Contents Of Melt Inclusions Research Articles

Magma recharge in persistently active basaltic–andesite systems and its geohazards implications: the case of Villarrica volcano, Chile

We report whole-rock chemistry, mineral chemistry, and volatile content from Villarrica volcano’s major recent paroxysms and background activity. Composition of the volcanic products are basalt to basaltic andesite with whole-rock SiO2 content between 50 and 56 wt%, and a mineralogy dominated by olivine (Fo71-80), clinopyroxene (Mg# ~ 50) and plagioclase (An60–80). Volatile contents in melt inclusions are up to 1.5 wt% H2O, 500 ppm CO2, 1230 ppm sulfur and 580 ppm chlorine. Regardless of the type of activity, there are no substantial variations in whole-rock composition or the volatile content when the activity switches from background activity to a major paroxysm, strongly suggesting that this shift does not just depend on the arrival of new magma in the shallow magmatic system. Geothermobarometry constrains crystallization of the major mineral phases at various depths between 3 and 12.7 km, suggesting that degassing of a volatile-rich recharge magma occurs deeper than 12 km, producing efficient mixing throughout the whole system, and sustaining the lava lake activity in Villarrica’s summit crater. The occurrence of a permanent lava lake also suggests that the magma recharge must be close to continuous and therefore sudden changes between background and paroxysmal volcanic activity are likely controlled by relatively small changes in the rate of recharge and/or the volatile release rate in the magmatic system. This has important implications for the understanding of eruption triggers and the forecasting of volcanic eruptions.Graphical abstract

Constraining the volatile evolution of mafic melts at Mt. Somma–Vesuvius, Italy, based on the composition of reheated melt inclusions and their olivine hosts

Abstract. Mount Somma–Vesuvius is a stratovolcano that represents a geological hazard to the population of the city of Naples and surrounding towns in southern Italy. Historically, volcanic eruptions at Mt. Somma–Vesuvius (SV) include high-magnitude Plinian eruptions, such as the infamous 79 CE eruption that occurred after 295 years of quiescence and killed thousands of people in Pompeii and surrounding towns and villages. The last eruption at SV was in 1944 and showed a Volcanic Explosivity Index (VEI) of 3 (0.01 km3 of volcanic material erupted). Following the 1944 eruption, SV has been dormant for the past nearly 79 years, with only minor fumarolic and seismic activity. During its long history, centuries of dormancy at SV have ended with Plinian eruptions (VEI 6) that signal the beginning of a new cycle of eruptive activity. Thus, the current dormancy stage demands a need to better understand the mechanism involved in high-magnitude eruptions in order to better predict future eruption magnitude and style. Despite centuries of research on the SV volcanic system, many questions remain, including the evolution of magmatic volatiles from deep primitive magmas to shallower more evolved magmas. Developing a better understanding of the physical and chemical processes associated with volatile evolution at SV can provide insights into magma dynamics and the mechanisms that trigger highly explosive eruptions at SV. In this study, we present new data for the pre-eruptive volatile contents of magmas associated with four Plinian and two inter-Plinian eruptions at SV based on analyses of reheated melt inclusions (MIs) hosted in olivine. We correct the volatile contents of bubble-bearing MIs by taking into account the volatile contents of bubbles in the MIs. We recognize two groups of MIs: one group hosted in high-Fo olivine (Fo85–90) and relatively rich in volatiles and the other group hosted in low-Fo olivine (Fo70–69) and relatively depleted in volatiles. The correlation between volatile contents and compositions of host olivines suggests that magma fractionation took place under volatile-saturated conditions and that more differentiated magmas reside at shallower levels relative to less evolved/quasi-primitive magmas. Using the CO2 contents of corrected MIs hosted in Fo90 olivine from SV, we estimate that 347 to 686 t d−1 of magmatic CO2 exsolved from SV magmas during the last 3 centuries (38–75 Mt in total) of volcanic activity. Although this study is limited to only few SV magmas, we suggest that further study applying similar methods could shed light on the apparent lack of correlation between the volatile contents of MIs and the style and age of eruptions. Further, such studies could provide additional constraints on the origin of CO2 and the interaction between the carbonate platform and ascending magmas below SV.

Partitioning and isotopic fractionation of Li between mineral phases and alkaline to calc-alkaline melts of explosive and effusive eruptions

Evaluating the lithium (Li) budget of alkaline to calc-alkaline melts and their co-existing mineral phases is crucial to understanding which magmatic or post-eruptive processes can influence the Li contents and isotopic compositions of volcanic deposits. This is also important to be able to better constrain how economically relevant Li deposits form. The lithium contents of many different eruptions from different tectonic settings (intraplate, subduction zone, extensional environment) and cooling environments (quenched, quickly cooled and slowly cooled) were studied by combining new and literature data that were obtained using LA-ICP-MS. We found that the Li concentrations in melt inclusions and re-entrants are independent of tectonic setting (hotspot, subduction zone and rifting) and rarely exceed 100 ppm. When melt inclusions are re-homogenised the Li concentration increases which can lead to the overestimation of the Li budget of a magma.The highest Li concentrations in the alkaline suite of samples were found in groundmass glass (reaching up to 318 ppm). The Li contents in the co-existing mineral phases decrease in the following order: biotite (0.8–99.3 ppm), plagioclase (0.9–25.1 ppm), clinopyroxene (0.6–31.0 ppm) and sanidine (0.5–38.7 ppm). The highest Li concentrations in the calc-alkaline suite of samples were found in the biotite crystals (10.2–2314 ppm) followed by the groundmass glass (2.1–176 ppm). The Li contents are lower in the other mineral phases, from plagioclase (1.4–35.0 ppm), quartz (3.2–21.1 ppm), olivine (1.0–16.3 ppm), sanidine (0.3–8.1 ppm) to clinopyroxene (0.7–21.8 ppm). The Li concentrations in groundmass glass and co-existing crystal phases are independent of the tectonic setting but appear to be more closely related to their cooling environment. To evaluate the influence of the different cooling environments the Li isotopic compositions (per mil deviation of 7Li/6Li ratio relative to L-SVEC reference material) of different eruptions were also determined. The Li isotopic compositions of groundmass glass and mineral phases reveal a range in δ7Li composition spanning up to 6 ‰ for alkaline samples and >10‰ between bulk sample and mineral phases for calc-alkaline samples. The cooling environment of the samples has a large impact on the Li isotopic compositions of the mineral phases with only quartz seemingly unaffected. Rapidly quenched samples may retain their magmatic δ7Li compositions or will only experience small amounts of diffusional overprinting (e.g., by degassing) whereas slowly cooled samples will experience a prolonged diffusional exchange between minerals and matrix overprinting their magmatic δ7Li compositions.The Li inventory in quartz crystals may be influenced by the tectonic setting and water concentration in the crystals. To further assess the influence of tectonic settings on Li behaviour we used a large data set which is based on the 13 alkaline to calc-alkaline eruptions from different tectonic settings and calculated the apparent partition coefficients (Kds) between groundmass glass and co-existing mineral phases. Biotite crystals (alkaline systems) return the highest Kds (subduction-extensional setting: 0.01–0.57, oceanic hotspot: 0.17–2.65) whereas sanidine crystals yield the lowest Kds. Sanidines from the subduction-extensional setting (alkaline trend) have the lowest Kds (0.01–0.07; one outlier ranging from 0.02 to 0.24), followed by the oceanic hotspot setting (alkaline trend, 0.02–0.09) and the extensional setting (calc-alkaline trend; 0.01–0.15). Plagioclase crystals from the subduction-extensional setting (alkaline trend) yield Kds between 0.01 and 0.29, crystals from subduction zone settings (0.06–0.81) and extensional settings (0.09–0.50) have larger ranges in their Kds. The clinopyroxene crystals from the subduction zone settings (calc-alkaline trend) have the largest range in Kds (0.03–0.99), followed by the oceanic hotspot setting (alkaline trend, 0.06–0.48) and subduction-extensional setting (alkaline trend, 0.01–0.22). Quartz crystals (only from the calc-alkaline trend) from subduction zone settings have lower Kds (0.09–0.13) compared to the extensional settings (Bishop Tuff: 0.14–0.28; Caetano Tuff: 0.73–0.94). The wide variety of processes that can affect final Li abundances in melt and crystals (degassing, scavenging into fluids, intra-crystal diffusion, hydration of glasses) makes estimating the partition coefficients particularly difficult and results in the wide ranges observed here. Targeted experimental work would thus help to better constrain the behaviour of Li in alkaline compositions.

Degassing from magma reservoir to eruption in silicic systems: The Li elemental and isotopic record from rhyolitic melt inclusions and host quartz in a Yellowstone rhyolite

Lithium and hydrogen are volatile elements which diffuse rapidly in crystals and melt, making them powerful geochemical tools to reconstruct geological processes that take place on short time scales, such as syn- and post-eruptive degassing. Although the dynamics of hydrogen are fairly well understood to better constrain such processes, the assessment of Li behaviour within the magma reservoir relevant for ascent-related degassing still lacks detailed evaluation. Here, the first in situ Li concentrations and isotopic compositions (using SIMS analysis) of rhyolitic quartz-hosted, naturally glassy and crystallised melt inclusions (MIs) and groundmass glass (Mesa Falls Tuff, Yellowstone) are used to reconstruct Li elemental and isotopic evolution in the magma reservoir. Lithium concentrations in quartz-hosted glassy MIs (10–61 ppm) from a fallout deposit overlap with their groundmass glass (32–46 ppm) and their host quartz (8–15 ppm). Crystallised MIs from a later erupted flow pumice clast sample have higher Li concentrations (8–190 ppm) compared to the groundmass glass (32–51 ppm) and their host quartz (15–24 ppm). Lithium content in quartz from the early erupted sample is relatively homogenous, whereas it is up to a factor of two higher and heterogeneous in the later erupted sample, with a simultaneous increase in Li versus a decrease in H towards crystal rims. The δ7Li difference (expressed as Δ7LiMI–glass) between MIs (−8.0‰ to + 12.3‰) and groundmass glass (+9.0‰ to + 20.5‰) of two pyroclastic deposits reaches up to 29‰. Glassy MIs are internally heterogeneous in δ7Li and Li abundance. The cores of the glassy MIs record the δ7Li of the least modified melt during entrapment and the data distribution can be modelled by equilibrium fractionation between the melt and vapour phase during early open-system degassing in the magma reservoir. Late degassing during eruption triggers Li–H diffusional exchange between quartz and melt, as the degassing of H2O and the accompanying pressure change trigger H diffusion out of the host quartz and the MIs, which is charge balanced by inward Li diffusion. This results in the modification of Li contents in quartz and δ7Li values in the rims of the glassy MIs. Crystallised MIs reflect the loss of H2O from the MIs and the resulting enrichment of Li during the crystallisation. Additionally, the variations of δ7Li in the groundmass glass can be explained through modelling by kinetic fractionation between the melt and vapour during late stage open-system degassing linked with magma ascent.

High CO2 content in magmas of the explosive andesitic Enco eruption of Mocho-Choshuenco volcano (Chile)

Mocho-Choshuenco volcano has produced several highly explosive eruptions during its history, which make it one of the most hazardous volcanoes in the southern volcanic zone of Chile, although it is still relatively little studied to date. We present a geochemical study of the products of the sub-Plinian, andesitic, Enco eruption that occurred about 1600 years ago. We determined the major and trace elements compositions, as well as the volatile (H2O, CO2, Cl, and S) contents of melt inclusions trapped in minerals (olivine, plagioclase, and pyroxene) using electron microprobe, ion microprobe (SIMS), and 3D confocal Raman mapping. Though the whole-rock composition of the Enco magma is andesitic (60.2 ± 1.1 wt.% SiO2), the melt inclusions have SiO2 contents ranging from 50.3 to 67.3 wt.%, following the magmatic series of Mocho-Choshuenco, and the compositions of the most mafic melt inclusions are close to those of the most mafic erupted magmas. Geochemical modeling indicates that mixing occurred between a mafic magma and an andesitic-to-dacitic magma. Glass analysis revealed typical parental arc magma values for H2O (2.6–3.8 wt.%), S (116–1936 ppm), and Cl (620–1439 ppm). However, CO2 contents are very high in some melt inclusions with concentrations above 4000 ppm (measured in the glass), suggesting trapping depths > ~ 17–22 km. Presence of solid carbonates inside inclusion-hosted bubbles clearly indicates that the CO2 contents measured in the glass phase were minimum values. We conclude that a CO2-rich basaltic magma ascended and mixed with a shallower andesitic magma. The magma cooled and exsolved high amounts of CO2, which may have dramatically increased the pressure and triggered the highly explosive Enco eruption.

Magma Mixing During Conduit Flow is Reflected in Melt‐Inclusion Data From Persistently Degassing Volcanoes

AbstractPersistent volcanic activity is thought to be linked to degassing, but volatile transport at depth cannot be observed directly. Instead, we rely on indirect constraints, such as CO2‐H2O concentrations in melt inclusions trapped at different depth, but this data is rarely straight‐forward to interpret. In this study, we integrate a multiscale conduit‐flow model for non‐eruptive conditions and a volatile‐concentration model to compute synthetic profiles of volatile concentrations for different flow conditions and CO2 fluxing. We find that actively segregating bubbles in the flow enhance the mixing of volatile‐poor and volatile‐rich magma in vertical conduit segments, even if the radius of these bubbles is several orders of magnitude smaller than the width of the conduit. This finding suggests that magma mixing is common in volcanic systems when magma viscosities are low enough to allow for bubble segregation as born out by our comparison with melt‐inclusion data: Our simulations show that even a small degree of mixing leads to volatile concentration profiles that are much more comparable to observations than either open‐ or closed‐system degassing trends for both Stromboli and Mount Erebus. Our results also show that two of the main processes affecting observed volatile concentrations, magma mixing and CO2 fluxing, leave distinct observational signatures, suggesting that tracking them jointly could help better constrain changes in conduit flow. We argue that disaggregating melt‐inclusion data based on the eruptive behavior at the time could advance our understanding of how conduit flow changes with eruptive regimes.

High Sulfur in Primitive Arc Magmas, Its Origin and Implications

Sulfur contents in 98.5% of melt inclusions (MI) from calc-alkaline subduction basalts do not exceed 4000 ppm, whereas experimentally established limits of sulfur solubility in basaltic melts with high fO2 (characteristic of subduction zones, e.g., QFM + 2) surpass 14,000 ppm. Here we show that primitive (Mg# 62-64) subduction melts may contain high sulfur, approaching the experimental limit of sulfur solubility. Up to 11,700 ppm S was measured in olivine-hosted MI from primitive arc basalt from the 1941 eruption of the Tolbachik volcano, Kamchatka. These MI often contain magmatic sulfide globules (occasionally enriched in Cu, Ni, and platinum-group elements) and anhydrite enclosed within a brown, oxidized glass. We conclude that the ubiquitous low sulfur contents in MI may originate either from insufficient availability of sulfur in the magma generation zone or early magma degassing prior to inclusion entrapment. Our findings extend the measured range of sulfur concentrations in primitive calc-alkaline basaltic melts and demonstrate that no fundamental limit of 4000 ppm S exists for relatively oxidized subduction basalts, where the maximum sulfur content may approach the solubility limit determined by crystallization of magmatic anhydrite.

Efficient release of bromine by super-eruptions

AbstractBromine is a key halogen element in the quantification of volcanic volatiles, but analytical difficulties in measuring its very low abundances have prevented progress in understanding its behavior and its role in volcanic emissions. We present a new data set of bromine, chlorine, and fluorine concentrations in melt inclusions and matrix glasses for two rhyolitic super-eruptions from the Toledo and Valles calderas, New Mexico, USA. We show that before eruption, Br and Cl were efficiently partitioned from the gas-saturated magma into a separate fluid phase, and we calculate the mass of halogens in the fluid phase. We further demonstrate that syn-eruptive magma degassing was negligible during the super-eruptions, so that the main source of halogen emissions must have been the fluid phase. If the fluid phase were erupted, the large mass of Br and Cl could have severely impacted the atmospheric chemistry upon eruption.

Simultaneous In Situ Determination of Pb Isotope Ratios and Trace Element Concentrations in Melt Inclusions by LASS‐ICP‐MS

AbstractMelt inclusions provide unique insights into magmatic sources and processes. This study presents a method for simultaneously measuring Pb isotopes and trace elements in melt inclusions using a 193 nm excimer laser ablation system coupled to a multiple‐collector inductively coupled plasma‐mass spectrometer (MC‐ICP‐MS) and a sector‐field ICP‐MS (SF‐ICP‐MS). The ablated aerosol was split with a Y‐shaped connector into two gas streams, which were simultaneously transported into two mass spectrometers. Lead isotopes were measured on the MC‐ICP‐MS, while trace element concentrations were determined on the SF‐ICP‐MS. Little analytical bias was observed when the aerosol is divided into variable proportions (5:5, 6:4, 7:3, 8:2–9:1) to the two instruments. The precision and accuracy of this method were evaluated by analyzing five geological reference glasses with two laser spot diameters (33, and 45 μm). All the 33 measured trace element concentrations are consistent with the reference values (±10%) and have precisions less than 15% (2RSD). The precisions of the 208Pb/206Pb, and 207Pb/206Pb ratios correlate with Pb concentrations. From BHVO‐2G (Pb = 1.6 ppm) to TB‐1G (Pb = 16.1 ppm), the precision on these ratios improve from about 0.4%–0.2% with a 33 μm laser spot diameter. The mean Pb isotope ratios for the five reference glasses are within 0.15% of the reference values. Using our new method, Pb isotope ratios and trace element contents of plagioclase‐hosted melt inclusions in a mid‐ocean ridge basalt from the Indian Ocean were measured.

Effects of sulfide composition and melt H2O on sulfur content at sulfide saturation in basaltic melts

Sulfur content at sulfide saturation (SCSS) is a critical parameter that controls the consumption of sulfide during partial melting as well as potential exsolution of sulfide during magma evolution. Natural magmatic systems involve H2O in silicate melt and NiS in FeS-dominated sulfide, but the combined effects of melt H2O and sulfide composition on SCSS have not been well constrained. We carried out 18 phase equilibrium experiments between anhydrous or hydrous basaltic melt and FeS-NiS sulfide at 1.0–2.5 GPa and 1523–1673 K in a piston cylinder apparatus. Experimental results demonstrate an increase of SCSS of approximately 104 ppm per wt% melt H2O and a proportional increase of SCSS with the mole fraction of FeS in sulfide (XFeS). In combination with literature data, we find that the effects of melt H2O and sulfide composition are independent and additive. A new model of SCSS for natural basaltic melt is developed:SCSS(ppm)=XFeS⋅exp13.88−9744T−328PT+104wH2O,where T is temperature in K, P is pressure in GPa, and wH2O is melt H2O content in wt%. This model not only fits our own data well but also reproduces literature data of SCSS in basaltic melt within 20% off. Application of this SCSS model to ridge magmatism shows that the observed S contents in MORBs corresponds to ~15% melting of mantle (after sulfide exhaustion). The higher S contents in some sulfide-saturated MORBs require a mantle source containing high-NiS sulfide (XNiS ≈ 0.4). Comparison of SCSS with S contents in melt inclusions from primary arc magmas suggests that sulfate is not required to be present for a majority of sub-arc mantle.

Source and behavior of pyroclastic density currents generated by Vulcanian-style explosions of Popocatépetl volcano (Mexico) on 22 January 2001

Since 1994 Popocatépetl volcano has extruded 85 lava domes, each of which has been destroyed by Vulcanian-style explosions. The paroxysmal Vulcanian explosion of 22 January 2001 has been unique by generating multiple pyroclastic density currents (PDCs). Scientists and Civil Protection authorities have used those PDCs to remap the high-risk hazard zone of the volcano, although little study has been carried out to understand the generation and evolution of those PDCs. Here, we examine 38 stratigraphic sections and analyze the granulometry, componentry, vesicularity, and petrology (glass, melt inclusions in plagioclase) of collected samples to interpret the origins of the Vulcanian explosions and the PDC behavior. The 22 January event generated two series of dense PDCs that detached from the base of the eruption column. The first set of PDCs traveled northwest across a glacier on the volcano, partially scouring and melting the ice that hours later generated a lahar. Two pulses of these PDCs were confined within a large ravine, surpassed the slope break, and traveled up to 3 km away from the vent. The deposits are matrix-supported throughout their extent. The second set of PDCs traveled to the northeast, mainly as unconfined currents that split into two lobes prior to the slope break. The farthest traveled current followed a western ravine and reached the forest 3.5 km from the summit. The eastern lobe buried a local hiking hut and reached 2.1 km. The deposits from these northeastern currents are massive and block-supported. All deposits (volume of 5.71 × 105 m3) have finger-like morphology with levees and frontal snouts. All PDCs flowed as a dense basal granular current with turbulent upper clouds at minimum speeds of 7–10 m/s.The components of the deposits are, in order of abundance, black scoria (53–62 wt% SiO2), banded scoria, pumice (63 wt% SiO2), dome lithics (60–61 wt% SiO2), accidental lithics, and free crystals. Juvenile pumice and scoria contain phenocrysts of plagioclase (An47±10), clinopyroxene (En48±3), orthopyroxene (En64±1), ilmenite, and titanomagnetite set in a rhyolitic glass (pumice clast; 74.39 wt% SiO2) or microlite-rich matrix (scoria clast). The dome lithics also contain these phenocrysts as well as rare olivine phenocrysts (Fe2+/(Fe2 + Mg) = 0.1). Dissolved H2O-CO2 contents in melt inclusions hosted by pumice-plagioclase phenocrysts suggest the felsic magma was stored at depths of 2–4 km. Thus, the 2001 event was triggered by the injection of mafic magmas into a small volume of felsic magma, probably forming dikes left from the previous eruptive episode. We propose that the Vulcanian-explosions tap a column of magma (~2 km depth) equivalent to the volume that contained the felsic dikes. This mechanism has been acting during the destruction of all summit domes of the ongoing eruption of the volcano.

Chalcophile elements track the fate of sulfur at Kīlauea Volcano, Hawai’i

Chalcophile element concentrations in melt inclusions and matrix glasses may be used to investigate low pressure degassing processes, as well as sulfide saturation during crustal fractionation, and mantle melting. Erupted products from Kīlauea Volcano, Hawaiʻi, record three stages of sulfide saturation (in the mantle, crust, and within lava lakes), separated by episodes of sulfide resorption (i.e., sulfide undersaturation) during ascent through the thick Hawaiian lithosphere, and during syn-eruptive degassing. The presence of residual sulfides in the mantle source throughout the melting interval accounts for the high S concentrations of Kīlauean primary melts (1387–1600 ppm). Residual sulfides retain chalcophile elements during melting, decoupling the variability of these elements in high MgO melts from that of lithophile elements. Decompression associated with magma ascent through the thick Hawaiian lithosphere drives an increase in the sulfide concentration at sulfide saturation (SCSS2−), resulting in shallow storage reservoirs (∼1–5 km depth) being supplied with sulfide-undersaturated melts. A drop in temperature, coupled with major element changes during the fractionation of olivine, causes the SCSS2− to decrease. Combined with an increase in melt S contents during fractionation, this initiates a second stage of sulfide saturation at relatively high MgO contents (∼12 wt%). Syn-eruptive degassing of S drives the resorption of sulfides in contact with the carrier liquid. The covariance structure of Cu, MgO and Ni contents in melt inclusions and matrix glasses indicates that the dissolution of sulfides effectively liberates sulfide-hosted Cu and Ni back into the melt, rather than the vapour phase. The contrasting behaviour of Cu, Ni, Se and S during sulfide resorption indicates that the chalcophile element signature of the Kīlauean plume is largely controlled by silicate melt-vapour partitioning, rather than sulfide-vapour partitioning. The participation of dense sulfide liquids in shallow degassing processes may result from their direct attachment to buoyant vapour bubbles, or olivine crystals which were remobilized prior to eruption. Sulfide resorption obscures the textural and chemical record of sulfide saturation in matrix glasses, but not in melt inclusions, which are isolated from this late-stage release of chalcophile elements. The partitioning of S between the dissolving sulfide, melt and the vapour phase accounts for approximately 20% of the total S release into the atmosphere.

High fluxes of deep volatiles from ocean island volcanoes: Insights from El Hierro, Canary Islands

Abstract Basaltic volcanism contributes significant fluxes of volatiles (CO2, H2O, S, F, Cl) to the Earth’s surface environment. Quantifying volatile fluxes requires initial melt volatile concentrations to be determined, which can be accessed through crystal-hosted melt inclusions. However, melt inclusions in volatile-rich mafic alkaline basalts, such as those erupted at ocean islands, often trap partially degassed melts, meaning that magmatic volatile fluxes from these tectonic settings are often significantly underestimated. We have measured major, trace element and volatile concentrations in melt inclusions from a series of young ( 20 ka) basanites from El Hierro, Canary Islands. Our melt inclusions show some of the highest CO2 (up to 3600 ppm) and S (up to 4290 ppm) concentrations measured in ocean island basalts to date, in agreement with data from the recent 2011–2012 eruption. Volatile enrichment is observed in melt inclusions with crystallisation-controlled major element compositions and highly variable trace element ratios such as La/Yb. We use volatile-trace element ratios to calculate original magmatic CO2 contents up to 4.2 wt%, which indicates at least 65% of the original CO2 was degassed prior to melt inclusion trapping. The trace element contents and ratios of El Hierro magmas are best reproduced by 1–8% partial melting of a garnet lherzolite mantle source. Our projected CO2 (200–680 ppm) and S (265–450 ppm) concentrations for the source are consistent with upper estimates for primitive mantle. However, El Hierro magmas have elevated F/Nd and F/Cl in comparison with melts from a primitive mantle, indicating that the mantle must also contain a component enriched in F and other volatiles, most probably recycled oceanic lithosphere. Our modelled original magmatic CO2 contents indicates that, per mass unit, volatile fluxes from El Hierro magmas are up to two orders of magnitude greater than from typical mid-ocean ridge basalts and 1.5–7 times greater than from recent Icelandic eruptions, indicating large variability in the primary volatile content of magmas formed in different geodynamic settings, or even within different ocean islands. Our results highlight the importance of characterising mantle heterogeneity in order to accurately constrain both short- and long-term magmatic volatile emissions and fluxes from ocean island volcanoes.

Magma plumbing and hybrid magma formation at an active back-arc basin volcano: North Su, eastern Manus basin

North Su is the central volcanic edifice of the SuSu Knolls hydrothermal district, located in an extensional transform zone of the eastern Manus back-arc basin. The submarine volcano hosts a vigorous hydrothermal system with black and white smokers, diffuse fluid discharge, and accumulation of native sulfur. Among its dacitic to andesitic lavas are remarkable hybrid rocks, containing highly magnesian olivine (Fo81–94) phenocrysts with thin orthopyroxene reaction rims in a dacitic matrix. We investigated these rocks in order to understand their formation, to constrain temperature and depth of magma storage, and to identify links between magmatic processes and hydrothermal activity. The compositions, zonations and reaction rims of phenocrysts suggest that the hybrid lavas formed by mixing when dacitic to rhyodacitic magma was invaded by Mg-rich andesite to basaltic andesite with boninite affinities. The resulting melts show strong enrichment of large ion lithophile elements, moderate enrichment of light rare earth elements and depletion of Nb, Ta and Ti relative to N-MORB, characterizing them as back-arc basin magmas. Compositional zonations at the outermost rims of olivine phenocrysts suggest a period of no more than a few days between the inferred mixing event and eruption. Mineral-melt thermometers indicate a temperature of ∼1010 °C for the hybrid magma. Mineral-melt barometers give a wide range of pressures with a mean of 19 ± 217 MPa, reflecting equilibration at shallow depths as well as limited barometer performance at low pressures. In addition, many clinopyroxene phenocrysts lack equilibrium between their strongly zoned rims and the host melt. The H2O (0.9–5.3 wt%) and CO2 (70–4730 ppm) contents of melt inclusions in clinopyroxene and plagioclase phenocrysts provide better constraints on the pressure range for pre-eruptive magma storage. The data suggest that dacite phenocryst crystallization and magma mixing occurred at 40–140 MPa pressure, about 1–5 km beneath the summit of North Su. The sulfur–H2O relationship in melt inclusions reflect a degassing trend that is consistent with formation of an aqueous-sulfurous magmatic fluid, inferred to produce highly acidic vent fluids at North Su (Seewald et al., 2015). The melt inclusion data also indicate significant fractionation between F, Cl and H2O after magma degassing, possibly reflecting formation of a high-salinity brine at depth and fluorine fixation in minerals formed in the subsurface. Our results suggest that the dynamic magmatic system, with magma mixing occurring shortly before ascent and eruption, causes rapid changes in composition and quantity of exsolved fluids. Due to the short distance to the seafloor, such fluid fluctuations are rapidly transferred to the hydrothermal systems. This may be the main cause for the observed temporal, spatial and compositional variations in hydrothermal activity at North Su, and possibly at other vent fields of the Manus basin.

Origin of volatiles emitted by Plinian mafic eruptions of the Chikurachki volcano, Kurile arc, Russia: Trace element, boron and sulphur isotope constraints

Chikurachki is a 1816-m high stratovolcano on Paramushir Island, Kurile arc, Russia, which has repeatedly produced highly explosive eruptions of mafic composition. The present work is aimed at constraining the origin of volatile components (CO2, H2O, F, S, and Cl), along with B and S isotopic compositions in a series of phenocryst-hosted melt inclusions and groundmass glasses from basaltic andesite pyroclasts of the 1853, 1986, and prehistoric Plinian eruptions of the volcano. The ranges of volatile concentrations in melt inclusions (47–1580μg/g CO2, 0.4–4.2wt% H2O, 399–633μg/g F, 619–3402μg/g S and 805–1240μg/g Cl) imply a sudden pressure release from ~460 through ~35MPa that corresponds to ~1.2–16-km-depth range of magma ascent upon decompression. We conclude that rapid ascent of the volatile-rich basaltic magmas from ~16-km initial depth accompanied by near-surface bubble nucleation and growth, and subsequent magma fragmentation appear to be a primary reason for the Plinian character of the Chikurachki eruptions. Significant negative correlations of S with K, Zr, Nb, Ba, La, Ce, Pr (R=−0.8 to −0.9), no clear relationships of S with H2O, CO2 and Cl, but strong positive correlations of S/K2O with H2O/K2O, Cl/K2O and F/K2O preclude magma degassing to be the only process affecting volatile concentrations dissolved in the melt. The δ34S values of the studied inclusion and groundmass glasses range from −1.6 to +12.3‰, decrease with decreasing S, show significant positive correlations with H2O/K2O, Cl/K2O and F/Zr, and negative correlations with a number of incompatible trace elements. Neither open- nor close-system magma degassing can account for the observed range of δ34S. The δ11B values of the melt inclusions range from −7.0 to +2.4‰ with 13–23μg/g B. The relationships of δ11B with B/K2O and B/Nb are inconsistent with magma contamination at shallow crustal depths. Linear character of 1/S vs. δ34S relationship suggests two-component mixing. The possible mixing end-members could be the magmas having similar major and trace element compositions, but strongly contrasting volatile contents and S isotopes. Based on the behaviour of fluid-mobile vs. fluid-immobile incompatible trace elements, we conclude that the subduction component likely represents a mixture of subduction sediment-derived melt with up to 60% of slab-derived fluid. Admixture of ~1–8% of the inferred subduction component to the depleted mantle wedge source is required to account for the compositional range of the Chikurachki melt inclusions, and ~0.4–10% to constrain the composition of Kurile arc mafic magmas.

O-50 富士火山宝永スコリアの斑晶メルト包有物のイオウ組成

O-50 富士火山宝永スコリアの斑晶メルト包有物のイオウ組成

Volatile (Cl, F and S) and major element constraints on subduction-related mantle metasomatism along the alkaline basaltic backarc, Payenia, Argentina

We present data on volatile (S, F and Cl) and major element contents in olivine-hosted melt inclusions (MIs) from alkaline basaltic tephras along the Quaternary Payenia backarc volcanic province (~34°S–38°S) of the Andean Southern Volcanic Zone (SVZ). The composition of Cr-spinel inclusions and host olivines in Payenia are also included to constrain any variations in oxygen fugacity. The variation of potassium, fluorine and chlorine in MIs in Payenia can be modelled by partial melting (1–10%) of a variously metasomatised mantle. The high chlorine contents in MIs (up to 3200 ppm) from Northern Payenia require addition of subduction-related fluids to a mantle wedge, whereas volatile signatures in the southern Payenia are consistent with derivation from an enriched OIB source. Cl and Cl/K ratios define positive correlations with host olivine fosterite content (Fo80-90) that cannot be explained by olivine fractionation, degassing and/or degree of mantle melting. Neither can the correlation between SiO2 and TiO2 in the MIs and host olivine Fo-content be explained by magmatic differentiation processes. Instead these correlations essentially require a south to north mantle source transition from a low Mg# pyroxenite (from recycled eclogite) to a high Mg# fluid metasomatised peridotite. The Cl/K and S/K ratios in Payenia MIs extend from enriched OIB-like signatures (south) to Andean SVZ arc like signatures (north). We show that the northward increase in S, Cl and S/K is coupled to a northward increase in melt oxidation states and thus in Fe3+/Fetot ratios in the magmas. The increase in oxidation state also correlates with an increase of Mn/Fe (olivine) ratios. We calculate that 25% of the apparent north–south pyroxenite–peridotite source variation in Payenia (based on olivine Mn/Fe ratios) can be explained by the south to north variation in melt oxidation states.

Trace element diffusion and incorporation in quartz during heating experiments

Heating of quartz crystals in order to study melt and high-temperature fluid inclusions is a common practice to constrain major physical and chemical parameters of magmatic and hydrothermal processes. Diffusion and modification of trace element content in quartz and its hosted melt inclusions have been investigated through step-heating experiments of both matrix-free quartz crystals and quartz crystals associated with sulfides and other minerals using a Linkam TS1500 stage. Magmatic and hydrothermal quartz were successively analyzed after each heating step for Cu, Al, and Ti using electron probe micro-analyzer. After the last heating step, quartz crystals and their hosted melt inclusions were analyzed by laser ablation inductively coupled plasma mass spectrometry and compared to unheated samples. Heated samples reveal modification of Cu, Li, Na, and B contents in quartz and modification of Cu, Li, Ag, and K concentrations in melt inclusions. Our results show that different mechanisms of Cu, Li, and Na incorporation occur in magmatic and hydrothermal quartz. Heated magmatic quartz records only small, up to a few ppm, enrichment in Cu and Na, mostly substituting for Li. By contrast, heated hydrothermal quartz shows enrichment up to several hundreds of ppm in Cu, Li, and Na, which substitute for originally present H. This study reveals that the composition of both quartz and its hosted melt inclusions may be significantly modified upon heating experiments, leading to erroneous quantification of elemental concentrations. In addition, each quartz crystal also becomes significantly enriched in Cu in the sub-surface layer during heating. We propose that sub-surface Cu enrichment is a direct indication of Cu diffusion in quartz externally sourced from both the surrounding sulfides as well as the copper pins belonging to the heating device. Our study shows that the chemical compositions of both heated quartz and its hosted inclusions must be interpreted with great caution to avoid misleading geological interpretations.

Origin and fate of sulfide liquids in hotspot volcanism (La Réunion): Pb isotope constraints from residual Fe–Cu oxides

Immiscible sulfide liquids in basaltic magmas play an important role in trace metal transport and the sulfur budget of volcanic eruptions. However, sulfides are transient phases, whose origin and fate are poorly constrained. We address these issues by analyzing sulfide destabilization products preserved in lavas from La Réunion Island. Iron oxide globules and coatings, typically 20–80μm in size, were found to occur in vesicles of differentiated lavas from Piton des Neiges, and recent pumice samples from Piton de la Fournaise. Field and mineralogical evidence indicates that the iron oxides are syn-eruptive phases not resulting from hydrothermal processes. Samples were first studied by Scanning Electron Microscopy. The globules were separated, whereas the smaller spherules and coatings were concentrated by magnetic sorting and acid leaching, and samples were processed through wet chemistry. The Fe oxide phases comprise 49–74wt.% Fe, 26–40wt.% O, and up to 6wt.% Cu, 811ppm Ni, 140ppm Bi, and 8.5ppm Pb. Compared to the host lava, Cu, Ni, and Bi are enriched by a factor of 101–103. Systematic Pb isotope disequilibrium (between 500ppm and 2.9% for 206Pb/204Pb) exists between Fe oxides and host rocks, with Fe oxides generally displaying less radiogenic ratios. Unradiogenic Pb is a typical signature of sulfide, which tends to concentrate Pb, but not its parent elements U and Th. Thus, both the chemical and isotopic compositions of the vesicle-hosted Fe oxides suggest that they are more or less direct products of the destabilization of immiscible sulfide liquids. Although Pb dominantly partitions into the gas phase during sulfide breakdown, the original Pb isotope signature of sulfide is preserved in the residual oxide. The composition estimated for the parent sulfides (206Pb/204Pb=18.20–18.77, 207Pb/204Pb=15.575, and 208Pb/204Pb=38.2–38.8) precludes a genetic link with the La Réunion plume, and suggests a lithospheric or crustal origin.It is estimated that magma ascent velocities at Piton de la Fournaise are high enough to counterbalance the settling velocities of millimeter-size sulfides. Despite their high density, sulfide liquids are thus transferred upward during eruptions and their destabilization contributes to SO2 emanations. Assimilation of foreign sulfides from the lithosphere can explain why SO2 emissions sometimes (e.g., during the April 2007 eruption) exceed those predicted from the S content of melt inclusions.

Prolonged ascent and episodic venting of discrete magma batches at the onset of the Huckleberry Ridge supereruption, Yellowstone

How exceedingly large, volcanic supereruptions begin provides crucial information on the storage, ascent and release of silica-rich magma in catastrophic events. Initial fall deposits of the 2.08 Ma, 2500 km3 Huckleberry Ridge eruption are multiply bedded and in several places contain reworked intervals, indicating time breaks in the opening phases of the eruption. A 2.5 m section of these fall deposits was sampled at nine levels below the earliest ignimbrite (member A) at Mount Everts (Mammoth, Wyoming). We analyzed major and trace elements and volatiles in quartz-hosted melt inclusions (MIs), reentrants (REs; unsealed melt inclusions) and associated obsidian pyroclasts (thick-walled shards) to establish quartz crystallization and storage depths and melt compositional groupings. Systematic relationships between Rb and other incompatible elements (U, Cl, B) indicate ∼55% fractional crystallization between the least and most evolved glass compositions. In contrast, H2O concentrations in MIs show scattered relationships with trace elements and are interpreted to reflect variable loss of H2O by diffusion through the quartz host during magma ascent. The wide H2O variations (1.0–4.7 wt.%) in MIs from individual fall horizons imply as much as ∼14 days of diffusive loss, reflecting highly variable and surprisingly slow decompression conditions. Water and CO2 gradients in reentrants, however, are consistent with final ascent times of <1 to 4 h (ascent rates of ∼0.3–1.5 m/s), similar to those represented by MIs that we infer to have experienced little to no diffusive H2O loss. The wide range of ascent rates for co-erupted crystals mirrors that of intermittent explosive activity at Mount St. Helens in summer 1980, and implies that the Huckleberry Ridge magma body was not strongly overpressured at eruption onset. Restored entrapment pressures and geochemical data for MIs provide evidence for six distinct populations of quartz that originally crystallized in geochemically distinct magma domains. The compositions of REs and obsidian pyroclasts, by comparison, show that by the onset of eruption, the quartz had been brought together into three discrete magma bodies, which we interpret to have been cupolas on the roof of the main magma body. These cupolas were erupted sequentially and episodically from separate vents to generate the fall deposits before escalating activity led to generation of voluminous pyroclastic flows, and this pattern of activity suggests that tectonic triggering may have destabilized multiple magma bodies. Supereruptions as large as the Huckleberry Ridge event may start hesitatingly if the parental magma bodies are not strongly overpressured, with small-scale episodic activity that is modulated by external controls that may leave no other geological evidence for their presence.