| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Petrology Volume 42 Number 1 Pages 207-219 2001
© Oxford University Press 2001
Records of Crustal Metasomatism in the Garnet Peridotites of the Ulten Zone (Upper Austroalpine, Eastern Alps)
1DIPARTIMENTO DI SCIENZE DELLA TERRA, UNIVERSITÀ DEGLI STUDI DI MILANO, VIA BOTTICELLI 23, 20133 MILANO, ITALY
2DIPARTIMENTO DI SCIENZE DELLA TERRA E GEOLOGICO-AMBIENTALI, PIAZZA DI PORTA S. DONATO 1, 40126 BOLOGNA, ITALY
Received November 17, 1999; Revised typescript accepted July 4, 2000
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPETROLOGICAL OUTLINE OF THE...SAMPLES AND ANALYTICAL METHODSTRACE-ELEMENT MINERAL CHEMISTRYDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Peridotites in the Ulten Zone (Upper Austroalpine, Eastern Alps), occur as small bodies within lower-crustal rocks (gneisses and migmatites) subducted at eclogite-facies conditions during the Variscan orogeny. They record a complex metamorphic and deformation evolution as indicated by the transition from coarse-grained spinel-bearing peridotites to fine-grained garnet + amphibole-bearing peridotites, and are interpreted as portions of mantle wedge that were incorporated in a downgoing slab of cold continental crust. The transition from spinel- to garnet-bearing assemblage was accompanied by significant input of metasomatic agents, as shown by the crystallization of abundant amphibole. Here we present trace-element mineral chemistry data for selected Ulten peridotites, with the aim of unravelling the nature of the metasomatic processes. Amphiboles display significant light rare earth element (LREE) enrichment [CeN/YbN = 3·90–11·50; LREE up to (20–50) x C1], high Sr (150–250 ppm), K (1910–7280 ppm) and Ba (280–800 ppm) contents, and low concentrations of high field strength elements (HFSE) (Zr = 14–25 ppm, Y = 6·7–16 ppm, Ti = 1150–2500 ppm, Nb = 2–7 ppm). On the basis of (1) the evidence for modal orthopyroxene decrease as a result of the garnet-forming reaction rather than abundant orthopyroxene crystallization, (2) the high modal amounts of amphibole (up to 23%) in the most metasomatized peridotites and (3) the strong large ion lithophile element (LILE)/HFSE fractionation in amphiboles, we infer that the metasomatic agent was an H2O–CO2 fluid with a low CO2/H2O ratio. Petrological investigations and geochronological data indicate that the host metapelites experienced in situ partial melting and migmatization concomitantly with the garnet + amphibole-facies recrystallization in the enclosed peridotites. We infer that the metasomatizing hydrous fluids could represent the residual fluids left after the crystallization of leucosomes, starting from water-undersaturated melts produced during migmatization of the host gneisses.
KEY WORDS: garnet peridotite; crustal metasomatism; amphibole; hydrous fluids
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGICAL OUTLINE OF THE...SAMPLES AND ANALYTICAL METHODSTRACE-ELEMENT MINERAL CHEMISTRYDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Garnet peridotite bodies within medium- to high-grade metamorphic rocks of orogenic zones are widely considered as useful tools to decipher the geodynamic processes governing the subduction–collisional stages of an orogenic cycle (Medaris & Carswell, 1990
). Among these occurrences, peridotites of the Ulten Zone from the Nonsberg area consist of mantle rocks that were incorporated as tectonic lenses within a high-grade crustal basement during the subduction–collisional stages of the Variscan orogeny (in the literature, the names Ulten and Nonsberg peridotites have been used to indicate the same rocks). The Ulten Zone pertains to the Tonale nappe, which, in turn, belongs to the Upper Austroalpine domain of the Eastern Alps. It mostly consists of massive and foliated garnet–kyanite-bearing granulitic gneisses, migmatites and orthogneisses, which include small pods of eclogitic metabasites and small (metre-scale) lens-shaped bodies of peridotites (Obata & Morten, 1987; Godard et al., 1996). Where the high-grade contact is preserved, banding and foliation of the peridotites are nearly concordant with the foliation of the host migmatites. In places, amphibole + phlogopite-rich bands occur at the contact between peridotites and migmatites, clearly recording metasomatism during and after migmatization (Godard et al., 1996).
Previous petrological (Obata & Morten, 1987; Morten, 1993; Godard et al., 1996, and references cited therein) and geochemical (Morten & Obata, 1990; Bondi et al., 1992; Petrini & Morten, 1993) studies have outlined that the Ulten peridotites record a metamorphic evolution from mantle to crustal environments, which was accompanied by significant chemical modifications [e.g. enrichment in light rare earth elements (LREE), K, Sr and Li] possibly caused by the inflow of metasomatic agents from the country rocks (Morten & Obata, 1990; Bondi et al., 1992). Nevertheless, the nature and origin of the metasomatic agents are still uncertain.
In this paper, we present the results of in situ investigations of the trace-element composition of minerals from seven selected peridotite samples representative of the entire tectonometamorphic evolution. Major aims have been to investigate the geochemical signature of the metasomatic agents and to provide further constraints on the nature of the metasomatic processes. The data are also useful to characterize, by direct evidence, the trace-element chemistry of mantle amphibole at P–T conditions (800°C, 25 kbar) compatible with the slab–wedge interface.
|
PETROLOGICAL OUTLINE OF THE ULTEN PERIDOTITES |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGICAL OUTLINE OF THE...SAMPLES AND ANALYTICAL METHODSTRACE-ELEMENT MINERAL CHEMISTRYDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The Ulten peridotites record a complex metamorphic and deformation evolution documented by the transition from coarse-grained protogranular spinel-bearing peridotites (coarse-type peridotite) to fine-grained garnet- and amphibole- (pargasitic to edenitic hornblende)-bearing peridotites (fine-type peridotite) with porphyroclastic to mosaic granoblastic textures (Obata & Morten, 1987
). Thermometric estimates on the coarse-type spinel lherzolites have yielded high temperatures of equilibration in the range 1100–1300°C (Obata & Morten, 1987; see Fig. 1). P–T estimates of the high-pressure eclogitic recrystallization that produced the spinel- to garnet-facies transition are mostly in the range 20–27 kbar and 750–850°C (Obata & Morten, 1987; Godard et al., 1996; Nimis & Morten, 2000).
|
The occurrence of Ca-amphibole in equilibrium with the garnet-bearing assemblage indicates that the low-T high-P recrystallization was accompanied by concomitant input of H2O-rich metasomatic agents (Obata & Morten, 1987). In very rare samples, amphibole occurs together with other metasomatic minerals, such as phlogopite, apatite and dolomite. In some fine-type peridotites, the increase in the modal amount of amphibole (up to 20–25 vol. %) is accompanied by the progressive disappearance of clinopyroxene and garnet: these rocks consist of an (olivine + orthopyroxene + amphibole) granoblastic assemblage (Obata & Morten, 1987). The abundant amphibole crystallization has been considered the result of clinopyroxene- and garnet-consuming reactions under increasing partial pressure of H2O, which probably occurred at roughly constant P–T conditions (Obata & Morten, 1987) (see the ‘Samples and analytical methods’ section, and Table 1).
|
Chemical investigations on both coarse-type spinel-facies and fine-type garnet-facies peridotites have shown that the major- and trace-element whole-rock compositions of the two types are similar, except for higher LREE, K and Sr contents in the fine-type ultramafic rocks (Morten & Obata, 1990; Bondi et al., 1992). Furthermore, Sr and Nd isotope investigations have shown that the increase in alkali content is positively correlated with the Sr isotopic composition (Petrini & Morten, 1990). These features have been interpreted as evidence that the metamorphic recrystallization of the Nonsberg peridotites was accompanied by crustal metasomatism (Morten & Obata, 1990; Bondi et al., 1992; Petrini & Morten, 1993).
As stated above, the Nonsberg peridotites and associated metabasites occur as discrete bodies within gneisses and migmatites. Both the metabasites and the host gneisses record P–T equilibration conditions for the eclogite- to post-eclogite-facies recrystallization that are broadly consistent with those derived for the peridotites (Godard et al., 1996; Hauzenberger et al., 1996). The physical conditions of the eclogite-facies metamorphic peak for the gneisses and migmatites have been calculated at P = 15–20 kbar and T = 700–850°C (Benciolini & Poli, 1993; Hoinkes & Thöni, 1993; Godard et al., 1996; Hauzenberger et al., 1996). In particular, it has been inferred by Benciolini & Poli (1993) that migmatization by fluid-absent partial melting of metapelites probably occurred at T 
850°C and minimum pressure of 15 kbar. Precise pressure conditions for the migmatization event are therefore not available, but were probably confined to the range 15–25 kbar (i.e. below the albite 
jadeite + quartz reaction; see Fig. 1), considering the occurrence of albitic plagioclase in the leucosomes (Godard et al., 1996).
In synthesis, the P–T metamorphic changes experienced by the different lithologies of the Ulten Zone are similar (see Fig. 1), the peak, eclogite-facies pressure conditions being recorded only by the ultramafic rocks. Available geochronological data indicate ages of 330–340 Ma for the peridotite garnet-facies re-equilibration (Gebauer & Grünenfelder, 1979), and 300–360 Ma for the peak high-grade metamorphism and migmatization of the host metapelites (Hauzenberger et al., 1993). The roughly coincident ages of both rock types and their similar P–T path imply that mantle material was emplaced into the continental crust either before or during the high-P compressional stage. Thus, the metasomatic effects recorded by the peridotites were probably caused by fluids and/or melts deriving from the host migmatites.
On the basis of the above evidence, it has been inferred (Nimis & Morten, 2000) that the Ulten peridotites represent portions of mantle wedge that were sampled and incorporated in a downgoing slab of relatively cold continental crust, and this caused the re-equilibration of the peridotites at 850°C and 20–25 kbar. This evolution occurred in response to a plate convergence mechanism, related to the Variscan orogenic cycle (Godard et al., 1996). A similar geodynamic model has been recently proposed by Brueckner (1998) to explain the occurrence of garnet peridotites as tectonic lenses within high-grade gneisses in mountains belts formed by continent–continent collision.
|
SAMPLES AND ANALYTICAL METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGICAL OUTLINE OF THE...SAMPLES AND ANALYTICAL METHODSTRACE-ELEMENT MINERAL CHEMISTRYDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The modal compositions of the investigated samples are reported in Table 1, together with the metamorphic reactions (from Obata & Thompson, 1981
) that may account for their paragenetic evolution. All samples have already been characterized for whole-rock major- and trace-element composition (Obata & Morten, 1987; Morten & Obata, 1990) (data reported, for completeness, in Table 2). Representative major- and trace-element data for minerals are reported in Tables 3 and 4. Major-element mineral analyses were partly already available (Obata & Morten, 1987) and partly performed using the ARL SEMQ electron microprobe at the Dipartimento di Scienze della Terra, Milan. Trace-element mineral analyses were performed by the Cameca IMS 4f ion microprobe at CSCC, Pavia [see Rampone et al. (1995) for analytical details].
|
|
|
Samples ULT16 and 373/3 are coarse-type spinel lherzolites with protogranular textures. Pyroxenes (especially clinopyroxene) are mostly exsolved. In some clinopyroxenes, the exsolved phases (orthopyroxene and spinel) are partly replaced by Ca-hornblende amphibole. Ca-hornblende also develops as rims around clinopyroxene grains. The amphibole crystallization is more pronounced in sample ULT16 than in sample 373/3.
Sample MK5D is a coarse-type spinel lherzolite partly replaced by a garnet-bearing assemblage. Its texture is transitional from protogranular to porphyroclastic. The relict spinel-facies assemblage consists of porphyroclasts of ortho- and clinopyroxenes, olivine and spinel. Garnet crystallizes as coronas around spinel, as exsolution in pyroxenes, and in granoblastic aggregates in equilibrium with new pyroxenes and olivine.
Samples MK5C, MO11 and ULT12 are fine-type garnet-bearing peridotites, with porphyroclastic to mosaic equigranular textures. Sample MK5C, in particular, comes from the same outcrop (separated by <1 m) as the coarse peridotite MK5D, thus constituting its fine-type equivalent. All these samples are characterized by large amounts of modal amphibole (13 vol. % in sample MK5C; 21–23 vol. % in samples ULT12 and MO11; see Table 1), whose composition is pargasitic to edenitic hornblende (Table 4). Deformed olivine and pyroxene porphyroclasts are rarely preserved, the rocks mostly consisting of a granoblastic fine-grained aggregate of garnet, amphibole, new pyroxenes and olivine. As shown in Table 1, the metamorphic reactions leading to crystallization of the garnet + amphibole-bearing assemblage cause the progressive disappearance of clinopyroxene. As a result, clinopyroxene occurs in very low modal abundance in sample MK5C, and is absent in samples MO11 and ULT12.
Sample 300 B is a fine-type amphibole-bearing peridotite. It consists of a fine-grained granoblastic assemblage of olivine + amphibole + orthopyroxene (+ minor small rounded spinels). Both amphibole and orthopyroxene are compositionally similar to the corresponding minerals in the garnet-bearing peridotites [see Table 3 and Obata & Morten (1987)]. It has been inferred (Obata & Morten, 1987) that this sample equilibrated in roughly the same P–T regime as the garnet-bearing peridotites, and that the garnet-free assemblage resulted from high partial pressure of H2O causing the reaction garnet + olivine + H2O amphibole + orthopyroxene + spinel (Obata & Thompson, 1981; see Table 1).
The bulk-rock major-element compositions of the investigated peridotites are shown in Fig. 2. They display variations (i.e. Al2O3 and CaO decrease, and MgO increase), which are not correlated with different textures (coarse vs fine types) or metamorphic assemblage (spinel vs garnet vs amphibole bearing), and probably reflect primary compositional heterogeneity (different degrees of fertility of the mantle protoliths). In particular, the two samples coming from the same outcrop (the coarse peridotite MK5D and its fine-type equivalent MK5C) are the most depleted (lower Al2O3 and CaO, and higher MgO contents) and display very similar major-element compositions. By contrast, they show very different whole-rock REE spectra (Fig. 3). In terms of LREE concentrations, they plot at the extreme ends of the total compositional range defined by all samples studied; sample MK5D displays an LREE-depleted pattern at values lower than 1 x C1 (CeN/EuN = 0·77), whereas sample MK5C is significantly enriched in LREE [up to (8–10) x C1; CeN/EuN = 5·96]. Moreover, all the fine-type peridotites are enriched in LREE, Sr and K relative to the coarse type peridotites (Fig. 3 and Table 2).
|
|
|
TRACE-ELEMENT MINERAL CHEMISTRY |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGICAL OUTLINE OF THE...SAMPLES AND ANALYTICAL METHODSTRACE-ELEMENT MINERAL CHEMISTRYDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
The coarse-type peridotites display negligible or modest metasomatic effects and heterogeneous clinopyroxene trace-element compositions. In the spinel peridotite 373/3, clinopyroxenes show slightly fractionated REE spectra (CeN/YbN = 0·54–0·85) at (4–8) x C1 values, almost flat LREE (CeN/SmN = 0·76–1·26), and Zr and Ti negative anomalies (Fig. 4). Clinopyroxenes in the spinel peridotite ULT16 exhibit peculiar REE patterns, characterized by concave shape and selective LREE enrichment (CeN/SmN = 2·45–4·55); they also show pronounced Zr and Ti negative anomalies. No significant compositional differences are found in core to rim traverses. Edenitic amphibole crystallized as rims around clinopyroxene has REE spectra very similar to those of clinopyroxenes and slightly higher absolute concentrations.
|
The garnet peridotite MK5D, with no modal amphibole, does not display any metasomatic effect. Clinopyroxenes exhibit convex-upward REE patterns characterized by LREE and HREE depletion (CeN/SmN = 0·17–0·19; SmN/YbN = 3·10–6·20). The depletion in LREE, as well as the low Sr (2·5–4·2 ppm), Zr (1·4–1·6 ppm) and Ti (1100–1280 ppm) concentrations in clinopyroxene, are probably a primary feature, and reflect the depleted signature of this peridotite, already indicated by its bulk-rock major-element chemistry (see Fig. 2). By contrast, the low HREE abundances in clinopyroxene result from the equilibration with garnet, which, as expected, shows severely fractionated pattern [CeN/YbN < 0·001; HREE (20–30) x C1]. Garnet also displays strongly negative Ti anomaly and small negative Zr anomaly relative to the adjacent REE.
Consistent with the bulk-rock data (see Fig. 3), the trace-element compositions of minerals in the fine-type peridotites, characterized by the abundant crystallization of amphibole (amphibole modal amount in the range 13–23 vol. %), display the most evident metasomatic effects. In the garnet-bearing peridotite MK5C (fine-type equivalent of MK5D), both clinopyroxene and amphibole exhibit strong LREE enrichment (Fig. 5; CeN/EuN = 4·9–6·1 in clinopyroxene; CeN/EuN = 4·3–4·6 in amphibole); their REE patterns are similarly fractionated but amphibole displays higher absolute concentrations [LREE up to (40–50) x C1]. Recent literature data indicate that amphibole–clinopyroxene partition coefficients for the REE in mantle peridotites are not fractionated and are mostly between one and two (Ionov et al., 1997, and references cited therein). In sample MK5C, Damph/cpx for REE displays higher values, in the range 2·8–4 for the LREE and up to 7–13 for the heavy REE (HREE). This could indicate that, in sample MK5C, complete geochemical equilibrium between amphibole and clinopyroxene was not attained, probably because of the low temperature (T 850°C) at which the recrystallization occurred. Both amphibole and clinopyroxene also display high Sr contents (124–129 ppm in cpx; 228–239 ppm in amphibole), whereas they are characterized by negative Ti and Zr anomalies. Garnet in the peridotite MK5C displays a fractionated REE pattern (CeN/YbN = 0·04) and relative depletion in both Ti and Zr.
|
Amphiboles (pargasitic to edenitic hornblendes) crystallized in the garnet + amphibole-bearing peridotites MO11 and ULT12 also show LREE-enriched REE spectra [Fig. 6a and b; LREE up to (20–40) x C1; CeN/YbN = 5·8–11·5], negligible or slightly negative Ti anomalies, and pronounced Zr negative anomalies. The strong LREE–HREE fractionation in amphiboles is partly due to their equilibrium crystallization with garnet, responsible for their low HREE contents [in the range (1–5) x C1]. Garnets in sample ULT12 display, as expected, strong LREE–HREE fractionation (CeN/YbN = 0·003–0·007), significant relative depletion in Ti and slightly positive Zr anomalies.
The most pronounced LREE enrichment and REE–HFSE (high field strength element) fractionation are shown by amphiboles crystallized in the amphibole-bearing peridotite 300B (absent modal clinopyroxene and garnet); amphiboles are edenitic hornblendes, and display very high LREE absolute concentrations [Fig. 6c; LREE up to (40–70) x C1; CeN/YbN = 4·4–9·2], and severe relative depletion in Ti and Zr. In Fig. 6c, the REE–Sr–Zr–Ti compositions of amphiboles in sample 300B are compared with the compositions of edenitic hornblendes crystallized in the Finero peridotites (Ivrea Zone), which are interpreted as the product of metasomatism by Si-rich hydrous melts in a mantle wedge environment (Zanetti et al., 1999). It is worth noting that amphiboles in peridotite 300B display REE spectra rather similar to those of Finero amphiboles, but a more pronounced REE–HFSE (especially REE–Zr) fractionation.
|
The peculiar large ion lithophile element (LILE) vs HFSE fractionation of amphiboles crystallized in the fine-type Ulten peridotites is also evident in terms of other trace elements. In Fig. 7, the trace-element abundances of Ulten amphiboles are compared with those of amphiboles from mantle xenoliths and peridotite massifs [data from Ionov et al. (1997)
]. The compositions of the Finero amphiboles (same data as in Fig. 6c) are also shown. The Ulten amphiboles display very high Ba (280–800 ppm), Sr (150–250 ppm) and K (1910–7280 ppm) contents, and low Zr (14–25 ppm), Nb (2–7 ppm) and Ti (1150–2500 ppm) concentrations. Their compositions are similar to those of Finero amphiboles, except for much higher Ba and lower Zr contents. In particular, the Ulten amphiboles exhibit the lowest Zr coupled with the highest Ba concentrations of all amphiboles shown for comparison (see Fig. 7b).
|
|
DISCUSSION AND CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGICAL OUTLINE OF THE...SAMPLES AND ANALYTICAL METHODSTRACE-ELEMENT MINERAL CHEMISTRYDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Petrological constraints on the nature of the metasomatic agent
As stated in the ‘Petrological outline of the Ulten peridotites’ section, field evidence together with the P–T–t evolution reconstructed for both the peridotites and the host migmatites furnish some major constraints on the origin of the metasomatic process: (1) metasomatism in the Ulten peridotites occurred during their recrystallization at low-temperature (850°C) and high-pressure (20–25 kbar) conditions, leading to the development of a garnet + amphibole-bearing assemblage; (2) the low-T high-P recrystallization and metasomatism in the peridotitites was concomitant with migmatization of the host crustal rocks. Thus, metasomatism of the Ulten peridotites was probably caused by crustal fluids and/or Si-rich melts produced during the partial melting event that generated the host migmatites.
Further constraints on the nature of the metasomatic agent that affected the Ulten peridotites can be derived from experimental results on the system peridotite–granite–H2O (Sekine & Wyllie, 1982). These studies indicate that, at 30 kbar and 900°C (i.e. roughly, the P–T conditions of re-equilibration and metasomatism of the studied peridotites), the peridotite assemblage coexists with a free liquid only for fairly high granite/peridotite ratios (i.e. granite component >35%), which causes a fall in the solidus temperatures of the system. In this case, the interaction with granitic melt induces significant modification of the peridotite major-element composition, as a result of the very high melt/peridotite ratio. Moreover, the interaction of the peridotite with a granitic melt causes abundant crystallization of orthopyroxene with formation, at such high melt/peridotite ratios, of olivine-free, garnet pyroxenite assemblages.
In the Ulten peridotites, the infiltration of the metasomatic agents leading to amphibole crystallization did not cause significant modifications of the major-element compositions, and was not accompanied by abundant orthopyroxene crystallization. This is clearly indicated by the higher orthopyroxene modal proportion observed in the coarse-grained amphibole- and garnet-free peridotites (ULT16 and 373/3) compared with the other samples (Table 1). Actually, the modal variations observed in the investigated peridotites partly reflect their primary (mantle) compositional heterogeneity (see sample description) and partly result from the metamorphic reactions governing the transition from spinel- to garnet + amphibole-bearing assemblages [see reactions (1)–(3) reported in Table 1]. Reaction (1) is well studied experimentally and defines the spinel to garnet lherzolite transition [see references cited by Obata & Thompson (1981)]. Reactions (2) and (3) are taken from a study by Obata & Thompson (1981) concerning a theoretical reconstruction, in the CMASH model system, of amphibole stability in mantle rocks.
The above features indicate that the high melt/peridotite ratios required to have a free Si-rich melt percolating through mantle rocks at 900°C and 30 kbar (i.e. roughly the P–T conditions of re-equilibration and metasomatism of the studied peridotites) were not reached during the metasomatic process. Thus, we infer that the metasomatic agent that infiltrated the Ulten ultramafic rocks was probably a crustal-derived low-density fluid. This inference is further supported by direct investigations on the metasomatic processes induced in sub-arc mantle xenoliths by hydrous Si-rich melts produced by partial melting of the subducting slab (Schiano et al., 1995). In fact, these studies indicate that the crystallization of abundant Fe-enriched orthopyroxene (+ minor clinopyroxene, amphibole and phlogopite) is a ubiquitous feature of metasomatism caused by Si-rich melts.
The chemical signature of the metasomatic agent
Whole-rock major- and trace-element compositions of the Ulten peridotites have provided two main pieces of evidence: (1) the peridotites are characterized by a significant LILE (Sr, Ba, LREE) enrichment, which is mostly recorded by the recrystallized fine-type amphibole ± garnet-bearing rocks; (2) there is no correlation between the major-element chemistry and the trace-element enrichment. This latter feature is particularly evident in the coarse and fine peridotites MK5D and MK5C, which have the same depleted major-element composition but very different (respectively depleted and extremely LILE-enriched) trace-element chemistry (see Figs 2, 3 and 5).
Another important observation is that the most pronounced chemical modifications are shown by the amphibole-bearing fine-type peridotites. The coarse-type ultramafic rocks, indeed, record negligible (samples 373/3, MK5D) or modest (sample ULT16) metasomatic effects. In spinel peridotite ULT16, showing incipient development of amphibole rims around clinopyroxene, both amphibole and clinopyroxene display U-shaped LREE-enriched REE spectra (see Fig. 4); similar U-shaped REE patterns have been observed in metasomatized mantle peridotites, and have been interpreted as the result of metasomatism caused by percolation of the metasomatic agent at distance from its source (with consequent chromatographic fractionation of the REE; Bodinier et al., 1990). Thus, the above features indicate that migration of the metasomatic fluids, in the Ulten peridotites, was mostly channelled along localized shear zones, where it promoted the low-T, high-P fine-grained recrystallization leading to the development of garnet + amphibole-facies assemblages.
The infiltration of the metasomatizing fluids caused both (1) chemical modifications in clinopyroxene (selective LREE and Sr enrichment; compare clinopyroxene in samples MK5C and MK5D, Tables 2 and 3) and (2) crystallization of abundant hornblenditic amphibole, which is invariably characterized by strong LILE enrichment (high LREE, Sr, Ba and Rb concentrations) coupled with very low contents in HFSE such as Zr, Ti and Nb.
Amphibole and phlogopite are the most common and diffuse metasomatic minerals occurring in mantle peridotites. Amphibole, in particular, can provide information about the origin of the metasomatic process, because its trace-element composition (e.g. the LILE–HFSE fractionation) is dependent on the nature (fluid or melt) and the trace-element budget of the agent from which it crystallizes (Vannucci et al., 1995; Ionov et al., 1997). For instance, amphiboles crystallized from alkali-rich silicate melts usually do not display LILE–HFSE fractionation, and are often enriched in both LILE and HFSE (Ionov et al., 1997, and references cited therein). Amphiboles selectively enriched in LILE and depleted in HFSE (i.e. compositionally similar to those crystallized in the Ulten peridotites) are frequently documented in mantle xenoliths affected by carbonated metasomatism, as well as in orogenic peridotites and mantle xenoliths contaminated by crustal-derived metasomatic agents.
LILE-enriched HFSE-depleted amphiboles have been reported in the Finero peridotites, and interpreted as the product of metasomatism by Si-rich (crustal-derived) hydrous melts in a mantle wedge environment (Zanetti et al., 1999) (see data reported for comparison in Figs 6c and 7). Selective enrichment in LILE (Ba, K, Sr, LREE), recorded in both bulk rock and amphibole, has been documented in some hornblende-bearing mantle ultramafic rocks from Zabargad Island, and has been ascribed to metasomatism by hydrous fluids released during granulitic metamorphism of the adjacent crustal rocks related to the Red Sea rifting (Dupuy et al., 1991; Piccardo et al., 1993). Furthermore, LILE-enriched metasomatism is commonly observed in amphibole + phlogopite-bearing peridotite xenoliths from the sub-arc mantle, where it has been ascribed to interaction of the mantle wedge with slab-derived hydrous fluids (Vidal et al., 1989; Maury et al., 1992) or hydrous Si-rich melts (Schiano et al., 1995).
In synthesis, the trace-element composition of the Ulten amphiboles indicates that the metasomatizing fluid probably had an LILE-enriched, HFSE-depleted chemical signature. As stated above, such geochemical features are consistent with the crustal derivation (i.e. from the host migmatites) of the infiltrating fluid, already inferred from field and petrological constraints.
Amphibole, in the Ulten peridotites, is the most abundant and ubiquitous metasomatic mineral; this indicates that the infiltrating fluids were rich in H2O component. However, the rare occurrence in some Ulten peridotites (Obata & Morten, 1987) of dolomite and apatite as metasomatic minerals suggests that the metasomatic agent was probably an H2O–CO2 fluid with a low CO2/H2O ratio (possibly containing other volatile and dissolved silicate components). CO2 saturation in the fluid, causing dolomite precipitation, was probably a consequence of fractionation of the fluid composition (increase of the CO2/H2O ratio) driven by amphibole crystallization.
Experimental studies on the mobility of low-density fluids in the mantle (Watson et al., 1990, and references cited therein) also suggest that the fluids infiltrating the Ulten peridotites were probably characterized by a low CO2/H2O ratio. These studies have shown that, at common P–T mantle conditions, CO2–H2O fluids do not represent effective metasomatic agents, because they have high wetting angles (>60°), which preclude their forming an interconnected network between the mantle minerals. This is especially true at low temperature, and for CO2-rich compositions; CO2-rich fluids possess therefore much less mobility than H2O-rich fluids in the mantle. The above investigations also indicate that (1) the wetting angle is strongly dependent on the pressure conditions (i.e. it significantly decreases with increasing pressure) and (2) fluid mobility and interconnectivity can be enhanced by crystallographic anisotropy. Thus, in the case of the Ulten peridotites, mobility of H2O-rich fluids through the peridotites was probably favoured by both the high-pressure conditions at which the metasomatism occurred and deformation along shear zones, which caused anisotropy of mantle mineral lattices. This is confirmed by the striking correlation, in the Ulten peridotites, between the degree of deformation (i.e. coarse-type vs fine-type rocks) and the extent of metasomatic effects (see Fig. 3). A similar relationship between deformation and metasomatism has been already documented in mantle xenoliths (Downes, 1990).
Finally, recent experimental investigations on the partitioning behaviour of trace elements between minerals and H2O-rich fluids (Brenan et al., 1995; Keppler, 1996) have shown that fluids and melts have similar capabilities to transport LILE elements such as Sr and Ba but, in contrast, highly charged elements such as Nb are more strongly partitioned into silicate melts than into hydrous fluids. Thus, the LILE-enriched HFSE-depleted geochemical signature of the Ulten amphiboles is also consistent with the inferred nature of the metasomatic agent, i.e. a hydrous low-density fluid.
Inferences on the nature of the metasomatic process
As stated above, field evidence together with the P–T–t evolution reconstructed for both the peridotites and the host migmatites strongly indicate that peridotite metasomatism was concomitant with migmatization of the country rocks. Thus, the H2O–CO2 fluid that infiltrated the Ulten peridotites probably derived from the partial melting event that affected the host gneisses.
Various hypotheses have been proposed for the nature of the partial melting process that produced the Ulten migmatites. According to Hauzenberger et al. (1996), the gneisses started to melt during the prograde path at 630°C in a fluid-saturated system; the melting reactions consumed most of the H2O fluid causing, at the metamorphic peak, the separation of a CO2-dominated fluid. By contrast, Benciolini & Poli (1993) have inferred that partial melting of the metapelites occurred under water-absent conditions by biotite-dehydration melting reactions. This latter process is at present considered to prevail during the high-grade metamorphism and partial melting of the lower crust (Clemens & Vielzeuf, 1987; Vielzeuf & Holloway, 1988; and references cited therein). If partial melting of the Ulten gneisses occurred under water-absent conditions [as inferred by Benciolini & Poli (1993)], the melts produced were water undersaturated although containing a fairly high amount of dissolved water (10 wt % at 10 kbar) and a free fluid phase was therefore not directly available. However, experimental studies on melting relations in the pelitic systems in fluid-absent conditions have shown that progressive crystallization of such water-undersaturated melts (with fractionation of anhydrous minerals such as kyanite, K-feldspar, garnet, quartz and plagioclase) drives the composition of residual melts towards water-saturated conditions, until a free water-rich fluid phase is liberated from the system (Vielzeuf & Holloway, 1988).
In synthesis, we have shown that the metasomatic agent that infiltrated and metasomatized the Ulten peridotites was probably an H2O–CO2 fluid with a low CO2/H2O ratio (and possibly containing other volatile and dissolved silicate components), characterized by a striking LILE-enriched, HFSE-depleted signature. On the basis of the above evidence, we infer that such hydrous fluids could represent the residual fluids left after the crystallization of the host migmatite leucosomes.
|
ACKNOWLEDGEMENTS |
|---|
This work benefited from fruitful discussions with Stefano Poli. We thank Danilo Biondelli and Valeria Diella for assistance with the electron microprobe. Alberto Zanetti is gratefully acknowledged for help and technical assistance with the ion probe. S. O’Reilly and G. Chazot are thanked for their constructive reviews. We much appreciated the constructive criticism and suggestions by J.-L. Bodinier. Funding was provided by the Italian MURST and CNR within the projects ‘Transformation in subducted materials and mass transfer to the mantle wedge’ and ‘Ricerche multidisciplinari sui processi subsolidus in petrologia metamorfica’.
|
FOOTNOTES |
|---|
*Corresponding author. Present address: Dipartimento di Scienze Geologiche e Geotecnologie, Università di Milano Bicocca, Piazza della Scienza 4, 20126 Milano, Italy. Fax: 39-2-64484273. E-mail: elisabetta.rampone{at}unimib.it
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPETROLOGICAL OUTLINE OF THE...SAMPLES AND ANALYTICAL METHODSTRACE-ELEMENT MINERAL CHEMISTRYDISCUSSION AND CONCLUSIONSREFERENCES |
|---|
Anders, E. & Ebihara, M. (1982). Solar system abundances of the elements. Geochimica et Cosmochimica Acta 46, 2363–2380.
Benciolini, N. & Poli, S. (1993). The lower continental crust in the Tonale nappe (Upper Austroalpine, Ultental): new petrological constraints. Terra Nova 5(1), 398.
Bodinier, J.-L., Vasseur, G., Vernières, J., Dupuy, C. & Fabriès, J. (1990). Mechanisms of mantle metasomatism: geochemical evidence from the Lherz orogenic peridotite. Journal of Petrology 31, 597–628.
Bondi, M., De Francesco, A. M. & Morten, L. (1992). Major elements, 3d transition elements, Cu and Sr geochemistry of peridotitic rocks within the Austridic crystalline basement, Nonsberg area, Northern Italy. In: Carmignani, L. & Sassi, F. P. (eds) Contributions to the Geology of Italy with Special Regard to the Paleozoic Basements. A Volume Dedicated to Tommaso Cocozza. IGCP No. 276, Newsletter 5, 229–235.
Brenan, J. M., Shaw, H. F., Ryerson, F. J. & Phinney, D. L. (1995). Mineral–aqueous fluid partitioning of trace elements at 900°C and 2·0 GPa: constraints on the trace element chemistry of mantle and deep crustal fluids. Geochimica et Cosmochimica Acta 59, 3331–3350.[Web of Science]
Brueckner, H. K. (1998). Sinking intrusion model for the emplacement of garnet-bearing peridotites into continent collision orogens. Geology 26, 631–634.
Clemens, J. L. & Vielzeuf, D. (1987). Constraints on melting and magma production in the crust. Earth and Planetary Science Letters 86, 287–306.[Web of Science]
Downes, H. (1990). Shear zones in the upper mantle—relation between geochemical enrichment and deformation in mantle peridotites. Geology 18, 374–377.
Dupuy, C., Mével, C., Bodinier, J.-L. & Savoyant, L. (1991). Zabargad peridotite: evidence for multistage metasomatism during Red Sea rifting. Geology 19, 722–725.
Gebauer, D. & Grünenfelder, M. (1979). U–Pb zircon dating of mafic and ultramafic rocks. In: Panayotou, A. (ed.) Proceedings of the International Ophiolite Symposium, Cyprus, pp. 215–218.
Godard, G., Martin, S., Prosser, G., Kienast, J. R. & Morten, L. (1996). Variscan migmatites, eclogites and garnet-peridotites of the Ulten zone, Eastern Austroalpine system. Tectonophysics 259, 313–341.
Hauzenberger, C. A., Höller, W. & Hoinkes, G. (1996). Transition from eclogite to amphibolite-facies metamorphism in the Austroalpine Ulten Zone. Mineralogy and Petrology 58, 111–130.
Hauzenberger, C. A., Höller, W., Hoinkes, G., Thöni, M. & Klötzli, U. (1993). Metamorphic evolution of the Nonsberg area, Southern Tyrol. Terra Nova 5, 11–12.
Hoinkes, G. & Thöni, M. (1993). Evolution of the Ötztal–Stubai, Scarl–Campo and Ulten Basement Units. In: von Raumer, J. & Neubauer, F. (eds) The Pre-Mesozoic Geology in the Alps. Berlin: Springer, pp. 485–494.
Ionov, D. A., Griffin, W. L. & O’Reilly, S. (1997). Volatile-bearing minerals and lithophile trace elements in the upper mantle. Chemical Geology 141, 153–184.[Web of Science]
Keppler, H. (1996) Constraints from partitioning experiments on the composition of subduction zone fluids. Nature 380, 237–240.
Maury, R. C., Defant, M. J. & Joron, J.-L. (1992). Metasomatism of the sub-arc mantle inferred from trace elements in Philippine xenoliths. Nature 360, 661–663.
Medaris, L. G., Jr & Carswell, D. A. (1990). Petrogenesis of Mg–Cr garnet peridotites in European metamorphic belts. In: Carswell, D. A. (ed.) Eclogite Facies Rocks. New York: Chapman and Hall, pp. 260–290.
Morten, L. (1993). Ultramafic rocks in the eastern Alps. In: Morten, L. (ed.) Italian Eclogites and Related Rocks. Accademia Nazionale delle Scienze detta dei XL, XIII, 111–119.
Morten, L. & Obata, M. (1990). Rare earth abundances in the eastern Alpine peridotites, Nonsberg area, Northern Italy. European Journal of Mineralogy 2, 643–653.[Web of Science]
Nimis, P. & Morten, L. (2000). P–T evolution of ‘crustal’ garnet peridotites and included pyroxenites from Nonsberg area (Upper Austroalpine), NE Italy: from the wedge to the slab. Journal of Geodynamics 30(1/2), 93–115.
Obata, M. & Morten, L. (1987). Transformation of spinel lherzolite to garnet lherzolite in ultramafic lenses of the Austridic crystalline complex, Northern Italy. Journal of Petrology 28(3), 599–623.
Obata, M. & Thompson, A. B. (1981). Amphibole and chlorite in mafic and ultramafic rocks in the lower crust and upper mantle. Contributions to Mineralogy and Petrology 77, 74–81.
Petrini, R. & Morten, L. (1990). Geochemistry and Sr–Nd isotopes on peridotitic rocks from the Nonsberg area, Italy, bearing on mantle heterogeneity and metasomatism. Terra Abstracts 2, 139.
Petrini, R. & Morten, L. (1993). Nd-isotopic evidence of enriched lithospheric domains: an example from the Nonsberg area, eastern Alps. Terra Abstracts 4, 19–20.
Piccardo, G. B., Rampone, E., Vannucci, R., Shimizu, N., Ottolini, L. & Bottazzi, P. (1993). Mantle processes in the sub-continental lithosphere: the case study of the rifted sp-lherzolites from Zabargad (Red Sea). European Journal of Mineralogy 5, 1039–1056.[Web of Science]
Rampone, E., Hofmann, A. W., Piccardo, G. B., Vannucci, R., Bottazzi, P. & Ottolini, L. (1995). Petrology, mineral and isotope geochemistry of the External Liguride peridotites (Northern Apennine, Italy). Journal of Petrology 36, 81–105.
Schiano, P., Clocchiatti, R., Shimizu, N., Maury, R. C., Jochum, K. P. & Hofmann, A. W. (1995). Hydrous, silica-rich melts in the sub-arc mantle and their relationship with erupted arc lavas. Nature 377, 595–600.
Sekine, T. & Wyllie, P. J. (1982). The system granite–peridotite–H2O at 30 kbar, with applications to hybridization in subduction zone magmatism. Contributions to Mineralogy and Petrology 81, 190–202.
Vannucci, R., Piccardo, G. B., Rivalenti, G., Zanetti, A., Rampone, E., Ottolini, L., Oberti, R., Mazzucchelli, M. & Bottazzi, P. (1995). Origin of LREE-depleted amphiboles in the subcontinental mantle. Geochimica et Cosmochimica Acta 59, 1763–1771.
Vidal, P., Dupuy, C., Maury, R. & Richard, M. (1989). Mantle metasomatism above subduction zones: trace-element and radiogenic isotope characteristics of peridotite xenoliths from Batan Island (Philippines). Geology 17, 1115–1118.
Vielzeuf, D. & Holloway, J. R. (1988). Experimental determination of the fluid-absent melting relations in the pelitic system. Contributions to Mineralogy and Petrology 98, 257–276.[Web of Science]
Watson, E. B., Brenan, J. M. & Baker, D. R. (1990). Distribution of fluids in the continental mantle. In: Menzies, A. (ed.) Continental Mantle. Oxford Monographs on Geology and Geophysics. Oxford: Oxford University Press, pp. 111–125.
Zanetti, A., Mazzucchelli, M., Rivalenti, G. & Vannucci, R. (1999). The Finero phlogopite-peridotite massif: an example of subduction-related metasomatism. Contributions to Mineralogy and Petrology 134, 107–122.[Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  K. Naemura, T. Hirajima, and M. Svojtka The Pressure-Temperature Path and the Origin of Phlogopite in Spinel-Garnet Peridotites from the Blansky Les Massif of the Moldanubian Zone, Czech Republic J. Petrology, October 1, 2009; 50(10): 1795 - 1827. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Liati and D. Gebauer Crustal origin of zircon in a garnet peridotite: a study of U-Pb SHRIMP dating, mineral inclusions and REE geochemistry (Erzgebirge, Bohemian Massif) European Journal of Mineralogy, August 1, 2009; 21(4): 737 - 750. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Malaspina, S. Poli, and P. Fumagalli The Oxidation State of Metasomatized Mantle Wedge: Insights from C-O-H-bearing Garnet Peridotite J. Petrology, August 1, 2009; 50(8): 1533 - 1552. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Tumiati, G. Godard, S. Martin, P. Nimis, V. Mair, and B. Boyer Dissakisite-(La) from the Ulten zone peridotite (Italian Eastern Alps): A new end-member of the epidote group American Mineralogist, July 1, 2005; 90(7): 1177 - 1185. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. K. BRUECKNER, H. L. M. VAN ROERMUND, and N. J. PEARSON An Archean(?) to Paleozoic Evolution for a Garnet Peridotite Lens with Sub-Baltic Shield Affinity within the Seve Nappe Complex of Jamtland, Sweden, Central Scandinavian Caledonides J. Petrology, February 1, 2004; 45(2): 415 - 437. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||