Garnet re‐equilibration by coupled dissolution–reprecipitation: evidence from textural,major element and oxygen isotope zoning of ‘cloudy’ garnet

The analysis of texture, major element and oxygen isotope compositions of cloudy garnet crystals from a metapelite sampled on Ikaria Island (Greece) is used to assess the model of growth and re‐equilibration of these garnet crystals and to reconstruct the pressure–temperature–fluid history of the sample. Garnet crystals show complex textural and chemical zoning. Garnet cores (100–200 μm) are devoid of fluid inclusions. They are characterized by growth zoning demonstrated by a bell‐shaped profile of spessartine component (7–3 mol.%), an increase in grossular from 14 to 22 mol.% and δ¹⁸O values between 9.5 ± 0.3‰ and 10.4 ± 0.2‰. Garnet inner rims (90–130 μm) are fluid inclusion‐rich and show a decreasing grossular component from 22 to 5 mol.%. The trend of the spessartine component observed in the inner rim allows two domains to be distinguished. In contrast to domain I, where the spessartine content shows the same trend as in the core, the spessartine content of domain II increases outwards from 2 to 14 mol.%. The δ¹⁸O values decrease towards the margins of the crystals to a lowest value of 7.4 ± 0.2‰. The outer rims (<10 μm) are devoid of fluid inclusions and have the same chemical composition as the outermost part of domain II of the inner rim. Garnet crystals underwent a four‐stage history. Stage 1: garnet growth during the prograde path in a closed system for oxygen. Garnet cores are remnants of this growth stage. Stage 2: garnet re‐equilibration by coupled dissolution–reprecipitation at the temperature peak (630 < T < 650 °C). This causes the creation of porosity as the coupled dissolution–reprecipitation process allows chemical (Ca) and isotopic (O) exchange between garnet inner rims and the matrix. The formation of the outer rim is related to the closure of porosity. Stage 3: garnet mode decreases during the early retrograde path, but garnet is still a stable phase. The resulting garnet composition is characterized by an increasing Mn content in the inner rim’s domain II caused by intracrystalline diffusion. Stage 4: dissolution of garnet during the late retrograde path as garnet is not a stable phase anymore. This last stage forms corroded garnet. This study shows that coupled dissolution–reprecipitation is a possible re‐equilibration process for garnet in metamorphic rocks and that intra‐mineral porosity is an efficient pathway for chemical and isotopic exchange between garnet and the matrix, even for otherwise slow diffusing elements.     

The partial equilibration of garnet porphyroblasts in pelitic schists and its control on prograde metamorphism,Glen Roy,Scotland

Garnet porphyroblasts in sillimanite‐bearing pelitic schists contain complex textural and compositional zoning, with considerable variation both within and between adjacent samples. The sillimanite‐bearing schists locally occur in regional Barrovian garnet zone assemblages and are indicative of a persistent lack of equilibrium during prograde metamorphism. Garnet in these Dalradian rocks from the Scottish Highlands preserves evidence of a range of metamorphic responses including initial growth and patchy coupled dissolution–reprecipitation followed by partial dissolution. Individual porphyroblasts each have a unique and variable response to prograde metamorphism and garnet with mainly flat compositional profiles co‐exists with those containing largely unmodified characteristic bell‐shaped Mn profiles. This highlights the need for caution in applying traditional interpretations of effective volume diffusion eliminating compositional variation. Cloudy garnet with abundant fluid inclusions is produced during incomplete modification of the initial porphyroblasts and these porous garnet are then particularly prone to partial replacement in sillimanite‐producing reactions. The modification of garnet via a dissolution–reprecipitation process releases Ca into the effective whole‐rock composition, displacing the pressure–temperature positions of subsequent isograd reactions. This represents the first report of internal metasomatism controlling reaction pathways. The behaviour of garnet highlights the importance of kinetic factors, especially deformation and fluids, in controlling reaction progress and how the resulting variability influences subsequent prograde history. The lack of a consistent metamorphic response, within and between adjacent schists, suggests that on both local and regional scales these rocks have largely not equilibrated at peak metamorphic conditions.     

Multiple growth of garnet,sillimanite/kyanite and monazite during amphibolite facies metamorphism: implications for the P–T–t and tectonic evolution of the western Altai Range,Mongolia

Four amphibolite facies pelitic gneisses from the western Mongolian Altai Range exhibit multistage aluminosilicate formation and various chemical‐zoning patterns in garnet. Two of them contain kyanite in the matrix and sillimanite inclusions in garnet, and the others have kyanite inclusions in garnet with sillimanite or kyanite in the matrix. The Ca‐zoning patterns of the garnet are different in each rock type. U–Th–Pb monazite geochronology revealed that all rock units experienced a c. 360 Ma event, and three of them were also affected by a c. 260 Ma event. The variations in the microstructures and garnet‐zoning profiles are caused by the differences in the (i) whole‐rock chemistry, (ii) pressure conditions during garnet growth at c. 360 Ma and (iii) equilibrium temperatures at c. 260 Ma. The garnet with sillimanite inclusions records an increase in pressure at low‐P (~5.2–7.2 kbar) and moderate temperature conditions (~620–660 °C) at c. 360 Ma. The garnet with kyanite inclusions in the other rock types was also formed during an increase in pressure but at higher pressure conditions (~7.0–8.9 kbar at ~600–640 °C). The detrital zircon provenance of all the rock types is similar and is consistent with that from the sedimentary rocks in the Altai Range, suggesting that the provenance of all the rock types was a surrounding accretionary wedge. One possible scenario for the different thermal gradient is Devonian ridge subduction beneath the Altai Range, as proposed by several researchers. The subducting ridge could have supplied heat to the accretionary wedge and elevated the geotherm at c. 360 Ma. The differences in the thermal gradients that resulted in varying prograde P–T paths might be due to variations in the thermal regimes in the upper plate that were generated by the subducting ridge. The c. 260 Ma event is characterized by a relatively high‐T/P gradient (~25 °C km^?1) and may be due to collision‐related granitic activity and re‐equilibrium at middle crustal depths, which caused the variations in the aluminosilicates in the matrix between the rock units.     

Deformation of amphibolites via dissolution–precipitation creep in the middle and lower crust

Continuous compositional zoning in amphibole grains in strongly deformed and lineated amphibolites from the Eastern Blue Ridge, North Carolina indicates that most of the deformation was accommodated by dissolution–precipitation creep. Amphibole in most samples shows moderate prograde and/or retrograde zoning parallel to the long‐axis with compositions ranging between magnesiohornblende and tschermakite. In one sample, grains are zoned from actinolitic (Si = 7.9 p.f.u.) cores to tschermakitic (Si = 6.2 p.f.u) rims. Amphibole‐plagioclase thermometry suggests prograde growth temperatures as low as 400 °C, but typically range from 650 to 730 °C and retrograde growth temperatures <700 °C. These estimates are corroborated quantitatively with amphibole‐garnet‐plagioclase thermobarometry and qualitatively with a positive correlation between TiO₂ concentration in amphibole and calculated temperature. This growth zoning provides persuasive evidence that amphibole precipitation produced the fabric, but evidence for dissolution is less common. It is present, however in the form of truncations of complicated zoning patterns produced by healed fractures and overgrowths in low‐temperature cores by high‐temperature tschermakitic grains lacking similar internal structures. The preservation of this network of straight cracks filled with optically continuous amphibole also provides evidence against the operation of dislocation creep even to temperatures >700 °C because dislocation‐creep would have deformed the fracture network. Thus, these amphibolites deformed by dissolution–precipitation creep that produced a strong linear fabric under upper amphibolite facies, middle‐to‐lower crustal conditions. The significance of this discovery is that dissolution–precipitation creep is activated at lower stresses than dislocation creep and that the strength of the lower crust, where amphibole is the dominant mineral is probably lower than that derived from experimental studies.     

Reaction softening by dissolution–precipitation creep in a retrograde greenschist facies ductile shear zone,New Hampshire,USA

We describe strain localization by a mixed process of reaction and microstructural softening in a lower greenschist facies ductile fault zone that transposes and replaces middle to upper amphibolite facies fabrics and mineral assemblages in the host schist of the Littleton Formation near Claremont, New Hampshire. Here, Na‐poor muscovite and chlorite progressively replace first staurolite, then garnet, and finally biotite porphyroblasts as the core of the fault zone is approached. Across the transect, higher grade fabric‐forming Na‐rich muscovite is also progressively replaced by fabric‐forming Na‐poor muscovite. The mineralogy of the new phyllonitic fault‐rock produced is dominated by Na‐poor muscovite and chlorite together with late albite porphyroblasts. The replacement of the amphibolite facies porphyroblasts by muscovite and chlorite is pseudomorphic in some samples and shows that the chemical metastability of the porphyroblasts is sufficient to drive replacement. In contrast, element mapping shows that fabric‐forming Na‐rich muscovite is selectively replaced at high‐strain microstructural sites, indicating that strain energy played an important role in activating the dissolution of the compositionally metastable muscovite. The replacement of strong, high‐grade porphyroblasts by weaker Na‐poor muscovite and chlorite constitutes reaction softening. The crystallization of parallel and contiguous mica in the retrograde foliation at the expense of the earlier and locally crenulated Na‐rich muscovite‐defined foliation destroys not only the metastable high‐grade mineralogy, but also its stronger geometry. This process constitutes both reaction and microstructural softening. The deformation mechanism here was thus one of dissolution–precipitation creep, activated at considerably lower stresses than might be predicted in quartzofeldspathic rocks at the same lower greenschist facies conditions.     

The duration of near‐peak metamorphism from diffusion modelling of garnet zoning

Garnet in a staurolite–kyanite zone sample from central Vermont displays a bell‐shaped Mn growth zoning with diffusional modification over the outer 100 μm. The diffusion is driven by the prograde net transfer reaction garnet + chlorite = kyanite + biotite as is evidenced by a well‐defined resorption zone on the rim. Analysis of the reaction history and resorbed garnet composition suggests that the peak temperature attained was 620–660 °C. Diffusional modelling of the rim diffusion provides an estimate of the duration of the metamorphic episode over which significant garnet diffusion occurs. The duration is a function of the assumed peak temperature and garnet diffusivities and range from a few hundred thousand years to a few million years. Such short durations require rapid tectonic burial and exhumation of relatively thin tectonic slices.     

The origin of chloritoid – 3‐mica pseudomorph growth in staurolite–muscovite schist,Bangriposi (Eastern India)

At Bangriposi, variable stages in replacement of staurolite by chloritoid – Na–K–Ca mica shimmer aggregates in muscovite schists provides insight into the complex interplay between fluid flow, mass transfer, and dissolution–precipitation during pseudomorph growth. Idioblastic chloritoid growing into mica caps without causing visible deformation, and monomineralic chloritoid veins (up to 300 μm wide) within shimmer aggregates replacing staurolite attest to chloritoid nucleation in fluid‐filled conduits along staurolite grain boundaries and crystallographic planes. The growth of shimmer aggregates initiated along staurolite margins, and advanced inwards into decomposing staurolite along networks of crystallographically controlled fluid‐filled conduits. Coalescence among alteration zones adjacent to channel fills led to dismemberment and the eventual demise of staurolite. Mass balance calculation within a volume‐fixed, silica‐conserved reference frame indicate the shimmer aggregates grew via precipitation from fluids in response to mass transport that led to the addition of H₂O, K₂O, Na₂O and CaO in the reaction zone, and Al₂O₃ was transported outward from the inward‐retreating margin of decomposing staurolite. This aided precipitation of chloritoid in veins and in the outer collars, and as disseminated grains in the shimmer aggregates at mid‐crustal condition (~520 ± 20 °C, 5.5 ± 2.0 kbar). Computation using one‐dimensional transport equation suggests that staurolite decomposition involved advection dominating over diffusive transport; the permeation of externally derived H₂O caused flattening of chemical potential gradients in H₂O and aqueous species, for example, and , computed using the Gibbs method. This suggests that staurolite decomposition was promoted by the infiltration of a large volume of H₂O that flattened existing chemical potential gradients. In the initial stages of replacement, chloritoid super‐saturation in fluid caused preferential nucleation and growth of chloritoid at staurolite grain boundaries and in crystallographic planes. As reaction progressed, further chloritoid nucleation was halted, but chloritoid continued to grow as the 3‐mica aggregates continued to replace the remaining staurolite in situ, while the chloritoid‐compatible elements were transported in the water‐rich phase facilitating continued growth of the existing chloritoid grains.     

Jadeite–garnet glaucophane schists in the Bizan area,Sambagawa metamorphic belt,eastern Shikoku,Japan: significance and extent of eclogite facies metamorphism

Eclogite facies metamorphic rocks have been discovered from the Bizan area of eastern Shikoku, Sambagawa metamorphic belt. The eclogitic jadeite–garnet glaucophane schists occur as lenticular or sheet‐like bodies in the pelitic schist matrix, with the peak mineral assemblage of garnet + glaucophane + jadeite + phengite + quartz. The jadeitic clinopyroxene (X_Jd 0.46–0.75) is found exclusively as inclusions in porphyroblastic garnet. The eclogite metamorphism is characterized by prograde development from epidote–blueschist to eclogite facies. Metamorphic P–T conditions estimated using pseudosection modelling are 580–600 °C and 18–20 kbar for eclogite facies. Compared with common mafic eclogites, the jadeite–garnet glaucophane schists have low CaO (4.4–4.5 wt%) and MgO (2.1–2.3 wt%) bulk‐rock compositions. The P–T– pseudosections show that low X_Ca bulk‐rock compositions favour the appearance of jadeite instead of omphacite under eclogite facies conditions. This is a unique example of low X_Ca bulk‐rock composition triggered to form jadeite at eclogite facies conditions. Two significant types of eclogitic metamorphism have been distinguished in the Sambagawa metamorphic belt, that is, a low‐T type and subsequent high‐T type eclogitic metamorphic events. The jadeite–garnet glaucophane schists experienced low‐T type eclogite facies metamorphism, and the P–T path is similar to lawsonite‐bearing eclogites recently reported from the Kotsu area in eastern Shikoku. During subduction of the oceanic plate (Izanagi plate), the hangingwall cooled gradually, and the geothermal gradient along the subduction zone progressively decreased and formed low‐T type eclogitic metamorphic rocks. A subsequent warm subduction event associated with an approaching spreading ridge caused the high‐T type eclogitic metamorphism within a single subduction zone.     

The initial garnet‐in reaction involving siderite–rhodochrosite,garnet re‐equilibration and P–T–t paths of graphitic schists in the Black Hills orogen,South Dakota,USA

It is generally thought that garnet in metapelites is produced by continuous reactions involving chlorite or chloritoid. Recent publications have suggested that the equilibrium temperatures of garnet‐in reactions may be significantly overstepped in regionally metamorphosed terranes. The growth of small spessartine–almandine garnet crystals on Mn‐siderite at the garnet isograd in graphitic metapelites in the Proterozoic Black Hills orogen, South Dakota, demonstrates that Mn‐siderite was the principal reactant that produced the initial garnet in the schists. Moreover, the positions of garnet compositions in isobaric, T–(C/H) pseudosections for the schists show that the temperature of the garnet‐in reaction from Mn‐siderite was overstepped minimally at the most. In the Black Hills, garnet was initially produced during regional metamorphism beginning at c. 1755 Ma due to the collision of Wyoming and Superior cratons, and was subsequently partially or fully re‐equilibrated at more elevated temperatures and pressures during intrusion of the Harney Peak Granite (HPG) at c. 1715 Ma. Garnet occurs in graphitic schists in garnet, staurolite and sillimanite zones, the latter being a product of contact metamorphism by HPG. During metamorphism, coexisting fluid contained both CO₂ and CH₄. In the garnet zone, garnet crystals contain petrographically distinct cores with inclusions of quartz, graphite and other minerals. Centres of the cores have distinctly elevated Y concentrations that mark the positions of garnet nucleation. The elevated Y is thought to have come from the Mn‐siderite onto which Y was probably absorbed during precipitation in an ocean. In the upper garnet and staurolite zones, the cores were overgrown by inclusion‐poor mantles. Mantles are highly zoned and have more elevated Fe and Mg and lower Mn and Ca than cores. The growth of mantles is attributed to late‐orogenic heating by leucogranite magmas and attendant influx of H₂O that caused consumption of graphite in rock matrices. A portion of the Proterozoic terrane that includes the HPG is surrounded by four large faults. In this ‘HPG block’, garnet is inclusion‐poor and its composition does not preserve its early growth history. This garnet appears to have re‐equilibrated by internal diffusion of its major components and/or recrystallization of an earlier inclusion‐rich garnet. It has equilibrated within the kyanite stability range, and together with remnant kyanite in the high‐strain aureole of the HPG, indicates that the HPG block had a ≥6 kbar history. The HPG block has undergone decompression during emplacement of the HPG. The decompression is evident in occurrences of retrograde andalusite and cordierite in the thermal aureole of the HPG. The data support a polybaric metamorphic history of the Black Hills orogen with different segments of the orogen having their own clockwise P–T–t paths.     

Integrated garnet and zircon–titanite geochronology constrains the evolution of ultra‐high–pressure terranes: An example from the Sulu orogen

Dating ultra‐high–pressure (UHP) metamorphic rocks provides important timing constraints on deep subduction zone processes. Eclogites, deeply subducted rocks now exposed at the surface, undergo a wide range of metamorphic conditions (i.e. deep subduction and exhumation) and their mineralogy can preserve a detailed record of chronologic information of these dynamic processes. Here, we present an approach that integrates multiple radiogenic isotope systems in the same sample to provide a more complete timeline for the subduction–collision–exhumation processes, based on eclogites from the Dabie–Sulu orogenic belt in eastern China, one of the largest UHP terranes on Earth. In this study, we integrate garnet Lu–Hf and Sm–Nd ages with zircon and titanite U–Pb ages for three eclogite samples from the Sulu UHP terrane. We combine this age information with Zr‐in‐rutile temperature estimates, and relate these multiple chronometers to different P–T conditions. Two types of rutile, one present as inclusions in garnet and the other in the matrix, record the temperatures of UHP conditions and a hotter stage, subsequent to the peak pressure (‘hot exhumation') respectively. Garnet Lu–Hf ages (c. 238–235 Ma) record the initial prograde growth of garnet, while coupled Sm–Nd ages (c. 219–213 Ma) reflect cooling following hot exhumation. The maximum duration of UHP conditions is constrained by the age difference of these two systems in garnet (c. 235–220 Ma). Complementary zircon and titanite U–Pb ages of c. 235–230 Ma and c. 216–206 Ma provide further constraints on the timing of prograde metamorphism and the ‘cold exhumation' respectively. We demonstrate that timing of various metamorphic stages can thus be determined by employing complementary chronometers from the same samples. These age results, combined with published data from adjacent areas, show lateral diachroneity in the Dabie–Sulu orogeny. Three sub‐blocks are thus defined by progressively younger garnet ages: western Dabie (243–238 Ma), eastern Dabie–northern Sulu (238–235 Ma) and southern Sulu terranes (225–220 Ma), which possibly correlate to different crustal slices in the recently proposed subduction channel model. These observed lateral chronologic variations in a large UHP terrane can possibly be extended to other suture zones.     

Epitaxial quartz inclusions in corundum from a sapphirine–garnet boudin,Bamble Sector,SE Norway: SiO2–Al2O3 miscibility at high P–T dry granulite facies conditions

An Al‐rich, SiO₂‐deficient sapphirine–garnet‐bearing rock occurs as a metapelitic boudin within granulite facies Proterozoic charnockitic gneisses and migmatites on the island of Hisøy, Bamble Sector, SE Norway. The boudin is made up of peraluminous sapphirine, garnet, corundum, spinel, orthopyroxene, sillimanite, cordierite, staurolite and biotite in a variety of assemblages. Thermobarometric calculations based on coexisting sapphirine–spinel, garnet–corundum–spinel–sillimanite, sapphirine–orthopyroxene, and garnet–orthopyroxene indicate peak‐metamorphic conditions near to 930 °C at 10 kbar. Corundum occurs as single 200 to 3000 micron sized skeletal crystal intergrowths in cores of optically continuous pristine garnet porphyroblasts. Quartz occurs as 5–60 micron‐sized euhedral to lobate inclusions in the corundum where it is in direct contact with the corundum with no evidence of a reaction texture. Some crystal inclusions exhibit growth zoning, which indicates that textural equilibrium was achieved. Electron Back‐Scatter Diffraction (EBSD) studies reveal that the quartz inclusions share a common c‐axis with the host corundum crystal. The origin of the quartz inclusions in corundum is enigmatic as recent experimental studies have confirmed the instability of quartz–corundum over geologically realistic P–T ranges. The combined EBSD and textural observations suggest the presence of a former silica‐bearing proto‐corundum, which underwent exsolution during post‐peak‐metamorphic uplift and cooling. Exsolution of quartz in corundum is probably confined to fluid‐absent conditions where phase transitions by coupled dissolution–precipitation mechanisms are prevented.     

Linking thermodynamic modelling,Lu–Hf geochronology and trace elements in garnet: new P–T–t paths from the Sevier hinterland

Major element, trace element and Lu–Hf geochronological data from amphibolite facies pelitic schist in the Raft River and Albion Mountains of northwest Utah and southern Idaho indicate that garnet grew during increasing pressure, interpreted to be the result of tectonic burial and crustal thickening during Sevier orogenesis. Garnet growth was interrupted by hiatuses interpreted from discontinuities in major element zonation. Pressure–temperature paths were determined from the pre‐hiatus portions of the garnet chemical zoning profiles and indicate an increase of ~2 kbar and ~50 °C in the western Raft River Mountains. Garnet Lu–Hf dates of 150 ± 1 Ma in the western Raft River Mountains and 138.7 ± 0.7 Ma and 132 ± 5 Ma in the southern Albion Mountains indicate the timing of garnet growth. Lutetium garnet zoning profiles indicate that the Lu–Hf ages are biased towards the post‐hiatus or outer pre‐hiatus segments, indicating that the determined ages likely post‐date the recorded P–T path history or date the tail end of the paths. Crustal thickening associated with Sevier orogenesis in the western Raft River Mountains thus began slightly before 150 ± 1 Ma, in the Late Jurassic. This study shows that integrating P–T paths determined from garnet growth zoning with Lu–Hf garnet geochronology and in situ garnet trace element analyses is an effective approach for interpreting and dating deformation events in orogenic belts.     

P–T–t evolution of garnet amphibolites in the Wutai–Hengshan area,North China Craton: insights from phase equilibria and geochronology

Garnet amphibolites can provide valuable insights into geological processes of orogenic belts, but their metamorphic evolution is still poorly constrained. Garnet amphibolites from the Wutai–Hengshan area of the North China Craton mainly consist of garnet, hornblende, plagioclase, quartz, rutile and ilmenite, with or without titanite and epidote. Four samples selected in a south–north profile were studied by the pseudosection approach in order to elucidate the characteristics of their metamorphic evolution, and to better reveal the northwards prograde change in P–T conditions as established previously. For the sample from the lower Wutai Subgroup, garnet exhibits obvious two‐substage growth zoning characteristic of pyrope (X_py) increasing but grossular (X_gr) decreasing outwards in the core, and both X_py and X_gr increasing outwards in the rim. Phase modelling using thermocalc suggests that the garnet cores were formed by chlorite breakdown over 7–9 kbar at 530–600 °C, and rims grew from hornblende and epidote breakdown over 9.5–11.5 kbar at 600–670 °C. The isopleths of the minimum An in plagioclase and maximum X_py in garnet were used to constrain the peak P–T conditions of ~11.5 kbar/670 °C. The modelled peak assemblage garnet + hornblende + epidote+ plagioclase + rutile + quartz matches well the observed one. Plagioclase–hornblende coronae around garnet indicate post‐peak decompression and fluid ingress. For the samples from the south Hengshan Complex, the garnet zoning weaken gradually, reflecting modifications during decompression of the rocks. Using the same approach, the rocks are inferred to have suprasolidus peak conditions, increasing northwards from 11.5 kbar/745 °C, 12.5 kbar/780 °C to 13 kbar/800 °C. Their modelled peak assemblages involve diopside, garnet, hornblende, plagioclase, rutile and quartz, yet diopside is not observed petrographically. The post‐peak decompression is characterized by diopside + garnet + quartz + melt = hornblende + plagioclase, causing the diopside consumption and garnet compositions to be largely modified. Thus, the pesudosection approach is expected to provide better pressure results than conventional thermobarometry, because the later approach cannot be applied with confidence to rocks with multi‐generation assemblages. U–Pb dating of zircon in the Wutai sample records a protolith age of c. 2.50 Ga, and a metamorphic age of c. 1.95 Ga, while zircon in the Hengshan samples records metamorphic ages of c. 1.92 Ga. The c. 1.95 Ga is interpreted to represent the pre‐peak or peak metamorphic stages, and the ages of c. 1.92 Ga are assigned to represent the cooling stages. All rocks in the Wutai–Hengshan area share similar clockwise P–T morphologies. They may represent metamorphic products at different crustal depths in one orogenic event, which included a main thickening stage at c. 1.95 Ga followed by a prolonged uplift and cooling after 1.92 Ga.     

Petrogenetic significance of orthopyroxene-free garnet + clinopyroxene + plagioclase ± quartz-bearing metabasites with respect to the amphibolite and granulite facies

Orthopyroxene‐free garnet + clinopyroxene + plagioclase ± quartz‐bearing mineral assemblages represent the paragenetic link between plagioclase‐free eclogite facies metabasites and orthopyroxene‐bearing granulite facies metabasites. Although these assemblages are most commonly developed under P–T conditions consistent with high pressure granulite facies, they sometimes occur at lower grade in the amphibolite facies. Thus, these assemblages are characteristic but not definitive of high pressure granulite facies. Compositional factors favouring their development at amphibolite grade include Fe‐rich mineral compositions, Ca‐rich garnet and plagioclase, and Ti‐poor hornblende. The generalized reaction that accounts for the prograde development of garnet + clinopyroxene + plagioclase ± quartz from a hornblende + plagioclase + quartz‐bearing (amphibolite) precursor is Hbl + Pl + Qtz=Grt + Cpx + liquid or vapour, depending on whether the reaction occurs above or below the solidus. There are significant discrepancies between experimental and natural constraints on the P–T conditions of orthopyroxene‐free garnet + clinopyroxene + plagioclase ± quartz‐bearing mineral assemblages and therefore on the P–T position of this reaction. Semi‐quantitative thermodynamic modelling of this reaction is hampered by the lack of a melt model and gives results that are only moderately successful in rationalizing the natural and experimental data.     

Coupled garnet Lu–Hf and monazite U–Pb geochronology constrain early convergent margin dynamics in the Ross orogen,Antarctica

The Ross orogen of Antarctica is an extensive (>3000 km‐long) belt of deformed and metamorphosed sedimentary rocks and granitoid batholiths, which formed during convergence and subduction of palaeo‐Pacific lithosphere beneath East Gondwana in the Neoproterozoic–early Palaeozoic. Despite its prominent role in Gondwanan convergent tectonics, and a well‐established magmatic record, relatively little is known about the metamorphic rocks in the Ross orogen. A combination of garnet Lu–Hf and monazite U–Pb (measured by laser‐ablation split‐stream ICP‐MS) geochronology reveals a protracted metamorphic history of metapelites and garnet amphibolites from a major segment of the orogen. Additionally, direct dating of a common rock‐forming mineral (garnet) and accessory mineral (monazite) allows us to test assumptions that are commonly used when linking accessory mineral geochronology to rock‐forming mineral reactions. Petrography, mineral zoning, thermobarometry and pseudosection modelling reveal a Barrovian‐style prograde path, reaching temperatures of ~610–680 °C. Despite near‐complete diffusional resetting of garnet major element zoning, the garnet retains strong rare earth element zoning and preserves Lu–Hf dates that range from c. 616–572 Ma. Conversely, monazite in the rocks was extensively recrystallized, with concordant dates that span from c. 610–500 Ma, and retain only vestigial cores. Monazite cores yield dates that overlap with the garnet Lu–Hf dates and typically have low‐Y and heavy rare earth element (HREE) concentrations, corroborating interpretations of low‐Y and low‐HREE monazite domains as records of synchronous garnet growth. However, ratios of REE concentrations in garnet and monazite do not consistently match previously reported partition coefficients for the REE between these two minerals. High‐Y monazite inclusions within pristine, crack‐free garnet yield U–Pb dates significantly younger than the Lu–Hf dates for the same samples, indicating recrystallization of monazite within garnet. The recrystallization of high‐Y and high‐HREE monazite domains over >50 Ma likely records either punctuated thermal pulses or prolonged residence at relatively high temperatures (up to ~610–680 °C) driving monazite recrystallization. One c. 616 Ma garnet Lu–Hf date and several c. 610–600 Ma monazite U–Pb dates are tentatively interpreted as records of the onset of tectonism metamorphism in the Ross orogeny, with a more robust constraint from the other Lu–Hf dates (c. 588–572 Ma) and numerous c. 590–570 Ma monazite U–Pb dates. The data are consistent with a tectonic model that involves shortening and thickening prior to widespread magmatism in the vicinity of the study area. The early tectonic history of the Ross orogen, recorded in metamorphic rocks, was broadly synchronous with Gondwana‐wide collisional Pan‐African orogenies.     

Eclogite and garnet pyroxenite from Stor Jougdan,Seve Nappe Complex,Sweden: implications for UHP metamorphism of allochthons in the Scandinavian Caledonides

Ultrahigh‐pressure metamorphism (UHPM) has recently been discovered in far‐travelled allochthons of the Scandinavian Caledonides, including finding of diamond in the Seve Nappe Complex. This UHPM of Late Ordovician age is older and less recognized than that in the Western Gneiss Region of southwestern Norway, which was related to terminal collision between Baltica and Laurentia. Here we report new evidence of UHPM in the Lower Seve Nappe, recorded by eclogite and garnet pyroxenite from the area of Stor Jougdan in northern Jämtland, central Sweden. Peak‐metamorphic assemblage of eclogite, garnet + omphacite + phengite + rutile + coesite? yields P–T conditions of 2.8–4.0 GPa and 750–900 °C, constrained by conventional geothermobarometry and thermodynamic modelling in the NCKFMTASH system. The prograde metamorphic evolution of the eclogite is inferred from inclusions of zoisite and amphibole in garnet, which are stable at lower pressure, whereas the retrograde evolution is recorded by formation of diopsidic clinopyroxene + plagioclase symplectites after omphacite, growth of amphibole replacing these symplectites, and of titanite around rutile. In garnet pyroxenite the peak‐metamorphic assemblage consists of garnet + orthopyroxene + clinopyroxene + olivine. P–T conditions of 2.3–3.8 GPa and 810–960 °C have been derived based on the conventional geothermobarometry and thermodynamic modelling in the CFMASH and CFMAS systems. Retrograde evolution has been recognized from replacement of pyroxene and garnet by amphibole. The results show that eclogite was metamorphosed during deep subduction of continental crust, most probably derived from the continental margin of Baltica, whereas the origin and tectonic setting of the garnet pyroxenite is ambiguous. The studied pyroxenite/peridotite of Baltican subcontinental affinity could have been metamorphosed as a part of the subducting plate and exhumed due to the downward extraction of a forearc lithospheric block.     

Record of plate boundary metamorphism during Gondwana breakup from Lu–Hf garnet geochronology of the Alpine Schist,New Zealand

The Zealandia portion of the Pacific–Gondwana margin underwent widespread extension, fragmentation, separation and subsidence during the final stages in the breakup of Gondwana. Although these processes shaped the geology of New Zealand, their timing and the timing of subduction cessation in the region remain unclear. To investigate the timing of these processes, we used Lu–Hf garnet geochronology to date six samples of the Alpine Schist, which represents the metamorphic section of the former Zealandia margin. The garnet dates range from 97.3 ± 0.3 to 75.4 ± 1.3 Ma. Compositional zoning in garnet indicates that the spread in ages results from diachronous metamorphism in the upper plate at the Pacific–Gondwana margin, occurring concurrently with rifting of Zealandia from East Gondwana via opening of the Tasman Sea. Clear spatial trends in the timing of garnet growth throughout the Alpine Schist are absent, indicating that either regional age trends were offset by post‐metamorphic deformation, or that metamorphism did not result from a single regional heat source, and was instead driven by short‐duration, spatially dispersed processes such as episodic fluid‐fluxing or mechanical heating. Diachronous metamorphism of the Alpine Schist can be attributed to heat conduction from the rising upper mantle during widespread extension, progressive burial and heating of accretionary wedge sediments during ongoing horizontal shortening, or fluid‐fluxing sourced from a subducting and dehydrating Hikurangi Plateau. These results indicate that during separation of Zealandia from East Gondwana in the late Cretaceous, the crust at the Pacific–Gondwana margin remained hot, potentially facilitating the extensive thinning of the Zealandia lithosphere during this time.     

Time‐scale of deformation and intertectonic phases revealed by P–T–D–t relationships in the orogenic middle crust of the Orlica‐Śnieżnik Dome,Polish/Czech Central Sudetes

A section of the orogenic middle crust (Orlica‐?nie?nik Dome, Polish/Czech Central Sudetes) was examined to constrain the duration and significance of deformation (D) and intertectonic (I) phases. In the studied metasedimentary synform, three deformation events produced an initial subhorizontal foliation S1 (D₁), a subsequent subvertical foliation S2 (D₂) and a late subhorizontal axial planar cleavage S3 (D₃). The synform was intruded by pre‐, syn‐ and post‐D₂ granitoid sheets. Crystallization–deformation relationships in mica schist samples document I_1–2 garnet–staurolite growth, syn‐D₂ staurolite breakdown to garnet–biotite–sillimanite/andalusite, I_2–3 cordierite blastesis and late‐D₃ chlorite growth. Garnet porphyroblasts show a linear Mn–Ca decrease from the core to the inner rim, a zone of alternating Ca–Y‐ and P‐rich annuli in the inner rim, and a Ca‐poor outer rim. The Ca–Y‐rich annuli probably reflect the occurrence of the allanite‐to‐monazite transition at conditions of the staurolite isograd, whereas the Ca‐poor outer rim is ascribed to staurolite demise. The reconstructed P–T path, obtained by modelling the stability of parageneses and garnet zoning, documents near‐isobaric heating from ~4 kbar/485 °C to ~4.75 kbar/575 °C during I_1–2. This was followed by a progression to 4–5 kbar/580–625 °C and a subsequent pressure decrease to 3–4 kbar during D₂. Pressure decrease below 3 kbar is ascribed to I_2–3, whereas cooling below ~500 °C occurred during D₃. In the dated mica schist sample, garnet rims show strong Lu enrichment, oscillatory Lu zoning and a slight Ca increase. These features are also related to allanite breakdown coeval with staurolite appearance. As Lu‐rich garnet rims dominate the Lu–Hf budget, the 344 ± 3 Ma isochron age is ascribed to garnet crystallization at staurolite grade, near the end of I_1–2. For the dated sample of amphibole–biotite granitoid sheet, a Pb–Pb single zircon evaporation age of 353 ± 1 Ma is related to the onset of plutonic activity. The results suggest a possible Devonian age for D₁, and a Carboniferous burial‐exhumation cycle in mid‐crustal rocks that is broadly coeval with the exhumation of neighbouring HP rocks during D₂. In the light of published ages, a succession of telescoping stages with time spans decreasing from c. 10 to 2–3 Ma is proposed. The initially long period of tectonic quiescence (I_1–2 phase, c. 10 Ma) inferred in the middle crust contrasts with contemporaneous deformation at deeper levels and points to decoupled P–T–D histories within the orogenic wedge. An elevated gradient of ~30 °C km^?1 and assumed high heating rates of c. 20 °C Ma^?1 are explained by the protracted intrusion of granitoid sheets, with or without deformation, whereas fast vertical movements (2–3 Ma, D₂ phase) in the crust require the activity of deformation phases.     

The nature and distribution of fluids during amphibolite facies metamorphism, Naxos (Greece)

On the basis of fluid inclusion evidence, pervasive influx of deep-seated CO₂-rich fluids has been invoked to account for mid- to upper amphibolite facies (M2_B) metamorphism on the island of Naxos (Cyclades, Greece). In this paper, mineral devolatilization and melt equilibria are used to constrain the composition of both syn- and post-peak-M2_B fluids in the deepest exposed levels of the metamorphic complex. The results indicate that peak-M2_B fluids were spatially and compositionally heterogeneous throughout the high-grade core of the complex, whereas post-peak-M2_B fluids were generally water-rich. The observed heterogeneities in syn-M2_B fluid composition are inconsistent with pervasive CO₂-flushing models invoked by previous workers on the basis of fluid inclusion evidence. It is likely that few CO₂-rich fluid inclusions on Naxos preserve fluids trapped under peak metamorphic conditions. It is suggested that many of these inclusions have behaved as chemically open systems during the intense deformation that accompanied the uplift of the metamorphic complex. A similar process may explain the occurrence of some CO₂-rich fluid inclusions in granulite facies rocks.     

Protracted garnet growth in high‐P eclogite: constraints from multiple geochronology and P–T pseudosection

Understanding convergent margin processes requires determination of the onset and the termination of subduction, the duration of subduction‐zone metamorphism, and the subduction zone polarity. Garnet growth and intracrystalline zonation can be used to constrain the timing, duration and kinetics of tectonometamorphic processes. An eclogite from the Huwan shear zone in the Hong'an orogen was investigated with combined pseudosection analysis and multiple geochronologies. The pseudosection analysis illustrates that garnet growth is continuous and along an early near‐isothermal trajectory followed by a near‐isobaric heating path from 1.9 GPa/500 °C to 2.4 GPa/575 °C and subsequent near‐isothermal decompression. ⁴⁰Ar/³⁹Ar dating of an amphibole inclusion in garnet from the eclogite yielded an age of 310 ± 5 Ma, which is consistent with a U–Pb age of 305 ± 3 Ma for the metamorphic zircon within uncertainty. Garnet core and rim material produced Lu–Hf ages of 296.9 ± 3.8 and 256.9 ± 3.9 Ma respectively; the latter is consistent with its Sm–Nd age of 254.3 ± 4.6 Ma for the same aliquots. Similarly, limited zircon U–Pb ages of c. 257 Ma were obtained in zircon rims with garnet inclusions. These ages were interpreted to bracket the period of garnet growth and the difference of up to c. 40 Ma is best explained by protracted garnet growth. We propose that the rocks represent detachment of part of the downgoing slab and remained free of significant compression/decompression or heating/cooling close to the subduction channel, most likely underplating the mantle wedge, for a long time. These rocks were incorporated into the following subduction channel due to the successive entry of the buoyant materials, and exhumed at some time later than c. 254 Ma. The increasing observations of protracted garnet growth and long‐lived subduction in various orogens worldwide demand more sophisticated geodynamic models.