| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology | Volume 44 | Number 10 | Pages 1917-1936 | 2003
© Oxford University Press 2003; all rights reserved
Mingling of Immiscible Dolomite Carbonatite and Trachyte in Tuffs from the Massif Central, France
1 LABORATOIRE DE GÉOLOGIE, UNIVERSITÉ BLAISE PASCAL, CNRS UMR 6524 ET OPGC, 5 RUE KESSLER, 63038 CLERMONT-FERRAND, FRANCE
2 ÉCOLE NORMALE SUPÉRIEURE DE LYON, LABORATOIRE DE SCIENCES DE LA TERRE, CNRS UMR 5570, ENS ET UCBL, 46 ALLÉE D'ITALIE, 69364 LYON CEDEX 07, FRANCE
* Corresponding author. E-mail: g.chazot{at}opgc.univ-bpclermont.fr
RECEIVED JUNE 13, 2002; ACCEPTED APRIL 23, 2003
 |
ABSTRACT |
|---|
TOP
ABSTRACT
INTRODUCTIONA pyroclastic tuff from the Velay volcanic province in the French Massif Central contains blocks up to 30 cm long of local basement rocks, lava clasts, coarse-grained cumulates and pyroclastic fragments, with more or less diffuse boundaries with the host tuff, which probably represent more consolidated parts of the tuff. All of the pyroclastics examined and approximately 10% of the cumulate xenoliths contain carbonates in variable amounts, textures and mineralogy. In some of the tuff samples, dolomite occurs in large amounts (up to 57%), principally as immiscible globules in trachytic melt (now glass), and represents the first occurrence of carbonatite reported from the Massif Central. The other carbonates, magnesiosiderite in the mafic cumulates or occasionally in some tuffs, and calcite in the felsic cumulates, are always associated with a silicate glass of trachytic composition. Coexisting feldspars and carbonates in the various types of sample are approximately in Sr isotopic equilibrium with an initial ratio of about 0·7042. C- and O-isotopic compositions of the carbonates covary and cover a very wide range of composition from -2·9 to 3·9
(PDB) and from 8·7 to 24·5 (SMOW), respectively; the globular dolomites have primary igneous isotopic signatures and plot within the primary igneous carbonatite field. The combined C–O–Sr isotopic data indicate that both the dolomitic carbonatitic and silicate magmas came from a mantle source with very similar isotopic characteristics. On compositional arguments, injection of a non-cogenetic carbonatite magma into a differentiating body of felsic magma within the crust is preferred to the unmixing of a relatively late, fractionated melt to carbonatitic and trachytic magmas. The textures, high 
13C and 18O values of Mg-siderite and calcite and their trend are consistent with post-magmatic precipitation of these carbonates, probably by interaction between a CO2–H2O-bearing fluid (<20 mol % CO2) undergoing Rayleigh distillation processes and the minerals and glass in the cumulates. Although minimum isotopic temperatures are >100°C for the calcites and >275°C for the magnesiosiderites, the isotopic data are compatible with the chemical, solvus temperatures of >500°C for the calcites and >500–700°C for the magnesiosiderites, if the CO2 content of the fluid decreased during the distillation processes. The highly variable K2O/Na2O ratios of some of the glasses on the scale of a few microns may be another consequence of H2O–CO2 metasomatic processes. Explosive eruption of the two immiscible magmas that entrained the xenoliths probably occurred as a result of CO2–H2O degassing processes soon after the arrival of the carbonatitic magma in the trachytic system. KEY WORDS: dolomite; carbonatite; trachyte; stable isotopes; Massif Central; France
|
INTRODUCTION |
|---|
Despite increasing evidence for carbonate metasomatism in the upper mantle (e.g. Hauri et al., 1993; Ionov et al., 1993; Schiano et al., 1994), occurrences of carbonatite at the surface of the Earth are relatively rare and many aspects of their petrogenesis are still the subject of debate (e.g. Harmer & Gittins, 1997; Bell et al., 1999). Although they frequently occur in close association with igneous silicate rocks, the genetic link between the two types of magma is controversial. Additionally, the conditions of formation of the various types of carbonates (dolomite, calcite and Mg-siderite) and their mutual association in carbonatite complexes is not well understood. Therefore, every new discovery of carbonatite or carbonate-bearing magmatic rock presents an opportunity for a better understanding of the complex processes leading to their formation and emplacement.
The frequent association of carbonatites with silica-undersaturated silicate igneous rocks has been widely described worldwide (Bailey, 1993). However, their association with saturated, mainly trachytic, silicate rocks is not so common (e.g. Macdonald et al., 1993; Cooper & Reid, 2000). Similarly, a wide range of chemical compositions and mineralogies of carbonates has been recognized in carbonatites worldwide but only rarely associated in the same eruption.
Various mechanisms have been proposed to explain the association of carbonate and silicate magmas: (1) as independent liquids produced in the mantle that are either immiscible or remain associated but unmixed in the crust; (2) as the products of immiscibility by unmixing of a common carbonated silicate parental magma. A genetic link between silicate and carbonate magmas by unmixing of a carbonate melt in the upper part of a magma chamber has been argued for the Suswa volcano in Kenya (Macdonald et al., 1993), despite evidence for Sr–Nd disequilibrium between carbonate and silicate phases. In another study, Cooper & Reid (2000) demonstrated that trachytes and associated carbonatites from the Dicker Willem complex in Namibia were immiscible, non-cogenetic, magmas, although linked spatially and temporally. In this case, they argued that the trachyte was formed by partial melting of a high-grade fenite during uprising of the carbonatitic magma. These two contrasting studies emphasize the difficulties in interpreting silicate–carbonate relationships.
No direct evidence of carbonatite magmatism has been reported from the French Massif Central. The existence of carbonated mantle beneath the region has, however, been suggested by Liotard et al. (1995), based on geochemical studies of nephelinites from the Craponne-sur-Arzon district, NW of the Velay province (about 50 km NNW of the present study area; Fig. 1), and by Jakni et al. (1996), who studied peridotite mantle xenoliths from a basanitic to nephelinitic volcanic complex near Escandorgue (south of the Massif Central, about 150 km south of the study area), which had been metasomatized by a carbonate-rich fluid. The involvement of CO2 and carbonate-rich fluids has also been proposed to explain the presence of calcite within feldspars in a trachyte from the Velay volcanic province (Batard et al., 1977). Here we report on the occurrence of a xenolith-bearing carbonate-rich tuff sampled in one of the few examples of differentiated pyroclastic deposits known in the Velay area (Mergoil & Boivin, 1993).
|
|
GEOLOGICAL SETTING |
|---|
Cenozoic volcanism in the French Massif Central occurred in the interval 60 Ma to 6 ka, with peak activity around 10–5 Ma, and is associated with the development of the western and central European rift system (Wilson & Downes, 1991; Wilson & Patterson, 2001). The study area is located in the SE part of the Velay area (Fig. 1), where two volcanic provinces are partly superposed: (1) the eastern Velay province, which was active from 14 to 6 Ma and produced a Na-alkaline series, from basalt to trachytic or phonolitic differentiates (Mergoil & Boivin, 1993); (2) the Vivarais province, which was active during the Late Pleistocene and is characterized by homogeneous undifferentiated alkali basalts (Rochette et al., 1993), similar to the western part of the Velay province (Devès), where basaltic volcanism occurred from 7 to 1 Ma.
The pyroclastic deposit of Chabrières is located 2 km SE of Mount Mézenc. The major and upper area of exposure is a landslip lying close to the SE flank of the phonolitic dome of Chabrières. Further down, the outcrop extends more than 1 km along the Saliouse stream. The apparent thickness of the deposit decreases from the landslip (
100 m) towards the opposite NE end, in accordance with a centre of eruption presumably located just to the east of the dome of Chabrières (Hodges, 1991). The deposit is formed of a pale yellowish to brown weakly to strongly welded tuff, locally rich in blocks up to 30 cm long, which are mainly concentrated in the upper part of the exposure. The blocks consist of: (1) local basement rocks (Hercynian granites and metamorphic rocks); (2) lava clasts (basalts, trachytes and phonolites); (3) various coarse-grained rocks displaying cumulate textures; (4) various pyroclastic fragments with more or less diffuse boundaries with the host tuff. These last generally form indurated blocks, in relief, within the host tuff, and it is not always clear in the field whether they represent enclaves or more indurated parts of the host tuff. The age of this pyroclastic deposit is not well constrained, but a whole-rock K–Ar age of 8·5 ± 0·2 Ma was obtained on a cumulate xenolith from the deposit (Hodges, 1991). Similar ages (7·6–9 Ma, Mergoil & Boivin, 1993) for nearby trachy-phonolitic domes suggest the contemporaneity of these volcanic events.
All of the pyroclastics examined and 10% of the cumulate xenoliths contain carbonates in variable amounts, textures and mineralogy. This study focuses on these carbonate-bearing rocks, with special emphasis on their chemical and isotopic compositions and the mineralogy of the carbonates in relation to the nature of their host-rock xenoliths and tuffs.
|
DESCRIPTION OF THE SAMPLES |
|---|
Carbonate-bearing cumulates and pyroclastics may be divided into several types, each of them being defined by its mineralogy and/or texture.
Cumulate xenoliths
These constitute a suite from mafic to felsic in composition, with mafic samples being more abundant. Evidence for cumulate origin comes from the presence of cumulus minerals (amphibole, apatite, biotite, magnetite) with inter-cumulus crystallization of plagioclase, or, in the most mafic samples, cumulus amphibole and clinopyroxene with opaque minerals growing in the interstices. The cumulate origin of these rocks as well as their formation in an evolving magma chamber have already been studied in detail (Hodges, 1991).
(1) Mafic cumulates (CHAB 22, 32, 87, 115) contain amphibole, Fe–Ti oxides, apatite, and often plagioclase and clinopyroxene. All the minerals are fresh and in textural equilibrium. The carbonates are magnesiosiderite and occur as sparse, euhedral, tooth-shaped crystals of millimetre size and are frequently included in silicate glass in spaces between the other minerals (Fig. 2a) or in veins cross-cutting the minerals. The glass is brown and often looks altered. Small skeletal plagioclases have sometimes crystallized in the glass.
|
(2) Intermediate to felsic cumulates (CHAB 35, 104, 112) contain alkali feldspar (CHAB 112) associated with plagioclase (CHAB 35, CHAB 104) with subordinate biotite, Fe–Ti oxides, apatite and sometimes clinopyroxene. The carbonates are pure calcite and may be observed in three settings: (1) as a replacement of clinopyroxene and/or amphibole; (2) as crystals, associated with apatite, oxides and silicate glass, in spaces between feldspars or in cracks in the feldspars; (3) as rounded crystals associated with optically fresh silicate glass (Fig. 2b and c). The felsic cumulate rocks (e.g. CHAB 112) are the most differentiated among the xenolith suite.
Pyroclastics
The investigated rocks have heterogeneous textures and compositions, but always contain carbonate phases. According to the size criteria of Schmid (1981), they can be classified as tuffs (average pyroclast size less than 2 mm) or lapilli-tuffs (average size from 2 to 64 mm). They will be referred to as tuffs hereafter and can be subdivided into three groups, according to their texture and composition.
(1) Polygenic lithic tuffs (CHAB 13, 133, 134, 136) are the most heterogeneous pyroclastics and the most common in the deposit. Depending on the degree of welding, they appear either as the host tuff or as indurated blocks that mimic (but are not) xenoliths within the tuff. They contain abundant lithic and crystal clasts of older lavas and basement rocks, as well as fragments of mafic and felsic cumulates. Pieces of carbonate-rich tuffs described hereafter can also be found as rounded fragments up to several millimetres in size. The matrix of the tuff is a fine- to very fine-scale association of carbonate (dolomite) and altered silicate glass. In the tuff CHAB136, which comes from the upper part of the actual exposure, smectite (identified by X-ray diffraction) and hydrous iron oxides are abundant. In a few cases (e.g. CHAB 133), rounded to ovoid pieces (from 200 µm to 2 mm in size) made of carbonate and K-feldspar microlites have been observed; they are interpreted as lapilli (Fig. 2h).
(2) Vitric tuffs (CHAB 66, 30-78) are characterized by large amounts (up to 50%) of fresh, dark brown silicate glass displaying weak to strong banding (Fig. 2d). In places, the glass is more or less altered into smectite. This glass encloses phenocrysts such as amphibole, clinopyroxene, feldspar and biotite, and lithic or crystal clasts of lavas and basement rocks. Carbonates are rare and mainly occur as small grains with euhedral tooth-shape, and are always associated with silicate minerals (mainly clinopyroxene) surrounded by the glass. As in the mafic cumulates, the carbonates are magnesiosiderites. In these samples, feldspars are corroded and fragmented.
(3) Carbonate-rich tuffs (CHAB 40, 122) occur as xenoliths enclosed in the previous tuffs and contain more than 50% of carbonates associated with either silicate glass or silicate minerals and oxides. The silicate paragenesis (K-feldspar, plagioclase, biotite) is that of the felsic cumulates. Scarce clasts of lavas (trachyte) and basement rocks are also observed (CHAB 122). Carbonates are calcian and ferroan dolomites. In places, they occur as large crystals filling all the spaces between the other minerals and disrupting them (Fig. 2e). The fragmented feldspars are partially resorbed at the contact with the carbonates. Some biotites are strongly disrupted, exfoliated and invaded by the carbonates between the exfoliated sheets. In other places, the silicate minerals are less abundant and the rock is mainly a carbonate–silicate glass association (Fig. 2f and g). The fresh brown to more altered silicate glass (in CHAB 40 and CHAB 122, respectively) encloses abundant carbonate globules of various sizes (<100 µm to 3–4 mm). The carbonate globules may vary from rounded to ovoid and amoeboid shape, and may be coalescent or budding. The carbonate–silicate glass contacts are sharp, and curved menisci of silicate glass can be observed between the carbonate globules. Occasionally, the carbonate globules are flattened and elongated in the banding plane of the silicate glass (not shown). These geometrical relationships are similar to those produced in experiments that illustrate intermingling of two immiscible silicate and carbonate liquids (e.g. Brooker, 1998). Therefore, these carbonate-rich tuffs correspond probably to felsic cumulates strongly fragmented and invaded by a mixture of silicate and carbonate magmas.
|
ANALYTICAL METHODS |
|---|
Electron microprobe analyses, X-ray diffraction (XRD) and Sr isotopic analyses were performed at the Laboratoire Magmas et Volcans (Clermont-Ferrand).
Major element analyses were obtained with a Cameca SX 100 electron microprobe. Na and K were always counted in the first sequence to minimize alkali loss under the beam. The peak counting time was 10 s for most of the elements. For silicate minerals, operating beam conditions were 15 kV and 15 nA, with a beam size about 1 µm. The detection limits vary from 0·04 to 0·07 wt % depending on the element. For glass analyses, we used 15 kV and 8 nA for the beam, with a defocused beam of 10 or 5 µm according to the size of the glass zone. For carbonates, beam conditions were 15 kV and 6 nA with a beam size of 10 µm. Both dolomite and siderite were used as standards for carbonate analyses.
X-ray diffraction was performed on a CGR S 2080 diffractometer using 
Cu K
1 (focusing quartz-monochromator on incident beam) and numerical acquisition (100 steps/°
with 5 s counting time by step). We analysed only the carbonate-bearing tuffs in which the carbonates are abundant but sometimes too small for microprobe analyses, so as to identify all the carbonate phases present in these rocks to estimate the representativeness of the microprobe analyses. The detection limit for calcite is estimated to be about 1 or 2%.
Sr isotopic compositions were measured on a Micromass VG54E mass spectrometer using single Ta filaments. Values obtained for the NBS 987 standard ranged from 0·710226 to 0·710255 during the period of analysis.
C and O isotopic compositions were determined at École Normale Supérieure (ENS) in Lyon. The stable isotope composition of carbonates from 10 samples from three of the groups (mafic cumulates, felsic cumulates and carbonate-bearing tuffs) were determined from the CO2 liberated during reaction of the powdered whole rock with 100% H3PO4. Because of the variable chemistry and/or mineralogy of the carbonates and the impracticability of physically separating them in milligram amounts, extractions were carried out after various times of reaction with H3PO4 (1 h, 3 h and up to 13 days) at temperatures of 30°C and, for the siderites, at 150°C for 2 h, to obtain semi-quantitative chemical separations (Epstein et al., 1964; Rosenbaum & Sheppard, 1986; Al-Aasm et al., 1990). In principle, if calcite is present in the sample, it is essentially completely reacted after 1 h (Epstein et al., 1964). However, for samples dominated by dolomite or Mg-siderite, the 1 h extractions are referred to as carbonate, if no calcite was evident from other techniques, because they must contain a major contribution from the dominant dolomite or Mg-siderite. Yields were measured at each step. Carbonates were not analysed from the vitric tuffs because their abundance was so low. The 13C and 18O values in per mil are reported relative to PDB and SMOW, respectively, using the CO2–carbonate fractionation factors for oxygen given by Rosenbaum & Sheppard (1986).
|
COMPOSITION OF THE CARBONATES |
|---|
The carbonate phases have been studied both by XRD and electron microprobe. Their nature and chemical composition are variable and directly correlated with the rock type in which they occur (Table 1 and Fig. 2). According to the standard classification of rhombohedral carbonates (Goldsmith, 1990; Reeder, 1990; Chang, 1998), the analysed carbonates belong to the calcite group, the dolomite group and the magnesite–siderite series (Fig. 3).
|
|
The magnesiosiderite occurs in small amounts (<3%; see Table 4, below) in the mafic cumulates and sometimes in the vitric tuffs. It is often euhedral and always in contact with silicate glass, in which apatite, small skeletal feldspars and Fe–Ti oxides have crystallized. Compositions range widely from Mg-rich to Fe-rich crystals, which are all named magnesiosiderite here. Most of the samples contain only magnesiosiderite of intermediate composition, with Mg/(Mg + Fe) ranging from 0·37 to 0·50. Such intermediate compositions (but with lower Ca contents) have been described by Buckley & Woolley (1990) from a number of carbonatites. In one sample (CHAB 87), two compositions are present: (1) magnesiosiderite with high Mg/(Mg + Fe) values (0·60–0·77) and low Ca content (CaCO3 component between 3·5 and 9%); (2) magnesiosiderite with low Mg/(Mg + Fe) values (between 0·04 and 0·12) and higher Ca content (CaCO3 component up to 17%). The MnO content ranges from 0·15 to 1·68% and is not correlated with any other major element in these carbonates.
|
The highest CaCO3 content is generally found in the magnesiosiderites of intermediate composition (Fig. 3). In their study of phase equilibrium in the CaCO3–MgCO3–FeCO3 system, Anovitz & Essene (1987) showed that the composition field of Fe–Mg-siderites expands with rise in temperature; the higher the temperature, the higher the Ca content. At temperatures near 700°C, siderite can contain more than 15% CaCO3 component. The high CaCO3 component contents of the magnesiosiderites analysed in this study are thus indicative of relatively high temperatures of formation, at least higher than 500°C and most probably around or above 700°C (Fig. 3).
Calcite is the only carbonate phase present (in small amounts; see Table 4) in the felsic cumulates. It may also occur in the polygenic tuffs (such as CHAB 13) but is located in clasts of felsic cumulates entrained in the tuff. The calcites have very low MgO and FeO contents (Fig. 3) but variable MnO content ranging from 0 to 3 wt % and positively correlated with FeO. Their usually very low MgO contents prevent the general application of the calcite–dolomite solvus thermometer (Anovitz & Essene, 1987); however, two samples that plot close to the calcite–dolomite join (Fig. 3) imply minimum temperatures around 500°C.
Dolomite is the most common carbonate phase present at Chabrières. It occurs only in the carbonate-rich tuffs, where it can represent more than 50% of the rock, and is often associated with silicate glass. It can be Ca-rich or Fe-rich dolomite (Fig. 3, Table 1) and both coexist (as confirmed by XRD) in the same samples. No systematic study has been performed to locate the two types of dolomite, but in some places Ca-rich dolomite forms globules of millimetre size, surrounded by a matrix formed of intimately mingled silicate glass and Fe-rich dolomite. In other places, mingling of the two phases at a very fine scale does not allow a precise survey of the composition of the carbonate in the matrix.
In the Ca-rich dolomite, expressed as Ca(Mg1–xCax) (CO3)2, the maximum x value for the Ca 
Mg substitution is 0·24. Calcite is not detected by XRD, indicating that Ca-dolomite is not a cryptocrystalline intergrowth of dolomite sensu stricto and calcite: this carbonate is a calcian dolomite. CHAB 40 contains dolomite globules that are low in iron (Table 1) and therefore plot very close to the CaMg(CO3)2–CaCO3 join of Fig. 4. Application of the experimentally determined calcite–dolomite solvus to the dolomite compositions [data summarized by Anovitz & Essene (1987)] gives estimated temperatures of 1000–900°C for these calcian dolomites.
|
For the Fe-rich dolomites [Ca(Mg1–yFey)(CO3)2], there is a usual limit between ferroan dolomite and ankerite at y = 0·20 (Chang, 1998). Fe-rich dolomites from Chabrières mainly cover the compositional field of ferroan dolomite and slightly overlap the ankerite field (Fig. 3), with y values up to 0·32. They are all named ferroan dolomites here, as also are the Ca–Fe-rich dolomites from CHAB 13 where both Ca and Fe are substituting for Mg.
MnO content in the dolomite is variable, and is generally higher in the ferroan dolomites than in the Ca-rich dolomites. It ranges from 0 to 1·6% and is well correlated with the FeO content. In general, dolomites have lower MnO content than calcites and magnesiosiderites.
Sr content has been analysed by electron microprobe in the various types of carbonates. In dolomites from CHAB 40, SrO content ranges from 0·024 to 0·054 wt %. In the other carbonates, Sr content is very low: always below the detection limit for the magnesiosiderites in the sample CHAB 115, and from below detection limit to 0·014 wt % SrO in the calcites from sample CHAB 104.
|
COMPOSITION OF THE SILICATE PHASES |
|---|
Feldspars of volcanic origin vary widely in composition. In the mafic cumulates they are labradorite to andesine in composition and are in textural equilibrium with the other minerals. In some samples, quenched skeletal crystals of andesine have grown in the glass.
In the felsic cumulates and in all the tuffs, the feldspars are mainly ternary and have similar ranges of composition. The crystals are not zoned and are often corroded in the tuffs, where in contact with glass or carbonates. The composition (in mol %) of the potassic oligoclase ranges from An29Ab63Or8 to An20Ab68Or12 and the composition of the calcic anorthoclase ranges from Or19Ab68An13 to Or32Ab63An5. Compared with the carbonate-bearing tuffs in which some andesine crystals also occur, the felsic cumulates and the vitric tuffs contain mainly Ca-anorthoclase and also some sodic sanidine crystals.
Ternay feldspars are typically observed in the trachytic lavas from the Velay province (Batard, 1974) as microlites, phenocrysts or large cores inside sanidine phenocrysts. As observed in these lavas (Batard, 1974), the FeO content of the feldspars in the Chabrières samples decreases from the basic plagioclases (0·9 to 0·4 from labrador to andesine) to the ternary feldspars (0·3 in K-oligoclase and 0·2 in Ca-anorthoclase). This is consistent with the work of Hodges (1991) relating all the cumulates from Chabrières to a differentiation series produced by fractional crystallization.
Pyroxenes can be subdivided into two groups, Al-augites (with up to 9·7% Al2O3) and low-Al pyroxenes (<2% Al2O3). The Al-augites are present in the mafic cumulates, are rich in TiO2 and are typically associated with amphiboles. They are also present and partially replaced by carbonates in the tuffs. Low-Al pyroxenes are found in the felsic cumulates and are enriched in Fe and Mn compared with the Al-augite group.
Amphiboles occur in the mafic cumulates and in the vitric tuffs. They are homogeneous kaersutites or silica-rich kaersutites according to the classification of Leake et al. (1997).
Other minerals present include titaniferous magnetite with up to 14% TiO2, and biotite, which occurs only in the felsic cumulates and in the tuffs, and has a high Ti and Fe content.
|
GLASS COMPOSITIONS |
|---|
Some electron microprobe analyses of glasses do not sum to 100%. The difference can probably be interpreted in terms of water or CO2 content except those (mostly not presented here) with very low totals (<95%). For these latter analyses, higher contents of CaO in the glasses from the felsic cumulates, and of FeO and MgO in the mafic cumulates and the vitric tuffs suggest the presence of carbonates at the micron scale.
Quenched silicate glasses are present in both cumulates and tuffs, and are frequently associated with carbonates and other minerals (e.g. feldspars, oxides, apatite). The analysed glasses are all silica rich, with SiO2 contents ranging from 57 to 66% (Table 2, Fig. 4a). In the total alkali–silica diagram (Fig. 4a), the glasses plot mainly in the trachyte field and show a large range in total alkali contents (from 8 to 16%) for a more restricted range of silica variation. They are divided into two groups: the first group (Na2O + K2O <13%) displays the same range of total alkali contents as the Velay trachytes sensu lato, whereas the second group has higher alkali contents (Na2O + K2O >13%) and straddles the boundary between the trachyte and the phonolite fields. This second group is slightly shifted away from any known composition of differentiated lavas in the Velay province. All the glasses from the mafic cumulates belong to the first group, whereas most of the glasses from the felsic cumulates belong to the second group. The glasses from the tuffs are distributed within the two compositional fields.
|
Both Na2O and K2O contents are highly variable, with Na2O ranging from 2·13 to 8·08% and K2O ranging from 1·24 to 12·90%. These variations are not correlated with the inferred water and CO2 content of the glasses. The K2O/Na2O ratios of the glasses are also highly variable, and when plotted against their Na2O contents define two trends (Fig. 4b), corresponding respectively to the two groups identified in Fig. 4a: the low-alkali group has a low and rather constant K2O/Na2O ratio, whereas the high-alkali group displays a very variable K2O/Na2O ratio, which is negatively correlated with the Na2O content. The low K/Na ratio group, dominated by glasses from the mafic cumulates and the vitric tuff CHAB 30-78, is similar to the compositions of trachytes from the Velay province (Mergoil & Boivin, 1993). These glass compositions thus probably represent the compositions of the melts in the magma chamber at the time of the pyroclastic eruption. The glasses from the second group (high K/Na ratios) can be found in close association with low-K/Na glasses, at a scale of a few microns, in felsic cumulates and tuffs. Their compositions deviate from those of the Velay trachytes and are not representative of any magma erupted in this province. The observed trend, characterized by sodium loss and potassium gain (up to twice the highest K2O content observed in the trachytes) is reproducible (and not an analytical bias) and may reflect processes remobilizing the alkali content. These modifications may represent local resorption of small alkali feldspar crystals, or a selective interaction between the magma and a K-enriched fluid phase, as previously proposed by Villemant & Treuil (1983). In the latter case, this fluid phase interacted preferentially with the felsic cumulates in the magma chamber (upper zone of the chamber?). The first hypothesis is favoured by the close superposition between this glass trend and the compositional range of alkali feldspars (Fig. 4b).
The similarity in composition between the unmodified glasses and the differentiated lavas from the eastern Velay province strongly suggests that the pyroclastic flow of Chabrières is both coeval and cogenetic with this Miocene volcanic province. A younger age is precluded, because the Quaternary Vivarais province did not produce differentiated lavas.
|
Sr–C–O ISOTOPIC DATA |
|---|
Sr isotopes
Sr isotopic compositions have been determined for carbonates and, when possible, biotite and feldspar in one mafic cumulate (CHAB 115), one felsic cumulate (CHAB 104) and two carbonate-rich tuffs (CHAB 40 and CHAB 122), to assess the degree of isotopic equilibrium among the mineral phases. The results are presented in Table 3. The Sr isotopic composition of associated carbonates and feldspars can be compared in three samples: one mafic cumulate, one felsic rock and one carbonate-bearing tuff. In samples CHAB 40 and CHAB 104, carbonates and feldspars have 87Sr/86Sr values ranging from 0·70392 to 0·70416 and 0·70409 to 0·70411, respectively, which are considered to be in Sr isotopic equilibrium with each other. Furthermore, carbonates in another tuff (CHAB 122) have an 87Sr/86Sr value of 0·70432, very close to the other samples. The only observed difference between carbonate and feldspar is in the mafic cumulate, in which the feldspars have a slightly lower 87Sr/86Sr ratio than the other samples (0·70353) whereas the carbonates are significantly more radiogenic (0·70511). Except for this sample, in which the carbonate appears texturally to have formed after the crystallization of the main silicate minerals, a genetic link between carbonates and feldspars can be inferred. As the feldspars and carbonates are approximately in Sr isotopic equilibrium and both are assumed to have low Rb/Sr ratios, we can consider their Sr isotopic ratio as approximately representative of the initial value. Biotites have radiogenic 87Sr/86Sr > 0·708 as a result of their high Rb/Sr ratios. In the tuff CHAB 40, the trace element composition of the biotite has been determined on separated grains by inductively coupled plasma mass spectrometry (ICP-MS), indicating an Rb/Sr ratio of 12·57. Assuming an age of 8·5 Ma, as measured by Hodges (1991) on a cumulate xenolith from the same outcrop, gives an initial Sr isotopic composition of 0·70404, indicating Sr isotopic equilibrium among carbonate, feldspar and biotite in this rock at the time of its formation.
|
C and O isotopes
The stable isotope data are summarized in Table 4, and Fig. 5 shows the relationship between the
18O and 13C values. Data on the carbonate extracted from reaction with H3PO4 for 1 h are given in Table 4 only if (1) the yield represents more than 1% of the total carbonate in the sample, and (2) both the 13C and 18O values are different from those of the major carbonate present. Thus such data are not considered for the mafic and felsic cumulate xenoliths. For the carbonates as a whole, 13C values range from -3·3 to +3·9 and 18O values range from 8·6 to 24·5; except for CHAB 136, 13C correlates positively with 18O with a slope of about 0·4 (Fig. 5). The low-13C and low-18O end of the range is defined by the two dolomite-rich (50–60%) xenoliths where much of the dolomite is globular. The high-13C and high-18O end of the correlation is constrained by the three calcite-bearing felsic cumulates. The magnesiosiderites from the two mafic cumulates and the two dolomite-bearing (<20%) tuffs plot in an intermediate position. The minor carbonate fraction (calcite and/or dolomite) from three of the four dolomitic samples, measured after the first hour of the phosphoric acid reaction (Table 4), systematically has both higher 13C and 18O values than the associated abundant dolomite; these data also plot within the 13C and 18O trend defined by all the carbonates except CHAB 136. Of these three samples available, chemical (Table 1) and XRD data support the presence of minor calcite in CHAB 13. Although not confirmed by XRD or other non-isotopic data, these isotopic data indicate the possible presence of trace to minor quantities of calcite in CHAB 40 and CHAB 122.
|
The two dolomite-rich xenoliths CHAB 40 and CHAB 122, which contain more than 50% carbonate, can be considered as carbonatites, according to the criteria of Le Maitre et al. (1989). Also, isotopically, these dolomites, the low-
13C–18O end-member of the carbonate data array, plot (Fig. 5) within the primary igneous carbonatite field of Taylor et al. (1967) as modified by Sheppard & Dawson (1973) to include additional carbonatites, including those from both subvolcanic and plutonic associations (e.g. Bukusu, Chilwa, Mbeya, Phalaborwa, Tororo), with no petrographic or other geochemical evidence for post-magmatic modifications of the samples analysed. Although a field with an upper 13C boundary at -5 (Taylor et al., 1967) or -4 (Deines, 1989) is often used, the larger range of 13C values and an upper boundary at -2 of the primary igneous carbonatite field used here is supported by the data on carbonatites in general [see figs 13.25 and 13.27 in the review by Deines (1989)] and data on, for example, the ankerite carbonatite from Swartbooisdrif, Namibia by Thompson et al. (2002).
Of the three polygenic tuff samples, CHAB 134 and CHAB 13 follow the 13C–18O trend with a slope of about 0·4, but CHAB 136 appears to be apart both mineralogically, being dominantly calcitic, and isotopically. It comes from near the top of the main exposure close to the present surface and is markedly limonitized. The 13C values of its calcite are, however, similar to that of the non-limonitized tuff CHAB 134, as if its 18O value records a subsequent modification process. Thus the isotopic data as a whole (Fig. 5) are comparable with those for carbonatites in general (Deines, 1989).
|
DISCUSSION |
|---|
Any explanation of the petrogenesis of the carbonates needs to take into account the following observations:
- the composition of the carbonate is linked to the mineralogy and texture of the tuff or the host xenolith.
- The carbonates span a large range of chemical compositions (dolomite, Mg-siderite and calcite). In most cases, and independent of their composition, the carbonates are associated with silicate glass of trachytic composition.
- In the dolomite-rich (>20%) samples, dolomite always occurs, in part, as globules in the silicate glass (Fig. 2f and g), consistent with the association of two immiscible liquids.
- The Mg-siderites with their high Ca contents formed at temperatures >500°C, and most probably around or above 700°C. At least some of the calcites formed at >500°C.
- The Sr isotopic compositions of associated carbonate and silicate phases are consistent with equilibrium.
- The carbonates as a whole have a very wide range of 13C and 18O values, which define a systematic trend. The low-13C–18O end-member of the globular dolomitic samples is within the primary igneous carbonatite ‘box’, consistent with a dominantly mantle origin for these carbonates.
Magmatic processes
Although carbonatites are often associated with silica-undersaturated rocks, the carbonates in Chabrières are clearly associated with the emplacement of a silica-saturated trachyte. This is illustrated by the chemical composition of the glasses found in the various samples (Fig. 4a and b) and by the mineralogical and chemical composition of the cumulate series found in the pyroclastic tuff (Hodges, 1991). This carbonatite–trachyte association is unusual and is discussed further below.
For the Chabrières volcanic tuffs, the presence of globules of dolomite within silicate glass suggests that two immiscible liquids coexisted. However, this immiscibility does not prove that the two liquids formed from the same parental magma; they may have been generated from two different sources and then became mingled without being able to homogenize. The 13C–18O data (Fig. 5) are consistent with the dolomite globules in the xenoliths being primary magmatic carbonate crystallized from a carbonatite magma derived from the mantle or that had equilibrated at high temperature with a mantle-derived silicate magma (Taylor et al., 1967). Although the C- and O-isotope data cannot be used to determine whether there was necessarily a cogenetic relationship between the carbonate and silicate magmas, they do argue, together with the similar Sr isotopic compositions of the carbonates (Table 3), which crystallized in association with trachytic silicate glass in various types of host rock, for a common source for the three types of carbonates. The basalts and phonolites from the Velay province also have mantle-like 18O values between 5·2 and 6·6 (Javoy, 1970).
None of the other 18O/16O ratios of carbonates, all with 18O > 10 and 13C higher than -2, could have equilibrated isotopically with this mantle reservoir at magmatic temperatures, based on the available experimental fractionation data (e.g. Chacko et al., 1991; Clayton & Kieffer, 1991). Equally, the carbonates as a whole do not record isotopically that they are part of an assemblage produced during fractional crystallization processes of a primary carbonatite (± silicate) magma. Also, in the absence of a known high-13C–18O reservoir such as sedimentary carbonates, either in the region or as xenoliths, the 13C–18O trend cannot reflect mixing or contamination processes between a high-13C–18O end-member and the low-13C–18O magmatic end-member near the site of emplacement. Such a mixing hypothesis is also not supported by the Sr isotope data (Table 3) and is not discussed further.
The discussion starts with the premise that the low-13C–18O end-member of the C–O data array is fundamental to the evolution of the isotopic systems and represents our best guide to the initial isotopic composition of the magmatic carbonate system before later processes operated. Dolomites of CHAB 13 and CHAB 134 plot just outside the carbonatite box, but represent particularly complex tuffs with a variety of carbonate textures.
Two possibilities can be envisaged for the formation of an immiscible trachyte–carbonate magmatic association: (1) unmixing of a carbonated silicate magma in a crustal chamber to give the associated dolomitic carbonatite and felsic (trachyte) magmas; (2) the intrusion of one magma into another without mixing (i.e. without homogenization). The apparent Sr isotopic equilibrium between the silicate and carbonate phases argues for either a cogenetic origin of the two magmas (Bell & Blenkinsop, 1989; Riley et al., 1996, 1999; Harmer & Gittins, 1998), or that the magmas were generated from mantle sources with essentially identical Sr isotopic compositions and their isotopic compositions remained unmodified during subsequent differentiation processes. If the trachytic magmas were produced by crystal fractionation from a more basic parent magma, as argued by Villemant & Treuil (1983) for the volcanic rocks of the Velay province, then the carbonatite melts must have been produced by immiscibility from the silicate magma at the end of its differentiation sequence. The formation of carbonatites, and especially magnesian carbonatites, by immiscibility from a silicate melt has often been debated, but remains contentious [see review by Harmer & Gittins (1997)]. This implies a very unusual chemical composition (high Mg and Ca contents) for the magma immediately before unmixing.
Many experimental studies have demonstrated that partial melting of a carbonated spinel lherzolite in the mantle can produce magnesian dolomitic melts (Wallace & Green, 1988; Dalton & Wood, 1993; Sweeney, 1994; Harmer & Gittins, 1997). In the case of Chabrières, such a liquid may have been produced from a mantle source with an Sr isotopic composition similar to the source of the primitive silicate magmas. If this carbonatitic magma escaped from the mantle, it could have ascended through the crust and been injected into a high-level chamber containing trachytic magma. As observed in the dolomite-bearing samples, the two liquids did not mix. This second possibility is the preferred scenario because it can explain why the carbonate magma was dolomitic in composition and it does not require the presence of a chemically unusual silicate magma that is not represented in either the volcanic edifice or the xenoliths.
Both the chemical data on the glasses and associated dolomites (Tables 1 and 2), and the isotopic compositions of the dolomites indicate that the tuff samples and many xenoliths record multiple processes of which only a part can be strictly magmatic. For example, in CHAB 40, a xenolith with dominantly ‘primary magmatic’ dolomite (Fig. 5), the contents of Na2O, K2O or CaO in the glass are not only variable, but are approximately bimodal, with the high-CaO, high-Na2O, low-K2O end-member being indistinguishable from unaltered Velay trachytes in general (Fig. 4b). Similarly, the principal dolomites in CHAB 40 and 122, with dominantly globular textures, have higher CaO and MgO but lower FeO contents than CHAB 13, with rare globular dolomite, which plots far to the right of the carbonatite box in Fig. 5. The magmatic signatures of the Chabrières system seem to be best preserved in the carbonatitic xenoliths (>50% carbonate).
Late and post-magmatic processes
Carbonates with C–O isotopic compositions plotting outside the primary carbonatite box (Fig. 5) must have been affected by late and/or post-magmatic processes. Late magmatic processes refer here to the reaction between a magmatic fluid, liberated during the unmixing or degassing processes in an associated magma, with the trachytic magma chamber with its mafic and felsic cumulates. This magmatic fluid is taken to be CO2–H2O bearing and, at least initially, in C- and O-isotope equilibrium with the carbonatite magma. Potential post-magmatic processes include the interaction between the tuff system and circulating meteoric waters during the cooling history and during subsequent supergene alteration.
A number of temperature and composition constraints can be placed on this magmatic carbo-hydrothermal fluid model if the system is treated isotopically as a simple mixture of CO2 and H2O, and the magma is modelled as mineral carbonate. This is justified because (1) experimental carbon isotope fractionations in the system CO2 vapour, silicate and carbonate melt (Mattey et al., 1990; Mattey, 1991), and (2) measured fractionations between gases (49 wt % CO2 and 49 wt % H2O) and coexisting carbonatitic magma (Javoy et al., 1989), are so similar to the experimental CO2–calcite fractionation data (Chacko et al., 1991; Scheele & Hoefs, 1992; Rosenbaum, 1994), or their values modified for dolomite (Sheppard & Schwarcz, 1970), over the temperature range 1400–500°C. Thus the above fractionation data are applied together with those for silicate systems (e.g. Clayton & Keiffer, 1991) and calcite–H2O (O'Neil et al., 1969). Also, calcite is taken to represent the carbonates because both the quality and temperature range of experimental data are far superior to those for dolomite or siderite, and the C- and O-isotope fractionation factors between dolomite and calcite are very small (<0·5 for T > 500°C) compared with those between CO2 or H2O and carbonate.
The slope of the trend of increasing 13C with increasing 18O for the carbonates, anchored in the primary igneous carbonatite box, is about 0·4 (Fig. 5) and is the same as the mean value of the range (0·4 ± 0·2) observed in many carbonatites worldwide for 18O values between the primary igneous carbonatite box and 14 (Pineau et al., 1973; Deines, 1989). It implies that the carbon and oxygen isotope compositions of the carbonates are somehow coupled. Although Deines (1989) has emphasized that these trends are too systematic to be coincidental and therefore they must reflect a fundamental process characteristic of carbonatite formation, there is no generally accepted process to explain them. The data presented here do not change this situation.
For the isotopic system carbonate–CO2–H2O, the fluid is overwhelmingly dominated by CO2 for carbon but the oxygen system is very sensitive to the CO2/H2O ratio for T < 700°C because, relative to carbonate, CO2 is enriched in 18O whereas H2O is depleted in 18O. Carbon isotope temperatures can thus be estimated from the CO2–carbonate fractionations, which are independent of the unknown CO2/H2O ratio of the fluid, and the experimental fractionation data (Chacko et al., 1991). The 13C value of the initial carbo-hydrothermal fluid is taken to have 13CCO2 = +0·5, i.e. CO2 in equilibrium with the carbonatite (dolomite or its magma) at about 800°C. Estimates are 275°C for the Mg-siderites, 200°C for CHAB 13 dolomite and 140–100°C for the calcites of CHAB 35, 104 and 112. These model isotopic temperatures are minimum values because the 13C/12C ratio of the fluid is assumed to be constant and therefore buffers the system. If the system is not buffered by the fluid and the mass of CO2 in the fluid decreases during carbonate formation, then the 13C–CO2 value would increase for all temperatures higher than about 175°C as a result of a Rayleigh distillation effect. The higher solvus temperatures of the Mg-siderites of >500°C argue in favour of a Rayleigh-type degassing process. Using a temperature of 500°C and the 13C value of the siderites (Table 4), a new 13C-CO2 value can be calculated. If this CO2 was also responsible for the calcites in the felsic xenoliths then the revised isotopic temperatures of their formation are 200-140°C. These temperatures are essentially independent of the temperature of formation of siderite between 350 and 800°C. Again, these isotopic temperatures are minimum values because the calcites in the felsic cumulates could have formed later than the siderites in the mafic cumulates and thus from a CO2 even more isotopically modified by Rayleigh distillation processes. Thus the higher temperatures derived from the chemical thermometers argue in favour of the 13C-CO2 value increasing on going from Mg-siderite to calcite, and thus a Rayleigh distillation type process with a limited reservoir of CO2. The 18O/16O ratios of the carbonates can also be explained by this Rayleigh distillation model of a magmatic CO2–H2O fluid at these same temperatures. However, because this isotopic ratio depends on temperature and the chemical (CO2/H2O) and isotopic composition of the fluid, all of which are not known with any precision, further analysis is not warranted here except to state that the CO2/H2O ratio has to be low (<20 mol % CO2) to account for the high 18O values.
Carbonates of CHAB 136 are interpreted to have formed initially under conditions comparable with those for CHAB 134. Subsequently, it underwent limonitization and isotopic exchange with meteoric waters, essentially free of carbon, either under late hydrothermal conditions during the cooling history or even during supergene alteration processes (the latter hypothesis would be supported by the upper position of CHAB 136 in the landslip exposure). From the available evidence, such late-stage processes, which have been reported from many other carbonatites (Deines, 1989), do not seem to have significantly affected the majority of our samples.
Except for the principal globular dolomite of the carbonatite tuffs (CHAB 40 and 122), formation of the other carbonates, including the trace carbonates of these two samples, occurred at high but sub-magmatic temperatures as indicated by the compositions and textures of the carbonates, including their presence as veinlets, small local variations in their chemical compositions and their association with optically and chemically altered silicate glass and pyroxenes. Modification of the glasses could represent closed-system post-formation exchange between the associated carbonate and glass during the cooling history of the system. This type of process is supported by the relatively rapid exchange between carbonate and silicates on a laboratory time scale, a property exploited for the experimental calibration of many isotopic fractionation factors (e.g. Chacko et al., 2001).
Eruption of the Chabrières pyroclastic flow probably occurred just after the injection of the carbonatite magma into the high-level trachyte magma chamber. Perhaps it was this event that triggered explosive eruption by the release of large amounts of H2O–CO2 volatiles during decompression including the rapid increase in specific volume of CO2-rich fluids (Kennedy & Holser, 1966) at depths of 2–3 km. In this context, it should be noted that the Chabrières pyroclastic deposit represents the best preserved example of an explosive trachytic eruption and the first occurrence of associated carbonatite in the Velay province. The observations and data are best explained by the carbonatitic and initial silicate magmas being generated from an isotopically similar mantle source, but the felsic magma, with a crustal differentiation history, was not cogenetic with the carbonatitic magma. They thus do not satisfy the basic requirement of the immiscibility hypothesis. The apparent immiscibility between the two magmas, observed in the tuffs as an emulsion in which carbonates form globules in the differentiated silicate magma, reflects their inability to mix under crustal conditions.
Well-documented CO2-rich mineral springs are widely associated with the volcanic regions of the French Massif Central (e.g. Fouillac, 1983; Matthews et al., 1987; Marty et al., 1992; Arthaud et al., 1994). In the Velay area, these springs are more closely associated with the Quaternary (Vivarais province) rather than the Miocene (eastern Velay province) volcanic rocks. No such CO2-bearing springs are known within 7 km of Chabrières.
The origin of the CO2, which may constitute >95% of the gas phase, has been debated for a long time. However, from rare gas data (3He/4He), Matthews et al. (1987) and Marty et al. (1992) have identified a dominant mantle He contribution in certain springs. These springs have 13C values of -4·1 to -6·1 and are therefore within the typical mantle range. However, as emphasized by Matthews et al. (1987), this does not prove that the carbon comes only from the mantle. A bulk crustal carbon provenance cannot be excluded because its range of 13C values must be similar to that of mantle CO2. Nevertheless, for these springs in the Massif Central, Matthews et al. presented good arguments based on the C/3He ratios that possibly up to 50% or more of the carbon comes from the mantle. We note, however, that the 13C values of these, at least in part, mantle-derived CO2 gases are very different from those inferred to have been associated with the Chabrières carbonatite (+0·5). These differences in values are consistent with the proposed magmatic provenances of the CO2. The Late Pleistocene volcanism of the Vivarais (Fig. 1) is composed only of undifferentiated alkali basalts (Rochette et al., 1993). This is in marked contrast to the wide range of magmas from alkali basalt to differentiated phonolites and trachytes of the eastern Velay Miocene volcanism. Thus the tectonic settings of these two volcanic provinces are distinct and the different values are considered to be another reflection of these magmatic differences.
|
CONCLUSION |
|---|
The eruption of the Chabrières trachytic tuff is considered to have taken place at 8·5 Ma, contemporaneous with the main period of eruption of phonolites and trachytes in the eastern Velay volcanic province. Mafic and felsic cumulates occur as blocks within the tuff. A minority of the xenoliths, but all of the tuff samples, contain carbonates with various chemical and mineralogical compositions and textures. Only some tuffs preserve textural evidence that the carbonate was present as an immiscible liquid in close association with a trachytic magma, now a glass. The similarity of the Sr isotopic composition of both feldspars and carbonates indicates that the parental silicate magma and the carbonatitic magma, with a primary igneous carbon and oxygen isotope composition, originated from an isotopically similar mantle source. Noting the dolomitic composition of the carbonatitic magma, an origin by unmixing of a late-stage differentiate that must have been high in Mg is considered unrealistic. Intrusion of a mantle-derived dolomitic liquid into a trachytic magma body is more consistent with all of the currently available data. The carbonatite and silicate magmas did not mix, resulting in the immiscible relationship between the two magmas.
The chemistry of the carbonates in the mafic and felsic cumulates and tuffs that have neither primary igneous magmatic isotopic compositions nor textural evidence for having been liquids suggests that they crystallized at high temperatures between 900°C and >500°C. Using these temperatures, the 13C–18O trend displayed by these carbonates (Fig. 5) is interpreted to reflect their formation from a magmatic carbo-hydrothermal fluid whose evolving C-isotope composition was principally controlled by a Rayleigh distillation type process. The high 18O values of, in particular, calcite in the felsic cumulates requires that the CO2/H2O ratio of the fluid was low (<20 mol % CO2). A calcite-bearing tuff from near the present surface records its formation either from a meteoric–hydrothermal fluid or during supergene alteration, and implies that most of our samples were not affected by such late-stage post-magmatic processes.
Eruption probably occurred very shortly after the arrival of the carbonatitic magma in the trachytic magma chamber but probably after the formation of the post-magmatic carbonates in the cumulates. This eruption generated a pyroclastic deposit with a carbonated silicate matrix containing occasional carbonate-bearing cumulates and carbonatite tuff xenoliths.
|
ACKNOWLEDGEMENTS |
|---|
We would like to acknowledge Frances Wall, Teal Riley and an anonymous reviewer, as well as Marjorie Wilson for their careful and very constructive reviews, which greatly improved the paper. We are also very grateful to Ken Bailey and Peter Bowden for encouraging support and fruitful discussions. Michèle Veschambre, Chantal Bosq and Jean-Luc Devidal are thanked for their help in performing microprobe, Sr isotopic and XRD analyses, respectively. We are grateful to Pierre Rochette for his hospitality and stimulating discussions during reconnaissance fieldwork. ANDRA and the École des Mines de Fontainebleau are thanked for their financial support to geochemical work in the Massif Central.
|
REFERENCES |
|---|
Al-Aasm, I. S., Taylor, B. E. & South, B. (1990). Stable isotope analysis of multiple carbonate samples using selective acid extraction. Chemical Geology 80, 119–125.[Web of Science]
Anovitz, L. M. & Essene, E. J. (1987). Phase equilibria in the system CaCO3–MgCO3–FeCO3. Journal of Petrology 28, 389–414.
Arthaud, F., Dazy, J. & Grillot, J.-C. (1994). Distribution of deep carbon dioxide in relation to the structure and tectonic evolution of south-east France. Geodinamica Acta 7, 86–102.[Web of Science]
Bailey, D. K. (1993). Carbonate magmas. Journal of the Geological Society, London 150, 637–651.
Batard, F. (1974). Les feldspaths dans les roches volcaniques différenciées du Massif Central français. Thèse 3ème cycle, Université de Clermont-Ferrand, 176 pp.
Batard, F., Mergoil-Daniel, J. & Mergoil, J. (1977). Calcite et rôle possible du CO2 dans la genèse des roches hyperalcalines et agpaïtiques du Velay oriental (Haute-Loire, Massif Central français). Bulletin de la Société Française de Minéralogie et Cristallographie 100, 343–347.
Bell, K. & Blenkinsop, J. (1989). Neodymium and strontium isotope geochemistry of carbonatites. In: Bell, K. (ed.) Carbonatites: Genesis and Evolution. London: Unwin Hyman, pp. 70–88.
Bell, K., Kjarsgaard, B. A. & Simonetti, A. (1999). Carbonatites—into the twenty-first century. Journal of Petrology 39, 1830–1845.
Brooker, R. A. (1998). The effect of CO2 saturation on immiscibility between silicate and carbonate liquids: an experimental study. Journal of Petrology 39, 1905–1915.[CrossRef][Web of Science]
Buckley, H. A. & Woolley, A. R. (1990). Carbonates of the magnesite–siderite series from four carbonatite complexes. Mineralogical Magazine 54, 413–418.[Web of Science]
Chacko, T., Mayeda, T. K., Clayton, R. N. & Goldschmidt, J. R. (1991). Oxygen and carbon isotope fractionations between CO2 and calcite. Geochimica et Cosmochimica Acta 55, 2867–2882.[CrossRef][Web of Science]
Chacko, T., Cole, D. R. & Horita, J. (2001). Equilibrium oxygen, hydrogen and carbon isotope fractionation factors applicable to geological systems. In: Valley, J. W. & Cole, D. R. (eds) Stable Isotope Geochemistry. Reviews in Mineralogy and Geochemistry 43, 1–81.
Chang, L. L. Y. (1998). Carbonates. In: Chang, L. L. Y., Howie, R. A. & Zussman, J. (eds) Rock Forming Minerals, Non silicates. London: Geological Society, pp. 95–234.
Clayton, R. N. & Kieffer, S. W. (1991). Oxygen isotopic thermometer calibrations. Geochemical Society, Special Publication 3, 3–10.
Cooper, A. F. & Reid, D. L. (2000). The association of potassic trachytes and carbonatites at the Dicker Willem Complex, southwest Namibia: coexisting, immiscible, but not cogenetic magmas. Contributions to Mineralogy and Petrology 139, 570–583.[CrossRef]
Dalton, J. A. & Wood, B. J. (1993). The compositions of primary carbonate melts and their evolution through wallrock reaction in the mantle. Earth and Planetary Science Letters 119, 511–525.[CrossRef][Web of Science]
Deer, W. A., Howie, R. A. & Zussman, J. (1963). Rock Forming Minerals—Framework Silicates. Harlow: Longman, 435 pp.
Deines, P. (1989). Stable isotopes variations in carbonatites. In: Bell, K. (ed.) Carbonatites: Genesis and Evolution. London: Unwin Hyman, pp. 301–359.
Epstein, S., Graf, D. L. & Degens, E. T. (1964). Oxygen isotope studies on the origin of dolomites. In: Craig, H., Miller, S. L. & Wasserburg, G. J. (eds) Isotopic and Cosmic Chemistry. Amsterdam: North-Holland, pp. 169–180.
Fouillac, C. (1983). Chemical thermometry in CO2-rich thermal waters, example of the French Massif Central. Geothermics 12, 149–160.[CrossRef][Web of Science]
Goldsmith, J. R. (1990). Phase relations of rhombohedral carbonates. In: Reeder, R. J. (ed.) Carbonates: Mineralogy and Chemistry. Mineralogical Society of America, Reviews in Mineralogy 11, 49–76.
Harmer, R. E. & Gittins, J. (1997). The origin of dolomitic carbonatites: field and experimental constraints. Journal of African Earth Sciences 25, 5–28.[CrossRef]
Harmer, R. E. & Gittins, J. (1998). The case for primary, mantle-derived carbonatite magmas. Journal of Petrology 39, 1895–1903.[CrossRef][Web of Science]
Hauri, E. H., Shimizu, N., Dieu, J. J. & Hart, S. R. (1993). Evidence for hotspot-related carbonatite metasomatism in the oceanic upper mantle. Nature 365, 221–227.[CrossRef]
Hodges, S. P. (1991). Petrogenesis of the phonolitic rocks of the Velay oriental, France. Ph.D. thesis, University of Oxford, 205 pp.
Ionov, D. A., Dupuy, C., O'Reilly, S. Y., Kopylova, M. G. & Genshaft, Y. S. (1993). Carbonated peridotite xenoliths from Spitsbergen: implications for trace element signature of mantle carbonate metasomatism. Earth and Planetary Science Letters 119, 283–297.[CrossRef][Web of Science]
Jakni, B., Dautria, J. M., Liotard, J. M. & Briqueu, L. (1996). Mise en évidence d'un manteau carbonaté à l'aplomb du Bas-Languedoc: les xénolites péridotitiques du complexe volcanique de Grand-Magnon (Lodévois). Comptes Rendus de l'Académie des Sciences 323, 33–40.
Javoy, M. (1970). Utilisation des isotopes de l'oxygène en magmatologie. D.Sc. thesis, Université de Paris.
Javoy, M., Pineau, F., Staudacher, T., Cheminée, J. L. & Kraft, M. (1989). Mantle volatiles sampled from a continental rift: the 1988 eruption of Oldoinyo Lengai (Tanzania). Terra Abstracts 1, 324.
Keller, J. (1981). Carbonatitic volcanism in the Kaiserstuhl alkaline complex: evidence for highly fluid carbonatitic melts at the Earth surface. Journal of Volcanology and Geothermal Research 9, 423–431.[CrossRef][Web of Science]
Kennedy, G. C. & Holser, W. T. (1966). Pressure–volume–temperature and phase relations of water and carbon dioxide. Geological Society of America, Handbook of Physical Constants 16, 371–383.
Leake, B. E., Woolley, R. A., Arps, C. E., et al. (1997). Nomenclature of amphiboles: report of the Subcommittee on Amphiboles of the International Mineralogical Association, Commission on New Minerals and Mineral Names. Canadian Mineralogist 35, 219–246.[Web of Science]
Le Bas, M. J., Le Maitre, R. W., Streckeisen, A. & Zanettin, B. (1986). A chemical classification of volcanic rocks based on the total alkali–silica diagram. Journal of Petrology 27, 745–750.
Le Maitre, R. W., Bateman, P., Dudek, A., Keller, J., Lameyre, J., Le Bas, M. J., Sabine, P. A., Schmid, R., Sorensen, H., Streckeisen, A., Woolley, A. R. & Zanettin, B. (1989). A Classification of Igneous Rocks and Glossary of Terms. Oxford: Blackwell Scientific, 193 pp.
Liotard, J. M., Dautria, J. M. & Cantagrel, J. M. (1995). Pétrogenèse des néphélinites de la région de Craponne-sur-Arzon (Nord-Velay): intervention possible d'un composant carbonatitique. Comptes Rendus de l'Académie des Sciences 320, 1043–1050.
Macdonald, R., Kjarsgaard, B. A., Skilling, I. P., Davies, G. R., Hamilton, D. L. & Black, S. (1993). Liquid immiscibility between trachyte and carbonate in ash flow tuffs from Kenya. Contributions to Mineralogy and Petrology 114, 276–287.[CrossRef][Web of Science]
Marty, B., O'Nions, R. K., Oxburgh, E. R., Martel, D. & Lombardi, S. (1992). Helium isotopes in Alpine regions. Tectonophysics 206, 71–78.[CrossRef][Web of Science]
Mattey, D. P. (1991). Carbon dioxide solubility and carbon isotope fractionation in basaltic melt. Geochimica and Cosmochimica Acta 55, 3467–3473.[CrossRef]
Mattey, D. P., Taylor, W. R., Green, D. H. & Pillinger, C. T. (1990). Carbon isotopic fractionation between CO2 vapour, silicate and carbonate melts: an experimental study to 30 kbar. Contributions to Mineralogy and Petrology 104, 492–505.[CrossRef][Web of Science]
Matthews, A., Fouillac, C., Hill, R., O'Nions, R. K. & Oxburgh, E. R. (1987) Mantle-derived volatiles in continental crust: the Massif Central of France. Earth and Planetary Science Letters 85, 117–128.[CrossRef][Web of Science]
Mergoil, J. & Boivin, P. (1993). Le Velay. Son volcanisme et les formations associées. Géologie de la France 3, 3–96.
O'Neil, J. R., Clayton, R. N. & Mayeda, T. K. (1969). Oxygen isotope fractionation in divalent metal carbonates. Journal of Chemical Physics 51, 5547–5558.[CrossRef]
Pineau, F., Javoy, M. & Allègre, C. J. (1973). Étude systématique des isotopes de l'oxygène, du carbone et du strontium dans les carbonatites. Geochimica and Cosmochimica Acta 37, 2363–2377.
Reeder, R. J. (1990). Crystal chemistry of the rhombohedral carbonates. In: Reeder, R. J. (ed.) Carbonates: Mineralogy and Chemistry. Mineralogical Society of America, Reviews in Mineralogy 11, 1–47.
Riley, T. R., Bailey, D. K. & Lloyd, F. E. (1996). Extrusive carbonatite from the Quaternary Rockeskyll complex, West Eifel, Germany. Canadian Mineralogist 34, 389–401.[Web of Science]
Riley, T. R., Bailey, D. K., Harmer, R. E., Liebsch, H., Lloyd, F. E. & Palmer, M. R. (1999). Isotopic and geochemical investigation of a carbonatite–syenite–phonolite diatreme, West Eifel (Germany). Mineralogical Magazine 63, 615–631.
Rochette, P., Bertrand, H., Braun, C. & Berger, E. (1993). La province volcanique Pléistocène Supérieur du Bas-Vivarais (Ardèche, France): propagation de fentes crustales en échelons? Comptes Rendus de l'Académie des Sciences 316, 913–920.
Rosenbaum, J. M. (1994). Stable isotope fractionation between carbon dioxide and calcite at 900 degrees C. Geochimica and Cosmochimica Acta 58, 3747–3753.[CrossRef]
Rosenbaum, J. M. & Sheppard, S. M. F. (1986). An isotopic study of siderites, dolomites and ankerites at high temperatures. Geochimica et Cosmochimica Acta 50, 1147–1150.[CrossRef][Web of Science]
Scheele, N. & Hoefs, J. (1992). Carbon isotope fractionation between calcite, graphite and CO2; an experimental study. Contributions to Mineralogy and Petrology 112, 35–45.[CrossRef][Web of Science]
Schiano, P., Clocchiatti, R., Shimizu, N., Weis, D. & Mattielli, N. (1994) Cogenetic silica-rich and carbonate-rich melts trapped in mantle minerals in Kerguelen ultramafic xenoliths: implications for metasomatism in the oceanic upper mantle. Earth and Planetary Science Letters 123, 167–178.[CrossRef][Web of Science]
Schmid, R. (1981). Descriptive nomenclature and classification of pyroclastic deposits and fragments: recommendations of the IUGS subcommission on the systematics of igneous rocks. Geology 9, 41–43.
Sheppard, S. M. F. & Dawson, J. B. (1973). 13C/12C and D/H isotope variations in ‘primary igneous carbonatite’. Fortschritte der Mineralogie 50, 128–129.
Sheppard, S. M. F. & Schwarcz, H. P. (1970). Fractionation of carbon and oxygen isotopes and magnesium between coexisting metamorphic calcite and dolomite. Contributions to Mineralogy and Petrology 26, 161–198.[CrossRef][Web of Science]
Sweeney, R. J. (1994). Carbonatite melt compositions in the Earth's mantle. Earth and Planetary Science Letters 128, 259–270.[CrossRef][Web of Science]
Taylor, H. P., Jr, Frechen, J. & Degens, E. T. (1967). Oxygen and carbon isotope studies of carbonatites from the Laacher See district, West Germany and the Alnö district, Sweden. Geochimica et Cosmochimica Acta 31, 407–430.[CrossRef][Web of Science]
Thompson, R. N., Smith, P. M., Gibson, S. A., Mattey, D. P. & Dickin, A. P. (2002). Ankerite carbonatite from Swartbooisdrif, Namibia: the first evidence for magmatic ferrocarbonatite. Contributions to Mineralogy and Petrology 143, 377–395.[Web of Science]
Villemant, B. & Treuil, M. (1983). Comportement des éléments traces et majeurs dans la série alcaline du Velay; comparaison avec la Chaîne des Puys (Massif Central, France). Bulletin of Mineralogy 106, 465–486.
Wallace, M. E. & Green, D. H. (1988). An experimental determination of primary carbonatite magma composition. Nature 335, 343–346.[CrossRef]
Wilson, M. & Downes, H. (1991). Tertiary–Quaternary extension-related alkaline magmatism in Western and Central Europe. Journal of Petrology 32, 811–849.
Wilson, M. & Patterson, R. (2001). Intraplate magmatism related to short-wavelength convective instabilities in the upper mantle: evidence from the Tertiary–Quaternary volcanic province of western and central Europe. In: Ernst, R. E. & Buchan, K. L. (eds) Mantle Plumes: their Identification through Time. Geological Society of America, Special Papers 352, 37–58.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  M. J. Le Bas FENITES ASSOCIATED WITH CARBONATITES Can Mineral, August 1, 2008; 46(4): 915 - 932. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Andersen COEXISTING SILICATE AND CARBONATITIC MAGMAS IN THE QASSIARSUK COMPLEX, GARDAR RIFT, SOUTHWEST GREENLAND Can Mineral, August 1, 2008; 46(4): 933 - 950. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Touron, C. Renac, S. Y. O'Reilly, J.-Y. Cottin, and W. L. Griffin Characterization of the metasomatic agent in mantle xenoliths from Deves, Massif Central (France) using coupled in situ trace-element and O, Sr and Nd isotopic compositions Geological Society, London, Special Publications, January 1, 2008; 293(1): 177 - 196. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Bailey, S. Kearns, J. Mergoil, J. M. Daniel, and B. Paterson Extensive dolomitic volcanism through the Limagne Basin, central France: a new form of carbonatite activity Mineralogical Magazine, April 1, 2006; 70(2): 231 - 236. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||