0.5–10 mm-wide mono-mineralic alternating fine-grained bands of omphacite II and garnet (Vi52d, Fig. 3f). Sample Vi17 (Lago Superiore area) is representative of these eclogite facies mylonites. In this sample, garnet is commonly strongly fractured and shows numerous atoll structures mainly filled by omphacite. Garnet cores are relatively enriched (Alm55Grs33Sps07Prp05) in Mn and Ca and depleted in Mg and Fe when compared with rim compositions (Alm67Prp19Grs13Sps01). Interestingly, three chemically different garnet generations have been identified on chemical maps and microprobe transects (Fig. 3g). Whereas boundaries between garnet II and garnet III show plain examples of embayments possibly because of corrosion, fractures and garnet blasts are healed by a later Ca-poor garnet III composition (Fig. 3g). In this sample, omphacite shows a clear ÔreverseÕ zonation (Fig. 3h) characterized by cores (Omp: Di43Jd42Acm15), dark in back-scattered electron images, mantled by thin rims (10–20 lm) enriched in Fe3+ and depleted in jadeite content (Omp III: Di45Jd26Acm29). Omphacite crystals included within lawsonite pseudomorphs or garnet mantles have a composition similar to dark cores (i.e. Omp II). Strong dynamic recrystallization of omphacite led to the crystallization of elongated (100–150 lm) omphacite II crystals. Somewhat coarser omphacite II grains are found in strain shadows around garnet. Phengite occurs as rare elongated (50–100 lm long) crystals along the eclogitic foliation. The Si content of peak phengite ranges from 3.55 to 3.66 p.f.u., while phengite within pseudomorphs after lawsonite has significantly lower values, from 3.34 to 3.42 p.f.u., suggesting late-stage breakdown of lawsonite at lower pressures (Fig. 4). Lawsonite pseudomorphs scattered along the main foliation occur either as milky several-millimetre-thick layers or as 1–5 mm patchy euhedral flakes of clinozoisite, white mica ± chlorite ± quartz (sample Vi21, sample Vi52d: Fig. 3f). Garnet and omphacite II are commonly included within pseudomorphs after lawsonite (Fig. 3i), while pseudomorphs after lawsonite are occasionally found included within garnet cores (Fig. 3j). The internal omphacitic fabric is often deflected around lawsonite-porphyroblast pseudomorphs. On the rim of some mylonitic blocks dispersed along the main shear zones, an extremely fine-grained (< 50 lm) mixture of epidote and paragonite fills veins (2–8 mm wide) cutting across the main mylonitic foliation (Vi21: Fig. 3l). Such veins already mentioned in the Zermatt-Saas ophiolite, 200 km northwards (Fry & Barnicoat, 1987; Angiboust & Agard, 2010), are interpreted to be derived from lawsonite. Rare (< 0.5 vol.%) and small (< 50 lm) glaucophane needles are dispersed in the matrix, in particular within lawsonite-rich domains. Occasionally, chlorite (XMg = [0.7–0.8]) is found stable together with the eclogite facies assemblage in the rim of some extremely

500

450

Monviso unit

Lago Superiore unit

Relative distance to lower shear zone

Fig. 5. Projection of RSCM maximum temperatures and associated errors for the calcschist samples. Horizontal position has been renormalized fixing western and eastern bounds of Lago Superiore Unit, materialized on the field by two shear zones (Fig. 2a). Two main temperature ranges can be identified (grey horizontal rectangles). The dashed circle shows T-max estimates from the lowermost calcschist body from Lower Shear Zone, where slightly higher temperatures were recorded.  2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 13

metagabbros and serpentinites in the Lago Superiore Unit display higher temperatures, generally falling between 520 and 555 C. Slightly higher peak temperatures (555–570 C) are also reported in a 50–100 m thick calcschist lens structurally below the main Lago Superiore Mg–Al-gabbro body (samples C18-C19, C25-C26; Fig. 2). Note that this layer contains some of the largest (up to 1 cm) porphyroblasts after chloritoid or lawsonite, generally pseudomorphed by white mica, chlorite or calcite. Internal dispersion of RSCM temperatures is generally unimodal and ranges from 11 to 20 C. Some samples display slightly higher dispersions (20–30 C) possibly in response to a greater concentration of structural defects caused by pervasive shear deformation (see Aoya et al., 2010 for further discussion). AVERAGE P–T THERMOCALC

Some of the studied rocks (Vi10, coronitic Mg–Al metagabbro; Vi56, low strain Fe–Ti metagabbro) show numerous disequilibrium textures, preventing reliable estimates of the bulk rock composition for pseudosection modelling (Rebay & Messiga, 2007). In such samples, it has been preferred to perform average P–T thermobarometric calculations using textural relationships between the different phases as a guide. The computer software THERMOCALC v.3.33 using an updated version (2003) of the internally consistent Holland & Powell (1998) data set was used in Ôaverage P–TÕ mode (Powell & Holland, 1994) to evaluate metamorphic conditions of peak burial of the Monviso ophiolite and to compare these results with previous works and with the other methods in use herein. Activities of different end-members were calculated using the software AX (Holland & Powell, 1998) running at 550 C and 25 kbar. Average P–T results on Lago Superiore Unit metagabbros are presented in Table 3. Some representative mineral compositions used for average P–T estimates are presented in Table 4. For each sample, several calculations were made using coexisting mineral compositions in different local domains considered to represent equilibrium associations at peak conditions. Calculated average P–T conditions for peak equilibration of these rocks are fairly homogeneous and span a range between 535 and 580 C and 25–28 kbar, for a unit water activity (Table 3). When decreasing water activity [a(H2O) = 0.8], temperature is generally lowered by 30 C while effect on pressure is minimal. Water activity was taken as unity in the calculations, but it can be substantially decreased in the case of an impure aqueous fluid, which in turn may affect the equilibrium temperature of the reactions (e.g. Selverstone et al., 1992; Scambelluri & Philippot, 2001; see also previous average P–T estimates). In Lago Superiore area, fluid inclusions trapped within clinopyroxene from an eclogite facies shear zone show moderate salinities (5–15% NaCl: Philippot & Selverstone, 1991).  2011 Blackwell Publishing Ltd

Given that carbonates are absent from these lithologies, dissolved halite is the substance most likely to decrease the water activity, and so the effect of decreasing water activity on average P–T estimates (Table 3) and pseudosections has been evaluated (see below). For average P–T THERMOCALC estimates, a value of a(H2O) = 0.8 has been chosen, following the approach described in Warren & Waters (2006) for fluid inclusions with comparable salinities (Table 3). Preliminary average P–T THERMOCALC calculations have been performed on a quartzite sampled in the northern Visolotto cliff (Fig. 2a). Peak equilibrium conditions of 487 ± 62 C and 21.3 ± 2.6 kbar were obtained on the assemblage Gln-Cpx-Ph-Grt-Qz-H2O [a(H2O) = 1, cor = )0.26, sigfit = 1]. Similar conditions, but with larger error bars on P–T, were derived from chemical analyses given in Kienast (1983) for the same quartzite level southwards (near Rifugio Sella). Large uncertainties on available riebeckite thermodynamic data are probably responsible for the large errors in temperature (±50–100 C) for these strongly oxidized lithologies. Thermodynamic modelling

A P–T pseudosection in the NCKFMASHTO system, using the latest version (v.3.33, 19 ⁄ 10 ⁄ 2009) of the software THERMOCALC (Holland & Powell, 1998), has been computed to evaluate more precisely the P–T conditions of peak metamorphism in Lago Superiore eclogites. Sample Vi17 is an eclogite facies mylonite almost completely recrystallized under peak conditions and the prograde part of the P–T path is only preserved in garnet-core fragments sealed or overgrown by a peak garnet generation. Phases considered in the modelling and references to the activity models used are clinopyroxene (Green et al., 2007), garnet (White et al., 2007), phengite and paragonite (Coggon & Holland, 2002), clinoamphibole (Diener et al., 2007), chlorite (Holland et al., 1998), epidote (Holland & Powell, 1998) and K-feldspar (Holland & Powell, 2003). Another P–T pseudosection has been computed in the KFMASH system on metapelitic sample (C32) from the Lago Superiore Unit. A bundle of activity models designed for metapelitic lithologies was downloaded from the THERMOCALC home website: Ôhttp://www.metamorph.geo.uni-mainz.de/thermocalc/ datafiles/index.htmlÕ (April 2010 version). The following activity models were used: biotite (White et al., 2007), chlorite, muscovite, garnet, chloritoid and staurolite (Holland & Powell, 1998). An additional pseudosection in the MnKFMASH system was calculated for sample C32, with the addition of 0.15 wt% MnO to the bulk composition to assess the effect of Mn on system topology close to peak conditions (between 16 and 28 kbar). Activity models chosen for this calculation are: chlorite (after Mahar et al., 1997 and Holland et al., 1998), garnet (White et al., 2005), muscovite (Holland & Powell, 1998) and chloritoid

14 S. ANGIBOUST ET AL.

Table 3. Summary of selected assemblages and average P–T Sample Location Protolith Strain

VI05 Punta Forcione Fe–Ti gabbro High

VI10 Lago Superiore Mg–Al gabbro (troctolite) Low

Garnet Prp Grs Alm Omphacite Di Hd Jd Acm Phengite Ms Cel Fcel Pg Others

af 126–144 – rim 0.0216 0.0094 0.18 af64 – rim 0.43 – 0.44 0.10 ag15 – core 0.30 0.27 0.036 0.139 Chl af94 – core clc 0.15 dph 0.00086

v119 – rim 0.057 0.099 0.033 v114 – inc. Grt 0.62 0.03 0.35 – v149 – core 0.49 0.17 0.023

Qz Lws Representative calculation aH2O 1 T (C) 560 SD (T) 17 P (kbar) 25.8 SD (P) 1.3 correl. 0.56 sigfit 1.51 Numb. Ind. Rea 6 end member eliminated – Average result (aH2O=1) T (C) ⁄ DT 557 SD (T) ⁄ D SD 17.8 P (kbar) ⁄ D P 25.8 SD (P) ⁄ D SD 1.4 correl. ⁄ D corr. 0.55 sigfit ⁄ D sigfit 1.63 Number of calc. 6

Tic v132 – core Tlc 0.77 Mns 0.000017 Ctd v124 – core Mctd 0.75 Fctd 0.28

H2 O 0.8 530 14 25.5 1.2 0.55 1.41 6

H2O

THERMOCALC

estimates on metagabbros from the Lago Superiore Unit. VI21 Alpetto Fe–Ti gabbro High

VI17 Lago Superiore Fe–Ti gabbro High

VI14 Passo Gallarino Fe–Ti gabbro Low

VI42 Ghincia Pastour Fe–Ti gabbro Intermediate

VI56 Colle della Gianna Fe–Ti gabbro Low

af169–113 – rim 0.00134 0.0047 0.28 af44 – rim 0.44 0.16 0.22 0.19 af39 – core 0.21 0.34 0.14 – Gin af53 – core Gln 0.106 Fgln 0.046 Rbk 0.0117

ae45–9 – rim 0.0149 0.00196 0.27 r72 – inc. Lws ps 0.35 0.085 0.44 0.097 r94 – core 0.25 0.30 0.14 0.1114 –

af106 – rim 0.0233 0.021 0.12 af128-9 – rim 0.57 0.10 0.23 0.087 af111 – core 0.19 0.40 0.094 0.095 –

r82 – rim 0.03 0.0052 0.17 ae133 – rim 0.39 0.073 0.38 0.14 ae138 – core 0.24 0.34 0.104 0.1266 –

u135–199 – rim 0.0061 0.0038 0.30 u114 – rim 0.40 0.11 0.38 0.097 u115 – core 0.21 0.36 0.11 0.1058 Tic u104 – core Tlc 0.6 Mns 0.0026

Qz Lws H2O

Qz Lws H2O

Qz Lws H2O

Qz Lws H2O

Qz Lws H2O

1 546 34 27.8 2.4 0.898 1.28 5

1 546 25 26 1.8 0.899 0.98 5

0.8 518 25 25.7 1.8 0.897 1.08 5

1 553 26 26.7 1.8 0.894 0.67 5

0.8 524 24 27.2 1.7 0.892 0.91 5

1 546 20 27.3 1.6 0.81 1.17 7

0.8 518 20 27 1.6 0.805 1.25 7

–

–

–

–

–

–

550.5 33.3 27.1 2.4 0.90 1.22 6

8.7 5.4 1.1 0.4 0.0 0.3

546 35.0 27.2 2.4 0.89 1.34 9

13.1 5.8 1.1 0.4 0.00 0.34

547 21.6 27.0 1.7 0.81 1.28 8

3.9 2.1 0.8 0.2 0.01 0.10

1 580 17 26.6 2.4 0.34 0.72 7

0.8 561 16 26.4 2.4 0.332 0.74 7

0.6 538 15 26.1 2.3 0.322 0.85 7

1 527 16 26.3 1.4 0.707 1.06 7

0.8 503 15 26.4 1.3 0.703 1.08 7

–

–

–

–

Fgl

Fgl

6.4 1.5 0.9 0.1 0.0 0.0

581 17.3 26.4 2.5 0.35 0.85 6

3.06 0.82 0.75 0.05 0.01 0.14

535 23.57 27.1 2.06 0.70 1.56 7

19.54 5.38 0.95 0.41 0.03 0.34

(after Mahar et al., 1997 and Holland & Powell, 1998). For carpholite, for which no activity model is yet available, only the Mg-bearing end-member was taken into account. Bulk composition and assumptions

Determining the equilibrium volume during metamorphism requires considering caveats such as element mobility, compositional zoning and pseudomorphed minerals that may have been stable at peak assemblage (Marmo et al., 2002; Powell & Holland, 2008). To get a bulk composition representative of the actual equilibrium volume of our samples, EMP mineral analyses were combined with phase proportions determined by SEM. For each sample, relative proportions of phases were derived from the average of four high-resolution EDS X-ray chemical maps (512 · 384 pixels, or 11.76 · 2.2 mm) using a Matlab reconnaissance programme based on successive filters on elemental concentrations. This method is believed to produce more robust estimates than a whole-rock chemical analysis, which averages cm-scale heterogeneities.

0.8 516 35 27.4 2.5 0.896 1.42 5

– 539 29.9 27.2 2.1 0.90 1.21 7

7.4 4.4 0.4 0.3 0.00 0.15

Calculated mineral modes for metapelitic sample C32 are Ph (42) Qz (33.7) Chl (10.7) Grt (7.1) Pg (3.5) Cld (3). These proportions may have changed significantly during the retrograde evolution of the rock, however, making the estimation of peak phase proportions difficult. To draw a pseudosection in a simple chemical system representative of most of Monviso and Schistes Lustre´s metasedimentary rocks, the small amounts of CaO and Na2O were neglected and the FEG-SEM composition was normalized into the (Mn)KFMASH system (Table 5). Ferric iron has also been neglected as silicate minerals in HP metapelites usually do not incorporate large amounts of Fe3+ (e.g. Diener & Powell, 2010). Moreover, given the strong degree of retrogression of these rocks, subject to pervasive retrograde fluid flow, it is impossible to evaluate accurately the amount of ferric iron present under peak conditions for calculation purposes. Incidentally, this model composition is in fact not so different from the whole-rock analysis given for comparison in Table 5. Further tests on the pseudosection topology also revealed no significant difference with that presented in Fig. 8.  2011 Blackwell Publishing Ltd

 2011 Blackwell Publishing Ltd

VI14 af44

54.76 0.02 5.26 11.97 0.01 7.56 14.33 6.05 0.00 99.96

6 2.00 0.00 0.23 0.17 0.20 0.00 0.41 0.56 0.43 0.00 0.23 0.20

0.53

VI10 v114

55.80 0.06 8.59 1.03 0.00 12.19 17.49 4.33 0.00 99.47

6 1.99 0.00 0.36 0.03 0.00 0.00 0.65 0.67 0.30 0.00 0.31 0.00

0.96

Sample Ref.

SiO2 TiO2 AI2O3 FeOT MnO MgO CaO Na2O K2 O Sum

O Si Ti Al Fe2+ Fe3+ Mn Mg Ca Na K XJd XAcm XPrp XAlm XMg

Mineral

0.67

6 2.01 0.00 0.45 0.09 0.09 0.00 0.36 0.42 0.57 0.00 0.47 0.10

56.59 0.00 10.71 6.14 0.04 6.79 11.05 8.29 0.02 99.63

VI17 r72

0.75

6 1.98 0.00 0.26 0.10 0.09 0.00 0.57 0.69 0.31 0.00 0.23 0.08

55.09 0.05 6.07 6.23 0.05 10.64 17.93 4.45 0.00 100.51

VI21 af128–9

Omphacite

0.64

6 2.00 0.00 0.38 0.08 0.15 0.00 0.41 0.46 0.53 0.00 0.38 0.15

56.31 0.00 9.22 7.54 0.00 7.69 11.99 7.69 0.01 100.46

VI42 ae133

0.65

6 2.01 0.00 0.38 0.12 0.10 0.00 0.40 0.49 0.50 0.00 0.40 0.10

56.02 0.00 9.03 7.27 0.03 7.53 12.69 7.17 0.01 99.75

VI56 u114

12 3.02 0.00 1.90 2.16 0.00 0.06 0.24 0.51 0.09 0.00

0.08 0.72 0.10

0.21 0.60 0.26

38.38 0.07 19.98 31.95 0.91 2.03 5.93 0.59 0.00 99.84

V14 af169–113

12 2.95 0.00 2 1.83 0.04 0.05 0.63 0.49 0.00 0.00

37.60 0.06 21.68 28.47 0.79 5.39 5.79 0.00 0.00 99.79

VI05 af126–144

0.20 0.69 0.23

12 3.00 0.00 2.00 2.05 0.00 0.04 0.60 0.29 0.01 0.02

37.86 0.00 21.49 30.94 0.60 5.05 3.40 0.05 0.20 99.60

V17 ae45–9

Garnet

0.18 0.58 0.24

12 2.99 0.00 2.00 1.75 0.01 0.05 0.55 0.65 0.00 0.00

37.92 0.02 21.47 26.67 0.70 4.70 7.68 0.00 0.00 99.16

VI21 af106

0.25 0.59 0.29

12 2.96 0.00 1.98 1.81 0.05 0.05 0.75 0.39 0.01 0.00

38.06 0.00 21.60 28.60 0.77 6.47 4.67 0.05 0.02 100.24

VI42 r82

0.14 0.71 0.17

12 3.01 0.00 2.04 2.08 0.00 0.05 0.41 0.39 0.01 0.00

37.91 0.00 21.75 31.35 0.72 3.49 4.63 0.04 0.02 99.89

VI56 u135–199

Table 4. Representative EMPA mineral compositions used for average P–T estimations of the eclogite facies stage.

0.70

11 3.68 0.01 1.58 0.23 0.00 0.00 0.54 0.00 0.00 0.96

53.75 0.16 19.53 4.01 0.00 5.28 0.03 0.00 11.01 93.76

VI14 af39

0.68

11 3.64 0.00 1.66 0.23 0.00 0.00 0.51 0.00 0.01 0.96

53.71 0.00 20.80 4.10 0.04 5.01 0.00 0.06 11.05 94.77

VI17 r94

0.81

11 3.70 0.01 1.54 0.15 0.00 0.00 0.64 0.00 0.01 0.94

55.03 0.17 19.35 2.67 0.05 6.42 0.02 0.06 10.98 94.73

VI21 af111

Phengite

0.77

11 3.66 0 1.67 0.16 0.00 0.00 0.53 0.00 0.01 0.94

53.92 0.00 20.91 2.84 0.02 5.26 0.03 0.07 10.83 93.88

VI42 ae138

0.75

11 3.68 0.00 1.60 0.19 0.00 0.00 0.57 0.01 0.01 0.95

54.67 0.00 20.20 3.35 0.00 5.70 0.09 0.06 11.13 95.19

VI56 u115

0.59

23 8.01 0.01 1.46 1.21 0.68 0.00 1.75 0.06 1.94 0.00

55.80 0.06 8.77 16.01 0.02 8.29 0.41 7.07 0.02 96.45

VI14 af53

Gln

0.86

22 8.05 0.00 0.01 0.83 0.00 0.00 5.04 0.01 0.01 0.00

60.58 0.00 0.10 7.46 0.00 25.43 0.10 0.03 0.00 93.69

VI56 u104

Tlc

0.70

14 2.83 0.00 2.02 1.58 0.00 0.01 3.71 0.00 0.00 0.00

27.87 0.00 16.84 18.58 0.13 24.47 0.04 0.00 0.01 87.94

VI05 af94

Chl

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 15

16 S. ANGIBOUST ET AL.

Table 5. Bulk compositions (in molar proportions) normalized to 100% for sample C32 (calcschist, Lago Superiore Unit) and sample Vi17 (Fe–Ti mylonitized metagabbro, Lago Superiore Unit) derived from surface analysis calculation using a FEG (C32) or by a combination of calculated modal proportions and EMPA analyses (Vi17). Metapelite C32

SiO2 TiO2 A12O3 FeOT MnO MgO CaO Na2O K2O O

Fe–Ti metagabbro VI17

FEG

EMP-SEM

ICP-OES

EMP-SEM

ICP-OES

bulk

bulk

bulk

minus Grt cores

bulk

72.63 0.38 12.58 6.89 0.15 4.16 0.12 0.24 2.85 –

72.06 0.00 12.81 6.97 0.46 4.12 0.24 0.52 2.82 –

68.68 0.83 14.32 6.80 0.19 4.78 0.29 1.25 2.85 –

52.02 2.89 8.20 9.00 0.00 8.98 11.66 5.67 0.18 1.40

53.59 4.12 8.57 10.36 0.27 8.41 8.95 5.16 0.57 –

To estimate the actual equilibrium volume of sample VI17 (Fe–Ti metagabbro), garnet cores presumably isolated from the bulk chemical system during peak burial conditions were removed (i.e., areas characterized by a spessartine content higher than 5 mol.%; Fig. 3g, Table 5): garnet rim composition (Alm68Grs17Prp15: average of garnet II and III compositions in sample VI17, Fig. 3g) is indeed more representative of the equilibrium volume than the whole garnet. Other phase compositions chosen for calculation are: Di45Jd43Acm12 for dynamically recrystallized omphacite core (Lardeaux et al., 1987; Fig. 6), Ms65Cel23Fcel11Pa1 for muscovite (average phengite cores) and Ps16 for epidote (average epidote cores). Although lawsonite was arguably present in the peak assemblage, epidote core compositions were chosen instead of lawsonite because fresh lawsonite is lacking in mylonitic Fe–Ti metagabbros. This approximation is supported by the fact that epidote is by far the major phase crystallizing after lawsonite breakdown (over 90 vol.%). Modal volumes (minus garnet cores) derived from EMP-SEM estimates, which were used for calculation of peak bulk-rock equilibrium composition are: Omp (72.6) Grt (15.4) Ep (6.5) Rt (3.1) Ph (2.8). The amount of ferric iron in the rock was estimated from the Fe3+ content of omphacite and epidote (Droop, 1987). As a mean of comparison, the whole-rock analysis of sample VI17 is given in Table 5. Water is considered to be in excess as suggested by (i) the presence of numerous pseudomorphs after lawsonite (12 wt% H2O) aligned along the main eclogite facies mylonitic foliation, (ii) the abundance of atoll-shaped garnet whose formation may be facilitated by the presence of an aqueous phase (e.g. Cheng et al., 2007), (iii) the healing of garnet fractures, which is also strongly facilitated by a fluid phase under HP-LT conditions, and (iv) the presence of small lamellae of phengite (5 wt% H2O) along the main mylonite foliation. The latter point also suggests an external

source of K because Fe–Ti gabbros generally have extremely low K contents (< 0.2 wt%; Lombardo et al., 1978; Schwartz et al., 2000). Results of thermodynamic modelling

The potential advantage of pseudosection calculation over average P–T determination of peak conditions is to locate the stability field of the observed or inferred peak assemblage, whereas an average P–T result based on phase compositions alone may well fall outside this field. Within this stability field, the best-fit condition may be located from the intersection, within reasonable uncertainty, of composition isopleths. The accuracy and precision of the method rest largely on the quality of the thermodynamic data, but confidence in a P–T result is increased by the ability to match the phase compositions and proportions within the calculated stability field of the assemblage. The pseudosection for sample Vi17 is presented in Fig. 7a. The area showing the best agreement between modelled data and petrological observations lies in the pale box (i.e. 26–29 kbar and 550–570 C) which corresponds to the field where omphacite, garnet, lawsonite, phengite, glaucophane and rutile are stable. The position of the best-fit box is constrained on Fig. 7a by the intersection of the grossular and almandine contents of garnet and Si4+ content of phengite. Uncertainties for field boundaries propagating from errors on database enthalpies have been computed and plotted over a portion of the pseudosection (Fig. 7b). A comparison showing the good agreement between observed and calculated volume percentages under peak conditions is also shown (560 C, 27.5 kbar, aH2O = 1; Fig. 7c). The effect of water activity is shown in Fig. 7d, where the phase relations at 27 kbar are shown as a function

Quad Calculated peak composition

Alm

(pseudosection Vi17)

Selected analysis for average P-T estimate Rim

(Samp. Vi17: Grt: ae45-9, Omp: r72)

Rim

50

Core

Core

50

Jd

Acm Grs

50

Prp

Fig. 6. Plot showing zoning trends of omphacite (left triangle) and garnet (right triangle). Also shown is the selected analysis for average P–T estimate (star). Calculated peak (560 C, 27.5 kbar) composition based on pseudosection analysis VI17 corresponds to the circle.  2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 17

(a)

(b)

(c)

(d)

Fig. 7. (a) P–T pseudosection and relevant isopleths calculated with THERMOCALC in the system NCKFMASHTO for sample VI17. Darker field shading denotes higher variance. (b) Pseudosection showing absolute uncertainties associated with field boundaries calculated for sample VI17. (c) Comparison (vol.%) between calculated (left) and numerically estimated proportions on which sample VI17 composition is based (right). See also Table 6 for comparison between calculated and observed data in the best-fit region. (d) T-a(H2O) pseudosection showing the influence of decreasing water activity on garnet and phengite isopleths.

 2011 Blackwell Publishing Ltd

18 S. ANGIBOUST ET AL.

(b)

x

(a)

x x Fig. 8. Two P–T pseudosections calculated with THERMOCALC in the system KFMASH (a) and MnKFMASH (b) for the calcschist sample C32 with quartz, phengite and water in excess. Darker field shading indicates higher variance. The garnet-in line is emphasized with a thick black line. The dark rectangle shows the maximum temperature recorded on this sample using RSCM. Stripped areas locate the best-fit domain constrained by the intersection of XMg content of chloritoid, Si4+content of phengite (and spessartine content of garnet) for both peak (at 26 kbar) and retrograde conditions (at 10 kbar).

of temperature at water activity down to 0.75 (i.e. a salinity of 15 wt%). The modelled peak assemblage is shifted down-temperature with decreasing water activity, but with only small changes in calculated mineral compositions. At constant P–T, however, the effect of decreased water activity is to eliminate glaucophane with the formation of minor quartz and also to decrease the Si content of phengite. The area with the best agreement between calculated and analysed compositions lies at high water activities between 0.9 and 1. Comparison between calculated and observed mineral compositions is given in Table 6. The effect of decreasing water content may have a significant effect on topology and phase proportions in the NCKFMASHTO system. Rebay et al. (2010) and Angiboust & Agard (2010) have shown that lawsonite (or epidote) content generally decreases at the expense of garnet, omphacite and kyanite (i.e. a ÔdryÕ eclogite paragenesis; Smith & Lappin, 1989). The good agreement between calculated and observed paragenesis reinforce our conviction that the system was watersaturated under peak conditions. The pseudosection for sample C32 in the KFMASH system is presented in Fig. 8a. The peak assemblage

inferred from petrological observation is garnet + chloritoid + phengite + quartz ± chlorite. The best agreement between analysed compositional isopleths, petrological data and pseudosection modelling for peak conditions is obtained for temperatures in the order of 500–520 C and pressures between 24 and 26 kbar, at the entrance to the garnet stability field (Fig. 8a). Strong retrogression of the rock, marked by chemically different rims around phengite and chloritoid and by chloritization of garnet, happens under greenschist facies conditions at 380–430 C ⁄ 8–12 kbar as suggested by chloritoid and phengite isopleths (Table 6). Adding some minor MnO in the system mainly shifts the garnet-in line down temperature by 25– 30 C (Fig. 8b; see also Mahar et al., 1997; Tinkham et al., 2001). The best-fit area, inferred from isopleths intersection, is not changed significantly (< 10 C decrease). Calculated spessartine content for garnet cores (20 mol.%) is slightly higher than observed (15 mol.%), possibly because the small amount of Ca present in the rock was neglected. Note, however, that the calculated and observed garnet composition (Table 6) is closer when Mn is taken into account in the calculation.  2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 19

Table 6. Comparison between calculated and observed mineral data at best-fit temperatures and pressures (510 C, 26 kbar for sample C32; 560 C, 27.5 kbar for sample VI17). Lago Superiore Unit metapelite C32 Grt rim

Lago Superiore Unit eclogite VI17

Cld core

Ph core

Grt rim

Omp core

Ph core

Calc

Calc(Mn)

Obs

Calc

Obs

Calc

Obs

Calc

Obs

Calc

Obs

Calc

Obs

Si Al Fe2+ Fe3+ Mn Mg Ca Na K

3.00 2.00 2.55 0.00 0.00 0.45 0.00 0.00 0.00

3 2 2.32 0 0.24 0.44 0 0 0

2.99 2.04 2.31 0.00 0.19 0.26 0.19 0.00 0.00

1.00 2.00 0.73 0.00 0.00 0.27 0.00 0.00 0.00

1.02 1.99 0.74 0.01 0.01 0.23 0.00 0.00 0.00

3.58 1.84 0.20 0.00 0.00 0.38 0.00 0.00 1.00

3.60 1.86 0.19 0.00 0.00 0.36 0.00 0.01 0.89

3.00 1.99 1.98 0.00 0.00 0.61 0.41 0.00 0.00

2.99 2.01 1.98 0.00 0.07 0.58 0.37 0.00 0.00

2.00 0.41 0.10 0.13 0.00 0.36 0.45 0.54 0.00

2.02 0.40 0.06 0.15 0.00 0.40 0.47 0.54 0.00

3.58 1.84 0.13 0.00 0.00 0.45 0.00 0.01 0.99

3.62 1.68 0.23 0.00 0.00 0.53 0.00 0.01 0.94

XMg XFe XGrs

0.15 0.85

0.16 0.84

0.10 0.90 0.07

0.27 0.73

0.24 0.76

0.66 0.34

0.65 0.35

0.24 0.76 0.14

0.23 0.77 0.13

0.78 0.22

0.87 0.13

0.78 0.22

0.70 0.30

XGrs = Ca ⁄ (Ca + Fe2+ + Mg).

DISCUSSION Homogeneous P–T conditions for the eclogitization of the Monviso ophiolite? The record of HP ⁄ UHP conditions in mafic rocks

A compilation of our P–T estimates together with recent petrological data on Monviso ophiolite is shown in Fig. 9. Our thermobarometric data on the Lago Superiore Unit eclogites show a marked homogeneity with all methods and suggest peak equilibration at 550 C and 26–27 kbar in the lawsonite eclogite stability field (Fig. 9). As fluid activities in eclogites at such depths may slightly vary from one rock to the other, our temperature estimates provide reasonable upper bounds on the actual equilibration temperature, which was possibly slightly lower (by 10–20 C). These peak conditions are in remarkable agreement with recent work from Groppo & Castelli (2010) who deduced peak conditions of 25 kbar and 550 C from thermodynamic modelling (Perple_X; Connolly, 1990) on a low-strain Fe–Ti metagabbro block from the southern extent of our Lago Superiore Unit (sample OF2727 located in Fig. 2). The estimated peak conditions for sample Vi10 (581 ± 17 C, 26.4 ± 2.4 kbar) are consistent with estimates by Messiga et al. (1999) on the same type of Mg–Al gabbro from the same locality (620 ± 50 C, 24 kbar). Note that our pressure is significantly higher than of Schwartz et al. (2000) (Fig. 9). This difference can be explained by (i) the absence of lawsonite in their peak paragenesis, (ii) the use of phengite with much lower Si4+ content (3.27–3.41 p.f.u.) for average P–T estimates on Passo Gallarino unit (3.65–3.7 p.f.u. in this study), and (iii) by the use of an older phengite barometer (Massone & Schreyer, 1989) with significantly lower pressure values for Si4+ isopleths. Their P–T estimates, thus probably correspond to equilibrium conditions during the early  2011 Blackwell Publishing Ltd

part of the retrograde path, in the epidote domain, rather than to maximum burial conditions. Our estimates lie within the coesite stability field (using Bohlen & Boettcher, 1982) or slightly below (using Bose & Ganguly, 1995) depending on the chosen experimental curve. Despite extensive investigations, coesite has not been observed in our samples. Radiating tensional cracks in garnet around quartz (associated with omphacite, Fig. 3k) have been observed in sample Vi05 but it is not known if these cracks formed by polymorphic transformation after coesite (e.g. Van Der Molen & Van Roermund, 1986) or by dilation of a-quartz during isothermal decompression (e.g. Wendt et al., 1993). Widespread fluid infiltration along fractures and grain boundaries during early exhumation may have facilitated complete replacement of former coesite in Lago Superiore eclogites (Mosenfelder, 2000; Lathe et al., 2005). Insights from metasedimentary rocks

Pseudosection modelling of the P–T evolution of sample C32 suggests peak conditions of 500–520 C and 24–26 kbar. These P–T values are marginally lower (35 C, 2 kbar) than P–T conditions obtained using the other methods (530–560 C; 25–28 kbar). The effect of including Mn in the model improves the match between calculated and observed garnet compositions (Table 6). It does not significantly change the position of XMg isopleths of chloritoid, on which this cold temperature estimate is partly based. Calculated standard deviations of chloritoid XMg isopleths are on the order of 5–10 C. This difference in P–T conditions can be justified by several explanations such as: (i) a slight overestimation of Mg in the rock, thus leading to an artificially Mg-enriched chloritoid, (ii) retrograde overprint of peak composition of garnet rims and chloritoid porphyroblasts, and (iii) late opening of the system

20 S. ANGIBOUST ET AL.

Fig. 9. Summary of our average P–T results and previous results on the Monviso ophiolite. Ellipses represent average P–T results and associated errors for both Fe–Ti gabbros (samples Vi05,14,17,21,42,56), Mg–Al gabbros (Vi10) and Monviso metasedimentary rocks (Monviso quartzite, metapelite C32). The dark rectangles show the estimates based on isopleths for pseudosection VI17 and pseudosection C32. A dark dashed curved line represents the garnet-in reaction derived from pseudosection C32 on a calcschist sample. The two ranges of peak temperatures obtained by Raman spectroscopy are represented by the two thick grey bands. The different quartz–coesite transitions are from: B&G, Bose & Ganguly, 1995; K&K, Kitahara & Kennedy (1964); B&B, Bohlen & Boettcher (1982). Also shown are the prograde P–T path from Groppo & Castelli (2010) and previous peak conditions from Schwartz et al. (2000) for Lago Superiore (L.S) and Passo Gallarino (P.G). The peak estimate on a metatroctolite sample from Lago Superiore Unit (such as Vi10) from Messiga et al. (1999) is given. The background is a grid showing metamorphic facies for a mafic protolith derived from Bousquet et al. (1997).

during pervasive retrogression changing the original peak bulk-rock composition. Finally, it is emphasized that such a difference is close to the error bar on mineral isopleths in this calculation (± 10–20 C). Widespread sampling of metasedimentary rocks across the Monviso ophiolite allows for the first time an evaluation of the potential heterogeneity of maximum temperature conditions. It is pointed out that a relatively good agreement exists between RSCM data and pseudosection modelling on metapelite C32. Raman spectroscopy on organic matter also allows an

Fig. 10. Compilation of P–T paths for the two main units identified in the Monviso ophiolite (Monviso Unit, Lago Superiore Unit) and some other adjacent metamorphic terranes. Note that our new P–T estimates for Monviso ophiolite are in line with the cold subduction gradient identified in Western Alps (7–8 C km)1; Agard et al., 2001).

estimate of the gap in peak metamorphic temperature between the Monviso and Lago Superiore Units (i.e. 40–50 C). Note that this gap is much larger than the relative error bar of the method (10–15 C relative) and therefore representative of different P–T histories. Blake et al. (1995) reported peak temperatures of 510 ± 30 C for the Monviso Unit using Grt-Cpx thermometry on an eclogite boudin. Somewhat lower temperatures between 480 and 500 C for this unit are herein reported. A maximum pressure of 22–24 kbar can thus tentatively be proposed for the Monviso Unit, considering the regional metamorphic gradient for the nearby Schistes Lustre´s and the Western Alps (7–8 C km)1; Agard et al., 2001; Fig. 1b) or recent prograde P–T paths derived from eclogites in the Lago Superiore Unit (Groppo & Castelli, 2010). This estimate obviously relies, however, on the assumption that the metamorphic gradient was nearly identical throughout the area and that peak temperatures were reached under peak-pressure conditions rather than during exhumation. Note that such peak conditions for the Monviso Unit would accord with preliminary THERMOCALC average P–T estimates on a quartzite sample from the northern Visolotto cliff (480 C ⁄ 21 kbar).  2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 21

Finally, slightly higher peak temperatures (555– 570 C) are reported for the lowermost calcschist body embedded within the Lower Shear Zone serpentinites (dashed circle, Fig. 5). Even if these RSCM temperatures overlap with those of many Lago Superiore calcschists, this difference may reveal slightly deeper peak conditions for this relatively small boudin, on the order of 27–28 kbar assuming the above mentioned gradient. Implications for regional geodynamics

The Monviso ophiolite is frequently interpreted as a subduction serpentinite Ôme´langeÕ, i.e. as a chaotic unit gathering tectonic blocks (or tectonic slices) with contrasting P–T evolutions in a mechanically weak matrix (Blake et al., 1995; Schwartz et al., 2000; Guillot et al., 2004). However, early studies (e.g. Lombardo et al., 1978; Lagabrielle & Cannat, 1990) mentioned (i) a relatively well-preserved complete ophiolite magmatic sequence (ii) a similar metamorphic history characterized by the successive eclogiteblueschist–greenschist facies sequence. Our work suggests that (i) the Monviso ophiolite is composed of two main metamorphic sequences with slightly but distinctly different P–T histories and (ii) the Lago Superiore Unit underwent HP to UHP metamorphic conditions close to the coesite stability field. The latter unit, traditionally interpreted as a serpentinite me´lange (e.g. Nisio et al., 1987; Schwartz et al., 2001; Guillot et al., 2004) shows a striking homogeneity of peak P–T conditions, advocating against the possibility of a chaotic tectonic mixing with large P–T differences between adjacent blocks. If it is considered that peak metamorphism was synchronous across the Lago Superiore Unit (between 55 and 40 Ma), it is proposed that this unit corresponds to a more or less preserved portion of thinned oceanic crust detached from 80 km depth in the Alpine subduction zone with a seafloor structural pattern similar to that inferred by Lagabrielle & Cannat (1990) and Lagabrielle & Lemoine (1997). The main shear zone separating the Monviso and Lago Superiore Units (Fig. 2) is therefore interpreted as a key first-order tectonic contact recording deep underthrusting of the Lago Superiore Unit below Monviso Unit. This contact has been pervasively reactivated during late blueschist to greenschist facies exhumation (Schwartz et al., 2000). The other shear zones identified on the field (e.g., Lower Shear Zone, Fig. 2b; see also the cross-section from Lombardo et al., 1978) correspond to second-order tectonic contacts separating sub-units characterized by similar P–T conditions. These secondary contacts have generally been less subject to late retrogression and preserved eclogite facies mylonites, which potentially recorded extremely high strains during peak thrusting mechanisms. Note that such ductile shear zones and similar petrological features have also been reported within  2011 Blackwell Publishing Ltd

Rocciavre Fe–Ti eclogitized metagabbros 50 km to the north (Pognante, 1985). The apparent tectonic complexity observed today in the field is thus attributed to (i) large scale boudinage resulting from deformation of material having significant rheological contrasts (e.g. Fe–Ti gabbro v. serpentinite), (ii) the existence of initial crustal heterogeneities inherent to this slow-spreading oceanic lithosphere, and (iii) additional fracturing and folding during exhumation along several major tectonic contacts, such as with the overlying Schistes Lustre´s complex and the underlying Dora-Maira unit (Balle`vre et al., 1990; Blake & Jayko, 1990). It is herein emphasized that the exhumation pattern described for the Lago Superiore Unit is strikingly similar to those recently described for the Zermatt-Saas ophiolite (e.g. Angiboust et al., 2009) or for Alpine Corsica (e.g. Vitale-Brovarone et al., 2011b). An increasing number of reports showing large continuous crustal slices (e.g. Angiboust & Agard, 2010) and the preservation of almost undisturbed ocean-continent transitions (Zermatt-Saas, Beltrando et al., 2010; Corsica, Vitale-Brovarone et al., 2011a) suggests that Alpine HP deformation of oceanic units is not chaotic but rather localized along shear zones preserving km-scale, coherent volumes. Moreover, similar P–T conditions (510–550 C, 22–26 kbar) for these Alpine HP localities and ages of peak metamorphism (45–38 Ma; e.g. Rubatto et al., 1998; Brunet et al., 2000) may suggest a common detachment mechanism along strike from lawsonite-eclogite facies conditions (see also Agard et al., 2009), possibly in response to the entrance of the thinned continental margin in the subduction zone (e.g. Lapen et al., 2007) and ⁄ or key dehydration reactions at depth (Bucher et al., 2005). Once detached, these dense portions of eclogitized oceanic lithosphere may have been prevented from irreversibly sinking in the mantle by a high initial water budget and ⁄ or a relatively thick buoyant serpentinite sole (Hermann et al., 2000; Guillot et al., 2004; Angiboust & Agard, 2010). Implications for fluid processes in the subduction channel

Underformed Fe–Ti gabbros from the Lago Superiore Unit (Vi56) display a relatively dry peak paragenesis suggesting limited initial water input before subduction (Angiboust & Agard, 2010). By contrast, sills and dykes of Fe–Ti gabbros cross-cutting the metabasaltic upper sequence of the Lago Superiore Unit (e.g. Viso Mozzo, Passo Gallarino: Vi14) may be more hydrated because of the widespread presence of glaucophane associated with the peak assemblage. According to our average P–T estimates and phase diagram calculations, it is now clear that the stability of glaucophane under peak conditions does not point to lower P–T conditions (Schwartz et al., 2000) but reveals a higher initial water content of the protolith (Van Der Straaten et al., 2008; Angiboust & Agard, 2010; Vitale-Brovarone et al., 2011b).

22 S. ANGIBOUST ET AL.

It is also emphasized that lawsonite was stable in most lithologies within Fe–Ti metagabbros from burial to post-eclogitic conditions in the Lago Superiore Unit. This is suggested by (i) the recent finding of fresh small lawsonite inclusions shielded within garnet and omphacite cores in a low-strained Fe–Ti metagabbro sample (Groppo & Castelli, 2010), (ii) the observation of lawsonite pseudomorphs as inclusions in garnet or omphacite cores (Fig. 3j), (iii) lawsonite porphyroblastic growth within the eclogite facies matrix both in low and highly strained samples (Fig. 3b,f), (iv) the presence of veins filled by pseudomorphs after lawsonite crosscutting the eclogite facies mylonite fabric (Fig. 3l), and (v) the prediction of lawsonite stability with our pseudosections for Monviso eclogites (for further discussion, see also Clarke et al., 2006; Whitney & Davis, 2006; Angiboust & Agard, 2010; Rebay et al., 2010) Changes in oxygen fugacity during subduction of Monviso eclogites, in agreement with the observations of Groppo & Castelli (2010), are suspected by the presence of several omphacite generations with various aegirine components. While prograde aegirine-rich (20–30 mol.%) omphacite I after magmatic augite is preserved in low strained metagabbros, omphacite II in textural equilibrium with garnet rims, lawsonite and peak phengite is always relatively depleted in ferric iron. In highly strained recrystallized eclogites, omphacite I is rarely preserved and most of the matrix is constituted of Fe3+-poor omphacite II. Locally, a late Fe3+-rich omphacite III generation grows along joints and cracks between omphacite II (Vi17; Fig. 3h). The pattern of oxidation of omphacite II rims suggests a limited and local circulation of an oxidizing fluid, whose origin is uncertain. The zoning pattern described above is only observed in Fe–Ti metagabbros, whose initial Fe3+ content is by far the largest compared with other Monviso ophiolitic rocks. Similar trends were described above for epidote, which is also a significant ferric iron carrier. Further petrological investigations are needed from other localities to evaluate if aegirine content of clinopyroxene and pistachite content of epidote could be reliable monitors of oxidizing conditions and oxygen fugacity along the subduction channel. CONCLUSIONS

1. On the basis of field observations and RSCM results, two main subunits are proposed for the Monviso ophiolite: the Monviso s.s. Unit (Tmax = 480–500 C) and the Lago Superiore Unit (T-max = 520–550 C). Preliminary pressure estimates suggest a peak pressure gap between these two units on the order of 3–4 kbar. 2. Multi-method thermobarometric data show a good homogeneity for peak metamorphic conditions along the Lago Superiore Unit (530–550 C ⁄ 26– 27.5 kbar). These conditions point to subduction to 80 km at the entrance of the coesite stability field.

3. Lawsonite was stable and generally survived throughout peak conditions, as described in the Zermatt-Saas area northwards (Angiboust & Agard, 2010) and supported by conspicuous pseudomorphs in the rock matrix. 4. The Lago Superiore Unit detached from the downgoing slab as a large (20–30 km-long, 2–3 kmthick) coherent portion of the thinned heterogeneous Tethyan lithosphere. Eclogite mylonites occurring within several shear zones across the area correspond to high-strain zones associated with deep-seated exhumation processes. 5. These results show that the Lago Superiore Unit should not be viewed as a chaotic serpentinite me´lange but rather as a possible southern extension of the Zermatt-Saas ophiolite, recently interpreted as a relatively coherent slice stack detached from the slab under similar metamorphic conditions (Angiboust et al., 2009; Angiboust & Agard, 2010). ACKNOWLEDGEMENTS

M. Balle`vre, J-R. Kienast, S. Schwartz and S. Guillot are thanked for sharing their knowledge on the Monviso ophiolite. P. Yamato and H. Raimbourg are thanked for discussions on the field. B. Huet and P. Pitra are also thanked for insightful discussions on thermobarometry. N. Findling, E. Charon and E. Delairis are acknowledged for technical assistance. C. Groppo and an anonymous reviewer are acknowledged for thoughtful comments on the manuscript. This project has been funded by the programme Egide-PHC Alliance 19349F to P. Agard and C–J. De Hoog (Univ. Edinburgh). REFERENCES Agard, P., Jolivet, L. & Goffe, B., 2001. Tectonometamorphic evolution of the Schistes Lustre´s complex: implications for the exhumation of HP and UHP rocks in the western Alps. Bulletin de la Socie´te´ ge´ologique de France, 172, 617–636. Agard, P., Yamato, P., Jolivet, L. & Burov, E., 2009. Exhumation of oceanic blueschists and eclogites in subduction zones: timing and mechanisms. Earth Science Reviews, 92, 53–79. Angiboust, S. & Agard, P., 2010. Initial water budget: The key to detaching large volumes of eclogitized oceanic crust along the subduction channel? Lithos, 120, 453–474. Angiboust, S., Agard, P., Jolivet, L. & Beyssac, O., 2009. The Zermatt-Saas ophiolite: the largest (60-km wide) and deepest (c. 70-80km) continuous slice of oceanic lithosphere detached from a subduction zone? Terra Nova, 21, 171–180. Aoya, M., Kouketsu, Y., Endo, S. et al., 2010. Extending the applicability of the Raman carbonaceous-material geothermometer using data from contact metamorphic rocks. Journal of Metamorphic Geology, 28, 895–914. Auzende, A.-L., Guillot, S., Devouard, B. & Baronnet, A., 2006. Serpentinites in an Alpine convergent setting: Effects of metamorphic grade and deformation on microstructures. European Journal of Mineralogy, 18, 21–33. Balle`vre, M. & Merle, O., 1993. The Combin Fault - Compressional Reactivation of a Late Cretaceous-Early Tertiary Detachment Fault in the Western Alps. Schweizerische Mineralogische und Petrographische Mitteilungen, 73, 205–227.

 2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 23

Balle`vre, M., Lagabrielle, Y. & Merle, O., 1990. Tertiary ductile normal faulting as a consequence of lithospheric stacking in the western Alps. Me´moire de la Socie´te´ ge´ologique de France, 156, 227–236. Balle`vre, M., Pitra, P. & Bohn, M., 2003. Lawsonite growth in the epidote blueschists from the Ile de Groix (Armorican Massif, France): a potential geobarometer. Journal of Metamorphic Geology, 21, 723–735. Beltrando, M., Rubatto, D. & Manatschal, G., 2010. From passive margins to orogens: The link between ocean-continent transition zones and (ultra)high-pressure metamorphism. Geology, 38, 559–562. Beyssac, O., Goffe, B., Chopin, C. & Rouzaud, J.N., 2002. Raman spectra of carbonaceous material in metasediments: a new geothermometer. Journal of metamorphic Geology, 20, 859–871. Beyssac, O., Goffe, B., Petitet, J.P., Froigneux, E., Moreau, M. & Rouzaud, J.N., 2003. On the characterization of disordered and heterogeneous carbonaceous materials by Raman spectroscopy. Spectrochimica Acta Part a-Molecular and Biomolecular Spectroscopy, 59, 2267–2276. Beyssac, O., Bollinger, L., Avouac, J.P. & Goffe, B., 2004. Thermal metamorphism in the lesser Himalaya of Nepal determined from Raman spectroscopy of carbonaceous material. Earth and Planetary Science Letters, 225, 233– 241. Blake, M.C. & Jayko, A.S., 1990. Uplift of very high pressure rocks in the western Alps: evidence for structural attenuation along low-angle faults. Me´moire de la Socie´te´ Ge´ologique de France, 156, 237–246. Blake, M.C., Moore, D.E. & Jayko, A.S., 1995. The role of serpentinite melanges in the unroofing of ultrahigh-pressure metamorphic rocks: An example from the Western Alps in Italy. In: Ultrahigh pressure metamorphism (eds Coleman, R.G. & Wang, X.), pp. 182–205. Cambridge University Press, Cambridge. Bohlen, S.R. & Boettcher, A.L., 1982. The Quartz-ReversibleCoesite Transformation - a Precise Determination and the Effects of Other Components. Journal of Geophysical Research, 87, 7073–7078. Bose, K. & Ganguly, J., 1995. Quartz-coesite transition revisited; reversed experimental determination at 500-1200 degrees C and retrieved thermochemical properties. American Mineralogist, 80, 231–238. Bousquet, R., 2008. Metamorphic heterogeneities within a single HP unit: Overprint effect or metamorphic mix? Lithos, 103, 46–69. Bousquet, R., Goffe´, B., Henry, P., Le Pichon, X. & Chopin, C., 1997. Kinematic, thermal and petrological model of the Central Alps: Lepontine metamorphism in the upper crust and eclogitisation of the lower crust. Tectonophysics, 273, 105–127. Brunet, C., Monie´, P., Jolivet, L. & Cadet, J.P., 2000. Migration of compression and extension in the Tyrrhenian Sea, insights from 40Ar ⁄ 39Ar ages on micas along a transect from Corsica to Tuscany. Tectonophysics, 321, 127–155. Bucher, K. & Grapes, R., 2009. The Eclogite-facies Allalin Gabbro of the Zermatt-Saas Ophiolite,Western Alps: a Record of Subduction Zone Hydration. Journal of Petrology, 50, 1405–1442. Bucher, K., Fazis, Y., De Capitani, C. & Grapes, R., 2005. Blueschists, eclogites and decompression assemblages of the Zermatt-Saas ophiolite: high pressure metamorphism of the subducted Tethys lithosphere. American mineralogist, 90, 821– 835. Carignan, J., Hild, P., Mevelle, G., Morel, J. & Yeghicheyan, D., 2001. Routine Analyses of Trace Elements in Geological Samples using Flow Injection and Low Pressure On-Line Liquid Chromatography Coupled to ICP-MS: A Study of Geochemical Reference Materials BR, DR-N, UB-N, AN-G and GH. Geostandards Newsletter, 25, 187–198.

 2011 Blackwell Publishing Ltd

Carswell, D.A. & Zhang, R.Y., 1999. Petrographic characteristics and metamorphic evolution of ultrahigh-pressure eclogites in plate-collision belts. International Geology Reviews, 41, 781– 798. Castelli, D. & Lombardo, B., 2007. The plagiogranite-FeTioxide gabbro association of Verne` (Monviso metamorphic ophiolite, Western Alps). Ofioliti, 32, 1–14. Castelli, D., Rostagno, C. & Lombardo, B., 2002. Jd-Qtz-bearing metaplagiogranite from the Monviso meta-ophiolite (Western Alps). Ofioliti, 27, 81–90. Castelli, D., Rolfo, F., Groppo, C. & Compagnoni, R., 2007. Impure marbles from the UHP Brossasco-Isasca Unit (DoraMaira Massif, Western Alps): evidence for Alpine equilibration in the diamond stability field and evaluation of the X(CO2) fluid evolution. Journal of Metamorphic Geology, 25, 587–603. Cheng, H., Nakamura, E., Kobayashi, K. & Zhou, Z., 2007. Origin of atoll garnets in eclogites and implications for the redistribution of trace elements during slab exhumation in a continental subduction zone. American Mineralogist, 92, 1119–1129. Chopin, C., 1984. Coesite and pure pyrope in high-grade blueschists of the Western Alps - a first record and some consequences. Contributions to Mineralogy and Petrology, 86, 107–118. Chopin, C., 2003. Ultrahigh-pressure metamorphism: tracing continental crust into the mantle. Earth and Planetary Science Letters, 212, 1–14. Clarke, G.L., Powell, R. & Fitzherbert, J.A., 2006. The lawsonite paradox: a comparison of field evidence and mineral equilibria modelling. Journal of Metamorphic Geology, 24, 715–725. Cliff, R.A., Barnicoat, A.C. & Inger, S., 1998. Early Tertiary eclogite facies metamorphism in Monviso ophiolite. Journal of Metamorphic Geology, 16, 447–455. Cloos, M., 1982. Flow Melanges - Numerical Modeling and Geologic Constraints on Their Origin in the Franciscan Subduction Complex, California. Geological Society of America Bulletin, 93, 330–345. Coggon, R. & Holland, T.J.B., 2002. Mixing properties of phengitic micas and revised garnet-phengite thermobarometers. Journal of Metamorphic Geology, 20, 683–696. Connolly, J.A.D., 1990. Multivariable phase diagrams; an algorithm based on generalized thermodynamics. American Journal of Science, 290, 666–718. Davis, P. & Whitney, D., 2008. Petrogenesis and structural petrology of high-pressure metabasalt pods, Sivrihisar, Turkey. Contributions to Mineralogy and Petrology, 156, 217– 241. Deville, E., Fudral, S., Lagabrielle, Y., Marthaler, M. & Sartori, M., 1992. From oceanic closure to continental collision: A synthesis of the ‘‘Schistes lustre´s’’ metamorphic complex of the Western Alps. Geological Society of America Bulletin, 104, 127–139. Diener, J.F.A. & Powell, R., 2010. Influence of ferric iron on the stability of mineral assemblages. Journal of Metamorphic Geology, 28, 599–613. Diener, J.F.A., Powell, R., White, R.W. & Holland, T.J.B., 2007. A new thermodynamic model for clino- and orthoamphiboles in the system Na2O–CaO–FeO–MgO–Al2O3–SiO2–H2O–O. Journal of Metamorphic Geology, 25, 631–656. Droop, G.T.R., 1987. A general equation for estimating Fe[3+] concentrations in ferromagnesian silicates and oxides from microprobe analyses, using stoichiometric criteria. Mineralogical magazine, 51, 431–435. Ducheˆne, S., Lardeaux, J.M. & Albarede, F., 1997. Exhumation of eclogites: insights from depth-time path analysis. Tectonophysics, 280, 125–140. Ernst, W., 1999. Metamorphism, partial preservation, and exhumation of ultrahigh-pressure belts. The Island Arc, 8, 125–153.

24 S. ANGIBOUST ET AL.

Federico, L., Crispini, L., Scambelluri, M. & Capponi, G., 2007. Ophiolite melange zone records exhumation in a fossil subduction channel. Geology, 35, 499–502. Fry, N. & Barnicoat, A.C., 1987. The tectonic implications of high-pressure metamorphism in the western Alps. Journal of the Geological Society, 144, 653–659. Gabalda, S., Beyssac, O., Jolivet, L., Agard, P. & Chopin, C., 2009. Thermal structure of a fossil subduction wedge in the Western Alps. Terra Nova, 21, 28–34. Gabudianu Radulescu, I., Rubatto, D., Gregory, C. & Compagnoni, R., 2009. The age of HP metamorphism in the Gran Paradiso Massif, Western Alps: A petrological and geochronological study of ‘‘silvery micaschists’’. Lithos, 110, 95–108. Gebauer, D., Schertl, H.-P., Brix, M. & Schreyer, W., 1997. 35 Ma old ultrahigh-pressure metamorphism and evidence for very rapid exhumation in the Dora-Maira-massif, Western Alps. Lithos, 41, 5–24. Goffe´, B. & Chopin, C., 1986. High-pressure metamorphism in the Western Alps: zoneography of metapelites, chronology and consequences. Schweizerische Mineralogische und Petrographische Mitteilungen, 66, 41–52. Green, E., Holland, T. & Powell, R., 2007. An order-disorder model for omphacitic pyroxenes in the system jadeite-diopside-hedenbergite-acmite, with applications to eclogitic rocks. American Mineralogist, 92, 1181–1189. Groppo, C. & Castelli, D., 2010. Prograde P-T evolution of a lawsonite eclogite from the Monviso metaophiolite (Western Alps): dehydration and redox reactions during subduction of oceanic FeTI-oxide gabbro. Journal of Petrology, 51, 2489– 2514. Groppo, C., Lombardo, B., Castelli, D. & Compagnoni, R., 2007. Exhumation history of the UHPM Brossasco-Isasca Unit, Dora-Maira Massif, as inferred from a phengiteamphibole eclogite. International Geology Review, 49, 142–168. Groppo, C., Beltrando, M. & Compagnoni, R., 2009. The PT path of the ultra-high pressure Lago Di Cignana and adjoining high-pressure meta-ophiolitic units: insights into the evolution of the subducting Tethyan slab. Journal of Metamorphic Geology, 27, 207–231. Guillot, S., Schwartz, S., Hattori, K., Auzende, A. & Lardeaux, J.M., 2004. The Monviso ophiolitic Massif (Western Alps), a section through a serpentinite subduction channel. Evolution of the western Alps: insights from metamorphism, structural geology, tectonics and geochronology. The Virtual Explorer, 16. Paper 6. Hermann, J., Mu¨ntener, O. & Scambelluri, M., 2000. The importance of serpentinite mylonites for subduction and exhumation of oceanic crust. Tectonophysics, 327, 225–238. Holland, T.J.B. & Powell, R., 1990. An Enlarged and Updated Internally Consistent Thermodynamic Dataset with Uncertainties and Correlations in the System K2O-Na2O-CaOMgO-MnO-FeO-Fe2O3-Al2O3-TiO2-SiO2-H2O. Journal of Metamorphic Geology, 8, 89–124. Holland, T.J.B. & Powell, R., 1998. An internally consistent thermodynamic data set for phases of petrological interest. Journal of Metamorphic Geology, 16, 309–343. Holland, T. & Powell, R., 2003. Activity–composition relations for phases in petrological calculations: an asymmetric multicomponent formulation. Contributions to Mineralogy and Petrology, 145, 492–501. Holland, T.J.B., Baker, J.M. & Powell, R., 1998. Mixing properties and activity-composition relationships of chlorites in the system MgO-FeO-Al2O3-SiO2-H2O. European Journal of Mineralogy, 10, 395–406. Kienast, J., 1983. Le me´tamorphisme de haute pression et basse tempe´rature (e´clogites et schistes bleus) : donne´es nouvelles sur la pe´trologie des roches de la crouˆte oce´anique subducte´e et des se´diments associe´s. PhD Universite´ Pierre et Marie Curie, Paris. Kienast, J.R. & Messiga, B., 1987. Cr-rich Mg-chloritoid, a first record in high-pressure metagabbros from Monviso (Cottian Alps), Italy. Mineralogical magazine, 51, 681–687.

Kitahara, S. & Kennedy, G.C., 1964. The quartz-coesite transition. Journal of Geophysical Research, 69, 5395–5400. Kurz, W., Neubauer, F. & Unzog, W., 1999. Evolution of Alpine eclogites in the Eastern Alps: Implications for Alpine geodynamics. Physics and Chemistry of the Earth, Part A: Solid Earth and Geodesy, 24, 667–674. Lagabrielle, Y. & Cannat, M., 1990. Alpine Jurassic ophiolites resemble the modern central Atlantic basement. Geology, 18, 319–322. Lagabrielle, Y. & Lemoine, M., 1997. Alpine, Corsican and Apennine ophiolites: the slow-spreading ridge model. Comptes Rendus de lÕAcade´mie des Sciences - Series IIA - Earth and Planetary Science, 325, 909–920. Lapen, T., Johnson, C., Baumgartner, L., Piaz, G., Skora, S. & Beard, B., 2007. Coupling of oceanic and continental crust during Eocene eclogite-facies metamorphism: evidence from the Monte Rosa nappe, western Alps. Contributions to Mineralogy and Petrology, 153, 139–157. Lardeaux, J.M., Nisio, P. & Boudeulle, M., 1987. Deformational and metamorphic history at the Lago Superiore area of the Monviso ophiolitic complex (Italian Western Alps): a record of subduction–collision cycle. Ofioliti, 12, 479–502. Lathe, C., Koch-Muller, M., Wirth, R. et al., 2005. The influence of OH in coesite on the kinetics of the coesite-quartz phase transition. European Journal of Mineralogy, 90, 36–43. Leake, B.E., Wooley, A.R., Arps, C.E.S. et al., 1997. Nomenclature of amphiboles: Report of the subcommittee on amphiboles of the International Mineralogical Association, commission on new minerals and mineral names. American Mineralogist, 82, 1019–1037. Lemoine, M., Tricart, P. & Boillot, G., 1987. Ultramafic and gabbroic ocean floor of the Ligurian Tethys (Alps, Corsica, Apennines): In search of a genetic model. Geology, 15, 622– 625. Lombardo, B. & Pognante, U., 1982. Tectonic implications in the evolution of the Western Alps ophiolite metagabbros. Ofioliti, 7, 371–394. Lombardo, B., Nervo, R., Compagnoni, R. et al., 1978. Osservazioni preliminari sulle ofioliti metamorfiche del Monviso (Alpi occidentali). Rendiconti Societa Italiana Di Mineralogia e Petrologia, 34, 253–305. Mahar, E.M., Baker, J.M., Powell, R., Holland, T.J.B. & Howell, N., 1997. The effect of Mn on mineral stability in metapelites. Journal of Metamorphic Geology, 15, 223–238. Marmo, B.A., Clarke, G.L. & Powell, R., 2002. Fractionation of bulk rock composition due to porphyroblast growth: effects on eclogite facies mineral equilibria, Pam Peninsula, New Caledonia. Journal of Metamorphic Geology, 20, 151– 165. Martin, S., Rebay, G., Kienast, J.R. & Mevel, C., 2008. An eclogitized oceanic palaeo-hydrothermal field from the St Marcel valley (Italian Western Alps). Ofioliti, 33, 49–63. Massone, H.J. & Schreyer, W., 1989. Stability field of the highpressure assemblage, talc + phengite and two new phengite barometers. European Journal of Mineralogy, 1, 391–410. Meffan-Main, S., Cliff, R.A., Barnicoat, A.C., Lombardo, B. & Compagnoni, R., 2004. A Tertiary age for Alpine highpressure metamorphism in the Gran Paradiso massif, Western Alps: a Rb–Sr microsampling study. Journal of Metamorphic Geology, 22, 267–281. Messiga, B., Kienast, J.R., Rebay, G., Riccardi, M.P. & Tribuzio, R., 1999. Cr-rich magnesiochloritoid eclogites from the Monviso ophiolites (Western Alps, Italy). Journal of Metamorphic Geology, 17, 287–299. Monviso, 1980. The Monviso ophiolite complex. Extract from ‘‘Ophiolites: Proceedings International Ophiolite Symposium’’, Cyprus 1979, 332–340. Mosenfelder, J.L., 2000. Pressure dependence of hydroxyl solubility in coesite. Physics and Chemistry of Minerals, 27, 610–617. Nisio, P., Lardeaux, J.M. & Boudeulle, M., 1987. E´volutions tectonome´tamorphiques contraste´es des e´clogites dans le  2011 Blackwell Publishing Ltd

MONVISO ECLOGITES AND SUBDUCTION DYNAMICS 25

massif du Viso; conse´quences de la fragmentation de la crouˆte oce´anique lors de lÕorogene`se alpine. Comptes Rendus de lÕAcademie des Sciences, 304, 355–360. Oberha¨nsli, R. & Goffe´, B., 2004. Metamorphic Structure of the Alps. C.C.G.M, Paris. Philippot, P., 1990. Opposite vergence of nappes and crustal extension in the French-italian Western Alps. Tectonics, 9, 1143–1164. Philippot, P. & Kienast, J.-R., 1989. Chemical-microstructural changes in eclogite-facies shear zones (Monviso, Western Alps, north Italy) as indicators of strain history and the mechanism and scale of mass transfer. Lithos, 23, 179–200. Philippot, P. & Selverstone, J., 1991. Trace-Element-Rich Brines in Eclogitic Veins - Implications for Fluid Composition and Transport During Subduction. Contributions to Mineralogy and Petrology, 106, 417–430. Philippot, P. & van Roermund, H.L.M., 1992. Deformation processes in eclogitic rocks: evidence for the rheological delamination of the oceanic crust in deeper levels of subduction zones. Journal of Structural Geology, 14, 1059–1077. Platt, J.P., 1986. Dynamics of Orogenic Wedges and the Uplift of High-Pressure Metamorphic Rocks. Geological Society of America Bulletin, 97, 1037–1053. Pognante, U., 1985. Coronitic reactions and ductile shear zones in eclogitised ophiolite metagabbro, Western Alps, North Italy. Chemical Geology, 50, 99–109. Powell, R. & Holland, T., 1994. Optimal geothermometry and geobarometry. American Mineralogist, 79, 120–133. Powell, R. & Holland, T.J.B., 2008. On thermobarometry. Journal of Metamorphic Geology, 26, 155–179. Ravna, E.J.K. & Paquin, J., 2004. Thermobarometric methodologies applicable to eclogites and garnet ultrabasites. EMU Notes in Mineralogy, 5, 229–259. Ravna, E.J.K., Andersen, T.B., Jolivet, L. & De Capitani, C., 2010. Cold subduction and the formation of lawsonite eclogite – constraints from prograde evolution of eclogitized pillow lava from Corsica. Journal of Metamorphic Geology, 28, 381– 395. Rebay, G. & Messiga, B., 2007. Prograde metamorphic evolution and development of chloritoid-bearing eclogitic assemblages in subcontinental metagabbro (Sesia-Lanzo zone, Italy). Lithos, 98, 275–291. Rebay, G., Powell, R. & Diener, J.F.A., 2010. Calculated phase equilibria for a morb composition in a P–T range, 450–650 C and 18–28 kbar: the stability of eclogite. Journal of Metamorphic Geology, 28, 635–645. Reinecke, T., 1998. Prograde high- to ultrahigh-pressure metamorphism and exhumation of oceanic sediments at Lago di Cignana, Zermatt-Saas Zone, western Alps. Lithos, 42, 147– 189. Rubatto, D. & Hermann, J., 2003. Zircon formation during fluid circulation in eclogites (Monviso, Western Alps): implications for Zr and Hf budget in subduction zones. Geochimica et Cosmochimica Acta, 67, 2173–2187. Rubatto, D., Gebauer, D. & Fanning, M., 1998. Jurassic formation and Eocene subduction of the Zermatt-Saas-Fee ophiolites: implications for the geodynamic evolution of the Central and Western Alps. Contributions to Mineralogy and Petrology, 132, 269–287. Scambelluri, M. & Philippot, P., 2001. Deep fluids in subduction zones. Lithos, 55, 213–227. Schertl, H., Schreyer, W. & Chopin, C., 1991. The pyrope-coesite rocks and their country rocks at Parigi, Dora Maira Massif, Western Alps: detailed petrography, mineral chemistry and PT-path. Contributions to Mineralogy and Petrology, 108, 1–21. Schwartz, S., Lardeaux, J.M., Guillot, S. & Tricart, P., 2000. The diversity of eclogitic metamorphism in the Monviso ophiolitic complex, western Alps, Italy. Geodinamica Acta, 13, 169–188.

 2011 Blackwell Publishing Ltd

Schwartz, S., Allemand, P. & Guillot, S., 2001. Numerical model of the effect of serpentinites on the exhumation of eclogitic rocks: insights from the Monviso ophiolitic massif (Western Alps). Tectonophysics, 342, 193–206. Searle, M.P., Waters, D.J., Martin, H.N. & Rex, D.C., 1994. Structure and metamorphism of blueschist-eclogite facies rocks from the northeastern Oman Mountains. Journal of the Geological Society, 151, 555–576. Selverstone, J., Franz, G., Thomas, S. & Getty, S., 1992. Fluid variability in 2 GPa eclogites as an indicator of fluid behavior during subduction. Contributions to Mineralogy and Petrology, 112, 341–357. Sharp, Z.D., Essene, E.J. & Hunziker, J.C., 1993. Stable isotope geochemistry and phase equilibria of coesite-bearing whiteschists, Dora Maira Massif, western Alps. Contributions to Mineralogy and Petrology, 114, 1–12. Smith, D.C. & Lappin, M.A., 1989. Coesite in the Straumen kyanite-eclogite pod, Norway. Terra Nova, 1, 47–56. Tilton, G.R., Schreyer, W. & Schertl, H.P., 1991. Pb- Sr- Nd isotopic behavior of deeply subducted crustal rocks from the Dora Maira Massif, Western Alps, Italy-II: what is the age of the ultrahigh-pressure metamorphism? Contributions to Mineralogy and Petrology, 108, 22–33. Tinkham, D.K., Zuluaga, C.A. & Stowell, H.H., 2001. Metapelite phase equilibria modeling in MnNCKFMASH: The effect of variable Al2O3 and MgO ⁄ (MgO+FeO) on mineral stability. Geological Materials Research, 3, 1–42. Van Der Molen, I. & Van Roermund, H.L.M., 1986. The pressure path of solid inclusions in minerals: the retention of coesite inclusions during uplift. Lithos, 19, 317–324. Van Der Straaten, F., Schenk, V., John, T. & Gao, J., 2008. Blueschist-facies rehydration of eclogites (Tian Shan, NW-China): Implications for fluid–rock interaction in the subduction channel. Chemical Geology, 255, 195–219. Vignaroli, G., Rossetti, F., Bouybaouene, M. et al., 2005. A counter-clockwise P–T path for the Voltri Massif eclogites (Ligurian Alps, Italy). Journal of Metamorphic Geology, 23, 533–555. Vitale-Brovarone, A., Beltrando, M., Malavieille, J. et al., 2011a. Inherited Ocean-Continent Transition zones in deeply subducted terranes: Insights from Alpine Corsica. Lithos, 124, 273–290. Vitale-Brovarone, A., Groppo, C., Hete´nyi, G., Compagnoni, R. & Malavieille, J., 2011b. Coexistence of lawsonite-bearing eclogite and blueschist: phase equilibria modelling of Alpine Corsica metabasalts and petrological evolution of subducting slabs. Journal of Metamorphic Geology, 29, 583–600. Warren, C.J. & Waters, D.J., 2006. Oxidized eclogites and garnet-blueschists from Oman: P–T path modelling in the NCFMASHO system. Journal of Metamorphic Geology, 24, 783–802. Wendt, A.S., DÕArco, P., Goffe´, B. & Oberha¨nsli, R., 1993. Radial cracks around [alpha]-quartz inclusions in almandine: Constraints on the metamorphic history of the Oman mountains. Earth and Planetary Science Letters, 114, 449–461. White, R.W., Pomroy, N.E. & Powell, R., 2005. An in-situ metatexite-diatexite transition in upper amphibolite facies rocks from Broken Hill, Australia. Journal of Metamorphic Geology, 23, 579–602. White, R.W., Powell, R. & Holland, T.J.B., 2007. Progress relating to calculation of partial melting equilibria for metapelites. Journal of Metamorphic Geology, 25, 511–527. Whitney, D.L. & Davis, P.B., 2006. Why is lawsonite eclogite so rare? Metamorphism and preservation of lawsonite eclogite, Sivrihisar, Turkey. Geology, 34, 473–476. Whitney, D.L. & Evans, B.W., 2010. Abbreviations for names of rock-forming minerals. American Mineralogist, 95, 185–187. Received 18 January 2011; revision accepted 6 June 2011.