Showing papers in "Contributions to Mineralogy and Petrology in 1998"

Zircon megacrysts from kimberlite: oxygen isotope variability among mantle melts

Abstract: The oxygen isotope ratios of Phanerozoic zircons from kimberlite pipes in the Kaapvaal Craton of southern Africa and the Siberian Platform vary from 4.7 to 5.9& VSMOW. High precision, accurate analyses by laser reveal subtle pipe-to-pipe diAerences not previously suspected. These zircons have distinctive chemical and physical characteristics identifying them as mantle-de- rived megacrysts similar to zircons found associated with diamond, coesite, MARID xenoliths, Cr-diopside, K-richterite, or Mg-rich ilmenite. Several lines of evi- dence indicate that these d 18 O values are unaltered by kimberlite magmas during eruption and represent com- positions preserved since crystallization in the mantle, including: U/Pb age, large crystal size, and the slow rate of oxygen exchange in non-metamict zircon. The aver- age d 18 O of mantle zircons is 5.3&,0:1& higher and in equilibrium with values for olivine in peridotite xenoliths and oceanic basalts. Zircon megacrysts from within 250 km of Kimberley, South Africa have average d 18 Oa 5:32 0:17&Ona 28U. Small, but significant, diAerences among other kimberlite pipes or groups of pipes may indicate isotopically distinct reservoirs in the sub-continental lithosphere or asthenosphere, some of which are anomalous with respect to normal mantle values of 5:3 0:3&. Precambrian zircons (2.1-2.7 Ga) from Jwaneng, Botswana have the lowest values yet measured in a mantle zircon, d 18 Oa 3:4 to 4.7&. These zircon megacrysts originally crystallized in mafic or ul- tramafic rocks either through melting and metasomat- ism associated with kimberlite magmatism or during metamorphism. The low d 18 O zircons are best explained by subduction of late Archean ocean crust that ex- changed with heated seawater prior to underplating as eclogite and to associated metasomatism of the mantle wedge. Smaller diAerences among other pipes and dis- tricts may result from variable temperatures of equili- bration, mafic versus ultramafic hosts, or variable un- derplating. The narrow range in zircon compositions found in most pipes suggests magmatic homogenization. If this is correct, these zircons document the existence of sig- nificant quantities of magma in the sub-continental mantle that was regionally variable in d 18 O and this information restricts theories about the nature of ancient subduction.

The influence of water on melting of mantle peridotite

Abstract: This experimental study examines the effects of variable concentrations of dissolved H2O on the compositions of silicate melts and their coexisting mineral assemblage of olivine + orthopyroxene ± clinopyroxene ± spinel ± garnet. Experiments were performed at pressures of 1.2 to 2.0 GPa and temperatures of 1100 to 1345 °C, with up to ∼12 wt% H2O dissolved in the liquid. The effects of increasing the concentration of dissolved H2O on the major element compositions of melts in equilibrium with a spinel lherzolite mineral assemblage are to decrease the concentrations of SiO2, FeO, MgO, and CaO. The concentration of Al2O3 is unaffected. The lower SiO2 contents of the hydrous melts result from an increase in the activity coefficient for SiO2 with increasing dissolved H2O. The lower concentrations of FeO and MgO result from the lower temperatures at which H2O-bearing melts coexist with mantle minerals as compared to anhydrous melts. These compositional changes produce an elevated SiO2/(MgO + FeO) ratio in hydrous peridotite partial melts, making them relatively SiO2 rich when compared to anhydrous melts on a volatile-free basis. Hydrous peridotite melting reactions are affected primarily by the lowered mantle solidus. Temperature-induced compositional variations in coexisting pyroxenes lower the proportion of clinopyroxene entering the melt relative to orthopyroxene. Isobaric batch melting calculations indicate that fluid-undersaturated peridotite melting is characterized by significantly lower melt productivity than anhydrous peridotite melting, and that the peridotite melting process in subduction zones is strongly influenced by the composition of the H2O-rich component introduced into the mantle wedge from the subducted slab.

Experimental determination of partition coefficients for rare earth and high-field-strength elements between clinopyroxene, garnet, and basaltic melt at high pressures

Abstract: Clinopyroxene/melt and garnet/melt partition coefficients have been determined for Ti, Sr, Y, Zr, Nb, Hf, and rare earth elements from 19 doped experiments on 1921 Kilauea basalt The experiments were carried out from 20 to 30 GPa and 1310° to 1470 °C The purpose was to derive a set of partition coefficients for high-field-strength elements (HFSE) and rare earth elements (REE) in a systematic, linked set of experiments at P and T conditions relevant to basalt petrogenesis These data are used in melting models to understand the development of negative HFSE anomalies observed in many abyssal peridotite clinopyroxenes It is shown that melting can account for the observed trace element patterns in some residual peridotites, but that other processes may also be needed to account for most residual mantle compositions in mid-ocean ridge systems It is also shown that REE are more strongly fractionated by garnet at these P-T conditions than previously thought

The hydrous phase equilibria (to 3 kbar) of an andesite and basaltic andesite from western Mexico: constraints on water content and conditions of phenocryst growth

Abstract: We have conducted high pressure (to 3 kbar), water saturated melting experiments on an andesite (62 wt% SiO2) and a basaltic andesite (55 wt% SiO2) from western Mexico. A close comparison between the experimental phase assemblages and their compositions, and the phenocryst assemblages of the lavas, is found in water saturated liquids, suggesting that the CO2 content was minimal in the fluid phase. Thus the historic lavas from Volcan Colima (with phenocrysts of orthopyroxene, augite, plagioclase, and hornblende) were stored at a temperature between 950–975 °C, at a pressure between 700–1500 bars, and with a water content of 3.0–5.0 wt%. A hornblende andesite (spessartite) from Mascota, of nearly identical composition but with only amphibole phenocrysts, had a similar temperature but equilibrated at a minimum of 2000 bars pressure with a dissolved water content of at least 5.5 wt% in the liquid. Experiments on the basaltic andesite show that the most common natural phenocryst assemblages (olivine, ±augite, ±plagioclase) could have precipitated at temperatures from 1000–1150 °C, in liquids with a wide range of dissolved water content (∼2.0–6.0 wt%) and a corresponding pressure range. A lava of the same bulk composition with phenocrysts of hornblende, olivine, plagioclase, and augite is restricted to temperatures below 1000 °C and pressures below 2500 bars, corresponding to <5.5 wt% water in the residual liquid. Although there is some evidence for mixing in the andesites (sporadic olivine phenocrysts), the broad theme of the history of both lava types is that the phenocryst assemblages for both the andesitic magmas and basaltic andesitic magmas are generated from degassing and reequilibration on ascent of initially hydrous parents containing greater than 6 wt% water. Indeed andesitic magmas could be related to a basaltic andesite parent by hornblende-plagioclase fractionation under the same hydrous conditions.

Jurassic formation and Eocene subduction of the Zermatt-Saas-Fee ophiolites: implications for the geodynamic evolution of the Central and Western Alps

Abstract: The Zermatt–Saas-Fee ophiolites (ZSFO) are one of the best preserved slices of eclogitic oceanic crust in the Alpine chain. They formed during the opening of the Mesozoic Tethys and underwent subduction to HP/UHP conditions during Alpine compression. A cathodoluminescence-based ion microprobe (SHRIMP) dating of different zircon domains from metagabbros and oceanic metasediments was carried out to constrain the timing of formation and subduction of this ophiolite, two fundamental questions in Alpine geodynamics. The formation of the ophiolitic sequence is constrained by the intrusion ages of the Mellichen and the Allalin metagabbros (164.0 ± 2.7 Ma and 163.5 ± 1.8 Ma) obtained on magmatic zircon domains. These data are in line with the maximum deposition age for Mn-rich metasediments which overlie the mafic rocks at Lago di Cignana (161 ± 11 Ma) and at Sparrenflue (ca. 153–154 Ma). An Eocene age of 44.1 ± 0.7 Ma was obtained for whole zircons and zircon rims from an UHP eclogite and two metasediments at Lago di Cignana. One of the Eocene zircons contains a rutile inclusion indicating formation at HP conditions. As the temperature and pressure peak of these rocks nearly coincide, the Eocene zircons probably constrain the age for the deepest subduction of the ZSFO. This Eocene age for the UHP metamorphism implies that the ZSFO were subducted later than the Adriatic margin (Sesia-Lanzo Zone) and before the Late Eocene subduction of the European continental crust below Apulia. A scenario with three subduction episodes propagating in time from SE to NW is proposed for the geological evolution of the Central and Western Alps.

Trace element modeling of aqueous fluid - peridotite interaction in the mantle wedge of subduction zones

Abstract: Recently measured partition coefficients for Rb, Th, U, Nb, La (Ce), Pb, Sr, Sm, Zr, and Y between lherzolite assemblage minerals and H2O-rich fluid (Ayers et al 1997; Brenan et al 1995a,b) are used in a two-component local equilibrium model to assess the effects of interaction between slab-derived aqueous fluids and wedge lherzolite on the trace element and isotopic composition of island arc basalts (IAB) The model includes four steps representing chemical processes, with each process represented by one equation with one adjustable parameter, in which aqueous fluid: (1) separates from eclogite in the subducted slab (Rayleigh distillation, mass fraction of fluid released F fluid); (2) ascends through the mantle wedge in isolated packets, exchanging elements and isotopes with depleted lherzolite (zone refining, the rock/fluid mass ratio n); (3) mixes with depleted lherzolite (physical mixing, the mass fraction of fluid in the mixture X fluid); (4) induces melting to form primitive IAB (batch melting, mass fraction of melt F melt) The amount of mantle lherzolite processed by the fluid in step (2) determines its isotopic and trace element signature and the relative contributions of slab and wedge to primitive IAB Assuming an average depleted lherzolite composition and mineralogy (70% olivine, 26% orthopyroxene, 3% clinopyroxene and 1% ilmenite) and using nonlinear regression to adjust parameter values to obtain an optimal fit to the average composition of IAB (McCulloch and Gamble 1991) yields values of F fluid= 020, n= 26, X fluid= 017, and F melt= 015, with r 2= 0995 and the average relative error in trace element concentration = 6% The average composition of IAB can also effectively be modeled with no contribution from the slab other than H2O (ie, skip model step 1): n= 27, X fluid= 021, F melt= 017, with r 2= 0992 By the time the fluid reaches the IAB source, exchange with depleted wedge lherzolite reduces the 87Sr/86Sr ratio isotopic composition to near-mantle values and the slab contribution to <50% for all but the most incompatible elements (eg, Pb) The IAB may retain the slab signature for elements such as B and Be that are highly incompatible and that have very low concentrations in the depleted mantle wedge The relatively high equilibrium D mineral / fluid values measured by Ayers et al (1997), Brenan et al (1995a) and Stalder et al (1998) suggest that large amounts of fluid (>5 wt%) must be added to lherzolite in the IAB source Decreasing X fluid below 005 causes model results to have unacceptably high levels of error and petrologically unreasonable values of F melt That H2O contents of IAB are generally <6 wt% suggests that not all of the H2O that metasomatizes the IAB source remains in the source to dissolve in the subsequently formed melt Modeling of the compositions of specific primitive IAB from oceanic settings with low sediment input and depleted mantle wedges (Tonga, Marianas) shows a generally lower level of fluid-wedge interaction (low n), and therefore a larger slab component in primitive IAB

Clinopyroxene geobarometry of magmatic rocks Part 1: An expanded structural geobarometer for anhydrous and hydrous, basic and ultrabasic systems

Abstract: Crystal-structure modeling of experimental Ca-rich clinopyroxenes [Ca + Na > 0.5 apfu; Mg/(Mg + Fe2+) > 0.7] coexisting with basic and ultrabasic melts was utilized for calibration of geobarometers based on unit-cell volume (Vcell) vs M1-site volume (VM1). The clinopyroxene database includes over one hundred experiments from literature and sixteen previously unpublished experiments on basanite and picrobasalt starting materials. The coexisting melts span a wide range of petrologically relevant anhydrous and hydrous compositions (from quartz-normative basalt to nephelinite, excluding high-Al basalts and melts coexisting with garnet or melilite) at pressure conditions pertinent to the earth's crust and uppermost mantle (P= 0–24 kbar) in a variety of fO 2 conditions (from CCO-buffered to air-buffered) and mineral assemblages (Cpx ± Opx ± Pig ± Ol ± Plag ± Lc ± Ne ± Spl ± Amp ± Ilm). As previously found for near-liquidus products of basaltic melts, the experimental clinopyroxenes follow two distinct trends: (i) at a given P, Vcell is linearly and negatively correlated with VM1. This corresponds with the extent of Tschermak-type substitutions, which depends strongly on aSiO2 and a CaO; (ii) for a fixed melt composition, Vcell and VM1 decrease linearly as P increases, due to a combination of M1, M2 and T site exchanges. Despite the chemical complexity of these relationships, P could be modeled as a linear function of Vcell and VM1. A simplified solution for anhydrous magmas reproduced the experimental pressures with an uncertainty of 1.75 kbar (=1; max. dev. = 5.5 kbar; N = 135). An expanded T-dependent solution capable of recovering the measured pressures of both anhydrous and hydrous experiments with an uncertainty of 1.70 kbar (=1; max. dev. = 5.4 kbar; N = 157) was obtained by correcting unit-cell and M1-site volumes for thermal expansivity and compressibility. The corrected formulation is more resistant to the effects of temperature variations and is therefore recommended. Nevertheless, it requires an independent, accurate estimate of crystallization T. Underestimating T by 20 °C propagates into a 1-kbar increase of calculated P. The applicability of the T-dependent formulation was tested on hydrous ultramafic to gabbroic rocks of the southern Adamello batholith for which P-T evolution could independently be constrained by field observation, petrography and experimentally determined phase relations. The pressure estimates obtained by clinopyroxene structural geobarometry closely matched those predicted by phase equilibria of a picrobasaltic melt parental to the investigated magmatic rocks. To facilitate application of the present geobarometers, both anhydrous and corrected solutions were implemented as MS-DOS® and UNIX® software programs (CpxBar) designed to permit retrieval of the pressure of crystallization directly from a chemical analysis or from uncorrected unit-cell and M1-site volume X-ray data.

Internal structures of zircons from Archaean granites from the Darling Range batholith: implications for zircon stability and the interpretation of zircon U-Pb ages

Abstract: Internal structures in zircons from granitoids from the late Archaean Darling Range Batholith show secondary features revealed by HF etching, which record reconstitution of the zircons and modification of the distribution of trace elements during post crystallisation cooling of the granitoid. Zircons from the granites commonly contain unzoned to weakly zoned cores surrounded by rims showing oscillatory zoning which has been modified by recrystallisation. The most striking feature is the development of high trace element concentration areas found in zircons from a number of granites. These structures range from enhanced trace element concentrations in primary zones to a single accumulation of most trace elements in one band, about half way between the outer edge and the centre of the zircon. In any zircon the extent of the concentration of trace elements towards the formation of a single trace element band appears to be inversely related to the fading and broadening of primary oscillatory zones in the outer rim. This suggests that the trace element bands formed by migration of trace elements from the outer primary zones to new concentration sites on an inner set of primary zones. This explanation is supported by the formation of multiple curved trace element bands that transgress primary zoning and the determination of younger SHRIMP ages on depleted zircon outer rims compared to remnant primary oscillatory zoned areas of the zircon and unzoned centres. Also observed in some granite zircons is a finely convoluted zoning which overprints oscillatory zoning in parts of a zoned zircon and in rare cases occurs throughout the zircon. This structure is explained in terms of secondary migration and reconcentration of trace elements in curved bands. All structures can be transgressed by generally rounded lobes and patches of low U, weakly nebulously zoned zircon. This is interpreted as a late stage interaction between the zircon and fluids formed during cooling and crystallisation of the granitoid, resulting in recrystallisation of affected parts of the zircon with accompanying loss of trace elements from the zircon.

Carbonatitic melts along the solidus of model lherzolite in the system CaO-MgO-Al2O3-SiO2-CO2 from 3 to 7 GPa

Abstract: We have experimentally determined the solidus position of model lherzolite in the system CaO-MgO-Al2O3-SiO2-CO2 (CMAS.CO2) from 3 to 7 GPa by locating isobaric invariant points where liquid coexists with olivine, orthopyroxene, clinopyroxene, garnet and carbonate. The intersection of two subsolidus reactions at the solidus involving carbonate generates two invariant points, I1A and I2A, which mark the transition from CO2-bearing to dolomite-bearing and dolomite-bearing to magnesite-bearing lherzolite respectively. In CMAS.CO2, we find I1A at 2.6 GPa/1230 °C and I2A at 4.8 GPa/1320 °C. The variation of all phase compositions along the solidus has also been determined. In the pressure range investigated, solidus melts are carbonatitic with SiO2 contents of <6 wt%, CO2 contents of ˜45 wt%, and Ca/(Ca+Mg) ratios that range from 0.59 (3 GPa) to 0.45 (7 GPa); compositionally they resemble natural magnesiocarbonatites. Volcanic magnesiocarbonatites may well be an example of the eruption of such melts directly from their mantle source region as evidenced by their diatremic style of activity and lack of associated silicate magmas. Our data in the CMAS.CO2 system show that in a carbonate-bearing mantle, solidus and near-solidus melts will be CO2-rich and silica poor. The widespread evidence for the presence of CO2 in both the oceanic and continental upper mantle implies that such low degree SiO2-poor carbonatitic melts are common in the mantle, despite the rarity of carbonatites themselves at the Earth's surface.

Melting and metasomatism in the continental lithosphere: laser ablation ICPMS analysis of minerals in spinel lherzolites from eastern Australia

Abstract: Olivine, low-Ca pyroxene, diopside, and spinel from a suite of protogranular lherzolite xenoliths from southeastern Australia have been analysed for their major and trace element compositions using electron microprobe and laser ablation ICPMS. Bulk compositions of the lherzolites range from fertile (12–13% modal diopside) to depleted (2–3% modal diopside), with equilibration temperatures of 850–900 °C indicating entrainment of these lherzolites from relatively shallow depths (probably ≤ 35 km) within the lithosphere. Mineral compositions and abundances indicate a primary control by partial melting, with decreasing abundance of modal diopside accompanied by increasing Mg# of olivine and pyroxene, decreasing Al and Ti contents of diopside, increasing Ni contents of olivine, and increasing Cr/Al of spinel. HREE, Y, and Ga in diopside also follow melting trends, decreasing in concentration with increasing Mg#. In contrast, highly incompatible elements such as LREE, Nb, and Th reveal divergent behaviour that cannot be ascribed entirely to partial melting. Diopsides from the fertile lherzolites have mantle-normalized patterns that are depleted in Th, Nb, and the LREE relative to Y and the HREE, whereas, diopsides from the cpx-poor samples are strongly enriched in Th, Nb and the LREE, and have elevated Sm/Hf and Zr/Hf, and low Ti/Nb. All diopsides have strongly negative Nb anomalies relative to Th and the LREE. Trace element patterns of diopside in the fertile lherzolites can be reproduced by ≤ 5% batch melting of a primitive source. The negative Nb anomalies are a consequence of this melting, and do not require special conditions or tectonic environments. The low concentrations of Y and HREE in diopside from the cpx-poor lherzolites cannot be produced by realistic degrees of batch melting, but can be accomplished by up to ∼20% fractional melting, suggesting multiple episodes of melt depletion. Os isotopic compositions of these lherzolites show that the melt depletion events occurred in the middle and late Proterozoic, demonstrating the long-term stability of lithospheric mantle beneath regions of eastern Australia. The LREE-enriched diopsides are well equilibrated and record metasomatic enrichment events that pre-date the magmatism that entrained these xenoliths. Trace element patterns of these pyroxenes suggest a carbonatitic melt as the metasomatic agent.

U-Pb monazite ages in amphibolite- to granulite-facies orthogneiss reflect hydrous mineral breakdown reactions: Sveconorwegian Province of SW Norway

Abstract: In the Rogaland–Vest Agder terrain of the Sveconorwegian Province of SW Norway, two main Sveconorwegian metamorphic phases are reported: a phase of regional metamorphism linked to orogenic thickening (M1) and a phase of low-pressure thermal metamorphism associated with the intrusion of the 931 ± 2 Ma anorthosite-charnockite Rogaland igneous complex (M2). Phase M1 reached granulite facies to the west of the terrane and M2 culminated locally at 800–850 °C with the formation of dry osumilite-bearing mineral associations. Monazite and titanite U-Pb geochronology was conducted on 17 amphibolite- to granulite-facies orthogneiss samples, mainly from a suite of 1050 +2/−8 Ma calc-alkaline augen gneisses, the Feda suite. In these rocks, prograde negatively discordant monazite crystallized during breakdown of allanite and titanite in upper amphibolite facies at 1012–1006 Ma. In the Feda suite and other charnockitic gneisses, concordant to slightly discordant monazite at 1024–997 Ma probably reflects breakdown of biotite during granulite-facies M1 metamorphism. A spread of monazite ages down to 970 Ma in biotite ± hornblende samples possibly corresponds to the waning stage of this first event. In the Feda suite, a well defined monazite growth episode at 930–925 Ma in the amphibolite-facies domain corresponds to major clinopyroxene formation at the expense of hornblende during M2. Growth or resetting of monazite was extremely limited during this phase in the granulite-facies domain, up to the direct vicinity of the anorthosite complex. The M2 event was shortly followed by cooling through ca. 610 °C as indicated by tightly grouped U-Pb ages of accessory titanite and titanite relict inclusions at 918 ± 2 Ma over the entire region. A last generation of U-poor monazite formed during regional cooling below 610 °C, in hornblende-rich samples at 912–904 Ma. This study suggests: (1) that monazite formed during the prograde path of high-grade metamorphism may be preserved; (2) that monazite ages reflect primary or secondary growth of monazite linked to metamorphic reactions involving redistribution of REEs and Th, and/or fluid mobilisation; (3) that the U-Pb system in monazite is not affected by thermal events up to 800–850 °C, provided that conditions were dry during metamorphism.

The application of single zircon evaporation and model Nd ages to the interpretation of polymetamorphic terrains: an example from the Proterozoic mobile belt of south India

Abstract: The technique of single zircon dating from the thermal evaporation of 207Pb/206Pb (Kober 1986, 1987) provides a means of dating successive periods of growth and nucleation of zircons in polymetamorphic assemblages. In contrast Nd model ages may provide a measure of the period of crustal residency for the sample or its protolith. These two techniques have been combined to elucidate the tectonic history of the Proterozoic mobile belt of southern India, exposed south of the Palghat-Cauvery Shear Zone that marks the southern boundary of the Archaean craton of Karnataka. The two main tectonic units of this mobile belt comprise the Madurai and Trivandrum Blocks, both of which are characterised by massive charnockite uplands and low-lying polymetamorphic metasedimentary belts that have undergone a complex tectonic history throughout the Proterozoic. Evidence for early Palaeoproterozoic magmatism is restricted to the Madurai Block where single zircon evaporation ages from a metagranite (2436 ± 4 Ma) are similar to model Nd ages from a range of lithologies suggesting crustal growth at that time. The Trivandrum Block, to the south of the Achankovil shear zone, is comprised of the Kerala Khondalite Belt, the Nagercoil charnockites and the Achankovil metasediments. Single zircon evaporation ages, together with conventional zircon and garnet chronometry, suggest that all three units underwent upper-amphibolite facies metamorphism at ∼1800 Ma, an event unrecorded in the metagranite from the Madurai Block. This implies that the Madurai and Trivandrum blocks represent distinct terrains throughout the Palaeoproterozoic. Model Nd ages from the Achankovil metasediments are much younger (1500–1200 Ma) than those from the adjacent Kerala Khondalite Belt and Madurai Blocks (3000–2100 Ma), but there is no evidence for zircon growth in these metasediments during the Mesoproterozoic. Hence the comparatively young model Nd ages of the metasediments are indicative of a mixed provenance rather than a discrete period of crustal growth. Zircon overgrowths from the Madurai Block (547 ± 17 Ma) and Achankovil metasediments (530 ± 21 Ma) suggest that all tectonic units of the Proterozoic mobile belt of South India shared the same metamorphic history from the early Palaeozoic. This event has been recognised in the basement lithologies of Sri Lanka and East Antarctica, confirming that the constituent terrains of East Gondwana had assembled by this time.

High pressure breakdown of antigorite to spinifex-textured olivine and orthopyroxene, SE Spain

Abstract: The prograde, high pressure, transition from antigorite serpentinite to enstatite-olivine rock occurs along a tectonically undisturbed profile at Cerro del Almirez, SE Spain. The reactant assemblage is antigorite + olivine with tremolite rimming precursor diopside. The product assemblage of tremolite + chlorite + enstatite + olivine has a spinifex-like texture with arborescent or radiating olivine elongated parallel to [001] and with radially grown enstatite. Product enstatite is very poor in Al2O3. Due to numerous oriented submicroscopic inclusions of chromian magnetite, product olivine has a brownish pleochroism and a bulk chromium content similar to precursor antigorite. Titanian clinohumite with a fluorine content of 0.45–0.50 wt% persisted beyond the breakdown of antigorite. The partitioning of iron and magnesium amongst the silicate phases is almost identical to that at lower pressures. Average Kd values Mn/Mg and Ni/Mg are 0.17 and 0.70 for antigorite-olivine pairs and 1.83 and 0.22 for orthopyroxene-olivine pairs, respectively. These data are useful in discriminating generations of olivine grown on each other. From the field data a phase diagram topology for a portion of the system CaO-MgO-SiO2-H2O is derived. This topology forms the basis for extrapolations into inaccessible P-T regions.

Cation diffusion in aluminosilicate garnets: experimental determination in pyrope-almandine diffusion couples

Abstract: Diffusion couples made from homogeneous gem quality natural pyrope and almandine garnets were annealed within graphite capsules under anhydrous conditions at 22–40 kbar, 1057–1400 °C in a piston-cylinder apparatus. The concentration profiles that developed in each couple were modeled to retrieve the self diffusion coefficients [D(I)] of the divalent cations Fe, Mg, Mn and Ca. Because of their usually low concentrations and lack of sufficient compositional change across the interface of the diffusion couples, only a few reliable data can be obtained for D(Ca) and D(Mn) from these experiments. However, nine sets of D(Fe) and D(Mg) data were retrieved in the above P-T range, and cast in the form of Arrhenian relation, D=D 0exp{−[Q(1 bar)+PΔV +]/RT}. The values of the activation energy (Q) and activation volume (ΔV +) depend on whether f O2 is constrained by graphite in the system C-O or held constant. For the first case, we have for Fe:Q(1 bar)=65,532±10,111 cal/mol, D 0=3.50 (±2.30)×10−5 cm2/s, ΔV +=5.6(±2.9) cm3/mol, and for Mg:Q(1 bar)=60,760±8,257 cal/mol, D 0=4.66 (±2.48)×10−5 cm2/s, ΔV +=5.3(±3.0) cm3/mol. Here the ΔV + values have been taken from Chakraborty and Ganguly (1992). For the condition of constant f O2, the Q values are ∼9 kcal lower and ΔV + values are ∼4.9 cm3/mol larger than the above values. Lower temperature extrapolation of the Arrhenian relation for D(Mg) is in good agreement with the Mg tracer diffusion data (D * Mg) of Chakraborty and Rubie (1996) and Cygan and Lasaga (1985) at 1 bar, 750–900 °C, when all data are normalized to the same pressure and to f O2 defined by graphite in the system C-O. The D * Mg data of Schwandt et al. (1995), on the other hand, are lower by more than an order of magnitude than the low temperature extrapolation of the present data, when all data are normalized to the same pressure and to f O2 defined by the graphite buffer. Comparison of the D(Fe), D(Mg) and D(Mn) data in the pyrope-almandine diffusion couple with those in the spessartine-almandine diffusion couple of Chakraborty and Ganguly (1992) shows that the self diffusion of Fe and Mn are significantly enhanced with the increase in Mn/Mg ratio; the enhancement effect on D(Mg) is, however, relatively small. Proper application of the self diffusion data to calculate interdiffusion coefficient or D matrix elements for the purpose of modeling of diffusion processes in natural garnets must take into account these compositional effects on D(I) along with the effects of thermodynamic nonideality, f O2, and pressure.

The age and origin of a thick mafic–ultramafic keel from beneath the Sierra Nevada batholith

Abstract: We present evidence for a thick (∼100 km) sequence of cogenetic rocks which make up the root of the Sierra Nevada batholith of California. The Sierran magmatism produced tonalitic and granodioritic magmas which reside in the Sierra Nevada upper- to mid-crust, as well as deep eclogite facies crust/upper mantle mafic–ultramafic cumulates. Samples of the mafic–ultramafic sequence are preserved as xenoliths in Miocene volcanic rocks which erupted through the central part of the batholith. We have performed Rb-Sr and Sm-Nd mineral geochronologic analyses on seven fresh, cumulate textured, olivine-free mafic–ultramafic xenoliths with large grainsize, one garnet peridotite, and one high pressure metasedimentary rock. The garnet peridotite, which equilibrated at ∼130 km beneath the batholith, yields a Miocene (10 Ma) Nd age, indicating that in this sample, the Nd isotopes were maintained in equilibrium up to the time of entrainment. All other samples equilibrated between ∼35 and 100 km beneath the batholith and yield Sm-Nd mineral ages between 80 and 120 Ma, broadly coincident with the previously established period of most voluminous batholithic magmatism in the Sierra Nevada. The Rb-Sr ages are generally consistent with the Sm-Nd ages, but are more scattered. The 87Sr/86Sr and 143Nd/144Nd intercepts of the igneous-textured xenoliths are similar to the ratios published for rocks outcroping in the central Sierra Nevada. We interpret the mafic/ultramafic xenoliths to be magmatically related to the upper- and mid-crustal granitoids as cumulates and/or restites. This more complete view of the vertical dimension in a batholith indicates that there is a large mass of mafic–ultramafic rocks at depth which complement the granitic batholiths, as predicted by mass balance calculations and experimental studies. The Sierran magmatism was a large scale process responsible for segregating a column of ∼30 km thick granitoids from at least ∼70 km of mainly olivine free mafic–ultramafic residues/cumulates. These rocks have resided under the batholith as granulite and eclogite facies rocks for at least 70 million years. The presence of this thick mafic–ultramafic keel also calls into question the existence of a “flat” (i.e., shallowly subducted) slab at Central California latitudes during Late Cretaceous–Early Cenozoic, in contrast to the southernmost Sierra Nevada and Mojave regions.

Time of formation and peak of Variscan HP-HT metamorphism of quartz-feldspar rocks in the central Erzgebirge, Saxony, Germany

Abstract: The Variscan Erzgebirge represents an antiform with a core of gneisses and mica schists, surrounded by a phyllitic mantle. The Gneiss-Eclogite Unit (GEU), in the central part, is a composite tectonometamorphic assemblage characterized by a HP-HT imprint and comprises migmatitic para- and orthogneisses, HT mylonites, HP granulites, eclogites and garnet peridotites. It is tectonically sandwiched between two major units with distinctly lower PT histories. The GEU experienced a characteristic “kinked” retrograde PT path after HP-HT equilibration with: (1) strong near-isothermal decompression at high temperatures; (2) extensive re-equilibration at medium pressures, followed (3) by rapid cooling during continued uplift. We dated zircons (Pb-Pb evaporation) from granitoid orthogneisses and metapelites of the GEU. The orthogneisses contain euhedral, long-prismatic zircons of igneous origin that provided protolith ages between 470 and 524 Ma. Metapelites retain well-preserved granulite-facies mineral assemblages and contain spherical, multifaceted metamorphic zircons that grew near the peak of HP/HT metamorphism. Inclusions of prograde HP phengite (∼15 kbar) and rutile are included in one such zircon. Metamorphic zircons of three samples from different localities yielded identical 207Pb/206Pb ages of 340.5 ± 0.7 Ma, 341.2 ± 0.5 Ma and 341.6 ± 0.5 Ma respectively. Consideration of these zircon ages with published 39Ar/40Ar white mica ages suggests fast cooling and uplift rates in excess of 50 °C/Ma and 4 km/Ma. This is typical for large-scale extensional tectonic unroofing of the ultra-deep part of a fossil, thickened Variscan continental crust (>60 km) during continuing continental collision and orogenic collapse.

Rare and unusual mineral inclusions in diamonds from Mwadui, Tanzania

Abstract: Syngenetic diamond inclusions from the Mwadui kimberlite reveal that an unusually fertile section of lithospheric mantle beneath the Central African Craton was sampled. This is shown by a very high ratio of lherzolitic to harzburgitic garnet inclusions (1:2) and low Mg/Fe-ratios in olivine and orthopyroxene. Geothermometry applied to the peridotitic inclusions indicates disequilibrium between non-touching inclusion pairs to be common. Disequilibrium between garnet-olivine and garnet-orthopyroxene pairs suggests successive iron enrichment during diamond formation, e.g. leading to the presence of harzburgitic garnet and lherzolitic olivine in the same diamond. Apart from the dominant peridotitic inclusion suite (88%), rare eclogitic inclusions occur (2%) and a number of uncertain paragenesis. Two diamonds, one with eclogitic garnets with moderate pyroxene solid solution and the other with a single ferro-periclase inclusion, suggest the contribution of a small sub-lithospheric component. The finding of the association Fe-FeO-Fe3O4 in one single diamond indicates diamond formation over a large range of f O2 conditions, possibly along redox fronts. Steep compositional gradients may also be reflected by the joint occurrence of harzburgitic garnet and a SiO2-phase in the same diamond. Alternatively the formation of the SiO2-phase may be due to extreme carbonation of the peridotitic source. Further unusual findings include the exsolution of a silicate phase from magnetite inclusions, (i.e. primary solution of γ-olivine) and an ilmenite inclusion with an eskolaite (Cr2O3) component of 14.5 mol%, the latter together with harzburgitic paragenesis silicate inclusions.

Two mantle sources, two plumbing systems : tholeiitic and alkaline magmatism of the maymecha river basin, siberian flood volcanic province

Abstract: Rocks of two distinctly different magma series are found in a ∼4000-m-thick sequence of lavas and tuffs in the Maymecha River basin which is part of the Siberian flood-volcanic province. The tholeiites are typical low-Ti continental flood basalts with remarkably restricted, petrologically evolved compositions. They have basaltic MgO contents, moderate concentrations of incompatible trace elements, moderate fractionation of incompatible from compatible elements, distinct negative Ta(Nb) anomalies, and Nd values of 0 to +2. The primary magmas were derived from a relatively shallow mantle source, and evolved in large crustal magma chambers where they acquired their relatively uniform compositions and became contaminated with continental crust. An alkaline series, in contrast, contains a wide range of rock types, from meymechite and picrite to trachytes, with a wide range of compositions (MgO from 0.7 to 38 wt%, SiO2 from 40 to 69 wt%, Ce from 14 to 320 ppm), high concentrations of incompatible elements and extreme fractionation of incompatible from compatible elements (Al2O3/TiO2∼1; Sm/Yb up to 11). These rocks lack Ta(Nb) anomalies and have a broad range of Nd values, from −2 to +5. The parental magmas are believed to have formed by low-degree melting at extreme mantle depths (>200 km). They bypassed the large crustal magma chambers and ascended rapidly to the surface, a consequence, perhaps, of high volatile contents in the primary magmas. The tholeiitic series dominates the lower part of the sequence and the alkaline series the upper part; at the interface, the two types are interlayered. The succession thus provides evidence of a radical change in the site of mantle melting, and the simultaneous operation of two very different crustal plumbing systems, during the evolution of this flood-volcanic province.

Pressure-temperature paths from P-T pseudosections and zoned garnets: potential, limitations and examples from the Zanskar Himalaya, NW India

Abstract: The extraction of P-T histories from metamorphic rocks provides a valuable dataset for the elucidation of the tectonic mechanisms for orogeny. While continued re-equilibration frequently obliterates early information, garnet zonation and inclusion assemblages can often surmount this problem. The task is more difficult in high variance assemblages or if inclusions are not preserved, but one approach is to use pseudosections that are specific to the bulk composition of a given rock. In the latter case, the compositions and abundances of all the minerals are fixed at a given P-T point such that, if the effective bulk composition is known, the garnet composition alone can be used to reconstruct the history. Here, we explore this approach using examples from the Zanskar Himalaya, NW India. Pseudosections have been calculated for four pelitic to semipelitic rocks from the Zanskar Himalaya and have been contoured for garnet composition. The calculations adequately model the mineral assemblages in the rocks and predict the presence of chlorite in the early assemblage where chlorite is found as inclusions within garnet. Moreover, the pseudosections successfully model the garnet core compositions, with all three independent compositional contours overlapping at a single pressure and temperature. This occurs at ∼550 °C and at pressures varying from 3–7 kbar for the four rocks studied. We have been less successful, however, at modelling garnet compositions beyond the cores because fractionation of the effective bulk composition is caused by garnet growth itself. However, in this case, a combination of the␣pseudosection and conventional thermobarometry using␣Fe-Ti inclusions and matrix phases allows us to reconstruct␣the entire P-T history. The resulting P-T paths record burial of 3–5 kbar without significant temperature increase followed by isobaric heating of 50–100 °C. This evolution is consistent with Himalayan collision in the early Tertiary but a combination of the P-T data presented here and published geochronological data suggests renewed thrusting south of the suture zone in the Oligocene. In addition, the data demonstrate that no extra heat source is required to cause melting of the Himalayan crust in the Miocene. While melting could have occurred both by dehydration during decompression or in the presence of a fluid, the lack of garnet resorption does suggest decompression was rapid and followed quickly by cooling. This scenario favours melting by decompression.

Occurrence and distribution of “moganite” in agate/chalcedony: a combined micro-Raman, Rietveld, and cathodoluminescence study

Abstract: Agate/chalcedony samples of different origin were investigated by performing Raman, X-ray diffraction (using Rietveld refinement), and cathodoluminescence measurements. These analyses were performed to measure the content and spatial distribution of the silica polymorph moganite, which is considered to represent periodic Brazil-law twinning of α-quartz at the unit-cell scale in agate/chalcedonies. Homogeneous standard samples including the nearly α-quartz free moganite type material from Gran Canaria were analysed in order to compare results of the X-ray diffractometry and Raman spectroscopy techniques and to provide a calibration curve for the Raman results. However, due to the different length scales analysed by the two techniques, the “moganite content” in microcrystalline SiO2 samples measured by Raman spectroscopy (short-range order) was found to be considerably higher than the “moganite content” measured by X-ray diffractometry (long-range order). The difference is explained by the presence of moganite nanocrystals, nano-range moganite lamellae, and single Brazil-law twin-planes that are detected by vibrational spectroscopy but that are not large enough (in the sense of coherently scattering lattice domains) to be detected by X-ray diffractometry. High resolution Raman analysis provides a measure of the moganite content and its spatial variation in microcrystalline silica samples with a lateral resolution in the μm-range. Variations in the moganite-to-quartz ratio are revealed by varying intensity ratios of the main symmetric stretching-bending vibrations (A1 modes) of α-quartz (465 cm−1) and moganite (502 cm−1), respectively. Traces of Raman microprobe analyses perpendicular to the rhythmic zoning of agates revealed that the moganite-to-quartz ratio is often not uniform but shows a cyclic pattern that correlates with the observed cathodoluminescence pattern (colour and intensity). Data obtained from an agate sample from a fluorite deposit near Okorusu, Namibia and from a volcanic agate from Los Indios, Cuba were selected for detailed presentation. Variations of cathodoluminescence and Raman data between single bands in agates suggest alternating formation of fine-grained, highly defective chalcedony intergrown with moganite, and coarse-grained low-defect quartz. Multiple zones indicate dynamic internal growth during a self-organizational crystallization process from silica-rich fluids.

Syndeformational recrystallization – dynamic or compositionally induced?

Abstract: Dynamic recrystallization in the strict sense of the term is the reconstitution of crystalline material without a change in chemical composition, driven by strain energy in the form of dislocations. Driving potentials additional to internal strain energy may contribute to the recrystallization of naturally deformed minerals, which form solid solutions such as feldspar, amphiboles and pyroxenes, if they change their composition during recrystallization. To estimate the relative importance of these driving potentials, the chemical composition of porphyroclasts and recrystallized grains of plagioclase, clinopyroxene and hornblende have been investigated in samples from a high grade shear zone of the Ivrea Zone, Italy. The plagioclases show two different recrystallization microstructures: bulging recrystallization at grain boundaries and discrete zones of recrystallized grains across porphyroclasts probably involving fracturing. Deformation took place under amphibolite facies conditions on a retrograde P,T-path. Porphyroclast and recrystallized compositions from bulging recrystallization microstructures differ only in their Or-content and yield a ΔG between mean host grain and mean recrystallized grain composition at fixed P,T-conditions of approximately 5 Joules/10−4 m3. Extreme compositional variations yield approximately 60 J/10−4 m3. The increase of free energy due to dislocations calculated for common glide systems in plagioclase are on the order of 100 Joules/10−4 m3 for high values of dislocation densities of 1014 m−2. Thus, the effect of chemically induced driving energies on grain boundary velocity appears small for mean compositions but may be as great as that of deformational energies for larger chemical differences. In the other type of microstructure, porphyroclasts and recrystallized grains in discrete zones differ in their anorthite content. The maximum ΔG induced by the compositional disequilibrium is on the order of 100 J/10−4 m3. This maximum value is of the same magnitude as the ΔG derived from high dislocation densities of 1014 m−2. The resulting combined ΔG is approximately twice as high as for deformational ΔG alone, and heterogeneous nucleation may become a feasible recrystallization mechanism which is evident from the microstructures. The recrystallization mechanism depends on the nature of the driving potential. Grain boundary migration (GBM) and heterogeneous nucleation can release Gibbs free energy induced by compositional disequilibrium, whereas this is not likely for subgrain rotation. Therefore, only GBM and heterogeneous nucleation may link metamorphism and deformation, so that syndeformational recrystallization may represent a transitional process ranging from dynamic recrystallization to metamorphic reaction.

Petrology of mafic lavas within the Onega plateau, Central Karelia: evidence for 2.0 Ga plume-related continental crustal growth in the Baltic Shield

Abstract: The Onega plateau constitutes part of a vast continental flood basalt province in the SE Baltic Shield. It consists of Jatulian-Ludikovian submarine volcanic, volcaniclastic and sedimentary sequences attaining in places 4.5 km in thickness. The parental magmas of the lavas contained ∼10% MgO and were derived from melts generated in the garnet stability field at depths 80–100 km. The Sm-Nd mineral and Pb-Pb whole-rock isochron ages of 1975 ± 24 and 1980 ± 57 Ma for the upper part of the plateau and a SHRIMP U-Pb zircon age of 1976 ± 9 Ma for its lower part imply the formation of the entire sequence within a short time span. These ages coincide with those of picrites in the Pechenga-Imandra belt (the Kola Peninsula) and komatiites and basalts in the Karasjok-Kittila belt (Norway and Finnmark). Together with lithostratigraphic, chemical and isotope evidence, these ages suggest the derivation of the three provinces from a single large (∼2000 km in diameter) mantle plume. These plume-generated magmas covered ∼600,000 km2 of the Baltic Shield and represent a major contribution of juvenile material to the existing continental crust at 2.0 Ga. The uppermost Onega plateau lavas have high (Nb/Th)N = 1.4–2.4, (Nb/La)N= 1.1–1.3, positive ɛNd(T) of +3.2 and unradiogenic Pb-isotope composition (μ1 = 8.57), comparable with those of modern oceanic plume-derived magmas (oceanic flood basalt and ocean island basalt). These parameters are regarded as source characteristics. The lower sequences have (Nb/Th)N= 0.58–1.2, (Nb/La)N= 0.52–0.88 and ɛNd(T) =−2.6. They have experienced mixing with 10–30% of continental crust and resemble contaminated lavas from other continental flood basalt provinces. The estimated Nb/U ratios of 53 ± 4 in the uncontaminated rocks are similar to those found in the modern mantle (∼47) suggesting that by 2.0 Ga a volume of continental crust similar to the present-day value already existed.

A U-series study of lavas from Kamchatka and the Aleutians: constraints on source composition and melting processes

Abstract: New data are presented for lavas from the Kamchatka Peninsula and the Aleutian arc. Radiogenic isotopes are strikingly homogeneous in the Kamchatka lavas and although incompatible trace element ratios exhibit much greater variability, much of this appears to result from shallow level, crystal fractionation. The data reveal little systematic across-arc change in radiogenic isotopes or trace element ratios. The Nd and Pb isotope data overlap those for Pacific MORB and limit the amount of sediment that could be incorporated in the mantle source region to <1% which is insufficient to account for the observed La/Ta ratios (50–68) in the high-MgO lavas. The lack of a positive correlation between La/Ta and depth to the slab suggests that melt–wall rock interaction was not important in controlling this ratio. Instead it is inferred that La/Ta increased during partial melting and that DLa/DTa = 0.11–0.06, possibly due to residual amphibole. Ba, U, Sr and Pb were added to the source by an aqueous fluid from the subducting slab and its inferred isotopic composition indicates that this fluid was derived from the altered oceanic crust. The addition of U resulted in a large range of (238U/232Th) from 0.79–2.48 similar to that observed in the Mariana and Lesser Antilles island arcs. However, (230Th/232Th) = 0.79–2.34, and the majority of samples lie close to the equiline indicating that the time since U/Th fractionation is generally ≥150 thousand years. The large width of the volcanic zone is assumed to reflect protracted fluid release from the subducting slab over the depth interval 170–380 km possibly coupled with extension across the Central Kamchatka Depression. The data from the Aleutians contrast strongly with those from Kamchatka. Radiogenic isotope data indicate that the Aleutian lavas contain a significant recycled sedimentary component, consistent with elevated 10Be/9Be ratios. The Aleutian lavas have (230Th/232Th) = 0.79–2.34 and exhibit a significant range of U/Th disequilibria [(238U/230Th) = 0.75–1.01]. However, 10Be/9Be is positively correlated with (238U/230Th) suggesting that the 10Be signal was carried by the aqueous fluid from the slab. The U/Th disequilibria for the Aleutian lavas lie close to a 30 thousand year reference line suggesting that this fluid was released from the slab ∼30 thousand years ago similar to recent estimates from the Lesser Antilles, Marianas, and Tonga-Kermadec island arcs from which it is inferred that fluid addition was the trigger for partial melting. Given that the rate of convergence in Kamchatka is similar to that in the Aleutians, Marianas and Tonga-Kermadec the inferred greater time since␣fluid release in Kamchatka requires further investigation.

Infiltration of refractory melts into the lowermost oceanic crust: evidence from dunite- and gabbro-hosted clinopyroxenes in the Bay of Islands Ophiolite

Abstract: Up to 3 km of dunitic rocks occur below crustal gabbro in the Blow Me Down massif (Bay of Islands Ophiolite, Newfoundland). Analyses of dunite- and gabbro-hosted clinopyroxene grains (cpx) for rare earth elements (REE), Zr, and Ti reveal three types of chondrite-normalized patterns: N-group patterns are similar to cpx grains as they would form by fractionation from a range of mid ocean ridge basalts (MORB). They are typical for a few higher level dunitic samples as well as mafic cumulates. F-group patterns show light REE depletion, very strong middle REE fractionation and a positive Zr anomaly and occur in dunites only. R-group patterns are severely depleted in both light and heavy REEs relative to MORB-like cpx and two samples of the group display a positive Ti anomaly. They are also restricted to dunitic rocks. The patterns are explained in a two stage model in which an established dunite sequence, dominated by MORB-type cumulate signatures (N-group), was infiltrated by extremely refractory melts. During infiltration of the refractory melt chromatographic fractionation occurred, transforming N-group dunites into F-group and R-group dunites. The F-group patterns are composite patterns: heavy REE, Ti ± Zr reflect the original MORB-like cumulate dunite host, light REEs indicate equilibrium with the infiltrating, refractory melts. Steep slopes in the middle REEs reflect the position of the chromatographic front. For more intense percolation of refractory melts, R-group patterns with a positive Ti anomaly will form by the same process. The rest of the R-group patterns displaying no positive Ti anomaly may represent either the most intensely reacted host rocks or these dunites derive directly as cumulates from refractory melts. Only small volumes of refractory melt (a 5 m column) are required to imprint the observed trace element pattern on the thick original dunite sequence. One of several possible origins for the refractory melts is transformation of original MORB-type melts by way of chromatographic fractionation within the highly depleted, residual uppermost mantle. In the framework of an oceanic spreading centre, the migrating, refractory liquids are considered a late event following the main constructive stage dominated by aggregated melts. The study demonstrates that highly refractory melts can exist under oceanic spreading centres dominated by a MORB-like cumulate and volcanic sequence.

Multi-stage magma ascent beneath the Canary Islands: evidence from fluid inclusions

Abstract: Gabbroic and ultramafic xenoliths and olivine and clinopyroxene phenocrysts in basaltic rocks from Gran Canaria, La Palma, El Hierro, Lanzarote and La Gomera (Canary Islands) contain abundant CO2-dominated fluid inclusions. Inclusion densities are strikingly similar on a regional scale. Histogram maxima correspond to one or more of the following pressures: (1) minimum 0.55 to 1.0 GPa (within the upper mantle); (2) between 0.2 and 0.4 GPa (the Moho or the lower crust); (3) at about 0.1 GPa (upper crust). Fluid inclusions in several rocks show a bimodal density distribution, the lower-density maximum comprising both texturally early and late inclusions. This is taken as evidence for an incomplete resetting of inclusion densities, and simultaneous formation of young inclusions, at well-defined magma stagnation levels. For Gran Canaria, pressure estimates for early inclusions in harzburgite and dunite xenoliths and olivine phenocrysts in the host basanites overlap at 0.9 to 1.0 GPa, indicating that such magma reservoir depths coincide with levels of xenolith entrainment into the magmas. Magma chamber pressures within the mantle, inferred to represent levels of mantle xenolith entrainment, are 0.65–0.95 GPa for El Hierro, 0.60–0.68 GPa for La Palma, and 0.55–0.75 GPa for Lanzarote. The highest-density fluid inclusions in many Canary Island mantle xenoliths have probably survived in-situ near-isobaric heating at the depth of xenolith entrainment. Inclusion data from all islands indicate ponding of basaltic magmas at Moho or lower crustal depths, and possibly at an additional higher level, strongly suggestive of two main crustal accumulation levels beneath each island. We emphasize that repeated magmatic underplating of primitive magmas, and therefore intrusive accretion, are important growth mechanisms for the Canary Islands, and by analogy, for other ocean islands. Comparable fluid inclusion data from primitive rocks in other tectonic settings, including Iceland, Etna and continental rift systems (Hungary, South Norway), indicate that magma accumulation close to Moho depths shortly before eruption is not, however, restricted to oceanic intraplate volcanoes. Lower crustal ponding and crystallization prior to eruption may be the rule rather than the exception, independent of the tectonic setting.

Melt/harzburgite reaction in the petrogenesis of tholeiitic magma from Kilauea volcano, Hawaii

Abstract: We use the results of elevated pressure melting experiments to constrain the role of melt/mantle reaction in the formation of tholeiitic magma from Kilauea volcano, Hawaii. Trace element abundance data is commonly interpreted as evidence that Kilauea tholeiite is produced by partial melting of garnet lherzolite. We experimentally determine the liquidus relations of a tightly constrained estimate of primary tholeiite composition, and find that it is not in equilibrium on its liquidus with a garnet lherzolite assemblage at any pressure. The composition is, however, cosaturated on its liquidus with olivine and orthopyroxene at 1.4 GPa and 1425 °C, from which we infer that primary tholeiite is in equilibrium with harzburgite at lithospheric depths beneath Kilauea. These results are consistent with our observation that tholeiite primary magmas have higher normative silica contents than experimentally produced melts of garnet lherzolite. A model is presented whereby primary tholeiite forms via a two-stage process. In the first stage, magmas are generated by melting of garnet lherzolite in a mantle plume. In the second stage, the ascent and decompression of magmas causes them to react with harzburgite in the mantle by assimilating orthopyroxene and crystallizing olivine. This reaction can produce typical tholeiite primary magmas from significantly less siliceous garnet lherzolite melts, and is consistent with the shift in liquidus boundaries that accompanies decompression of an ascending magma. We determine the proportion of reactants by major element mass balance. The ratio of mass assimilated to mass crystallized (Ma/Mc) varies from 2.7 to 1.4, depending on the primary magma composition. We use an AFC calculation to model the effect of melt/harzburgite reaction on melt rare earth and high field strength element abundances, and find that reaction dilutes, but does not significantly fractionate, the abundances of these elements. Assuming olivine and orthopyroxene have similar heats of fusion, the Ma/Mc ratio indicates that reaction is endothermic. The additional thermal energy is supplied by the melt, which becomes superheated during adiabatic ascent and can provide more thermal energy than required. Melt/harzburgite reaction likely occurs over a range of depths, and we infer a mean depth of 42 km from our experimental results. This depth is well within the lithosphere beneath Kilauea. Since geochemical evidence indicates that melt/harzburgite reaction likely occurs in the top of the Hawaiian plume, the plume must be able to thin a significant portion of the lithosphere.

Reactions between mantle xenoliths and host magma beneath La Palma (Canary Islands): constraints on magma ascent rates and crustal reservoirs

Abstract: Spinel-bearing peridotitic mantle xenoliths from the 1949 eruption on La Palma were modified mineralogically and chemically during prolonged reaction with their host magma. The magmatism that brought the peridotites to the surface caused two distinct generations of xenolith fractures: (1) Old fractures are characterized by crystalline selvages with cumulus textures towards the host magma, or by polymineralic veins. They are accompanied by 0.9–2 mm wide diffusion zones where peridotite olivine became less forsteritic through diffusive exchange with the host magma. Old fractures represent most of each xenolith's surface. (2) Young fractures show no selvages and only narrow diffusion zones of <0.02 mm width. Calculations based on a model of Fe-Mg interdiffusion give an age of 6 to 83 years and <4 days for old and young fractures, respectively. A combination of these data with fluid inclusion barometry indicates that selvages and veins formed during xenolith transport rather than representing wall-rock reactions or mantle metasomatism. The results provide ample evidence for prolonged storage of the xenoliths in the crust, constraining a multi-stage magma ascent: Years to decades prior to eruption, ascending magma ruptured peridotitic wall-rock possibly through hydraulic fracturing and stoping around magma reservoirs. Magma batches transported the peridotite xenoliths to the crust at ascent rates exceeding 0.2 ms−1. The xenoliths and their host magma stagnated during at least 6 years in possibly sill-like reservoirs at 7–11 km depth. The xenoliths became deposited and subsequently embedded in a mush of settled phenocrysts, while selvages and veins crystallized until the eruption commenced. At the end of the eruption, the xenoliths were finally transported to the surface within hours to days. Decompression during the rapid ascent induced internal stresses and caused renewed fragmentation of the xenoliths, producing the young fractures.

The high-pressure stability of zoisite and phase relationships of zoisite-bearing assemblages

Abstract: The fluid-absent reaction 12 zoisite = 3 lawsonite + 7 grossular + 8 kyanite + 1 coesite was experimentally reversed in the model system CaO-Al2O3-SiO2-H2O (CASH) using a multi-anvil apparatus The upper pressure stability limit for zoisite was found to extend to 50 GPa at 700 °C and to 66 GPa at 950 °C Additional experiments both in the H2O-SiO2-saturated and in the H2O-Al2O3-saturated portions of CASH provide further constraints on high pressure phase relationships of lawsonite, zoisite, grossular, kyanite, coesite, and an aqueous fluid Consistency of the present experiments with the H2O-saturated breakdown of lawsonite is demonstrated by thermodynamic analysis using linear programming techniques Two sets of data consistent with databases of Berman (1988) and Holland and Powell (1990) were retrieved combining experimental phase relationships, calorimetric constraints, and recently measured elastic properties of solid phases The best fits result in G f ,1,298 ∘,zoisite=−6,499,400 J and S 1,298 ∘,zoisite=302 J/K, and G f ,1,298 ∘,lawsonite=−4,514,600 J and S 1,298 ∘,lawsonite=220 J/K for the dataset of Holland and Powell, and G f ,1,298 ∘,zoisite=−6,492,120 J and S 1,298 ∘,zoisite=304 J/K, and G f ,1,298 ∘,lawsonite=−4,513,000 J and S 1,298 ∘,lawsonite= 218 J/K for the dataset of Berman Examples of the usage of zoisite as a geohygrometer and as a geobarometer in rocks metamorphosed at eclogite facies conditions are worked, profiting from the thermodynamic properties retrieved here

Models of corundum origin from alkali basaltic terrains: a reappraisal

Abstract: Corundums from basalt fields, particularly in Australia and Asia, include a dominant blue-green-yellow zoned “magmatic” suite (BGY suite) and subsidiary vari-coloured “metamorphic” suites. The BGY corundums have distinctive trace element contents (up to 0.04 wt% Ga2O3 and low Cr/Ga and Ti/Ga ratios <1). Different melt origins for BGY corundums are considered here from their inclusion and intergrowth mineralogy, petrologic associations and tectonic setting. Analysed primary inclusion minerals (over 100 inclusions) cover typical feldspars, zircon and Nb-Ta oxides and also include hercynite-magnetite, gahnospinel, rutile-ilmenite solid solution, calcic plagioclase, Ni-rich pyrrhotite, thorite and low-Si and Fe-rich glassy inclusions. This widens a previous inclusion survey; New England, East Australia corundums contain the most diverse inclusion suite known from basalt fields (20 phases). Zircon inclusion, intergrowth and megacryst rare earth element data show similar patterns, except for Eu which shows variable depletion. Temperature estimates from magnetite exsolution, feldspar compositions and fluid inclusion homogenization suggest that some corundums crystallized between 685–900 °C. Overlap of inclusion Nb, Ta oxide compositions with new comparative data from niobium-yttrium-fluorine enriched granitic pegmatites favour a silicate melt origin for the corundums. The feasibility of crystallizing corundum from low-volume initial melting of amphibole-bearing mantle assemblages was tested using the MELTS program on amphibole-pyroxenite xenolith chemistry from basalts. Corundum appears in the calculations at 720–880 °C and 0.7–1.1 GPa with residual feldspathic assemblages that match mineral compositions found in corundums and their related xenoliths. A model that generates melts from amphibole-bearing lithospheric mantle during magmatic plume activity is proposed for BGY corundum formation.

A close look at dihedral angles and melt geometry in olivine-basalt aggregates: a TEM study

Abstract: Olivine-basalt aggregates sintered at high P/T have been used as a simplest approximation of partially molten upper mantle peridotite. In the past, geometry of partial melt in polycrystalline olivine (and other materials) has been characterised by dihedral (wetting) angles which depend upon surface free energy. However, since olivine (like most other crystalline materials) is distinctively anisotropic, the simple surface energy balance defining the dihedral angles cos(Θ/2)=gb/2sl is not valid and melt geometry is more complicated than can be expressed by a single dihedral angle value. We examine in detail melt geometry in aggregates held at high temperature and pressure for very long times (240–612 h). We show the simple dihedral angle concept to be invalid via transmission electron microscope images. Olivine-basalt interfaces are frequently planar crystal faces (F-faces) which are controlled by the crystal structure rather than the surface area minimisation used in the simple dihedral angle concept. Nevertheless, the dihedral angles may provide useful insights in some situations. They may give a rough estimation of the wetting behaviour of a system, or be used to approximate the melt distribution if F-faces are not present (possibly at large grain size and very low melt fraction). Our measurements, excluding F-faces, give a range of dihedral angle values from 0 to 10° which is significantly lower than reported previously (20–50°). The nature of 0° angles (films and layers up to 1 μm in thickness) is unclear but their frequency compared to dry grain boundaries depends on grain size and melt fraction (e.g. 70% for grain size 43 μm and melt fraction 2%).