Search

The numerous volcanic centres in the Main Ethiopian Rift (MER) present significant but poorly understood hazards to local populations. The MER is also an important site to gain insights into tectonic processes as it captures the transition from continental rifting (to the south) to incipient seafloor spreading (to the north). Peralkaline magmas account for around 90% of the volcanic products found in the MER. Determining the conditions under which these magmas evolve is critical to understanding rift-related volcanism and its associated hazards. Corbetti Caldera has an extensive record of large-scale, predominantly aphyric, peralkaline rhyolite eruptions. However, little is known about the mafic magmas from which these highly differentiated melts have evolved. Here we present data from the only basaltic deposit found within the caldera, coupled with whole rock, glass and mineral analysis of the peralkaline products, to investigate magma storage conditions at Corbetti. We demonstrate that magma mixing played a role in the evolution of the basaltic magmas and use RhyoliteMELTS modelling to show Corbetti's peralkaline magmas likely evolved at pressures between 100 and 250 MPa, from a magma with an initial water content of 0.5–1 wt%, at or below the QFM buffer. Mineral hygrometry on the sparse crystal populations corroborates the RhyoliteMELTS modelling, suggesting that the basaltic magma had 0.1–1.2 ± 0.32 wt% H 2 O, and the peralkaline magmas an average of ∼5.5 ± 1.25 wt% H 2 O. These results also match melt inclusion data for Corbetti and other peralkaline systems. We also provide new 40Ar/39Ar ages for two eruptions, a pre-caldera rhyolitic lava flow (206.7 ± 0.9 ka) and a post-caldera peralkaline ignimbrite (160 ± 0.8 ka). These results add to our understanding of the history of Corbetti and the storage conditions of peralkaline magmas within a continental rift setting and highlight the hydrous nature of Corbetti's magmas and the role that H 2 O plays during explosive eruptions. • We present new data from the only known basaltic units within Corbetti Caldera. • We determine shallow magma storage conditions between 100 and 250 MPa at temperatures ∼1100 °C for the basaltic magma and ∼ 793 °C for the peralkaline magma. • Corbetti's magmas are hydrous with ∼5.5 wt% H 2 O. • We date a pre-caldera lava flow to ∼207 ka and a post-caldera ignimbrite to 160 ka [ABSTRACT FROM AUTHOR]

The caldera-forming Samalas eruption of the Samalas–Rinjani volcanic complex on Lombok, Indonesia, in 1257 CE ranks as one of the most explosive and sulphur-rich eruptions of the past thousand years. Along with other significant events (e.g. Tambora 1815, El Chichón 1982), Samalas 1257 forms a class of powerful eruptions of magmas of intermediate alkaline composition, which remain relatively understudied. This study aims to better understand the pre-eruptive magma reservoir conditions and volatile storage capacities of intermediate alkaline systems, using the 1257 Samalas eruption as a case study. Systematic partial-equilibrium experiments were run to constrain potential magma reservoir conditions (25–200 MPa and 850–1000°C) by examining changes in the chemistry and textures of experimental glass and mineral products. Natural trachydacite pumice from the 1257 Samalas eruption was used as the experimental charge for a series of water-saturated experiments run in cold-seal pressure vessels, buffered at an oxygen fugacity ƒ O2 of NNO/NNO + 1 log units. Assessing the match between natural and experimental glass compositions and mineral phases and textures reveals that pre-eruptive magma reservoir conditions were between 875°C and 930°C and ca. 100–150 MPa (4.5 ± 1km depth). Breakdown of experimental amphibole at pressures below 75 MPa, and plagioclase instability at 950°C/100 MPa, provide the strongest constraints on phase stabilities that are consistent with the magma storage region. In the observed temperature and pressure range, the natural compositions of plagioclase, amphibole and orthopyroxene are replicated in experimentally precipitated rims. Clinopyroxene and high anorthite plagioclase (An>64) in the natural mineral assemblage could not be replicated in the experiments, implying that these compositions most likely did not grow in the shallow pre-eruptive reservoir of Mt Samalas. The pre-eruptive storage depths of trachydacite magma are significantly shallower than the postulated magma storage at the currently active basaltic-andesite magmatic system at Rinjani, and thus monitoring magma storage depth may be a useful indicator of changing magma composition in the future. Our findings show that highly explosive VEI 7 eruptions of intermediate alkaline magma can be fed from a relatively limited range of storage pressures (100–150 MPa), suggesting accumulation of magma in one place, rather than the rapid extraction of magma from a vertically extensive, transcrustal magma system. [ABSTRACT FROM AUTHOR]

For many magmatic systems, crystal compositions preserve a complex and protracted history, which may be largely decoupled from their carrier melts. The crystal cargo may hold clues to the physical distribution of melt and crystals in a magma reservoir and how magmas are assembled prior to eruptions. Here we present a geochemical study of a suite of samples from three peralkaline volcanoes in the Main Ethiopian Rift. While whole‐rock data show strong fractional crystallization signatures, the trace element systematics of feldspars, and their relationship to their host glasses, reveals complexity. Alkali feldspars, particularly those erupted during caldera‐forming episodes, have variable Ba concentrations, extending to high values that are not in equilibrium with the carrier liquids. Some of the feldspars are antecrysts, which we suggest are scavenged from a crystal‐rich mush. The antecrysts crystallized from a Ba‐enriched (more primitive) melt, before later entrainment into a Ba‐depleted residual liquid. Crystal‐melt segregation can occur on fast timescales in these magma reservoirs, owing to the low‐viscosity nature of peralkaline liquids. The separation of enough residual melt to feed a crystal‐poor postcaldera rhyolitic eruption may take as little as months to tens of years (much shorter than typical repose periods of 300–400 years). Our observations are consistent with these magmatic systems spending significant portions of their life cycle dominated by crystalline mushes containing ephemeral, small (< 1 km3) segregations of melt. This interpretation helps to reconcile observations of high crustal electrical resistivity beneath Aluto, despite seismicity and ground deformation consistent with a magma body. Key Points: Alkali feldspars have variable trace element (Ba) concentrations, some far from equilibrium with their carrier liquids (melts)Some feldspars are antecrysts (in equilibrium with a liquid on the line of descent, but not the host melt), picked up from a crystal mushIgnimbrites erupted during caldera‐forming events are dominated by antecrysts, perhaps scavenged from deep parts of the magma reservoir [ABSTRACT FROM AUTHOR]

Timescales of magma chamber assembly and recharge are investigated here by applying 1D and 2D diffusion modeling techniques to high-resolution maps of titanium in quartz from a large-volume ignimbrite eruption in the Taupo Volcanic Zone, New Zealand. We compare quartz zonation patterns and associated diffusion timescales from the ∼340 ka Whakamaru super-eruption (magma volume ∼1000 km3) with the Younger Toba Tuff super-eruption, 74 ka (2000 km3), Sumatra, and the smaller volume ∼50 ka Earthquake Flat eruption (10 km3), Okataina Caldera Complex, New Zealand. Two principal timescales are presented: that of chamber recharge and eruption triggering events, and that of magma generation (involving long-term assembly, stirring and reactivation). Synchrotron micro-X-ray fluorescence maps of core–rim quartz transects provide a high-resolution record of magma chamber conditions throughout quartz crystallization. Quartz crystals from the Whakamaru magma display complex zonation patterns indicating fluctuating pressure–temperature conditions throughout the crystallization history. Toba and Earthquake Flat, in contrast, display simple quartz-zoning patterns and record slightly longer periods of crystal residence in the chamber that fed the eruption. We apply Lattice Boltzmann 2D diffusion modeling to reconstruct the timescales of quartz crystal zonation, accounting for crystal boundary complexities. Quartz crystal orientation is also accounted for by using geometry constraints from the synchrotron data. Our calculations suggest that crystal-mush reactivation for the main Whakamaru magma reservoir occured over a period of the order of 103–104 years. Both the Earthquake Flat and Toba eruptions experienced a significant recharge event (causing a temperature and pressure change), which occurred within ∼100 years of eruption. In comparison, the complex Whakamaru quartz zoning patterns suggest that the magma body experienced numerous thermal and compositional fluctuations in the lead-up to eruption. The final magma recharge event, which most probably triggered the eruption, occurred within ∼10–60 years of the eruption. Even though the volume of these systems spans two orders of magnitude, there does not appear to be a relationship between magma volume and diffusion timescale, suggesting similar histories before eruption. [ABSTRACT FROM AUTHOR]

At Mount Etna, Italy, vigorous gas-rich eruptions in A.D. 2001, 2002, and 2003 were followed by gas-poor eruptions in 2004, 2006, and 2007. Analyses of volatile (CO2, H2O, S, Cl, F), semivolatile (Cu), and involatile (Nb, La) elements trapped in olivine-hosted melt inclusions from these latest eruptions reveal the effects of the sustained interaction between a percolating gas phase and the stored magma. Melt inclusion compositions indicate that magmas erupted from 2004 to 2007 were residual from the 2001-2003 eruptions, and show significant evolution in the volatile content of the melt. These melt inclusion observations, and variations in the C/S of volcanic gases, can be accounted for if melts reequilibrated with CO2-rich gases during storage and prior to entrapment as melt inclusions. Sustained gas percolation caused loss of water and enhancement of CO2 in the evolving melt and may strongly influence the behavior of Cu, which potentially partitions into the gas phase. Vapor-melt interactions during magma storage are important controls on magma evolution at persistently degassing volcanoes. [ABSTRACT FROM AUTHOR]

Abstract: The dacitic lava flows of Nea Kameni, Santorini, Greece, contain a variety of partially-crystalline mafic enclaves. Each enclave type is derived from the break-up of a layer of replenishing magma, emplaced in the dacite chamber prior to eruption. Crystallisation-induced vesiculation of this layer, driven by the temperature contrast between the two magmas, was sufficient to drive upwelling and overturn of the layer, resulting in the formation of mafic enclaves within the host. The replenishing magmas are preserved as enclaves due to the presence of a plagioclase crystal framework. Spatial distribution pattern (SDP) analysis and melting experiments demonstrate that the plagioclase crystallinity and crystal aspect ratios observed in the Kameni enclaves are sufficient to form a touching framework of crystals. Observations on the coherence of the framework during exsolution of volatiles and enclave transport suggest that the morphology of the dominant framework-forming phase places a significant control on the strength and viscosity of the crystal framework, which in turn determines the extent to which the replenishing magmas will fragment and mix with the dacite host. Frameworks composed of high-aspect ratio crystals are weak, and respond to rapid vesicle expansion in a brittle manner. The replenishing magma then disaggregates to become a source of crystals dispersed throughout the host magma. In contrast, low-aspect ratio crystals form strong crystal frameworks that are more resistant to rapid vesicle expansion, resulting in preservation of the replenishing magma as vesicular enclaves. [Copyright &y& Elsevier]

Post-caldera eruptions of Santorini, Greece, over the past 3000 years resulted in the formation of the Kameni Islands, which comprise a series of compositionally similar dacitic lava flows. Each lava flow has a distinct population of partially-crystalline mafic enclaves, which we propose were derived from the break-up of a layer of replenishing magma responsible for triggering the eruption. Five of the recent flows (erupted in 1570, 1939, 1940, 1941 and 1950) include enclaves of essentially identical andesitic bulk compositions, which formed by crystallization of originally aphyric melts prior to eruption. Detailed examination of angles subtended at the junctions between pairs of plagioclase grains demonstrates that enclaves from each flow have a characteristic textural signature, with distinct differences in the extent of quench-related modification of the original population of dihedral angles formed by impingement of growing grains. These variations suggest that the temperature difference between the host dacite and the replenishing andesite at the time of layer overturn and eruption differed between flows. The uniformity of major element compositions of both the replenishing magma and the host dacitic lava flows demonstrates that the critical parameter in determining the timing of layer-overturn is the pre-eruptive H2O content of both host dacite and replenishing magma. We suggest that the replenishing magma responsible for two of the three eruptions in the period 1939-1941 was significantly wetter than that responsible for the later 1950 eruption. The enclaves with the least amount of quench-related modification occur in the 1570 flow. We suggest that in this case the intruding magma was relatively dry. [ABSTRACT FROM AUTHOR]

Magma plays a vital role in the break-up of continental lithosphere. However, significant uncertainty remains about how magma-crust interactions and melt evolution vary during the development of a rift system. Ethiopia captures the transition from continental rifting to incipient sea-floor spreading and has witnessed the eruption of large volumes of silicic volcanic rocks across the region over ∼45 Ma. The petrogenesis of these silicic rocks sheds light on the role of magmatism in rift development, by providing information on crustal interactions, melt fluxes and magmatic differentiation. We report new trace element and Sr–Nd–O isotopic data for volcanic rocks, glasses and minerals along and across active segments of the Main Ethiopian (MER) and Afar Rifts. Most δ 18 O data for mineral and glass separates from these active rift zones fall within the bounds of modelled fractional crystallization trajectories from basaltic parent magmas (i.e., 5.5–6.5‰) with scant evidence for assimilation of Pan-African Precambrian crustal material ( δ 18 O of 7–18‰). Radiogenic isotopes ( ε Nd = 0.92 – 6.52 ; 87 Sr/ 86 Sr = 0.7037–0.7072) and incompatible trace element ratios (Rb/Nb <1.5) are consistent with δ 18 O data and emphasize limited interaction with Pan-African crust. However, there are important regional variations in melt evolution revealed by incompatible elements (e.g., Th and Zr) and peralkalinity (molar Na 2 O + K 2 O / Al 2 O 3 ). The most chemically-evolved peralkaline compositions are associated with the MER volcanoes (Aluto, Gedemsa and Kone) and an off-axis volcano of the Afar Rift (Badi). On-axis silicic volcanoes of the Afar Rift (e.g., Dabbahu) generate less-evolved melts. While at Erta Ale, the most mature rift setting, peralkaline magmas are rare. We find that melt evolution is enhanced in less mature continental rifts (where parental magmas are of transitional rather than tholeiitic composition) and regions of low magma flux (due to reduced mantle melt productivity or where crustal structure inhibits magma ascent). This has important implications for understanding the geotectonic settings that promote extreme melt evolution and, potentially, genesis of economically-valuable mineral deposits in ancient rift-settings. The limited isotopic evidence for assimilation of Pan-African crustal material in Ethiopia suggests that the pre-rift crust beneath the magmatic segments has been substantially modified by rift-related magmatism over the past ∼45 Ma; consistent with geophysical observations. We argue that considerable volumes of crystal cumulate are stored beneath silicic volcanic systems (>100 km 3 ), and estimate that crystal cumulates fill at least 16–30% of the volume generated by crustal extension under the axial volcanoes of the MER and Manda Hararo Rift Segment (MHRS) of Afar. At Erta Ale only ∼1% of the volume generated due to rift extension is filled by cumulates, supporting previous seismic evidence for a greater role of plate stretching in mature rifts at the onset of sea-floor spreading. We infer that ∼45 Ma of magmatism has left little fusible Pan-African material to be assimilated beneath the magmatic segments and the active segments are predominantly composed of magmatic cumulates with δ 18 O indistinguishable from mantle-derived melts. We predict that the δ 18 O of silicic magmas should converge to mantle values as the rift continues to evolve. Although current data are limited, a comparison with ∼30 Ma ignimbrites (with δ 18 O up to 8.9‰) supports this inference, evidencing greater crustal assimilation during initial stages of rifting and at times of heightened magmatic flux. [ABSTRACT FROM AUTHOR]

The Main Ethiopian Rift hosts a number of peralkaline volcanic centres, several of which show signs of recent unrest. Due to the low number of historical eruptions recorded in the region and lack of volcanic monitoring, conditions of magma storage in the Main Ethiopian Rift remain poorly constrained. Aluto is one of these restless volcanic centres and identifying magma storage conditions is vital for evaluating the significance of recent periods of unrest. Using Aluto as a case study, we explore magma storage conditions using Rhyolite-MELTS thermodynamic modelling software. We performed ~ 150 fractional crystallisation models using a primitive basalt as the starting composition, and for a range of pressures (50–300 MPa), initial H 2 O contents (0.5–3 wt%) and oxygen fugacities (QFM − 2–QFM + 1). Predicted liquid lines of descent from these models are compared with published whole-rock data and, together with new observations of natural phase assemblages and erupted mineral compositions, provide constraints on magma storage conditions. Using a statistical approach to compare empirical data and thermodynamic model outputs, we find that compositions of evolved peralkaline rhyolites from Aluto are best reproduced by protracted (90%) isobaric fractional crystallisation from a rift-related basaltic composition, without the need for significant crustal assimilation. The required extent of fractional crystallisation suggests that much of the magmatic system may exist as a highly crystalline mush with only a small lens of rhyolitic melt. The best agreement between models and natural samples is at low pressures (150 MPa), low initial H 2 O concentrations (0.5 wt%) and an oxygen fugacity near the QFM buffer. The depth of magma storage derived from these results (~ 5.6 ± 1 km) is consistent with the source depths modelled from measured ground deformation. Data from other peralkaline volcanic centres in the Main Ethiopian Rift (Boset and Gedemsa), and other locations globally (e.g. Pantelleria, Italy) suggest that these storage conditions are a common feature of many peralkaline volcanic centres. [ABSTRACT FROM AUTHOR]

Repetitive and violent Vulcanian explosion sequences are common hazards at many volcanoes. Statistical analyses of such sequences form the basis of forecasting models and reveal underlying explosive processes. However, no single statistical model describes interexplosion repose intervals in Vulcanian systems. Soufrière Hills, Montserrat, is best described by a log-logistic model, while Sakurajima, Japan, shows a transition from a log-logistic to a Weibull model as activity intensifies. Anak Krakatau, Indonesia, displays two failure modes on different time scales, both described by Weibull distributions. At the Kameni Islands, Santorini, Greece, model-fitting parameters vary between eruption cycles. Rates of magma rise and pressurization may be the most important controls in determining the statistical distribution of repose intervals in Vulcanian systems. [ABSTRACT FROM AUTHOR]

Magmas from continental arcs built on thick crust have elevated incompatible element abundances and "enriched" radiogenic isotope ratios compared to magmas erupted in island and continental arcs overlying thinner crust. The relative influence of the slab, mantle, and upper plate on this variability is heavily debated. The Andean Southern Volcanic Zone (SVZ; 33–46°S) is an ideal setting to investigate the production of enriched continental arc compositions, because both crustal thickness and magma chemistry vary coherently along strike. However, the scarcity of primitive magmas in the thick-crusted northern SVZ has hindered previous regional studies. To better address the origin of enriched continental compositions, we investigate the geochemistry (major and trace element abundances, 87Sr/86Sr and 143Nd/144Nd ratios) of new mafic samples from Don Casimiro and Maipo volcanoes within the Diamante-Maipo Caldera Complex of the northern SVZ. While evolved Diamante-Maipo samples show evidence for crustal assimilation, the trace element and isotopic enrichment of the most mafic samples cannot result from crustal processing, as no known regional or global basement lithologies are enriched in all of the necessary incompatible trace elements. Subduction erosion models similarly fail to account for the enriched isotopic and trace element signature of these samples. Instead, we suggest that the enrichment of northern SVZ magmas is derived from an enriched ambient mantle component (similar to EM1-type ocean island basalts), superimposed on a northward decline in melt extent. A substantial, but nearly uniform contribution of melts from subducting sediment and altered oceanic crust are required at all latitudes. The EM1-like enrichment may arise from recycling of metasomatized subcontinental lithospheric mantle (M-SCLM), as the isotopic trajectory of primitive rear-arc monogenetic cones trend towards the compositions of SCLM melts sampled across South America. Isotopic data from spatially distributed rear-arc centres demonstrate that the arc-parallel variations in the degree of EM1-type enrichment observed in arc-front samples are also present up to 600 km behind the trench in the rear-arc. Rear-arc trace element systematics require significant but variable quantities of slab melts to be transported to the mantle wedge at these large trench distances. Overall, we show that a unified model incorporating variable mantle enrichment, slab additions, and melt extents can account for along and across-arc trends within the SVZ. The recognition that mantle enrichment plays a key role in the production of enriched continental compositions in the SVZ has important implications for our understanding of the chemical evolution of the Earth. If ambient mantle enrichment is not taken into account, petrogenetic models of evolved lavas may overestimate the role of crustal assimilation, which, in turn, may lead models of continental crust growth to overestimate the amount of continental material that has been recycled back into the mantle. [ABSTRACT FROM AUTHOR]

Volcanogenic halogens, in particular bromine, potentially play an important role in the ozone depletion of the atmosphere. Understanding bromine behaviour in magmas is therefore crucial to properly evaluate the contribution of volcanic eruptions to atmospheric chemistry and their environmental impact. To date, bromine partitioning between silicate melts and the gas phase is very poorly constrained, with the only relevant experimental studies limited to investigation of synthetic melt with silicic compositions. In this study, fluid/melt partitioning experiments were performed using natural silicate glasses with mafic, intermediate and silicic compositions. For each composition, experiments were run with various Br contents in the initial fluid (H 2 O–NaBr), at T – P conditions representative of shallow magmatic reservoirs in volcanic arc contexts (100–200 MPa, 900–1200 °C). The resulting fluid/melt partition coefficients (D Br f/m ) are: 5.0 ± 0.3 at 1200 °C–100 MPa for the basalt, 9.1 ± 0.6 at 1060 °C–200 MPa for the andesite and 20.2 ± 1.2 at 900 °C–200 MPa for the rhyodacite. Our experiments show that D Br f/m increases with increasing SiO 2 content of the melt (as for chlorine) and suggest that it is also sensitive to melt temperature (increase of D Br f/m with decreasing temperature). We develop a simple model to predict the S–Cl–Br degassing behaviour in mafic systems, which accounts for the variability of S–Cl–Br compositions of volcanic gases from Etna and other mafic systems, and shows that coexisting magmatic gas and melt evolve from S-rich to Cl–Br enriched (relative to S) upon increasing degree of degassing. We also report first Br contents for melt inclusions from Etna, Stromboli, Merapi and Santorini eruptions and calculate the mass of bromine available in the magma reservoir prior to the eruptions under consideration. The discrepancy that we highlight between the mass of Br in the co-existing melt and fluid prior to the Merapi 2010 eruption (433 and 73 tons, respectively) and the lack of observed BrO (from space) hints at the need to investigate further Br speciation in ‘ash-rich’ volcanic plumes. Overall, our results suggest that the Br yield into the atmosphere of cold and silicic magmas will be much larger than that from hotter and more mafic magmas. [ABSTRACT FROM AUTHOR]

Cyclic behaviour over a range of timescales is a well-documented feature of many dome-forming volcanoes, but has not previously been identified in high resolution seismic data from Volcán de Colima (Mexico). Using daily seismic count datasets from Volcán de Colima and Soufrière Hills volcano (Montserrat), this study explores parallels in the long-term behaviour of seismicity at two long-lived systems. Datasets are examined using multiple techniques, including Fast-Fourier Transform, Detrended Fluctuation Analysis and Probabilistic Distribution Analysis, and the comparison of results from two systems reveals interesting parallels in sub-surface processes operating at both systems. Patterns of seismicity at both systems reveal complex but broadly similar long-term temporal patterns with cycles on the order of ~ 50- to ~ 200-days. These patterns are consistent with previously published spectral analyses of SO 2 flux time-series at Soufrière Hills volcano, and are attributed to variations in the movement of magma in each system. Detrended Fluctuation Analysis determined that both volcanic systems showed a systematic relationship between the number of seismic events and the relative ‘roughness’ of the time-series, and explosions at Volcán de Colima showed a 1.5–2 year cycle; neither observation has a clear explanatory mechanism. At Volcán de Colima, analysis of repose intervals between seismic events shows long-term behaviour that responds to changes in activity at the system. Similar patterns for both volcanic systems suggest a common process or processes driving the observed signal but it is not clear from these results alone what those processes may be. Further attempts to model conduit processes at each volcano must account for the similarities and differences in activity within each system. The identification of some commonalities in the patterns of behaviour during long-lived dome-forming eruptions at andesitic volcanoes provides a motivation for investigating further use of time-series analysis as a monitoring tool. [ABSTRACT FROM AUTHOR]

Abstract: Between January 2011 and April 2012, Santorini volcano (Greece) experienced a period of unrest characterised by the onset of detectable seismicity and caldera-wide uplift. This episode of inflation represented the first sizeable intrusion of magma beneath Santorini in the past 50 years. We employ a new approach using 222Rn– systematics to identify and quantify the source of diffuse degassing at Santorini during the period of renewed activity. Soil CO2 flux measurements were made across a network of sites on Nea Kameni between September 2010 and January 2012. Gas samples were collected in April and September 2011 for isotopic analysis of CO2 ( ), and radon detectors were deployed during September 2011 to measure (222Rn). Our results reveal a change in the pattern of degassing from the summit of the volcano (Nea Kameni) and suggest an increase in diffuse CO2 emissions between September 2010 and January 2012. High-CO2-flux soil gas samples have . Using this value and other evidence from the literature we conclude that these CO2 emissions from Santorini were a mixture between CO2 sourced from magma, and CO2 released by the thermal or metamorphic breakdown of crustal limestone. We suggest that this mixing of magmatic and crustal carbonate sources may account more broadly for the typical range of values of CO2 (from to ) in diffuse volcanic and fumarole gas emissions around the Mediterranean, without the need to invoke unusual mantle source compositions. At Santorini a mixing model involving magmatic CO2 (with of and elevated (222Rn)/CO2 ratios ) and CO2 released from decarbonation of crustal limestone (with (222Rn)/CO2 ∼ 30–300 Bq kg−1, and of ) can account for the and (222Rn)/CO2 characteristics of the ‘high flux’ gas source. This model suggests of the carbon in the high flux deep CO2 end member is of magmatic origin. This combination of and (222Rn) measurements has potential to quantify magmatic and crustal contributions to the diffuse outgassing of CO2 in volcanic areas, especially those where breakdown of crustal limestone is likely to contribute significantly to the CO2 flux. [Copyright &y& Elsevier]

Abstract: Voluminous rhyolitic eruptions from Toba, Indonesia, and Taupo Volcanic Zone (TVZ), New Zealand have dispersed volcanic ash over vast areas in the late Quaternary. The ∼74 ka Youngest Toba Tuff (YTT) eruption deposited ash over the Bay of Bengal and the Indian subcontinent to the west. The ∼340 ka Whakamaru eruption (TVZ) deposited the widespread Rangitawa Tephra, dominantly to the southeast (in addition to occurrences northwest of vent), extending across the landmass of New Zealand, and the South Pacific Ocean and Tasman Sea with distal terrestrial exposures on the Chatham Islands. These super-eruptions involved ∼2500 km3 and ∼1500 km3 of magma (dense-rock equivalent; DRE), respectively. Ultra-distal terrestrial exposures of YTT at two localities in India, Middle Son Valley, Madhya Pradesh, and Jurreru River Valley, Andhra Pradesh, at distances of >2000 km from the source caldera, show a basal ‘primary’ ashfall unit ∼4 cm thick, although deposits containing reworked ash are up to ∼3 m in total thickness. Exposures of Rangitawa Tephra on the Chatham Islands, >900 km from the source caldera, are ∼15–30 cm thick. At more proximal localities (∼200 km from source), Rangitawa Tephra is ∼55–70 cm thick and characterized by a crystal-rich basal layer and normal grading. Both distal tephra deposits are characterized by very-fine ash (with high PM10 fractions) and are crystal-poor. Glass chemistry, stratigraphy and grain-size data for these distal tephra deposits are presented with comparisons of their correlation, dispersal and preservation. Using field observations, ash transport and deposition were modeled for both eruptions using a semi-analytical model (HAZMAP), with assumptions concerning average wind direction and strength during eruption, column shape and vent size. Model outputs provide new insights into eruption dynamics and better estimates of eruption volumes associated with tephra fallout. Modeling based on observed YTT distal tephra thicknesses indicate a relatively low (<40 km high), very turbulent eruption column, consistent with deposition from a co-ignimbrite cloud extending over a broad region. Similarly, the Whakamaru eruption was modeled as producing a predominantly Plinian column (∼45 km high), with dispersal to the southeast by strong prevailing winds. Significant ash fallout of the main dispersal direction, to the northwest of source, cannot be replicated in this modeling. The widespread dispersal of large volumes of fine ash from both eruptions may have had global environmental consequences, acutely affecting areas up to thousands of kilometers from vent. [Copyright &y& Elsevier]

• Ambient mantle metasomatism and melting degree traced through Nb/Zr, Ce/B, and δ 11 B. • δ 11 B consistent with isotopically heavy slab component(s) and DMM-OIB mantle source. • Greater δ 11 B complexity in melt inclusion suites reveals melt mixing processes. • Combined melt inclusions and whole-rocks record assembly of diverse melt batches. • Contributions from shallow crustal repositories of B remain poorly constrained. The chemistry of primitive arc rocks provides a window into compositional variability in the mantle wedge, as well as slab-derived inputs to subduction-related magmatism. However, in the long-term cycling of elements between Earth's internal and external reservoirs, a key unknown is the importance of retaining mobile elements within the subduction system, through subduction-related metasomatism of the mantle. To address these questions, we have analysed olivine-hosted melt inclusions and corresponding bulk rocks from the Kamchatka arc. Suites of melt inclusions record evidence for entrapment along melt mixing arrays during assembly of diverse parental magma compositions. Systematic variations in parental magma B/Zr, Nb/Zr, Ce/B, and δ 11 B are also apparent among the different eruptive centres studied. These element ratios constrain the nature of subduction-related metasomatism and provide evidence for ambient mantle heterogeneity and variable degrees of mantle melting. High Nb/Zr and low B/Zr in back-arc rocks indicate smaller degree melts, lower slab-derived inputs, but relatively enriched mantle compositions. Similarly, small monogenetic eruptive centres located away from the main stratocones also tend to erupt magmas with relatively lower slab contribution and overall smaller melting degrees. Conversely, arc-front compositions reflect greater slab contributions and larger degree melts of a more depleted ambient mantle. Across-arc variations in δ 11 B (ranging from ca. − 6 ‰ in the rear-arc and Sredinny Ridge to + 7 ‰ in the Central Kamchatka Depression) are generally consistent with variable addition of an isotopically heavy slab-derived component to a depleted MORB mantle composition. However, individual volcanic centres (e.g. Bakening volcano) show correlations between melt inclusion δ 11 B and other geochemical indicators (e.g. Cl/K 2 O, Ce/B) that require mixing between isotopically distinct melt batches that have undergone different extents of crustal evolution and degassing processes. Our results show that while melt inclusion volatile inventories are largely overprinted during shallower melt storage and aggregation, incompatible trace element ratios and B isotope compositions more faithfully trace initial mantle compositions and subduction inputs. Furthermore, we suggest that the signals of compositional heterogeneity generated in the sub-arc mantle by protracted metasomatism during earlier phases of subduction can be preserved during later magma assembly and storage in the crust. [ABSTRACT FROM AUTHOR]