Deloule, Etienne

The late Pleistocene trachytic Campanian Ignimbrite (> 300 km3 DRE, ca. 39 ka) covers the Campanian Plain (Italy) around Naples. It is found behind ridges>1000m high at 80 km from the source in the Campi Flegrei caldera. This points towards very dilute currents, that together with the huge amount of discharged magmatic material, suggest a magma reservoir highly enriched in volatiles. The huge volume of magma that extruded during the Campanian Ignimbrite eruption differentiated and mixed at shallow depth (6–3 km), as shown by modelling of the petrologic and geochemical features of the erupted products together with melt inclusion-based studies of gas-melt saturation. With respect to compositionally similar but lower magnitude Phlegraean eruptions (e.g., Agnano-Monte Spina,>1 km3 DRE, ca. 4.6 ka), the large amount of volatiles discharged by Campanian Ignimbrite was likely due to fractional crystallization of the lowermost portion of its magmatic reservoir. Because of the long residence time of the Campanian Ignimbrite magmatic system within the crust, an efficient decoupling took place during the upward migration of volatile elements in response to a chromatographic fractionation between H2O and CO2. This caused early pulses of nearly pure CO2 (CO2-fluxing), followed by a long-lasting H2O enrichment which yielded dissolved H2O contents up to 6–7 wt%. An overpressurized CO2- dominated gas cap was consequently produced, uniformly distributed at the top of the magma chamber. The onset of the eruption tapped this cap and generated Plinian columns, causing depressurization and fast volume decrease that facilitated, or even drove, the caldera collapse. H2O-rich magma was discharged during the following phase, characterised by pyroclastic density currents. The uniform distribution of the high values of void fraction (> 70%) and the high degree of vesicle connectivity throughout the magma body testify to the huge abundance of volatiles. These percolate from the crystallizing basal layers and determine the volatile-melt fining process yielding an efficient separation between H2O and CO2. The gas saturation-based estimates of the tapped foamy magma are compatible with the extent of magma chamber roof collapse, the strong expansion revealed by textural data, and the transport and deposition mechanisms, reflecting depressurization and inflation of the volatile-rich magma within the collapsed and laterally confined caldera.

Measuring the low bromine abundances in Earth's materials remains an important challenge in order to constrain the geodynamical cycle of this element. Suitable standard materials are therefore required to establish reliable analytical methods to quantify Br abundances. In this study we characterise 21 Br-doped glasses synthesized from natural volcanic rocks of mafic to silicic compositions, in order to produce a new set of standards for Br analyses using various techniques. The nominal Br contents (amounts of Br loaded in the experimental samples) of 15 of 21 glasses were confirmed within 20% by instrumental neutron activation analysis (INAA). Using this newset of standards, we compare three micro-analytical approaches to measure Br contents in silicate glasses: synchrotron X-ray fluorescence (SR-XRF), laser ablation-inductively coupled plasma mass spectrometry (LA-ICP-MS), and secondary ion mass spectrometry (SIMS). With SR-XRF, the Br contents of the standard glasses were determined with the highest accuracy (b10% for Br ≥ 10 ppm; N25% for Br ≤ 5 ppm), and high precision (b10% for Br contents N10 ppm; 20–30% for Br ≤ 10 ppm). The detection limit was estimated to be b1 ppm Br. All those factors combined with a high spatial resolution (5 × 5 μm for the presented measurements), means that SR-XRF iswell suited to determine the lowBr abundance in natural volcanic glasses (crystal-hosted melt inclusions or matrix glasses of crystallized samples). At its current stage of development, the LA-ICP-MS method allows the measurement of hundreds to thousands ppm Br in silicate glasses with a precision and accuracy generally within 20%. The Br detection limit of this method has not been estimated but its low spatial resolution (90 μm) currently prevents its use to characterise natural volcanic glasses, however it is fully appropriate to analyse super liquidus or sparsely phyric, Br-rich experimental charges. Our study shows that SIMS appears to be a promising technique to measure the lowBr contents of natural volcanic glasses. Its spatial resolution is relatively good (~15 μm) and, similarly to SR-XRF, the detection limit is estimated to be ≤ 1 ppm. Using our new set of standards, the Br contents of two MPI-DING reference glasses containing ≤ 1.2 ppm of Br were reproduced with precision b5% and accuracy b20%. Moreover, SIMS presents the advantage of being a more accessible instrument than SR-XRF and data processing is more straightforward.

The 2007 caldera-forming eruption of Piton de la Fournaise (PdF) erupted the largest volume of magma (210 Mm3)recorded at this volcano in at least three centuries. Major and trace element and Sr^Nd isotope data for bulk-rocks, groundmasses and olivine phenocrysts have been combined with melt inclusion data (major, trace and volatile elements) to track magma evolution over the whole eruptive sequence. We show that each eruptive phase had a distinctive geochemical and petrological signature and that caldera collapse on 5 April was preceded by a marked shift in bulk magma composition and crystal content and size. Aphyric basalt erupted at the beginning of the sequence (February 2007) had relatively high Sr isotope ratio (87Sr/86Sr ¼ 0·70420^0·704180) and low Nd isotopic ratio(143Nd/144Nd ¼ 0·51285^0·51286). Olivine-basalts extruded on2^5 April just before caldera collapse are less enriched in radiogenic Sr (87Sr/86Sr ¼ 0·70412^0·70416), but characterized by the same Nd isotopic composition. This magma is interpreted as a new deep input, which pressurized the shallow PdF plumbing system and triggered the 2007 activity. Post-collapse oceanite lavas represent the main volume of magma extruded in 2007. Their bulk-rocks and groundmasses have 87Sr/86Sr (0·70418) intermediate between those of February and 5 April, and similar to those of the March 2007 and 2001^2006 lavas.We show that the Steady State Basalts (SSB) commonly erupted at PdF are hybrid melts, which result from multistep mixing between ‘alkaline’and ‘transitional’end-members. Our results lead us to propose a new model of the PdF plumbing system to reconcile the petrological, geochemical and geophysical observations: (1) the shallow portion (above sea level) of the PdF plumbing system hosts several small sills, in which magma experiences variable degrees of degassing, cooling and crystallization; (2) oceanite lavas result from the withdrawal of shallow harrisitic mushes stored at low pressures (548 MPa; 51800^2400 m depth) below both the volcano summit and its eastern flank; (3) water degassing plays a major role in fast magma crystallization at shallow depths. Multistep ascent and periodic extrusion of the shallow magmas is promoted by injections of deeper and hotter basaltic magma, containing up to 1·3 wt % H2O and 1630 ppm S. In 2007, the new deep input was the ultimate source of the large excess in sulfur degassing detected by satellites. Lateral draining and intrusion of magma below the eastern flank of the volcano are the cause of major volcano deformation, flank sliding and summit caldera collapse.

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 (H2O–NaBr), at T–Pconditions representative of shallow magmatic reservoirs in volcanic arc contexts (100–200MPa, 900–1200◦C). The resulting fluid/melt partition coefficients (DBrf/m) are: 5.0 ±0.3 at 1200◦C–100MPa for the basalt, 9.1 ±0.6 at 1060◦C–200MPa for the andesite and 20.2 ±1.2 at 900◦C–200MPa for the rhyodacite. Our experiments show that DBrf/mincreases with increasing SiO2content of the melt (as for chlorine) and suggest that it is also sensitive to melt temperature (increase of DBrf/mwith 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.