Fluid flow and mass transport at the Valentine wollastonite deposit, Adirondack Mountains, New York State

The Valentine wollastonite skarn in the north-west Adirondack Mountains, New York, is a seven million ton deposit which resulted from channellized infiltration of H₂O-rich, silica-bearing fluids. The wollastonite formed by reaction of these fluids with non-siliceous calcite marble. The skarn formed at the contact of the syenitic Diana Complex and was subsequently overprinted by Grenville-age granulite facies metamorphism and retrograde hydrothermal alteration during uplift. Calcite marbles adjacent to the deposit have generally high δ¹⁸O values (c. 21‰), typical of Grenville marbles which have not exchanged extensively with externally derived fluids. Carbon isotopic fractiona-tions between coexisting calcite and graphite in the marbles indicate equilibration at 675d? C, consistent with the conditions of regional metamorphism. Oxygen isotopic ratios from wollastonite skarn are lower than in the marbles and show a 14‰ variation (-1‰ to 13‰). Some isotopic heterogeneity is preserved from skarn formation, and some represents localized exchange with low-δ¹⁸O retrograde fluids. Detailed millimetre- to centimetre-scale isotopic profiles taken across skarn/marble contacts reveal steep δ¹⁸O gradients in the skarn, with values increasing towards the marble. The gradients reflect isotopic evolution of the fluid as it reacted with high δ¹⁸O calcite to form wollastonite. Calcite in the marble preserves high δ¹⁸O values to within <5 mm of the skarn contact. The preservation of high δ¹⁸O values in marbles at skarn contacts and the disequilibrium fractionation between wollastonite skarn and calcite marble across these contacts indicate that the marbles were not infiltrated with significant quantities of the fluid. Thus, the marbles were relatively impermeable during both the skarn formation and retrograde alteration. Skarn formation may have been episodic and fluid flow was either chaotic or dominantly parallel to lithological contacts. Although these steep isotope gradients resemble fluid infiltration fronts, they actually represent the sides of the major flow system. Because chromatographic infiltration models of mass transport require the assumption of pervasive fluid flow through a permeable rock, such models are not applicable to this hydrothermal system and, by extension, to many other metamorphic systems where low-permeability rocks restrict fluid migration pathways. Minimum time-integrated fluid fluxes have been calculated at the Valentine deposit using oxygen isotopic mass balance, reaction progress of fluid buffering reactions, and silica mass balance. All three approaches show that large volumes of fluid were necessary to produce the skarn, but silica mass balance calculations yield the largest minimum flux and are hence the most realistic.     

Fluid-rock history of granulite facies humite-marbles from Ambasamudram, southern India.

An extensive humite‐bearing marble horizon within a supracrustal sequence at Ambasamudram, southern India, was studied using petrological and stable isotopic techniques to define its metamorphic history and fluid characteristics. At peak metamorphic temperatures of 775±73°C, based on calcite‐graphite carbon isotope thermometry, the mineral assemblages suggest layer‐by‐layer control of fluid compositions. Clinohumite + calcite‐bearing assemblages suggest X_CO2 < 0.4 (at 700°C and 5 kbar), calcite + forsterite + K‐feldspar‐bearing assemblages suggest X_CO2>0.9 (at 790°C); and local wollastonite + scapolite + grossular‐bearing zones formed at X_CO2 of c. 0.3. Retrograde reaction textures such as scapolite + quartz symplectites after feldspar and calcite and replacement of dolomite + diopside or tremolite+dolomite after calcite+forsterite or calcite+clinohumite are indicative of retrogression under high X_CO2 conditions. Calcite preserves late Proterozoic carbon and oxygen isotopic signatures and the marble lacks evidence for extensive retrograde fluid infiltration, while during prograde metamorphism the possible infiltration of aqueous fluids did not produce significant isotopic resetting. Isotopic zonation of calcite and graphite grains was likely produced by localized CO₂ fluid infiltration during retrogression. Contrary to the widespread occurrence of humite‐marbles related to retrograde aqueous fluid infiltration, the Ambasamudram humite‐marbles record a prograde‐to‐peak metamorphic humite formation and retrogression under conditions of low X_H2O.     

Palæohydrology of the calcsilicate aureole of the Beinn an Dubhaich granite, Skye, Scotland: a stable isotopic study

The contact aureole developed in siliceous carbonates surrounding the Beinn an Dubhaich granite, Skye, shows textural and stable isotope evidence for infiltration of aqueous fluids during both prograde and retrograde metamorphism. Strongly depleted isotope compositions of reaction-product calcite correlate with high silica and fluorine contents, demonstrating a strong link between isotopic alteration and metasomatism by fluids with a significant magmatic component, even at the margins of the aureole. The oxygen and carbon isotope compositions of the carbonates form a linear cluster with a positive slope of about five, consistent with the depletion of isotope compositions by the infiltration of magmatic and/or meteoric fluids. Rayleigh fractionation during devolatilization played a minor role in determining the final isotope composition. Stable isotope compositions of coexisting calcite–dolomite pairs show varying amounts of isotopic disequilibrium, which correlate with the inferred fluid infiltration mechanism. Much of the calcite in dolostones is the product of infiltration-driven reactions along fractures, and is greatly depleted isotopically relative to the host dolomite, especially at talc grade. At higher grades the calcite–dolomite fractionation is smaller, probably due to both increased fluid–rock interaction and a greater tendency for fluid infiltration to be pervasive on the grain-scale. Limestones generally show near-equilibrium fractionation of oxygen and carbon owing to the overwhelming compositional influence of the host calcite. Veins formed during late-stage hydrothermal circulation have strongly ¹⁸O-depleted compositions relative to the host rock. No small-scale spatial patterns to the isotopic depletion were observed, but the extent of fluid infiltration was greatest in the west of the aureole. Fluid infiltration was clearly highly heterogeneous, with no evidence of a consistent flow direction. It is not possible to determine fluid fluxes or flow directions from one-dimensional flow models based on continuum flow in the Beinn an Dubhaich aureole.     

The correlation of reaction and isotope fronts and the mechanism of metamorphic fluid flow

 Infiltration of a metabasite sill from Islay, Scotland by an H₂O-CO₂ fluid caused (1) modification of δ¹⁸O and (2) carbonation at the sill margins. Maps of δ¹⁸O and reaction progress were constructed from a 20 × 47.7 metre sample grid across the sill. The grid consisted of 300 samples, spaced at m, dm and cm intervals, many of which were analysed for both δ¹⁸O and reaction progress. The δ¹⁸O was determined by laser fluorination of whole rock silicate powders and reaction progress was determined by rapid field-based measurement of % calcite (“fizz-o-meter”, Skelton et al. 1995). Reaction and isotope fronts outlined tube-like features that emanate from the sill margin and discrete nodes that, although detached from the sill margin in two dimensions, are thought to represent sections through similar tubes in three dimensions. We envisage that these protrusions are the fossil record of metamorphic “fluid pathways” whereby fluid permeated the sill. Isotope and reaction fronts are found to correlate spatially as predicted by a modified form of the chromatographic equation which describes this envisaged geometry, that is where isotopic and reactive transport in the fluid phase are facilitated by advection along specific fluid pathways and transverse diffusion in the surrounding rock. These fluid pathways consist of bundles of anastomosing grain boundary channels or micro-cracks, which are thought to propagate through transient cyclic infiltration, reaction, porosity enhancement and fracturing. This mechanism is self-perpetuating and accentuates random perturbations at the sill margin to form the observed tubes. We argue that this is the earliest stage of the infiltration process which has affected metabasites of the SW Scottish Highlands and that subsequent shear deformation of the reacted rims of these pathways, has caused their re-orientation and juxtaposition to form the reacted sill margins described by Skelton et al. (1995). Received: 17 February 1998 / Accepted: 6 December 1999     

Limited fluid-rock interaction at marble-gneiss contacts during Cretaceous granulite-facies metamorphism,Seward Peninsula,Alaska

Stable-isotope profiles show that flat-lying marble units acted as impermeable barriers to upward fluid flow in transitional amphibolite-granulite grade rocks of the Kigluaik Mountains, Seward Peninsula, Alaska. The degree of permeability is related to the composition of the marble. The margin of a thick pure dolomite marble chemically reacted with underlying metasyenite (aH₂O=0.2) to form a 2 cm boundary layer of calcite + forsterite by introduction of SiO₂. No fluid penetrated past this reaction front, although the high temperature of metamorphism (800°C) allowed transport of carbon and oxygen isotopes for an additional 2 cm by diffusion through the solid dolomite. A second marble with a higher silica content underwent more decarbonation, which enhanced porosity and lead to a greater extent of isotope transport (2–3 m) in contact with quartzo-feld-spathic gneiss below. An estimate of total fluid flux across the bottom of this marble layer based on the shape of the isotope profile is 1 cm³/cm² directed down, out of the marble. At two other marble-gneiss contacts steep isotopic gradients coincide with lithologic contacts, indicating very little cross-lithology fluid flow. The extent of diffusional transport of isotopes in the marbles is limited and interpreted as indicating the transient presence of a pore fluid, generated by thermally driven devolatilization reactions. No wholesale pervasive advection of C-O-H fluid occurred across the thick, continuous, marble units near the exposed base of the Kigluaik Group section during the entire regional metamorphic cycle. Activities of pore-fluid species were controlled by internal processes. Movement of volatiles and stable-isotopes between contrasting rock-types was dominantly diffusive. Channelized fluid pathways through the marble units developed during uplift and cooling but were not present during peak metamorphism. Heating of the section occurred by conduction, probably from an underlying magma source, and not by advection of a C-O-H fluid.     

The role of the grain boundary at chemical and isotopic fronts in marble during contact metamorphism

Carbon and oxygen isotopic profiles around a low pressure metasomatic wollastonite reaction front in a marble of the Hida metamorphic terrain, central Japan, display typical metamorphic fluid-enhanced isotopic zonations. Isotopic profiles obtained from detailed microscale analyses perpendicular to the chemical reaction front in calcite marble show that diffusion-enhanced isotopic exchange may control these profiles. Carbon and oxygen isotopic behaviour in grain boundaries is remarkably different. Oxygen isotopic troughs (¹⁸O depleted rims) around the calcite-grain boundaries are widely observed in this contact aureole, demonstrating that diffusion of oxygen in calcite grain boundary dominates over lattice diffusion in calcite. In contrast, no difference is observed in carbon isotopic profiles obtained from grain cores and rims. There is thus no specific role of the grain boundary for diffusion of carbonic species in the metamorphic fluid during transportation. Carbon chemical species such as CO₂ and CO₃ ions in metamorphic fluid migrate mainly through lattice diffusion. The carbon and oxygen isotope profiles may be modelled by diffusion into a semi-infinite medium. Empirically lattice diffusion of oxygen isotopes is almost six times faster than that of carbon isotopes, and oxygen grain-boundary diffusion is ten times faster than oxygen lattice diffusion. Oxygen isotopic results around the wollastonite vein indicate that migration of the metamorphic fluid into calcite marble was small and was parallel to the aquifer. From the stability of wollastonite and the attainment of oxygen isotopic equilibrium, we suggest that diffusion of oxygen occurred through an aqueous fluid phase. The timescale of formation of the oxygen isotopic profile around the wollastonite vein is calculated to be about 0.76 × 10⁶ years using the experimentally determined diffusion constant. Received: 14 January 1997 / Accepted: 23 April 1998     

Permeability generation and resetting of tracers during metamorphic fluid flow: implications for advection-dispersion models

Advection-dispersion fluid flow models implicitly assume that the infiltrating fluid flows through an already fluid-saturated medium. However, whether rocks contain a fluid depends on their reaction history, and whether any initial fluid escapes. The behaviour of different rocks may be illustrated using hypothetical marble compositions. Marbles with diverse chemistries (e.g. calcite + dolomite + quartz) are relatively reactive, and will generally produce a fluid during heating. By contrast, marbles with more restricted chemistries (e.g. calcite + quartz or calcite-only) may not. If the rock is not fluid bearing when fluid infiltration commences, mineralogical reactions may produce a reaction-enhanced permeability in calcite + dolomite + quartz or calcite + quartz, but not in calcite-only marbles. The permeability production controls the pattern of mineralogical, isotopic, and geochemical resetting during fluid flow. Tracers retarded behind the mineralogical fronts will probably be reset as predicted by the advection-dispersion models; however, tracers that are expected to be reset ahead of the mineralogical fronts cannot progress beyond the permeability generating reaction. In the case of very unreactive lithologies (e.g. pure calcite marbles, cherts, and quartzites), the first reaction to affect the rocks may be a metasomatic one ahead of which there is little pervasive resetting of any tracer. Centimetre-scale layering may lead to the formation of self-perpetuating fluid channels in rocks that are not fluid saturated due to the juxtaposition of reactants. Such layered rocks may show patterns of mineralogical resetting that are not predicted by advection-dispersion models. Patterns of mineralogical and isotopic resetting in marbles from a number of terrains, for example: Chillagoe, Marulan South, Reynolds Range (Australia); Adirondack Mountains, Old Woman Mountains, Notch Peak (USA); and Stephen Cross Quarry (Canada) vary as predicted by these models. Received: 3 February 1997 / Accepted: 26 June 1997     

Prograde and retrograde fluid flow during contact metamorphism of siliceous carbonate rocks from the Ballachulish aureole,Scotland

 Siliceous dolomites and limestones contain abundant retrograde minerals produced by hydration-carbonation reactions as the aureole cooled. Marbles that contained periclase at the peak of metamorphism bear secondary brucite, dolomite, and serpentine; forsterite-dolomite marbles have retrograde tremolite and serpentine; wollastonite limestones contain secondary calcite and quartz; and wollastonite-free limestones have retrograde tremolite. Secondary tremolite never appears in marbles where brucite has replaced periclase or in wollastonite-bearing limestones. A model for infiltration of siliceous carbonates by CO₂-H₂O fluid that assumes (a) vertical upwardly-directed flow, (b) fluid flux proportional to cooling rate, and (c) flow and reaction under conditions of local equilibrium between peak temperatures and ≈400 °C, reproduces the modes of altered carbonate rocks, observed reaction textures, and the incompatibility between tremolite and brucite and between tremolite and wollastonite. Except for samples from a dolomite xenolith, retrograde time-integrated flux recorded by reaction progress is on the order of 1000 mol fluid/cm² rock. Local focusing of flow near the contact is indicated by samples from the xenolith that record values an order of magnitude greater. Formation of periclase, forsterite, and wollastonite at the peak of metamorphism also required infiltration with prograde time-integrated flux approximately 100–1000 mol/cm². The comparatively small values of prograde and retrograde time-integrated flux are consistent with lack of stable isotope alteration of the carbonates and with the success of conductive thermal models in reproducing peak metamorphic temperatures recorded by mineral equilibria. Although isobaric univariant assemblages are ubiquitous in the carbonates, most formed during retrograde metamorphism. Isobaric univariant assemblages observed in metacarbonates from contact aureoles may not record physical conditions at the peak of metamorphism as is commonly assumed. Received: 19 September 1995 / Accepted: 14 March 1996     

Fluid-rock interaction and thermal evolution during thrusting of an Alpine metamorphic complex (Tinos island, Greece)

This study examines the fluid-rock interaction and thermal evolution along a thrust that juxtaposes calcite-rich marbles of high P-T metamorphic unit of the Attic-Cycladic Massif (Greece) on top of a lower-grade dolomite marble unit. The Tertiary thrust represents a major phase of tectonic movement related to the decompression of the Alpine orogen in the Hellenides. The stable isotope signatures of the thrust plane and adjacent sections of the footwall and hanging wall rocks are characterized by significant carbon and oxygen isotope depletions. The depletion is most pronounced in calcite, but is almost entirely missing in coexisting dolomite. The isotopic patterns in the thrust zone can be explained by the infiltration of an externally derived water-rich H₂O-CO₂-CH₄ fluid [X _C (=X _{CO 2}+X _{CH 4})<0.05] at water-rock ratios on the order of 0.1 to 0.5 by weight. The fluid-induced calcite recrystallization is viewed as an important rheological control during thrusting. The temperature evolution of the footwall, hanging wall and mylonitic tectonic contact was determined by calcite-dolomite solvus thermometry. Histograms of calcite-dolomite temperatures are interpreted as indicating a heating of the footwall dolomite marble during the thrusting of the hotter upper plate. Conversely, the hanging wall marble unit was cooled during the thrusting. The calcite-dolomite thermometry of the thrust plane gives temperatures intermediate between the initial temperatures of the lower and upper marble units, and this leads to the conclusion that conductive heat transfer rather than fluid infiltration controlled the thermal evolution during thrusting. Received: 14 April 1998 / Accepted: 9 December 1998     

Carbonates in CM2 chondrites: constraints on alteration conditions from oxygen isotopic compositions and petrographic observations

High-precision measurements of the oxygen isotopic compositions of carbonates (calcite and dolomite) from five CM2 chondrites are presented and put into context of the previously determined mineralogic alteration index (MAI), which places these meteorites into an alteration sequence. The carbonate oxygen isotopic compositions range from +20.0 to +35.7‰ for δ¹⁸O, +8.0 to +17.7‰ for δ¹⁷O, and −0.7 to −2.7‰ for Δ¹⁷O. Carbonate Δ¹⁷O values are inversely correlated with MAI and track the evolution of fluid composition from higher to lower Δ¹⁷O values with increasing alteration on the CM parent body. Similar Δ¹⁷O values for calcite and dolomite fractions from the same splits of the same meteorites indicate that calcite and dolomite in each split precipitated from a single fluid reservoir. However, reversed calcite dolomite fractionations (δ¹⁸O_dol − δ¹⁸O_cc) indicate that the fluid was subject to processes, such as freeze-thaw or evaporation, that fractionated isotopes in a mass-dependent way. Consideration of the carbonate isotopic data in the context of previously proposed models for aqueous alteration of carbonaceous chondrites has provided important insights into both the evolving alteration conditions and the utility of the models themselves. The data as a whole indicate that the isotopic evolution of the fluid was similar to that predicted by the closed-system, two-reservoir models, but that a slightly larger matrix-water fractionation factor may apply. In the context of this model, more altered samples largely reflect greater reaction progress and thus probably indicate more extended times of fluid exposure. Petrographic observations of carbonates reveal a trend of variable carbonate morphology correlated with alteration that is also consistent with changes in the duration of fluid-rock interaction. The data can also be reconciled with fluid-flow models in a restricted region of the parent body, which is consistent with assertions that the different types of carbonaceous chondrites derive from different regions of their parent bodies. In this case, the model results for a 9-km-radius body, and our data place the location of the CM chondrite formation in a 100-m-thick zone 1 km from the surface. The size of this zone could be increased if the model parameters were adjusted.     

Contrasting fluid evolution of granulite-facies marbles: implications for a high-T intermediate-P terrain in the Famatinian Range, San-Juan, Argentina

High-temperature, intermediate-pressure calc-silicate marbles occur in the granulite-facies terrain of the La Huerta Range in the Province of San Juan, NW-Argentina, in three bulk-compositional varieties: Type (1) dolomite-absent scapolite-wollastonite-grandite-clinopyroxene-quartz—calcite marbles; Type (2) diopside-forsterite-spinel-corundum—calcite marbles with dolomite exolution lamellae in calcite; Type (3) serpentinized forsterite-spinel-dolomite marbles. An isobaric cooling path from peak-metamorphic conditions of 860°C to 750°C at 6.5 kbar is inferred from scapolite-wollastonite-grandite reaction textures in Type (1) and is consistent with cooling after an advective heat input from related gabbroic and tonalitic intrusive bodies. Stable carbon and oxygen isotope geochemistry was used to decipher the fluid/rock evolution of the three marble types. An interpreted four-stage temperature-time-fluid flow path comprises: (1) infiltration of pre-peak-metamorphic fluids, depleted in δ¹⁸O, that caused a shift of primary sedimentary δ¹⁸O ratios to lower values (19.6–20.0); (2) syn-metamorphic fluid liberation from Type (1) marbles with evidence for processes close to batch devolatilization that caused a weak coupled ¹³C and ¹⁸O depletion during prograde metamorphism. A different devolatilization behaviour, close to Rayleigh fractionation, texturally associated with fold settings indicates that granulite-facies fluid flow was focused rather than pervasive; (3) H₂O-absent conditions were dominant when coronal grandite formed during incipient high-temperature isobaric cooling at the expense of scapolite and wollastonite in the Type (1) marbles; (4) intense post-peak- hydration of Type (2) and Type (3) marbles is the last recognizable metasomatic event. In combination, the three marble types record fluid infiltration both before and after the metamorphic peak.     

P-T-X effects on equilibrium carbonate-H2O-CO2-NaCl dihedral angles: constraints on carbonate permeability and the role of deformation during fluid infiltration

Fluid-solid-solid dihedral angles in the NaCl-H₂O-CO₂-calcite-dolomite-magnesite system have been determined at pressures ranging from 0.5 to 7 kbar and temperatures from 450°C to 750°C. At 1 kbar and 650°C, both dolomite and magnesite exhibit a dihedral angle minimum for intermediate H₂O-CO₂ fluids similar to that previously determined by the present authors for calcite, but the depth of the minimum is smaller, being above the critical value of 60° for both dolomite and magnesite for all fluid compositions. Calcite-calcite-brine dihedral angles at 650°C have been determined in the pressure range 1–5 kbar. Angles decrease with increasing salt content of the fluid, tending towards a constant value of about 65° for strong brines at pressures above 2 kbar. There is a general increase of angle with increasing pressure which is most marked for strong brines. A positive correlation of angle with pressure is also observed in calcite-H₂O-CO₂ fluids, the position of the minimum moving towards higher angles and towards H₂O-rich fluids with increasing pressure. The permeability window previously observed by the present authors at 1 kbar and intermediate fluid compositions closes at about 1.5 kbar. The results demonstrate that the permeability of carbonates to grain edge fluid flow is only possible at low pressures and for fluids of restricted H₂O-CO₂-NaCl compositions. However, geochemical evidence from metamorphic terrains suggests that pervasive infiltration does occur under conditions where impermeability is predicted. From examination of published studies of infiltrated carbonates we conclude that deformation plays a critical role in enhancing carbonate permeability. Possible mechanisms for this include shear-enhanced dilatancy (micro-cracking), fluid inclusion drag by deformation-controlled grain boundary migration, and dynamically maintained transient grain boundary fluid films.     

An empirical estimate of the diffusion rate of oxygen in diopside

The intracrystalline diffusion rate of oxygen in diopside was constrained based on natural isotopic variations from a granulite facies marble from Cascade Slide, Adirondacks (New York, USA). The oxygen isotope compositions of the diopsides, measured as a function of grain size, are nearly constant (20.9 ± 0.3‰ vs. SMOW) over the entire measured size range (0.3–3.2 mm diameter). The δ¹⁸O values of the cores of calcite grains are 23.0‰. Temperature estimates based on the Δ¹⁸O(calcite-diopside) are 800d?C, in agreement with the highest previous thermometric estimates for these rocks. The lack of isotopic variation in the diopsides as a function of grain size requires that the oxygen intracrystalline diffusion rate in diopside from the Adirondack samples was very slow. The maximum diffusion rates (D_800d?C parallel to the c-axis) were calculated with an infinite reservoir model (IRM) and a finite reservoir model (FRM) that incorporates mineral modal abundances and initial isotopic variations. For an assumed activation energy (Q) = 100 kJ/mol, the IRM diffusion rate estimate of 1.6 times 10^-20cm²/s is two orders of magnitude faster than from the FRM; at Q=500kJ/mol, the D_800d?C estimate for both methods is c. 5.6 times 10^-20 cm²/s. The present results require that a hydrothermal fluid significantly enhances the diffusion rate of oxygen in diopside if previous data are correct. The δ¹⁸O(SMOW) and δ¹³C(PDB) values of the calcite, measured in situ with a CO₂ laser, are 22.9 ± 0.3, 0.1±0.3‰ in the grain cores, 22.1 ±0.3, 0.2 ±0.1‰ at the grain boundaries and 21.7 ±0.4, -0.6±0.1‰ abutting diopside grains. The δ¹⁸O and δ¹³δC values measured conventionally are: crystal cores, 22.96, -0.95‰; abutting diopside grains, 22.38, -0.93‰; bulk, 22.79, -0.95%. Use of the bulk δ¹⁸O(calcite) values for thermometry yields unreasonably high temperatures. The lower δ¹⁸O values at the calcite grain boundaries are not due to retrograde diffusional exchange with the diopside, they are thought to be a result of a late retrograde fluid infiltration.     

Phase equilibrium and stable isotope constraints on the formation of metasomatic garnet-vesuvianite veins (SW Adamello,N Italy)

 Metasomatic garnet-vesuvianite veins occur within the contact metamorphic marble sequence of the Lower Triassic Prezzo formation in a narrow, 1–5 m wide zone along an intrusive marble-granodiorite contact at the southwestern border of the Tertiary Adamello batholith. The metasomatic mineral assemblage is comprised of garnet, vesuvianite, clinopyroxene, wollastonite, and pyrrhotite, which were precipitated from the vein-forming fluid in a preexisting calcite matrix at conditions of about 2800 bars and 630° C. The veins are enriched in silicon, aluminum, iron, magnesium, titanium and depleted in calcium with respect to the unaltered contact metamorphic marble. Graphite, which is present in the unaltered Prezzo Marble is absent in the veins. Irregularly shaped mineralogically distinct zones with different degrees of silicification can be distinguished within the veins. The isotopic compositions of calcite (cc) in the unaltered marble are about δ¹⁸O (SMOW; Standard mean Ocean Water)=21.0‰ and δ¹³C(PDB; Peedee belemnite)=0.0‰. They are reset to significantly lower values within the veins, where δ¹⁸O_cc is 15.0 to 16.0‰ and δ¹³C_cc is −4.5 to −3.5‰. The isotopic front coincides with an abrupt change in the microscopic texture of matrix carbonate which occurs at the sharp boundary between graphite-bearing and graphite-free material. Within the veins the oxygen isotope fractionation between calcite and garnet (gar) varies systematically with distance from highly silicified zones. The variations in Δ¹⁸O_cc-gar are as large as 2‰, on a millimeter scale, indicating garnet-calcite isotopic disequilibrium. Vein formation was due to the infiltration of a water rich fluid of magmatic provenance into the carbonate country rock along fractures. Removal of graphite from the wall rock by dissolution through the metasomatic fluid induced recrystallization of matrix calcite. Permeability was enhanced during calcite recrystallization facilitating material transport into the wall rock and metasomatic alteration. Vein garnet was precipitated in isotopic equilibrium with the metasomatic fluid. The isotopic composition of preexisting calcite was initially out of equilibrium with the vein-forming fluid and it was shifted towards equilibrium by surface-reaction controlled calcite-fluid isotopic exchange during calcite recrystallization. Due to the short lifetime of the metasomatic system, calcite-fluid isotopic equilibrium was generally not attained. Within the veins, oxygen and carbon transport was fast relative to mineral-fluid exchange of their isotopes and the geometry of the isotopic pattern is largely controlled by the kinetics of mineral-fluid exchange. Received: 16 June 1994/Accepted: 20 May 1995     

Constraints on the application of carbon isotope thermometry in high- to ultrahigh-temperature metamorphic terranes

Nine marble horizons from the granulite facies terrane of southern India were examined in detail for stable carbon and oxygen isotopes in calcite and carbon isotopes in graphite. The marbles in Trivandrum Block show coupled lowering of δ¹³C and δ¹⁸O values in calcite and heterogeneous single crystal δ¹³C values (? 1 to ? 10‰) for graphite indicating varying carbon isotope fractionation between calcite and graphite, despite the granulite facies regional metamorphic conditions. The stable isotope patterns suggest alteration of δ¹³C and δ¹⁸O values in marbles by infiltration of low δ¹³C–δ¹⁸O‐bearing fluids, the extent of alteration being a direct function of the fluid‐rock ratio. The carbon isotope zonation preserved in graphite suggests that the graphite crystals precipitated/recrystallized in the presence of an externally derived CO₂‐rich fluid, and that the infiltration had occurred under high temperature and low f_O2 conditions during metamorphism. The onset of graphite precipitation resulted in a depletion of the carbon isotope values of the remaining fluid+calcite carbon reservoir, following a Rayleigh‐type distillation process within fluid‐rich pockets/pathways in marbles resulting in the observed zonation. The results suggest that calcite–graphite thermometry cannot be applied in marbles that are affected by external carbonic fluid infiltration. However, marble horizons in the Madurai Block, where the effect of fluid infiltration is not detected, record clear imprints of ultrahigh temperature metamorphism (800–1000 °C), with fractionations reaching <2‰. Zonation studies on graphite show a nominal rimward lowering δ¹³C on the order of 1 to 2‰. The zonation carries the imprint of fluid deficient/absent UHT metamorphism. Commonly, calculated core temperatures are > 1000 °C and would be consistent with UHT metamorphism.     

Ultra-high-pressure metamorphism of carbonate rocks in the Dabie Mountains, central China

Abstract Widespread ultra-high-P assemblages including coesite, quartz pseudomorphs after coesite, aragonite, and calcite pseudomorphs after aragonite in marble, gneiss and phengite schist are present in the Dabie Mountains eclogite terrane. These assemblages indicate that the ultra-high-P metamorphic event occurred on a regional scale during Triassic collision between the Sino-Korean and Yangtze cratons. Marble in the Dabie Mountains is interlayered with coesite-bearing eclogite and gneiss and as blocks of various size within gneiss. Discontinuous boudins of eclogite occur within marble layers. Marble contains an ultra-high-P assemblage of calcite/aragonite, dolomite, clinopyroxene, garnet, phengite, epidote, rutile and quartz/coesite. Coesite, quartz pseudomorphs after coesite, aragonite and calcite pseudomorphs after aragonite occur as fine-grained inclusions in garnet and omphacite. Phengites contain about 3.6 Si atoms per formula unit (based on 11 oxygens). Similar to the coesite-bearing eclogite, marble exhibits retrograde recrystallization under amphibolite–greenschist facies conditions generated during uplift of the ultra-high-P metamorphic terrane. Retrograde minerals are fine grained and replace coarse-grained peak metamorphic phases. The most typical replacements are: symplectic pargasitic hornblende + epidote after garnet, diopside + plagioclase (An₁₈) after omphacite, and fibrous phlogopite after phengite. Ferroan pargasite + plagioclase, and actinolite formed along grain boundaries between garnet and calcite, and calcite and quartz, respectively. The estimated peak P–T conditions for marble are comparable to those for eclogite: garnet–clinopyroxene geothermometry yields temperatures of 630–760°C; the garnet–phengite thermometer gives somewhat lower temperatures. The minimum pressure of peak metamorphism is 27 kbar based on the occurrence of coesite. Such estimates of ultra-high-P conditions are consistent with the coexistence of grossular-rich garnet + rutile, and the high jadeite content of omphacite in marble. The fluid for the peak metamorphism was calculated to have a very low X_CO2 (<0.03). The P–T conditions for retrograde metamorphism were estimated to be 475–550°C at <7 kbar.     

Grain growth kinetics of dolomite,magnesite and calcite: a comparative study

The rates of grain growth of stoichiometric dolomite [CaMg(CO₃)₂] and magnesite (MgCO₃) have been measured at temperatures T of 700–800°C at a confining pressure P _c of 300 MPa, and compared with growth rates of calcite (CaCO₃). Dry, fine-grained aggregates of the three carbonates were synthesized from high purity powders by hot isostatic pressing (HIP); initial mean grain sizes of HIP-synthesized carbonates were 1.4, 1.1, and 17 μm, respectively, for CaMg(CO₃)₂, MgCO₃, and CaCO₃, with porosities of 2, 28, and 0.04% by volume. Grain sizes of all carbonates coarsened during subsequent isostatic annealing, with mean values reaching 3.9, 5.1, and 27 μm for CaMg(CO₃)₂, MgCO₃, and CaCO₃, respectively, in 1 week. Grain growth of dolomite is much slower than the growth rates of magnesite or calcite; assuming normal grain growth and n = 3 for all three carbonates, the rate constant K for dolomite (≃5 × 10⁻⁵ μm³/s) at T = 800°C is less than that for magnesite by a factor of ~30 and less than that for calcite by three orders of magnitude. Variations in carbonate grain growth may be affected by differences in cation composition and densities of pores at grain boundaries that decrease grain boundary mobility. However, rates of coarsening correlate best with the extent of solid solution; K is the largest for calcite with extensive Mg substitution for Ca, while K is the smallest for dolomite with negligible solid solution. Secondary phases may nucleate at advancing dolomite grain boundaries, with implications for deformation processes, rheology, and reaction kinetics of carbonates.     

Fluid infiltration during contact metamorphism of interbedded marble and calc-silicate hornfels, Twin Lakes area, central Sierra Nevada, California

Abstract In the Twin Lakes area, central Sierra Nevada, California, most contact metamorphosed marbles contain calcite + dolomite + forsterite ± diopside ± phlogopite ± tremolite, and most calc-silicate hornfelses contain calcite + diopside + wollastonite + quartz ± anorthite ± K-feldspar ± grossular ± titanite. Mineral-fluid equilibria involving calcite + dolomite + tremolite + diopside + forsterite in two marble samples and wollastonite + anorthite + quartz + grossular in three hornfels samples record P± 3 kbar and T± 630° C. Various isobaric univariant assemblages record CO₂-H₂O fluid compositions of χCO₂= 0.61–0.74 in the marbles and χCO₂= 0.11 in the hornfelses. Assuming a siliceous dolomitic limestone protolith consisting of dolomite + quartz ° Calcite ± K-feldspar ± muscovite ± rutile, all plausible prograde reaction pathways were deduced for marble and hornfels on isobaric T-XCO₂ diagrams in the model system K₂O-CaO-MgO-Al₂O₃-SiO₂-H₂O-CO₂. Progress of the prograde reactions was estimated from measured modes and mass-balance calculations. Time-integrated fluxes of reactive fluid which infiltrated samples were computed for a temperature gradient of 150 °C/km along the fluid flow path, calculated fluid compositions, and estimated reaction progress using the mass-continuity equation. Marbles and hornfelses record values in the range 0.1–3.6 × 10⁴ cm³/cm² and 4.8–12.9 × 10⁴ cm³/cm², respectively. For an estimated duration of metamorphism of 10⁵ years, average in situ metamorphic rock permeabilities, calculated from Darcy's Law, are 0.1–8 × 10^?6 D in the marbles and 10–27 × 10^?6 D in the hornfelses. Reactive metamorphic fluids flowed up-temperature, and were preferentially channellized in hornfelses relative to the marbles. These results appear to give a general characterization of hydrothermal activity during contact metamorphism of small pendants and screens (dimensions ± 1 km or less) associated with emplacement of the Sierra Nevada batholith.     

Assessment of grain-scale homogeneity and equilibration of carbon and oxygen isotope compositions of minerals in carbonate-bearing metamorphic rocks by ion microprobe

Nineteen samples of metamorphosed carbonate-bearing rocks were analyzed for carbon and oxygen isotope ratios by ion microprobe with a ∼5-15 μm spot, three from a regional terrain and 16 from five different contact aureoles. Contact metamorphic rocks further represent four groups: calc-silicate marble and hornfels (6), brucite marble (2), samples that contain a reaction front (4), and samples with a pervasive distribution of reactants and products of a decarbonation reaction (4). The average spot-to-spot reproducibility of standard calcite analyses is ±0.37‰ (2 standard deviations, SD) for δ¹⁸O and ±0.71‰ for δ¹³C. Ten or more measurements of a mineral in a sample that has uniform isotope composition within error of measurement can routinely return a weighted mean with a 95% confidence interval of 0.09-0.16‰ for δ¹⁸O and 0.10-0.29‰ for δ¹³C. Using a difference of >6SD as the criterion, only four of 19 analyzed samples exhibit significant intracrystalline and/or intercrystalline inhomogeneity in δ¹³C at the 100-500 μm scale, with differences within individual grains up to 3.7‰. Measurements are consistent with carbon isotope exchange equilibrium between calcite and dolomite in five of six analyzed samples at the same scale. Because of relatively slow carbon isotope diffusion in calcite and dolomite, differences in δ¹³C can survive intracrystalline homogenization by diffusion during cooling after peak metamorphism and likely represent the effects of prograde decarbonation and infiltration. All but 2 of 11 analyzed samples exhibit intracrystalline differences in δ¹⁸O (up to 9.4‰), intercrystalline inhomogeneity in δ¹⁸O (up to 12.5‰), and/or disequilibrium oxygen isotope fractionations among calcite-dolomite, calcite-quartz, and calcite-forsterite pairs at the 100-500 μm scale. Inhomogeneities in δ¹⁸O and δ¹³C are poorly correlated with only a single mineral (dolomite) in a single sample exhibiting both. Because of relatively rapid oxygen isotope diffusion in calcite, intracrystalline inhomogeneities in δ¹⁸O likely represent partial equilibration between calcite and fluid during retrograde metamorphism. Calcite is in oxygen isotope exchange equilibrium with forsterite in one of four analyzed samples, in equilibrium with dolomite in none of six analyzed samples, and in equilibrium with quartz in neither of two analyzed samples. There are no samples of contact metamorphic rock with analyzed reactants and products of an arrested metamorphic reaction that are in oxygen isotope equilibrium with each other. The degree of departure from equilibrium in analyzed samples is variable and is often related, at least in part, to alteration of δ¹⁸O of calcite during retrograde fluid-rock reaction. In situ sub-grain-scale carbon and oxygen isotope analyses of minerals are advisable in the common applications of stable isotope geochemistry to metamorphic petrology. Correlation of sub-mm scale stable isotope data with imaging will lead to improved understanding of reaction kinetics, reactive fluid flow, and thermal histories during metamorphism.     

Stable isotopic signatures of superposed fluid events in granulite facies marbles of the Rauer Group, East Antarctica

The role of volatiles in the stabilization of the lower (granulite facies) crust is contentious. Opposing models invoke infiltration of CO₂-rich fluids or generally vapour-absent conditions during granulite facies metamorphism. Stable isotope and petrological studies of granulite facies metacarbonates can provide constraints on these models. In this study data are presented from metre-scale forsteritic marble boudins within Archaean intermediate to felsic orthogneisses from the Rauer Group, East Antarctica. Forsteritic marble layers and associated calcsilicates preserve a range of ¹³C- and ¹⁸O-depleted calcite isotope values (δ¹³C= -9.9 to -3.0% PDB, δ¹⁸O = 4.0 to 12.1% SMOW). A coupled trend of ¹³C and ¹⁸O depletion (～2%, ～5%, respectively) from core to rim across one marble layer is inconsistent with pervasive CO₂ infiltration during granulite facies metamorphism, but does indicate localized fluid-rock interaction. At another locality, more pervasive fluid infiltration has resulted in calcite having uniformly low, carbonatite-like δ¹⁸O and δ¹³C values. A favoured mechanism for the low δ¹⁸O and δ¹³C values of the marbles is infiltration by fluids that were derived from, or equilibrated with, a magmatic source. It is likely that this fluid-rock interaction occurred prior to high-grade metamorphism; other fluid-rock histories are not, however, ruled out by the available data. Coupled trends of ¹³C and ¹⁸O depletion are modified to even lower values by the superposed development of small-scale metasomatic reaction zones between marbles and internally folded mafic (?) interlayers. The timing of development of these layers is uncertain, but may be related to Archaean high-temperature (>1000d?C) granulite facies metamorphism.