Plagioclase-hosted Melt Inclusions Research Articles

Europium in plagioclase-hosted melt inclusions reveals mantle melting modulates oxygen fugacity

To gain insights into the composition and heterogeneity of Earth’s interior, the partial pressure of oxygen (oxygen fugacity, or fO2) in igneous rocks is characterized. A surprising observation is that relative to reference buffers, fO2s of mantle melts (mid-ocean ridge basalts, or MORBs) and their presumed mantle sources (abyssal peridotites) differ. Globally, MORBs have near-uniform fO2s, whereas abyssal peridotites vary by about three orders of magnitude, suggesting these intimately related geologic reservoirs are out of equilibrium. Here, we characterize fO2s of mantle melting increments represented by plagioclase-hosted melt inclusions, which were entrapped as basaltic melts migrated from their sources toward the seafloor. At temperatures and fO2s constrained by rare earth element distributions, a range of fO2s consistent with the abyssal peridotites is recovered. The fO2s are correlated with geochemical proxies for mantle melting, suggesting partial melting of Earth’s mantle decreases its fO2, and that the uniformity of MORB fO2s is a consequence of the melting process and plate tectonic cycling.

Amphibole recycling and polybaric crystallization in rhyolitic lava domes from melt inclusion geochemistry at Cerro La Torta, Altiplano-Puna Volcanic Complex

La Torta is a rhyolitic lava dome located in the El Tatio volcanic complex, Región de Antofagasta (northern Chile), to the west of the Cerros de Tocorpuri volcanoes. La Torta is considered to be an important part of the central zone of the Altiplano Puna Magma Body (APMB) based on chemical and petrological signatures that distinguish it from other highly silicic volcanic systems in the area (i.e., Chao dacite). The petrological and geochemical characteristics, along with parental composition and magma evolution, were determined by analyzing amphibole and plagioclase-hosted melt inclusions using electron microprobe (EMP) and laser ablation ICP-MS (LA-ICP-MS). Our results show that La Torta dome was formed from a rhyolitic Ca- and K-rich, calc-alkaline magma, with an average SiO2 content of 76.8 wt % and H2O content of about 5.4 wt %, which was emplaced in the upper continental crust between 3.9 and 7.6 km below the paleosurface. The parental source of La Torta dome was a highly hydrous mush of andesitic composition, with plagioclase + amphibole ± quartz ± pyroxene crystal assemblage.Heating by an influx of magmas ascending from deeper sources to upper crustal depths (i.e., Altiplano-Puna Magma Body) caused the partial melting of amphibole and pyroxene in the source, evidenced by amphibole dissolution textures, enriching the melt in Cr, Ni, HFSE and HREE. As the system starts to cool and ascend, new amphibole crystallized from this mush, trapping the enriched residual melt as enriched melt inclusions in its cores, along with crystallization of biotite and plagioclase. We refer to this process as amphibole recycling. At the same time, the new melt drags the plagioclase and quartz inherited from the parental mush, forming the magma that emplaced at shallower levels in the crust to eventually erupt as La Torta.

Arc–Backarc Exchange Along the Tonga–Lau System: Constraints From Volatile Elements

Abstract Analyses of olivine-, pyroxene- and plagioclase-hosted melt (now preserved as glass) inclusions (MI) in submarine glasses and subaerial scoria from seven volcanoes along the volcanic front of the Tonga arc provide the first comprehensive dataset including volatile contents (H2O, CO2, S, Cl and F) for this arc. Maximum water contents at each volcano are typical (3–5 wt %) of other arc volcanoes worldwide, and within each volcano, water generally correlates with sulfur, consistent with degassing but not diffusive re-equilibration. The Tonga arc is notable for the eruption of magmas sourced from strongly depleted upper mantle, including boninites in the northern half of the active arc. A key question has been whether such boninites are derived from high degrees of melting actively occurring under the arc, driven by high mantle temperatures and water contents, or from high integrated degrees of melting including melting of depleted mantle in the backarc. This study supports the latter view, based on sub-arc mantle melt fractions of 17–23%, calculated from primary melt water contents and mantle-melt thermobarometry. Such fractions are not high enough to generate boninites from a typical, fertile mantle source in a single melting stage, and melt fractions do not increase north of 22°S, where the boninitic magmas appear. Instead, the northern Tonga arc boninites reflect high cumulative degrees of melting (>30%, with respect to fertile mantle) and occur along the volcanic front where adjacent backarc spreading is fast, proximal and mature. Slab tracers such as the Ba/La ratio peak at around 20–21°S, in the central part of the Tonga arc. A Ba/La peak also appears at about the same latitude in the Lau backarc basin. This peak is consistent with the coolest slab temperatures beneath the arc and backarc at this latitude, as the slab warms to the north due to an increase in mantle potential temperature, and to the south due to less heat extraction from melting. A new concept developed here is cooling of the mantle wedge by melting, due to the supply of the heat of fusion and melt removal. Such a process can cool the mantle by more than 100 degrees, and thus affect slab heating as well. We find a critical distance of slab influence in the Lau spreading centers; back-arc magmas erupted >70 km from the arc are >210 km above the slab, beyond the point of slab dehydration, and show no enrichment in Ba/La. The coupled dynamics of the Tonga arc-Lau basin are rich in their connections, and demonstrate how melting processes beneath the back-arc and arc can affect slab processes, which in turn can affect the composition of fluids that feed the arc and backarc.

Magma evolution of the South China Sea basin from continental-margin rifting to oceanic crustal spreading: Constraints from In-situ trace elements and Sr isotope of minerals

The South China Sea (SCS) has undergone a complete Wilson cycle. We studied the volcanic rocks from IODP Sites U1502, U1500, U1431, U1433, and U1434 that record the exact mantle dynamics of the SCS basin from continental rifting to oceanic crustal spreading. We present in-situ major and trace element data for minerals (olivine, clinopyroxene, and plagioclase) and melt inclusions; and in-situ Sr isotopic ratios for plagioclases from the volcanic samples. All SCS volcanic rocks belong to the tholeiite series. Most Site U1431 basalts are slightly depleted in light rare earth elements (LREEs) and large-ion lithophile elements (LILEs) and similar to that of normal mid-oceanic ridge basalts (N-MORB). All clinopyroxenes and olivines, and the calculated equilibrium melts of clinopyroxene cores display depletion of LREEs and enrichment of heavy rare earth elements (HREEs) similar to N-MORB-type primary magma. Thus, these volcanic rocks are likely formed by the extension decompression partial melting of shallow depleted magma. The Sites U1502, U1500, U1433, and U1434 bulk-rocks and melt inclusions show flat LREE and HREE patterns similar to that of enriched MORB (E-MORB), likely attributed to the partial melting of a depleted mantle source with different degrees of lower continental crustal contamination. The increase in 87Sr/86Sr ratios from core to rim in Site U1500 plagioclases further prove the crustal contamination during magmatism. Site U1431 clinopyroxene and Site U1500 plagioclase show normal zoning textures and slight enrichment of incompatible elements in their rims but depletion in their cores, likely indicating significant crystallization of late-stage differentiated melts. Although some clinopyroxene and plagioclase were likely crystallized in a relatively closed environment, the existence of melt inclusions and the large Sr isotopic variations between the plagioclase rim and groundmass suggest a dynamic and open magmatic system that likely experienced melt–rock interactions and magma mixing processes. Melt supply rate changes from low to high during the SCS evolution process from initial rifting (Site U1500) to later spreading (Sites U1431 and U1433) indicating that: (1) the number of high-An plagioclase-hosted melt inclusions has gradually decreased; and (2) the zoning textures of clinopyroxene and plagioclase have gradually changed from simple to complex. The influence of mantle plumes and the upwelling of fertile subduction-related asthenospheric magma were negligible during the continental-margin breakup and initial SCS spreading, but played an important role during the late SCS spreading period, as revealed by the variation in Sr isotopic compositions of high-An plagioclase cores, the different mantle potential temperatures and source lithologies.

Experimental Constraints on Homogenization of Plagioclase-Hosted Melt Inclusions From Plagioclase Ultraphyric Basalts

Study of melt inclusions (MIs) is a commonly applied method for defining the composition of magmas present at depth prior to mixing, fractionation, and degassing. Our ability to use data from MIs is complicated by post-entrapment processes (PEP) that can modify their composition during transport and eruption. Many of the PEP can be reversed by heating the MIs to temperatures near those at which the MI and its host were formed. However, the process of reversing PEP by homogenization may introduce changes in MI compositions, making interpretation difficult. We present a series of low and high pressure homogenization experiments on plagioclase-hosted MIs from Plagioclase Ultraphyric Basalts (PUBs) designed to develop a methodology for recovering the composition at the time of entrapment of plagioclase-hosted MIs. These experiments included low pressure (1 bar) homogenization experiments conducted as a time series for 30 min, 4 h, 1 day, 4 days, and 8 days), and at 7.5 kbar for 2 and 4 days. The 7.5 kbar pressure used for the high pressure experiments was based on the CO2-based entrapment pressures determined from MI from this sample. Experiments run at low pressure and run times of 4 and 8 days exhibited compositional drift, most notably in the form of increasing MgO in MIs. This drift was not observed at 7.5 kbars or for the shorter run time 1 atm experiments. These results are consistent with a model where drift in composition with time is caused by crystal relaxation driven by the high internal pressure within the MI (the pressure at which the MI formed), together with the lower confining pressure during homogenization (1 bar). Therefore, MI homogenization will produce the least amount of drift if runs are made for short time periods (∼30 min) or at the pressure of entrapment.

LA-ICP-MS Analysis of Crystallized Melt Inclusions in Olivine, Plagioclase, Apatite and Pyroxene: Quantification Strategies and Effects of Post-Entrapment Modifications

Abstract Melt inclusions represent a unique tool to reconstruct the composition and chemical evolution of silicate melts in magmatic systems. Laser-ablation inductively-coupled-plasma mass-spectrometry (LA-ICP-MS) is the most commonly used microanalytical technique to analyse crystallized melt inclusions without prior re-homogenization. Well-preserved melt inclusions can be quantified by subtracting the contribution of co-ablated host with a carefully selected internal standard. However, post-entrapment compositional re-equilibration commonly renders this task difficult, to the same degree as it would affect any quantification after prior re-homogenization. In this study, we first examine well-preserved, crystallized melt inclusions hosted in olivine, plagioclase, apatite, clinopyroxene and orthopyroxene from porphyry dikes and volcanic rocks to test various quantification strategies and evaluate the associated uncertainties, and then we use these strategies to quantify coarsely crystallized melt inclusions from gabbroic rocks at Marble Canyon (USA) and Laiyuan (China) that experienced severe post-entrapment modifications due to relatively slow cooling rates. The results demonstrate that even for well-preserved melt inclusions hosted in chemically complex minerals the uncertainty related to inclusion–host deconvolution can be rather high (up to 30‒50% for host-incompatible trace elements significantly above their limits of detection), though other uncertainties inherent to LA-ICP-MS analysis are relatively small (typically ≤5‒10%). The deconvolution-related uncertainty can be minimized to ca. 10% by (1) choosing whole rocks that are fresh and representative of magmatic liquids, (2) choosing the smallest possible spot size to ablate the melt inclusions, and (3) choosing a host endmember that is compositionally as similar as possible to the one ablated together with the inclusion. Results of coarsely crystallized melt inclusions from gabbroic rocks suggest that the range of elements affected by post-entrapment re-equilibration varies from intrusion to intrusion. Olivine-hosted melt inclusions from Marble Canyon appear to have diffusively lost Fe, Ti and Ca, whereas those from Laiyuan lost Fe, Na, Al, Ca, Ti and Y and gained V. However, the relative abundances of K, P, Rb, Sr, Zr, Nb, Mo, Cs, Ba, Ce, Ta, Pb, Th, U and ±Cu appear unchanged. Plagioclase-hosted melt inclusions from Marble Canyon are relatively well-preserved, whereas those from Laiyuan lost significant amounts of Fe, K, Mg, Mn, Rb and Co. Apatite-hosted melt inclusions seem well preserved with regard to most elements except for Cu. These results suggest that despite the post-entrapment modification of certain element concentrations and the associated difficulties in melt inclusion quantification, information on the approximate abundances of other elements that are invaluable for petrogenetic and metallogenic studies can still be retrieved from melt inclusions in gabbroic rocks using the LA-ICP-MS technique.

Experimental determinations of trace element partitioning between plagioclase, pigeonite, olivine, and lunar basaltic melts and an fO2 dependent model for plagioclase-melt Eu partitioning

We report partition coefficients for Sr, Sc, La, Ce, Pr, Nd, Pm, Sm, Eu, Gd, Tb, Dy, Ho, Er, Tm, Yb, Y, and Th for plagioclase, pigeonite, and olivine experimentally grown from a lunar basaltic liquid at 1 bar and two different oxygen fugacities (fO2; fayalite-magnetite-quartz (FMQ) and the iron-wüstite (IW) oxygen buffers) at temperatures (Ts) of 1100–1125 °C. These results expand the partitioning database for plagioclase-silicate melt partitioning at lunar relevant conditions and compositions. Variations of partition coefficients for rare earth elements (REEs) are highly systematic with respect to ionic radii, in agreement with lattice strain theory. We use the lattice strain model to evaluate partition coefficients for highly incompatible elements such as the light (or in the case of plagioclase, heavy) REEs, for which direct determinations are experimentally and analytically difficult. Our experiments suggest that within analytical and experimental uncertainty, over the range of explored fO2 (FMQ to IW), T (1100–1125 °C) and An# conditions (An# >94.5, where An#=Ca/(Ca + Na) × 100, in moles), fO2 exerts an important but secondary control on Eu partitioning in anorthitic plagioclase. In contrast, the global plagioclase-melt partitioning dataset highlights the first-order role of fO2 in determining plagioclase-melt Eu partition coefficients over a wide range of An#s. Using experimental results and predictive models for divalent and trivalent element plagioclase-melt partitioning in the plagioclase ring site, we parameterized an fO2-dependent plagioclase-melt Eu partitioning model. Within a factor of two, the model reproduces measured Eu partition coefficients for a wide range of experimental conditions (750–1400 °C) and compositional systems (basalts to dacites) over > 15 orders of magnitude variation in fO2. When equilibrium partitioning of Eu between plagioclase and melt can be established, the model may be applied as a plagioclase-melt oxybarometer. Preliminary application to plagioclase-hosted melt inclusions in mid-ocean ridge basalts suggests they equilibrated near the FMQ buffer. We used the new Eu partitioning model to evaluate the petrogenesis of a lunar anorthositic impactite that hosts plagioclase with both positive and negative Eu anomalies. Our investigation suggests the distribution of REEs and Eu in this sample is consistent with subsolidus reequilibration after impact-induced heating.

Plagioclase–hosted melt inclusions as indicators of inhibited rhyolitic melt beneath a mafic volcano: a case study of the Izu–Omuroyama monogenetic volcano, Japan

Plagioclase–hosted melt inclusions as indicators of inhibited rhyolitic melt beneath a mafic volcano: a case study of the Izu–Omuroyama monogenetic volcano, Japan

Deep roots for mid-ocean-ridge volcanoes revealed by plagioclase-hosted melt inclusions.

The global mid-ocean ridge system is the most extensive magmatic system on our planet and is the site of 75 per cent of Earth's volcanism1. The vertical extent of mid-ocean-ridge magmatic systems has been considered to be restricted: even at the ultraslow-spreading Gakkel mid-ocean ridge under the Arctic Ocean, where the lithosphere is thickest, crystallization depths of magmas that feed eruptions are thought to be less than nine kilometres2. These depths were determined using the volatile-element contents of melt inclusions, which are small volumes of magma that become trapped within crystallizing minerals. In studies of basaltic magmatic systems, olivine is the mineral of choice for this approach2-6. However, pressures derived from olivine-hosted melt inclusions are at odds with pressures derived from basalt major-element barometers7 and geophysical measurements of lithospheric thickness8. Here we present a comparative study of olivine- and plagioclase-hosted melt inclusions from the Gakkel mid-ocean ridge. We show that the volatile contents of plagioclase-hosted melt inclusions correspond to much higher crystallization pressures (with a mean value of 270 megapascals) than olivine-hosted melt inclusions (with a mean value of 145 megapascals). The highest recorded pressure that we find equates to a depth 16.4 kilometres below the seafloor. Such higher depths are consistent with both the thickness of the Gakkel mid-ocean ridge lithosphere and with pressures reconstructed from glass compositions. In contrast to previous studies using olivine-hosted melt inclusions, our results demonstrate that mid-ocean-ridge volcanoes may have magmatic roots deep in the lithospheric mantle, at least at ultraslow-spreading ridges.

Copper and Li diffusion in plagioclase, pyroxenes, olivine and apatite, and consequences for the composition of melt inclusions

Silicate melt inclusions are an important tool to reconstruct original concentrations of volatiles and metals in silicate melts, as these constituents are partly lost during magma solidification. However, melt inclusions analyzed in co-precipitated minerals in volcanic rocks commonly show strongly divergent Cu contents, raising doubts about the validity of their composition. To understand the origin of this divergence a multiple approach was followed. First, the phenomenon was documented by recording petrographic details of five volcanic samples and analyzing 189 melt inclusions by laser-ablation ICP-MS. Second, diffusion experiments on gem-quality plagioclase, clinopyroxene, orthopyroxene, apatite and olivine were performed in a gas mixing furnace to constrain diffusion coefficients of Cu and to a lesser extent of Li. Third, re-equilibration experiments were performed on melt inclusions within plagioclase crystals to induce changes in their Cu and Li contents.The LA-ICP-MS analyses reveal that plagioclase-hosted and orthopyroxene-hosted melt inclusions in volcanic rocks commonly contain an order of magnitude more Cu than melt inclusions in co-precipitated clinopyroxene and olivine. Plagioclase-hosted melt inclusions in intrusive rocks, on the other hand, do not show this divergence. The diffusion experiments conducted on minerals find that Cu and Li diffusion in plagioclase is extremely fast (log D = −13.0 to −11.5 m2 s−1 at 1000 °C) and in both cases occurs via two separate diffusion mechanisms. Copper diffusion coefficients in apatite, clinopyroxene, orthopyroxene and olivine are 2–3 orders of magnitude lower but are still high compared to most other elements. Both the re-equilibration experiments on melt inclusions and quantitative modeling based on the measured diffusion coefficients demonstrate that at 1000 °C plagioclase-hosted melt inclusions can re-equilibrate their Cu and Li content with that of the surrounding magma within a few hours to a few weeks, whereas apatite-, clinopyroxene-, orthopyroxene- and olivine-hosted melt inclusions require tens of years to hundreds of years to do so. Abnormally high Cu contents of plagioclase- and orthopyroxene-hosted melt inclusions appear to result from postentrapmental Cu gain. In the case of orthopyroxene-hosted melt inclusions Cu seems to have been gained as a consequence of postentrapmental sidewall crystallization, which triggered the formation of sulfide globules that acted as a Cu sink. This process likely occurred at depth in magma chambers. Plagioclase-hosted melt inclusions, on the other hand, seem to have gained Cu in exchange for hydrogen that diffused out of the melt inclusions during or after volcanic eruption. Our results suggest that Cu (and also Li) concentrations measured in melt inclusions have to be generally treated with caution, and that particular prudence has to be used in the case of plagioclase and orthopyroxene hosts.

Volatile and light lithophile elements in high-anorthite plagioclase-hosted melt inclusions from Iceland

Melt inclusions formed during the early stages of magmatic evolution trap primitive melt compositions and enable the volatile contents of primary melts and the mantle to be estimated. However, the syn- and post-entrapment behaviour of volatiles in primitive high-anorthite plagioclase-hosted melt inclusions from oceanic basalts remains poorly constrained. To address this deficit, we present volatile and light lithophile element analyses from a well-characterised suite of nine matrix glasses and 102 melt inclusions from the 10ka Grímsvötn tephra series (i.e., Saksunarvatn ash) of Iceland’s Eastern Volcanic Zone (EVZ). High matrix glass H2O and S contents indicate that eruption-related exsolution was arrested by quenching in a phreatomagmatic setting; Li, B, F and Cl did not exsolve during eruption. The almost uniformly low CO2 content of plagioclase-hosted melt inclusions cannot be explained by either shallow entrapment or the sequestration of CO2 into shrinkage bubbles, suggesting that inclusion CO2 contents were controlled by decrepitation instead. High H2O/Ce values in primitive plagioclase-hosted inclusions (182–823) generally exceed values expected for EVZ primary melts (∼180), and can be accounted for by diffusive H2O gain following the entrainment of primitive macrocrysts into evolved and H2O-rich melts a few days before eruption. A strong positive correlation between H2O and Li in plagioclase-hosted inclusions suggests that diffusive Li gain may also have occurred. Extreme F enrichments in primitive plagioclase-hosted inclusions (F/Nd=51–216 versus ∼15 in matrix glasses) possibly reflect the entrapment of inclusions from high-Al/(Al+Si) melt pools formed by dissolution-crystallisation processes (as indicated by HFSE depletions in some inclusions), and into which F was concentrated by uphill diffusion since F is highly soluble in Al-rich melts. The high S/Dy of primitive inclusions (∼300) indicates that primary melts were S-rich in comparison with most oceanic basalts. Cl and B are unfractionated from similarly compatible trace elements, and preserve records of primary melt heterogeneity. Although primitive plagioclase-hosted melt inclusions from the 10ka Grímsvötn tephra series record few primary signals in their volatile contents, they nevertheless record information about crustal magma processing that is not captured in olivine-hosted melt inclusions suites.

Magma Storage and Extraction Associated with Plinian and Interplinian Activity at Santorini Caldera (Greece)

Santorini caldera has had a long history of plinian eruptions and caldera collapses, separated by20–40 kyr interplinian periods. We have carried out a study to constrain magma storage/extractiondepths beneath the caldera. We analysed H2O in 138 olivine-, pyroxene- and plagioclase-hostedmelt inclusions from plinian and interplinian products from the last 200 kyr, and CO2, S, Cl, F anddD in various subsets of these. The dataset includes 64 inclusions in products of the Minoan plinianeruption of the late 17th century BCE. All the melt inclusions were ellipsoidal and isolated, with notextural evidence for volatile leakage. Mafic melt inclusions contain 1–4 wt % H2O and up to1200ppm CO2, 1200ppm S, 2000ppm Cl and 400ppm F; silicic inclusions contain 2–7 wt % H2O, upto 150ppm CO2, up to 400ppm S, 2000–6000ppm Cl and 600–1000ppm F. The dD values of 27 representativeinclusions (–37 to –104%) are intermediate between mantle and slab values and ruleout significant H2O loss by hydrogen diffusion from olivine-hosted inclusions. H2O, S and Cl behavecompatibly in melt inclusion suites varying from mafic to silicic in composition, showing thatentrapment of many melt inclusions took place under volatile-saturated conditions. Most Santorinimelts are saturated in a free COHSCl vapour phase at depths of less than 10 km; the only exceptionsare basaltic melts from a single interplinian eruption, which were volatile-undersaturated upto K2O contents of 1 wt %. The rhyolitic melt of the Minoan eruption probably contained a freehypersaline liquid phase. H2OþCO2 saturation pressures were calculated using suitably calibratedsolubility models to estimate pre-eruptive magma storage depths. Magmas feeding plinian eruptionswere stored at >4km (>100 MPa) and extracted over depth intervals of several kilometres.Plagioclase phenocrysts in rhyodacitic pumice from the Minoan eruption have cores containingmelt inclusions trapped at depths up to 10–12km (320 MPa), and rims (also orthopyroxene andclinopyroxene) containing inclusions trapped at 4–6km (100–160 MPa). This records late-stage silicicreplenishment of a 4 km). Melt inclusions fromthe AD 726 eruption of Kameni Volcano yield a pre-eruptive storage depth of 4 km, which is similarto that estimated from geodetic data for the inflation source during the 2011–2012 period of caldera unrest; this supports a magmatic origin of the unrest. The level of pre-AD 726 magma storage beneathKameni was deeper than that of earlier silicic interplinian eruptions, perhaps owing tochanges in crustal stress caused by the Minoan eruption. Combined with previously published results,the melt inclusion data provide a time-integrated image of the crustal plumbing system.Mantle-derived basalts are injected into the lower crust, where they fractionate to produce evolvedmelts in bodies of hot crystal mush. Evolved residual melts separate from their parent mushes inthe 8 to >15km depth interval, then ascend rapidly into the upper crust, where they either crystallizeor accumulate as bodies of eruptible, crystal-poor magma.

Ca/Al of plagioclase-hosted melt inclusions as an indicator for post-entrapment processes at mid-ocean ridges?

The composition of melt inclusions in basalts erupted at mid-ocean ridges may be modified by post-entrapment processes, so the present composition of melt inclusions may not represent their original composition at the time of entrapment. By combining the melt inclusion composition in samples from the South Mid-Atlantic Ridge at 19oS analyzed in this study, and from the Petrological Database, we found that post-entrapment crystallization processes resulted in higher Ca/Al, Mg#[100°—atomic Mg 2+ /(Mg 2+ +Fe 2+ )], MgO and FeO contents, and lower CaO and Al 2 O 3 contents of plagioclase-hosted melt inclusions relative to those hosted in olivine. In addition, melt inclusions hosted in plagioclase with anorthite content larger than 80mol.% had been modified more readily than others. By discussing the relationships between Ca/Al and fractional crystallization, post-entrapment crystallization, and the original melt composition, we propose that Ca/Al can be regarded as an indicator of the effect of post-entrapment processes on melt inclusion composition. Specifically, i) when Ca/Al 1.0, the compositions of melt inclusions do not reflect the original melt composition nor preserve information about the mantle source. According to these criteria, plagioclase-hosted melt inclusions with Ca/Al>1.0 in basalts from the South Mid-Atlantic Ridge at 19oS cannot represent the composition of the melt at the moment of their entrapment.

The evolution and storage of primitive melts in the Eastern Volcanic Zone of Iceland: the 10 ka Grímsvötn tephra series (i.e. the Saksunarvatn ash)

Major, trace and volatile elements were measured in a suite of primitive macrocrysts and melt inclusions from the thickest layer of the 10 ka Grímsvötn tephra series (i.e. Saksunarvatn ash) at Lake Hvítárvatn in central Iceland. In the absence of primitive tholeiitic eruptions (MgO > 7 wt%) within the Eastern Volcanic Zone (EVZ) of Iceland, these crystal and inclusion compositions provide an important insight into magmatic processes in this volcanically productive region. Matrix glass compositions show strong similarities with glass compositions from the AD 1783–1784 Laki eruption, confirming the affinity of the tephra series with the Grímsvötn volcanic system. Macrocrysts can be divided into a primitive assemblage of zoned macrocryst cores (An78–An92, Mg#cpx = 82–87, Fo79.5–Fo87) and an evolved assemblage consisting of unzoned macrocrysts and the rims of zoned macrocrysts (An60–An68, Mg#cpx = 71–78, Fo70–Fo76). Although the evolved assemblage is close to being in equilibrium with the matrix glass, trace element disequilibrium between primitive and evolved assemblages indicates that they were derived from different distributions of mantle melt compositions. Juxtaposition of disequilibrium assemblages probably occurred during disaggregation of incompatible trace element-depleted mushes (mean La/Ybmelt = 2.1) into aphyric and incompatible trace element-enriched liquids (La/Ybmelt = 3.6) shortly before the growth of the evolved macrocryst assemblage. Post-entrapment modification of plagioclase-hosted melt inclusions has been minimal and high-Mg# inclusions record differentiation and mixing of compositionally variable mantle melts that are amongst the most primitive liquids known from the EVZ. Coupled high-field strength element (HFSE) depletion and incompatible trace element enrichment in a subset of primitive plagioclase-hosted melt inclusions can be accounted for by inclusion formation following plagioclase dissolution driven by interaction with plagioclase-undersaturated melts. Thermobarometric calculations indicate that final crystal–melt equilibration within the evolved assemblage occurred at ~1140 °C and 0.0–1.5 kbar. Considering the large volume of the erupted tephra and textural evidence for rapid crystallisation of the evolved assemblage, 0.0–1.5 kbar is considered unlikely to represent a pressure of long-term magma accumulation and storage. Multiple thermometers indicate that the primitive assemblage crystallised at high temperatures of 1240–1300 °C. Different barometers, however, return markedly different crystallisation depth estimates. Raw clinopyroxene–melt pressures of 5.5–7.5 kbar conflict with apparent melt inclusion entrapment pressures of 1.4 kbar. After applying a correction derived from published experimental data, clinopyroxene–melt equilibria return mid-crustal pressures of 4 ± 1.5 kbar, which are consistent with pressures estimated from the major element content of primitive melt inclusions. Long-term storage of primitive magmas in the mid-crust implies that low CO2 concentrations measured in primitive plagioclase-hosted inclusions (262–800 ppm) result from post-entrapment CO2 loss during transport through the shallow crust. In order to reconstruct basaltic plumbing system geometries from petrological data with greater confidence, mineral–melt equilibrium models require refinement at pressures of magma storage in Iceland. Further basalt phase equilibria experiments are thus needed within the crucial 1–7 kbar range.

Discussion on ‘Magma storage region processes of the Soufrière Hills Volcano, Montserrat’, Geological Society, London, Memoirs , 39, 361-381

The recent study by Devine & Rutherford (2014), ‘Magma storage region processes of the Soufriere Hills Volcano, Montserrat’, published as Chapter 19 of the Geological Society of London Memoir, Volume 39, focused primarily on updating the record of magmatic temperatures recorded by Fe–Ti oxides in the andesite erupted from Soufriere Hills Volcano. However, a key result of the paper was that the compositions of some plagioclase-hosted melt inclusions indicate mixing of mafic magma components into the host andesite and trapping of the mixed melts within phenocrysts. This interpretation is consistent with our earlier work on the chemistry of melt inclusions (Humphreys et al . 2010) and of microlite crystal populations (Humphreys et al . 2009 a , 2013). Humphreys et al . (2010) proposed this interpretation by showing that a subset of Soufriere Hills melt inclusions and matrix glasses has anomalous concentrations of K2O and/or TiO2, as have mafic inclusion matrix glasses (see Humphreys et al . 2010, fig. 2). However, in their Appendix, Devine & Rutherford (2014) suggest that enrichment in K2O can result only from decompression crystallization and not from magma mingling. They also assert that our melt inclusion dataset is fundamentally flawed and subject to faulty corrections for post-entrapment correction; and that our melt inclusions were actually matrix glasses. Finally, Devine & Rutherford (2014) doubt ‘whether or not one can look at melt-inclusion analyses and distinguish the effects of decompression crystallization … from the effects of mingling … with components derived from the injected mafic magma’. Here we address the criticisms of the Devine & Rutherford (2014) Appendix, drawing on additional published data to support our arguments and explanations. We were unable to reproduce some of the key figures of Devine & Rutherford (2014) …

Prevalence of growth twins among anhedral plagioclase microlites

Crystal textures of volcanic rocks record the processes involved in magma storage and eruptive ascent. Syn-eruptive crystallization, in which groundmass crystals form and grow according to environmental factors such as thermodynamic undercooling, strongly influences the texture of erupted magma. This stage is difficult to isolate for study in natural rocks, but well-suited for laboratory experiments because the chemical compositions and crystallization timescales of eruptive processes can be emulated. This study examines the incipient stages of plagioclase crystallization in hydrous rhyodacite magma undergoing decompression-driven degassing. Experimental samples in which crystal growth at both near-equilibrium and far-from equilibrium conditions were examined using electron backscatter diffraction (EBSD) analysis to ascertain crystallographic lattice orientations of individual crystals. The crystal orientation investigation affirms a common assumption invoked in textural studies of crystal number density: contiguous crystals with parallel faces are crystallographically continuous, whereas contiguous crystals with non-parallel faces have unrelated crystal lattice orientations and as such, represent separate crystals. In the highly undercooled sample, twinning is identified in ~87% of the crystals examined; in the near-equilibrium sample, 38% of the crystals are twinned. We find that the observed twinning is unlikely to be the result of deformation, transformation, or synneusis, but rather a result of growth defects introduced during the incipient stages of crystallization. We suggest internal structural defects (twins) control macroscopic morphological defects (embayments, swallowtails, and melt inclusions) as a result of the high energy of the twin plane boundary. Formation of twins during the incipient stages of plagioclase crystallization is the single most important factor contributing to anhedral morphologies of feldspar microlites growing during magma decompression, and plays a role in the development of some plagioclase-hosted melt inclusions.

Magmatic storage conditions, decompression rate, and incipient caldera collapse of the 1902 eruption of Santa Maria Volcano, Guatemala

Phase equilibria experiments and analysis of natural pumice and phenocryst compositions indicate the 1902 Santa Maria dacite was stored at ~140–170MPa and 840–850°C prior to eruption. H2O-saturated, cold-seal experiments conducted in vessels with an intrinsic log fO2 of NNO+1±0.5 show that the natural phase assemblage (melt+plagioclase+amphibole+orthopyroxene+Fe–Ti oxides+apatite) is stable from approximately 115–140MPa at temperatures below ~825°C, to ~840–860°C at 150MPa, to >850 and <875°C at 200MPa. Natural plagioclase phenocrysts have rim compositions that range from An40–An45; this range of compositions defines a ~20°C band that intersects the stable phase assemblage at ~150MPa and 850°C. Plagioclase hosted melt inclusions were analyzed using FTIR and commonly contain <5wt.% H2O, which corresponds to a pressure of ~170MPa at 850°C, under pure H2O saturated conditions. Amphibole geothermobarometry (Ridolfi et al., 2010) applied to experimental samples suggest two populations of amphiboles, phenocrysts grown during the experiments and inherited xenocrysts, but the pressure–temperature conditions returned by the geothermobarometer are routinely >50MPa and >50°C greater than experimental run conditions; precise estimates of magmatic conditions based solely upon amphibole composition are likely inaccurate. The experimental results and analysis of natural crystals suggest that although the natural amphiboles likely record a broad range of magmatic conditions, only the lower bounds of that range reflect pre-eruptive storage conditions. Comparison of Santa Maria microlite abundances with decompression experiments examining other silicic systems from the literature suggests that the 1902 dacite decompressed at the rate of ~0.005 to 0.01MPa/s during the eruption. Applying the decompression rate with the previously described eruption rate of approximately 2–3×108kg/s (Williams and Self, 1983; Carey and Sparks, 1986) to the conduit model CONFLOW reveals that the eruption conduit was dike-like with an along-strike length >1km. Despite depositing ~20km3 of dacite tephra (equivalent to ~8.5km3 magma), the 1902 eruption did not form an obvious caldera. This work suggests that collapse of the dike-like conduit terminated the eruption, preventing full caldera collapse.

Petrological imaging of an active pluton beneath Cerro Uturuncu, Bolivia

Uturuncu is a dormant volcano in the Altiplano of SW Bolivia. A present day ~70 km diameter interferometric synthetic aperture radar (InSAR) anomaly roughly centred on Uturuncu’s edifice is believed to be a result of magma intrusion into an active crustal pluton. Past activity at the volcano, spanning 0.89 to 0.27 Ma, is exclusively effusive and almost all lavas and domes are dacitic with phenocrysts of plagioclase, orthopyroxene, biotite, ilmenite and Ti-magnetite plus or minus quartz, and microlites of plagioclase and orthopyroxene set in rhyolitic groundmass glass. Plagioclase-hosted melt inclusions (MI) are rhyolitic with major element compositions that are similar to groundmass glasses. H2O concentrations plotted versus incompatible elements for individual samples describe a trend typical of near-isobaric, volatile-saturated crystallisation. At 870 °C, the average magma temperature calculated from Fe–Ti oxides, the average H2O of 3.2 ± 0.7 wt% and CO2 typically 0.5 Ma compared to those <0.5 Ma. High microlite contents of domes indicate that effusion rates were probably slowest in dome-forming eruptions. Linear trends in WR major and trace element chemistries, highly variable, bimodal mineral compositions, and the presence of mafic enclaves in lavas demonstrate that intrusion of more mafic magmas into the evolving, shallow plutonic mush also occurred further amplifying local temperature fluctuations. Crystallisation and resorption of accessory phases, particularly ilmenite and apatite, can be detected in MI and groundmass glass trace element covariation trends, which are oblique to WRs. Marked variability of Ba, Sr and La in MI can be attributed to temperature-controlled, localised crystallisation of plagioclase, orthopyroxene and biotite within the evolving mush.

Integrated volcanologic and petrologic analysis of the 1650 AD eruption of Kolumbo submarine volcano, Greece

Kolumbo submarine volcano, located 7km northeast of Santorini, Greece in the Aegean Sea, last erupted in 1650AD. Submarine and subaerial explosive activity lasted for a period of about four months and led to the formation of thick (~250m) highly stratified pumice deposits on the upper crater walls as well as extensive pumice rafts that were dispersed throughout the southern Aegean Sea. Subaerial tephra fallout from eruption columns that breached the surface occurred as far east as Turkey.Remotely operated vehicle (ROV) video observations from transects of the northern and southwestern crater walls have been used to create stratigraphic sequences of the 1650AD deposits. Petrographic and geochemical analyses reveal that the bulk of the deposits consists of white, highly vesicular, sparsely porphyritic, biotite-bearing rhyolite pumice. Pre-eruption volatile contents determined from plagioclase-hosted melt inclusions yield a median value of 6.5wt.%. This corresponds to a pre-eruption storage pressure of ~200MPa, or a depth of ~6km assuming a H2O-saturated magma. Comparison of the natural glass compositions and mineral assemblage of the Kolumbo samples with experimental results on other rhyolite magmas of similar composition in the modified haplogranite system supports the pressure and total volatile estimates. Pre-eruption temperature was calculated as 750°C based on plagioclase–melt geothermometry.The high volatile content of the Kolumbo magma and historical accounts of substantial subaerial eruption plumes suggest that the explosive eruption was driven by primary volatile degassing. Three phases of activity can be inferred from historical accounts and deposit stratigraphy; an initial submarine eruption during which time the vent became shallower as pyroclastic material accumulated, a transitional phase with substantial fallout from a hybrid submarine/subaerial eruption plume, and a final subaqueous phase as the eruption waned. Depositional processes included a complex interplay of submarine fallout, pyroclastic density currents, and clast segregation associated with the transition from submarine to subaerial plumes, producing a highly stratified pumiceous sequence. This is the first study of the in situ pyroclastic deposits of a well-documented silicic submarine explosive eruption in the relatively shallow (<500m below sea level) marine environment.

Oxygen isotopes in melt inclusions and glasses from the Askja volcanic system, North Iceland

Primitive melt inclusions trapped during the earliest stages of fractional crystallisation are able to preserve oxygen isotope ratios inherited from mantle-derived melts. However, assimilation of low-δ18O hydrothermally altered crustal material and mixing with magmas held in shallow reservoirs may exert a strong control on the δ18O of melt inclusions trapped during later stages of crystallisation. Oxygen isotope ratios in olivine- and plagioclase-hosted melt inclusions and glasses from tephra samples collected from the Askja central volcano and Askja volcanic system indicate significant differences in the mechanisms of magma supply and storage between the northern and southern segments of the Askja volcanic system. Melt inclusions from the Holuhraun fissure eruption, ∼20km south of Askja, mostly preserve δ18O signatures of +4.1‰ to +5.4‰, suggesting that this magma underwent minimal modification by magma mixing or crustal assimilation prior to its eruption. By contrast, melt inclusions and glasses from the Nýjahraun fissure eruption, ∼60km north of Askja, have δ18O between +3.1‰ and +4.0‰. These relatively evolved melt inclusions (∼3.9–4.3wt.% MgO) were probably trapped during late-stage fractional crystallisation in a shallow magma storage zone. Melt inclusions from two phreatomagmatic tuff sequences within the Askja caldera have δ18O between +2.1‰ and +5.2‰, and this variability cannot be explained by mixing with low-δ18O rhyolitic or andesitic contaminants in the upper crust. Instead, mixing of the ascending magmas with hydrated, low-δ18O basaltic magmas is invoked, thus acquiring a low δ18O signature with minimal modification to the magma’s bulk composition. Such magma bodies are likely to be found throughout the upper 11km of the crust beneath Askja. Assimilation of low-δ18O meta-basalt in the upper crust is also likely to affect the δ18O of ascending magmas.