Very low-grade metamorphism in the para-autochtonous sedimentary cover of the Pelvoux massif (Western Alps, France)

Abstract

The metamorphic grade from the para-autochtonous cover of the Pelvoux massif (PM, western Alp, France) was investigated in Priabonian metamarls, through mineral assemblages, simultaneous measurements of illite and chlorite “crystallinity”, K-white mica b cell dimension and K-white mica polytypes content. Kübler (KI) and Árkai (ÁI) indexes display values characteristic for uppermost diagenetic to low anchizonal conditions. These results are supported by the incomplete transformation of K-white mica polytypes from 1M _d and 2M ₁ and are in line with the not annealed status of zircon fission tracks in the southeast zone of the Pelvoux massif (Seward et al. 1999). Our findings differ from previous results, which suggested epizonal conditions at several places in the northern part of our field area, but are coherent with the temperature estimation suggested by the mineral assemblage described in the literature. These results are further supported by temperature values deduced from fluid inclusion investigations. A small decrease of KI and ÁI values is observed within the basal decollement level of the Priabonian cover, when going from the least deformed zones to the most deformed ones. K-white mica b cell dimensions in micas of the area are in the range 9.000–9.040 Å, which corresponds to an inferred geothermal gradient of 25–35 °C/km, similar to the values found further north in the Dauphinois domain (Ceriani et al. 2003). This indicates a long steady state heat flow corresponding to a long term metamorphic event like in the eastern Alps.

1 Introduction

In the western Alps, the illite “crystallinity” method has been used in order to differentiate the anchizone from diagenesis and epizone domains according to Kübler (1968). The general trend of the metamorphic grade, increasing from W to E, from diagenesis to epizone, is perturbed by tectonic duplexes and the influence of crystalline basement (Desmons et al. 1999).

In the northern part of the Pelvoux massif, Ceriani et al. (2003) studied the low-grade metamorphism in the region of the Frontal Penninic units (FPU) of the Western Alps between the Arc and Isère Valleys (Fig. 1a, b). The results of the study allowed relating the metamorphism to the different stages of deformation.

Fig. 1

a Simplified geological overview of the arc of the western Alps (A Argentera massif, CM Combeynot massif, MB Mont Blanc massif, P Pelvoux massif, R Rhone-Simplon line, T Tonale line); after Ferreiro Mählmann (1996), Ceriani et al. (2001) and Ravenne et al. (1987, 2004). b Tectonic map along the eastern rim of the Pelvoux massif compiled from (1) geological maps BRGM (1/50,000): La Grave, Briançon, St Christophe en Oisans, Guillestre, Orcières and Embrun and (2) Tricart (1986). Points Yr, Fo, Be, Cha, Do and DoF refer to samples studied, collected at the eastern rim of the Pelvoux massif, in the Yret, Fournel, Beassac, Chambran, Dormillouse and Dormillouse-Fangeas valleys, respectively. RT Roselend Thrust, RYSZ Rocher de l’Yret shear zone

Previous metamorphic studies described south of the Pelvoux massif a laumontite–prehnite–pumpellyite subfacies (zeolite facies) in the “grès de Champsaur” formation, with rock-forming laumontite (Aprahamian 1988) and pumpellyite–prehnite in veins (Waibel 1990). On the metamorphic map of the Alps (Frey et al. 1999) and the map of metamorphic structure of the Alps (Oberhänsli et al. 2004) the sub-greenschist facies (epizone) is found in the tiny basement of Fournel and Dormillouse valleys (ref. Fig. 1b).

The aim of this contribution is to present the results of a study on metamorphic grade in the southern footwall of the Penninic Basal Contact (PBC, Ceriani et al. 2001), within the Dauphinois cover of the southeastern rim of the Pelvoux massif, and to compare the results with those from Ceriani et al. (2003). The metamorphic grade was investigated in the globigerina metamarl level and in the Mesozoic cover by mineral assemblages, simultaneous measurements of illite and chlorite “crystallinity” using the KI- and ÁI-methods, K-white mica b cell dimension and K-white mica polytypes content. Within the Rocher de l’Yret shear zone (RYSZ, Fig. 1), results are completed by a fluid inclusion study. Quartz precipitates found along the strike slip fault planes dissecting the uppermost basement units from the RYSZ were analysed. All the methods were used for a better P–T estimate to get a higher precision on the metamorphic evolution in this region and to conclude better with some geodynamic implications. The results of this metamorphic study are compared to those from Ceriani et al. (2003) north of the Pelvoux massif, to the FT data available in the region (Seward et al. 1999; Fügenschuh and Schmid 2003) and previous studies describing mineral assemblage in the “grès de Champsaur” (Aprahamian 1988; Waibel 1990).

2 Regional geology

The Dauphinois domain is part of the European margin and composed of Variscan basement carrying a thick Permian to Tertiary sedimentary cover. During Mesozoic rifting, the eastern margin of the European plate was substantially thinned. Formation of major tilted crustal blocks allowed for the deposition of thick Mesozoic sedimentary series in asymmetric half grabens intensively described in literature (Tricart et al. 1988; Davies 1982; Gillcrist et al. 1987; Butler 1989; Coward et al. 1991; Huyghe and Mugnier 1995; Lazarre et al. 1996; Sue et al. 1997).

Mesozoic strata are discordantly overlain by a conglomerate formation (Gupta 1997), followed by the classical “Priabonian trilogy” (Ravenne et al. 1987; Apps et al. 2004), which comprises, from base to top: (1) shallow water nummulitic limestone of variable thickness (0–50 m), (2) hemipelagic globigerina marls (0–50 m), and (3) the “grès de Champsaur” formation, a regular alternance of turbiditic sandstones and shales (Perriaux and Uselle 1968; Waibel 1990), whose thickness varies from 700 to 1,200 m. In some areas immediately south of the Pelvoux massif, the silici-clastic deposits contain up to 50 % of volcanic detritus (Debrand-Passard et al. 1984; Bürgisser 1998).

Locally, three Alpine deformation phases (D₁, D₂ and D₃) can be recognized (Trullenque 2005). D₁, first phase of deformation that affected the Dauphinois units corresponds to the third phase of deformation described by Ceriani et al. 2001, i.e. Oligocene to lower Miocene out of sequence thrusting along the Roselend thrust (RT). It is the major deformation phase encountered in the investigated area since it is related to the thrusting along the RT (Ceriani et al. 2001). D₁ deformation structures constantly show WNW-directed kinematic indicators measured directly along tectonic contacts, i.e. consistent with activity along the RT.

The second D₂ deformation phase recorded within the investigated area results in SW-directed movements that overprint the top-WNW D₁ deformation features and which are missing N of the Pelvoux massif. Trullenque (2005) proposed that the D₂ top-SW deformation phase is linked to the formation of the Apeninnes chain, as a consequence of the opening of the Thyrenian basin during Langhian times.

D₃ structures, normal faults related to the Durance fault system (High Durance Faulted Zone, Tricart 2004) are the latest deformation features found in the investigated area. These normal faults are notably responsible for a strong geothermal gradient and seismicity recorded along the Durance valley. The timing of the onset of normal faulting, also recorded north of Pelvoux, remains a matter of debate (see Fügenschuh et al. 1999; Fügenschuh and Schmid 2003; Tricart et al. 2001). However, in this region, fission track data on zircon (Seward et al. 1999) reveal that maximum temperatures were well below greenschist facies conditions. Seward et al. (1999) interpreted the heating event as being due to the emplacement of internal thrust sheets, which thin out westwards.

3 Materials and methods

In marine pelites and carbonates, no diagnostic minerals and mineral assemblages form at conditions of very low-grade field. In these rocks, the transitions from non-metamorphic to very low-grade and from very low-grade to low-grade metamorphic domains take place through the diagenetic zone, the anchizone and the epizone, each zone being characterized by specific values of the illite Kübler Index (KI; Árkai et al. 2003). The illite “crystallinity” (IC or KI; Kübler 1964) is defined as the full width at half maximum of the first illite basal reflection in XRD patterns (Frey 1987; Guggenheim et al. 2002). Guggenheim et al. (2002) recommended that the use of a “crystallinity index” should be avoided, although it may be placed within quotation marks when referring in a limited way to previously referenced work. They also recommended to refer to an index by relating it to the author describing the procedures necessary to define the value, regardless of what the index may actually be describing. Therefore, we will refer for K-white mica to the illite “crystallinity” for raw data and to KI after calibration against Kübler’s scale.

Illite “crystallinity” is considered to be a function of crystallite thickness, the number of lattice defects (Merriman et al. 1990). Temperature is thought to be the main factor controlling the illite “crystallinity”, but other parameters like lithology, time, tectonic stress and fluid/rock ratio may probably have important effect (see Frey 1987). Árkai (1991) and Árkai et al. (1995) proposed a similar index the chlorite “crystallinity” (ChC) or Árkai index (ÁI) to monitor the reaction’s progress.

Twenty-four metamarl samples were collected within the Priabonian and Mesozoic cover formations of the Dauphinois. Sampling was restricted to these formations in order to minimize the petrological effect on the IC. The outcrops density also limited sampling around the Dormillouse and Fournel valleys outcrops are poor. Moreover, westward the Selle Fault cut out the Priabonian formation. The nature, age and tectonic setting of this fault are poorly constrained leading to uncertainties on its effects on the metamorphic pattern. Therefore, samples were taken in the Dormillouse, Fournel, Beassac and Chambran valleys and in the Yret zone (Figs. 1, 2).

Fig. 2

Distribution and values of Kübler index (KI) and Árkai index (ÁI) in the studied area. Due to the proximity of the samples Do4,4a,4b; Do5a,c; DoF1a,b; Fo1b,c,e; Fo2b,c and Fo3a,b,c the values are shown as averages. RT Roselend thrust

In the different samples, the main schistosity is related to the D₁ phase (Trullenque 2005), the D₂ phase only affects the area south of an E-W line of Beassac valley. The clay fraction analysed is generally related to D₁.

Samples with detrital mica visible in hand specimen and/or weathered specimens were avoided as far as possible to eliminate detrital contamination. Mineral abbreviations used are from Kretz (1983).

3.1 X-ray diffraction

Clay mineral separation was conducted using techniques described by Potel et al. (2006). Carbonate removal was done using a 5 % acetic acid (C₂H₄O₂) and washed after with deionised water. To minimize the effect of possible detrital clay minerals, we avoided long grinding processes (<15 s) and repeated the settling procedure for the ≤2 μm fraction five times. Illite and chlorite crystallinity was measured at the University of Giessen on air-dried preparates, using a D501 Bruker-AXS (Siemens) diffractometer, CuKα radiation at 40 kV and 30 mA and divergence slits of 0.5° with a secondary graphite monochromator. Two slices of each sample were prepared and each measured two times as air-dried and one time glycolated. The range of measurement, the time counting and the step size were as follow: for whole-rock paragenesis between 2 and 70 ∆°2Θ with 1 s and 0.02° step, for air-dried preparate between 2 and 70 ∆°2Θ with 2 s and 0.01° step.

Illite “crystallinity” was calculated using the software DIFFRACPlus (evaluation/release 2001 by ^©Bruker AXS) and MacDiff 4.25 (written by R. Petschick, 17 May 2001). IC measured in Giessen (IC_Giessen) values were transformed into KI values using a correlation with the SW standards (CIS standards) of Warr and Rice (1994) (KI_CIS = 1.2702 × IC_Giessen − 0.0314) (Table 1). The KI was used to define the limits of anchizone, and the transition values were chosen as follows: KI = 0.25 ∆°2Θ for the epizone to high anchizone boundary, KI = 0.30 ∆°2Θ for the high to low anchizone boundary and KI = 0.42 ∆°2Θ for the low anchizone to diagenetic zone. However, the use of the CIS standards is not universally accepted as giving Kübler-equivalent zone limits. Kisch et al. (2004) show that the CIS standard values are much broader than those obtained by all other laboratories and that the high- and low-grade boundaries of the anchizone of the raw values of Warr and Rice (1994) are much broader than the Kübler-equivalent. This discrepancy is likely to reflect errors in the conversion of the IC values into Kübler equivalent (Kisch et al. 2004; Ferreiro Mählmann and Frey this volume; Ferreiro Mählmann et al. this volume). Therefore, following the recommendations done by Kisch et al. (2004), we published in the Table 1 the IC_Giessen and ChC to allow comparison with other laboratories. The same experimental conditions were also used to determine chlorite “crystallinity” on the (002) peak of the second (7 Å) basal reflections of chlorite (Table 1). The ChC measurements were calibrated with those of Warr and Rice (1994) and expressed as the ÁI (Guggenheim et al. 2002): ÁI = 0.8775 × ChC + 0.0239. The anchizone boundaries for the ÁI were defined by correlation with the KI and are given as 0.24 ∆°2Θ for the epizone to anchizone boundary and 0.30 ∆°2Θ for the anchizone to diagenetic zone.

Table 1 Samples identification, stratigraphic age, Illite Kübler index and chlorite “crystallinity” values of the <2 μm grain-size fraction samples (data in Δ°2θ), percent of 2M ₁ polytypes and K-white mica b cell dimension data

Randomly oriented samples for K-white mica b cell dimension and polytype determination of K-white mica were prepared using wood glue on quartz sample holder. The K-white mica b cell dimension is based on the d_060,331 spacing and on the increasing celadonite substitution that occurs with pressure increase in white mica (Ernst 1963; Guidotti et al. 1989). Guidotti et al. (1989) presented linear regression equations that quantify the changes in the K-white mica b cell dimensions of muscovite 2M ₁ that result from cation substitutions on the interlayer and octahedral sites. The K-white mica b cell dimension value was determined by measurement of the (060) peak of the potassic white mica, if present (Sassi and Scolari 1974), by using the cell-refinement program WIN-METRIC V.3.0.7 (^©Bruker AXS).

Illite-muscovite polytype determination was done using a curve of 2M ₁/(2M ₁ + 1M) peaks ratio calibrated using different mixture of illite polytype following the technique described by Dalla Torre et al. (1994). Merriman and Peacor (1999) recognized that the wealth of data on variations in white mica polytypism as a function of temperature is generally consistent with predictable trends. However, they recommended that polytype sequences should not be used other than as indicators of reaction progress.

3.2 Microthermometry

Quartz-calcite veins were sampled along the Rocher de l’Yret shear zone (RYSZ) (Figs. 1, 2), in order to substantiate P–T estimation and to reconstruct the fluid evolution history during metamorphism (Frey et al. 1980). Microthermometry investigations were done at the University of Giessen on double polished thin sections and performed using a Linkam THM 600/S/Geo heating-freezing stage coupled to a TMS 94 temperature controller with an error of ±1 °C. The stage is mounted on an Olympus microscope with a 100× Objective. The heating and cooling stage was calibrated using synthetic fluid inclusion calibration standard: CO₂ and H₂O from ^©Bubbles Incorporation.

Two-phase (consisting of vapour and liquid at room temperature) fluid inclusions were identified. No dissolved volatile phase was observed, either by melting of CO₂ at or below its triple point of −56.6 °C, nor by formation or dissolution of clathrate or liquid–vapor equilibrium of a volatile component such as higher hydrocarbons (HHC), CH₄, CO₂, N₂ or H₂S. Thus, microthermometry was restricted to measuring the melting temperature of ice (Tm_ice) and the bulk homogenization temperature of the fluid inclusions (Th_i). As none of the investigated fluid inclusions contained any observable gas component, salinity was derived from the ice melting temperature in NaCl-equivalence after Hall et al. (1988). Density for fluid inclusions with homogenisation temperatures less than 200 °C are from the equations of state given by Brown and Lamb (1989) and for temperatures greater than 200 °C are from Zhang and Frantz (1987). The isochores were calculated from the equation of state given by Zhang and Frantz (1987).

3.3 Raman microspectrometry

Gas, liquid and solid phases were investigated at the University of Frankfurt am Main using a Raman microprobe Leica/Renishaw and the software Renishaw WiRE™ 2.0 and GRAMS. An Argon laser (green laser: λ = 514.5 nm) was used as excitation laser radiation. The Raman microspectrometry was used to identify the presence of CH₄, CO₂, HHC (C₂H₆ or C₃H₈) and N₂ in fluid inclusions. The relevant peak is positioned at 2,917 cm⁻¹ for CH₄, 1,285 and 1,388 cm⁻¹ for CO₂, 2,890 and 2,954 cm⁻¹ for C₂H₆ or C₃H₈, respectively and 2,331 cm⁻¹ for N₂ (Burke 2001). The small size of most fluid inclusions (<10 μm) and the large vertical dimension of the laser beam focus (4 μm in diameter) cause the presence of Raman lines of the enclosing quartz (1,160 cm⁻¹) in the spectra of fluids. This line does not interfere at all with the CH₄, CO₂ and N₂ lines.

4 Results

4.1 Mineralogy

The mineralogy of the studied samples is given in Table 1.

In the Priabonian and Mesozoic meta-marls, the mineral assemblages consist of quartz + K-white micas + chlorite and calcite with minor amounts of albite. Gypsum is detected in some samples from the Fournel and Dormillouse valleys (Do2, Fo3a and Fo3c). In the fraction <2 μm, illite-muscovite predominates, chlorite and quartz are significant, and feldspar, if detected, is only present in small quantities.

4.2 Characteristics of the phyllosilicates

Figure 2 shows the distribution of the KI data, the values of which are listed in Table 1.

The values in the Dormillouse valley vary between uppermost diagenetic conditions (0.43 ∆°2Θ) in the south and low anchizonal values (0.36 ∆°2Θ) in the north. In the Fournel, Beassac and Chambran Valleys, the KI data indicate low anchizonal conditions (0.35–0.41 ∆°2Θ). In the Yret Zone, illite crystallinity values correspond to the diagenesis/anchizone boundary (0.37–0.43 ∆°2Θ) (Fig. 2). No trend can be observed between KI and elevation of the sample. However, a trend can be observed in each valley or zone, with an increase from south to north of KI values. This is similar to that Aprahamian (1988) described.

The distribution of ÁI values are presented in Fig. 2. The pattern of very low-grade metamorphism shown by AÍ agrees well with the pattern of the KI (Fig. 2). A positive linear correlation (R² = 0.64) is found between KI and ÁI values (Fig. 3a). This gives confidence on the fact that illite and chlorite “crystallinities” actually refer to the same P–T conditions and thus the same experienced T–t history.

Fig. 3

Correlation between Kübler index (KI) and Árkai index (AÍ) (a) and the 2M ₁/1M _d illite polytype ratio (b), all derived from air-dried mounts

The percentage of 2M ₁ illite-muscovite polytype relative to the KI in the studied area shows a positive trend with increasing metamorphic grade (Fig. 3b).

Sassi and Scolari (1974) determined a semi quantitative relationship between the K-white mica b cell dimension and the metamorphic pressure gradient under greenschist facies condition. They plotted the K-white mica b-values as cumulative frequency curves in order to compare them with other metamorphic belts. Guidotti and Sassi (1986) collated published K-white mica b cell dimension data for greenschist and blueschist facies rocks. They presented a qualitative plot of b cell dimension as a family of curves in P–T space. The values presented by Guidotti and Sassi (1986) do not extend to the sub-greenschist facies. However, the considerable amount of K-white mica b cell dimension data from very low-grade metapelites accumulated in different metamorphic belt corroborate the groupings found by Guidotti and Sassi (1986), as shown by Merriman and Peacor (1999). K-white mica b cell dimensions were calculated for 21 samples from the anchizone (three samples were not integrated due to interferences and poor intensity signal), all fall in the same range with a minimum 8.998 and a maximum 9.028 Å values (Table 1). Specimens from the Mesozoic cover have lower values (at around 9.000 Å), while Priabonian are centred around 9.020 Å. However, both value groups are in the intermediate pressure facies (between 9.000 and 9.040 Å) as shown in Fig. 4 (the curve obtained is similar to the reference curve from Ryoke), suggesting an inferred geothermal gradient of 25–35 °C/km (Guidotti and Sassi 1986).

Fig. 4

Cumulative curves of K-white mica b cell dimensions for 21 anchizonal samples from our study area (in bold) with reference curves from Sassi and Scolari (1974) and Guidotti and Sassi (1986)

4.3 Vein mineralogy, textural relationships and microthermometry

Samples were collected within the RYSZ, previously decribed by Butler (1992), and consisting in an imbricate of basement slices and para-autochtonous sedimentary cover. This shear zone, lying directly below the frontal penninic nappe stack, has been considered as the map trace of the RT in this area (Trullenque 2005). All kinematic indicators measured in the field or deduced from microfabric analysis of calcite ultramylonites (Trullenque et al. 2006), consistently show a WNW directed sense of transport, consistent with activity along the RT. No evidence of D₂ structures are found in this area.

Samples have been collected along strike slip fault planes dissecting basement slices and indicating a clear WNW directed sense of transport. These faults are clearly related to the first deformation phase D₁. The chronology of the different fluid inclusion populations (Table 2) is a relative one with respect to their host mineral and their overgrowth (Mullis 1976).

Table 2 Fluid inclusion data of two quartz samples from the Yret Zone, Western French Alps

Veins are mainly composed of small quartz crystals associated sometimes with calcite crystals. Only two-phase fluid inclusions (consisting of vapour and liquid at room temperature) were identified with the microscope. Fibre quartz is observed in the sample F72. In F92, two fluid inclusion assemblages are identified. The first one is a two-phase type in the host quartz and probably of primary type, i.e. trapped during mineral growth. The second one is overprinting the first assemblage and probably of pseudo-secondary type. The true temperature of entrapment of the fluid inclusions, i.e. mineral growth, was constrained by intersecting the isochores of primary two-type phase fluid inclusions with the inferred geothermal gradient deduced from the K-white b cell dimension (Fig. 5). Isochores were calculated for the lower and upper homogenization temperature (Th_i) with an average salinity of 5.6 and 13.8 wt. % NaCl equivalent for samples F72 and F92, respectively.

Fig. 5

Depth-temperature diagrams with isochores representing the higher and the lower calculated molar volumes for vapour-rich fluid inclusions and the meam isochores in samples F72 (a) and F92 (b) from the Yret zone. The dashed lines represent the range of geothermal gradients from 25 and 35 °C/km as deduced from the K-white mica b cell dimensions. The thick curve represents the liquid–gas reaction curve of water. CP critical point of water

5 Discussion

The structural metamorphic map of the Alps (Oberhänsli et al. 2004) indicates lower sub-greenschist facies E of the RT and N of the Pelvoux massif. North of the Pelvoux massif, Ceriani et al. (2003) described an E–W metamorphic gradient on the E side of the RT. In the Briançonnais, they reported an increase from anchizonal to epizonal conditions from E to W. In the same area, the Dauphinois domain (in the footwall of the RT) shows epizonal metamorphic conditions. The Dauphinois domain in this area is only affected by one deformation phase, corresponding to the third deformation phase in the FPU (Ceriani 2001; Ceriani et al. 2001) and equivalent to D₁ in our studied area (Trullenque 2005). The K-white b cell dimensions obtained by Ceriani et al. (2003) in the Dauphinois Domain are ranging between 9.000 and 9.030 Å similar to those obtained in our study, indicating similar geothermal metamorphic gradient between the two areas.

Frey (1987) and Essene and Peacor (1995) showed that KI cannot be used as a precise geothermometer and only provide an approximate temperature range. In spite of this, attempts have been made to relate KI values to absolute temperatures. This was done in different low-grade metamorphic units by comparing the KI data with other geothermometers (Ferreiro Mählmann 1994, 1996; Merriman and Peacor 1999; Mullis et al. 2002; Potel et al. 2006). Mullis et al. (2002) have suggested a diagenesis—anchizone boundary in the Alps on the basis of a comparison of IC versus FI data. This allows to better define the range of temperature that the SE of the Pelvoux massif was subjected during tectonic evolution.

The KI values indicate that the Dauphinois domain in the SE of the Pelvoux massif underwent up to low anchizonal metamorphic conditions (Table 1; Fig. 2). According to Mullis et al. (2002), the boundary is linked to a temperature of around 230–240 °C (similar to the compilation results from Ferreiro Mählmann (1994)). This can be compared to the temperature estimates obtained west of our studied area in the “grès de Champsaur” (W of Prapic). Waibel (1990) described quartz–prehnite ± calcite veins in the “grès de Champsaur”. The country rock is dark green in colour, enriched in pumpellyite of variable composition, and to a lesser extent prehnite, at the expense of laumontite. Aprahamian (1988) described also mottled sandstone (faciès moucheté), the pale-coloured mottles being enriched in laumontite as cement and as partial replacement product of albitized clastic plagioclase and the darker areas being relatively enriched in chlorite cement. According to Frey et al. (1991) and Potel et al. (2002), the coexistence of zeolites (laumontite in this case) and prehnite–pumpellyite is restricted to a rather small P–T area below 240 °C and therefore in the range obtained by using the correlation from Mullis et al. (2002).

Zircon FT data were gained from the turbiditic “grès de Champsaur” and the underlying Pelvoux crystalline basement (Seward et al. 1999). These data provide important information on the peak temperatures reached during Alpine metamorphism. For the zircon FT system, several estimates on the temperature range of the partial annealing zone exist (e.g. Yamada et al. 1995, 2007; Tagami et al. 1998). Tagami et al. (1998) suggested a temperature range between 200 and 320 °C to be appropriate, which covers the temperature range relevant to the metamorphic zonation based on illite “crystallinity” (e.g. Mullis et al. 2002). The combination of both methods therefore provides additional constraints on the maximum temperature reached to the SE of the Pelvoux massif. In the Glarus Alps, the comparison between fluid inclusion data and zircon FT data suggests that zircon FT annealing in the Taveyanne sandstone (a lateral equivalent of the “grès de Champsaur”) only becomes detectable above 250 °C (Rahn 2001). The measured K-white mica b cell dimension values inferred an intermediate geothermal gradient with 25–35 °C/km (Merriman and Peacor 1999). This gradient seems to be a very conservative approach, however the result is similar to that found in the Dauphinois domain north of the Pelvoux massif (Ceriani et al. 2003). Combining this information with the fluid inclusion data obtained in the Yret zone, a trapping temperature for the fluid inclusions of 220 ± 10 °C can be obtained (Fig. 5a, b). Consequently, the comparison of KI, FT and fluid inclusion leads to the conclusion that the SE of the Pelvoux massif was subjected to temperatures higher than 200 °C, but lower than 250 °C. This is higher than the estimation of temperature by Seward et al. (1999) who estimated that the burial temperature of the “grès de Champsaur” was not higher than 200 °C. This implies metamorphic burial between 7 and 9 km. Neither our KI nor the FT data (Seward et al. 1999) reveal the presence of a metamorphic gradient from E to W in the SE of the Pelvoux massif as described by Aprahamian (1974). The absence of epizonal KI values in the Vallouise valley (Aprahamian 1974, 1988), moves the anchizone–epizone boundary to the N of the Pelvoux massif, where epizonal metamorphic conditions are observed (Ceriani et al. 2003). As mentioned by Ferreiro Mählmann and Frey (2012), one of the reasons of the lack of epizonal values, is that the KI values obtained by CIS calibration led to a broadening of the anchizone area.

The available metamorphic data from the south of the Pelvoux massif combined with the available deformation data allow us to deduce some conclusions concerning the tectono-metamorphic evolution in this area. A first important result provided by the KI, fluid inclusion and FT data is the fact that the Dauphinois domain exhibits lower metamorphic conditions compared to the Penninic domain. According to the metamorphic map of the Alps, the Penninic domain west of the RT is metamorphosed under lower greenschist facies (Oberhänsli et al. 2004). Field evidences SE of the Pelvoux massif (this paper; Trullenque 2005), indicate that D₁ corresponds to the first penetrative schistosity, marked by very fine-grained K-white mica, chlorite, quartz and albite. The metamorphic gradient observed and the formation of the fluid inclusions studied are related to this deformation phase and the burial heating (with a maximum of 9 km burial) was generated by the overthrusted Penninic domain along the RT onto the Dauphinois domain. As mentioned above, the K-white mica b cell dimension values in our studied area SE of the Pelvoux massif, are similar to those obtained in the north by Ceriani et al. (2003). Therefore, the metamorphic gradient observed in the Penninic domain must be established before D₁, as observed by Ceriani et al. (2003) further to the N (the D₁ observed SE of the Pelvoux massif correspond to the D₃, see above). So, we have no trace of an older metamorphic event.

North of the Pelvoux massif, the metamorphic grade is higher with epizonal conditions and may be explained by thicker Penninic units on top of the Dauphinois domain leading to higher degree of burial and therefore show a regional metamorphic gradient increasing from S to N. This is corroborated by the same pressure gradient observed north and south of the Pelvoux massif in the Dauphinois domain (same range of K-white mica b cell dimension).

Ferreiro Mählmann (2001) in the eastern Alps showed that in rocks submitted to a long metamorphic event (20 Ma) and a steady state heat flow, the smectite content in the anchizone is between 0 and 5 % (90 % of the samples without smectite), and that under stabile thermal conditions (equilibrium), KI can be used as a thermometer. In this region, Henrichs (1993) reported K-white mica b cell dimension values between 9.000 and 9.022 Å and the compilation of the Ferreiro Mählmann et al. (2012) shows a normal to low geothermal gradient (30 °C/km). Comparison of our results with KI and K-white mica b cell dimension values will suggest that the Dauphinois Domain has been submitted to a relative long heating time like in the eastern Alps.

6 Conclusion

The tectono-metamorphic history of the Dauphinois domain in the SE of the Pelvoux is similar that described by Ceriani et al. (2003) in the north. Peak metamorphic conditions in the Dauphinois domain are reached during the overthrusting of the Penninic units during D₁ top-WNW directed thrusting along the RT. The metamorphism is characterized by temperatures below 250 °C and higher than 200 °C and a metamorphic gradient between 25 and 35 °C, and a long time event. This range of temperature is compatible with the FT data published by Seward et al. (1999) indicating that the burial temperature in the region was lower than 250 °C. Based on our KI values and the mineral paragenese observed in the W of the “grès de Champsaur”, no E–W metamorphic zonation as described by Aprahamian (1974, 1988) is observed. But, in the studied area, the grade of metamorphism in the Dauphinois Domain is lower than in the northern of the Pelvoux massif (Ceriani et al. 2003) confirming the N–S metamorphic zonation described by Aprahamian (1974, 1988). This is also outlined by the zircon FT ages indicating an increase in age primarily from the north to the south (Fügenschuh and Schmid 2003) and supports their hypothesis that the more southerly located areas were lees deeply buried and hence less exhumed. Like proposed by Ceriani et al. (2001) and Fügenschuh and Schmid (2003), the overthrust along the RT of the Penninic units may caused burial and metamorphism in the external Dauphinois units.