Search

Containing one pdf file with Supplemental Information (Fig. S1–S4 and list of references from Table 1) and 3 Excel spreadsheets (Table S1–S3).

International audience; Volcanic ash exposure can lead to significant health risks. Damage to the respiratory and pulmonary systems are the most evident toxic side effects although the causes of these symptoms remain unclear. Conversely, the effects on other organs remain largely under-explored, limiting our understanding of the long-term volcanic ash-related risk at the whole-body scale. The metallome i.e. metal concentrations and isotopic compositions within the body, is suspected to be affected by volcanic ash exposure, having thus the potential for capturing some specificities of ash toxicity. However, the means by and extent to which the metallome is affected at the entire body scale and how the consequent chemical and isotopic deregulations correlate with pathophysiological dysfunctions are currently poorly understood. Here, we adopt a transdisciplinary approach combining high precision chemical analyses (major and trace element concentrations) and Cusingle bondZn isotope measurements in seven organs and two biological fluids of isogenic mice (C57BL/6) exposed to eruption products from La Soufrière de Guadeloupe (Eastern Carribean), in tandem with biological parameters including physiological and morphological data. Based on principal component analysis, we show that after one month of exposure to volcanic ash deposits, the mice metallome; originally organ-specific and isotopically-typified, is highly disrupted as shown for example by heavy metal accumulation in testis (e.g., Fe, Zn) and Cu, Zn isotopic divergence in liver, intestine and blood. These metallomic variations are correlated with early testicular defects and might reflect the warning signs of premature (entero)hepatic impairments that may seriously affect fertility and favor the emergence of liver diseases after prolonged exposure. Monitoring the temporal evolution of the Cu and Zn isotope compositions seems to be a promising technique to identify the main biological processes and vital functions that are vulnerable to environmental volcanogenic pollutants although this will require further validation on human subjects.

BackgroundTungurahua volcano (Ecuador) intermittently emitted ash between 1999 and 2016, enduringly affecting the surrounding rural area and its population, but its health impact remains poorly documented.ObjectivesWe aim at assessing the respiratory health hazard posed by the 16-17 August 2006 most intense eruptive phase of Tungurahua.MethodsBased on detailed field surveys and grain size analyses, we mapped the spatial distribution of the health-relevant ash size fractions produced by the eruption in the area impacted by ash fallout. We used Scanning Electron Microscopy and Raman Spectroscopy to quantify the mineralogy, composition, surface texture and morphology of a respirable ash sample isolated by aerodynamic separation. The cytotoxicity and pro-inflammatory potential of this respirable ash towards lung tissues was assessedin-vitrousing A549 alveolar epithelial cells, by Electron Microscopy and biochemical assays (LDH assay, RT-qPCR, multiplex immunoassays).ResultsThe eruption produced a high amount of inhalable and respirable ash (12.0-0.04 kg/m2of sub-10 µm and 5.3-0.02 kg/m2of sub-4 µm ash deposited). Their abundance and proportion vary greatly across the deposit within the first 20 km from the volcano. The respirable ash is characteristic of an andesitic magma and no crystalline silica is detected. Morphological features and surface textures are complex and highly variable, with few fibres observed.In-vitroexperiments show that respirable volcanic ash are internalized by A549 cells and processed in the endosomal pathway, causing little cell damage, but some changes in cell morphology and membrane texture. The ash trigger a weak pro-inflammatory response.DiscussionThese data provide the first understanding of the respirable ash hazard near Tungurahua, and the extent to which it varies spatially in a fallout deposit. Given the long exposure duration of the surrounding population, the chronic effects of this inhalable, weakly bio-reactive ash on health could be further investigated.

International audience; Basaltic eruptions are commonly associated with lava emissions and relatively weak explosive activities, but they can sometimes produce strong explosive eruptive phases. In April and November 2005, two paroxysmal eruptive events occurred within the summit crater of Karthala basaltic shield volcano (Grande Comore Island, Comoros), which hosted a water lake before each of these events. Both 2005 ash plumes spread across the Comoros Archipelago and heavily impacted the whole Grande Comore Island. Associated deposits on the volcano summit are extremely fine-grained (up to 50 wt% of fine ash

Co-auteur étranger; International audience; Future occurrence of explosive eruptive activity at Cotopaxi and Guagua Pichincha volcanoes, Ecuador, is assessed probabilistically, utilizing expert elicitation. Eight eruption types were considered for each volcano. Type event probabilities were evaluated for the next eruption at each volcano and for at least one of each type within the next 100 years. For each type, we elicited relevant eruption source parameters (duration, average plume height, and total tephra mass). We investigated the robustness of these elicited evaluations by deriving probability uncertainties using three expert scoring methods. For Cotopaxi, we considered both rhyolitic and andesitic magmas. Elicitation findings indicate that the most probable next eruption type is an andesitic hydrovolcanic/ash-emission (~ 26–44% median probability), which has also the highest median probability of recurring over the next 100 years. However, for the next eruption at Cotopaxi, the average joint probabilities for sub-Plinian or Plinian type eruption is of order 30–40%—a significant chance of a violent explosive event. It is inferred that any Cotopaxi rhyolitic eruption could involve a longer duration and greater erupted mass than an andesitic event, likely producing a prolonged emergency. For Guagua Pichincha, future eruption types are expected to be andesitic/dacitic, and a vulcanian event is judged most probable for the next eruption (median probability ~40–55%); this type is expected to be most frequent over the next 100 years, too. However, there is a substantial probability (possibly >40% in average) that the next eruption could be sub-Plinian or Plinian, with all that implies for hazard levels.

International audience; Eruptive columnmodels are powerful tools for investigating the transport of volcanic gas and ash, reconstructing past explosive eruptions, and simulating future hazards. However, the evaluation of these models is challenging as it requires independent estimates of themainmodel inputs (e.g.mass eruption rate) and outputs (e.g. column height). There exists no database of independently estimated eruption source parameters (ESPs) that is extensive, standardized, maintained, and consensus-based. This paper introduces the Independent Volcanic Eruption Source Parameter Archive (IVESPA, ivespa.co.uk), a community effort endorsed by the International Association of Volcanology and Chemistry of the Earth’s Interior (IAVCEI) Commission on Tephra HazardModelling.We compiled data for 134 explosive eruptive events, spanning the 1902-2016 period, with independent estimates of: i) total erupted mass of fall deposits; ii) duration; iii) eruption column height; and iv) atmospheric conditions. Crucially, we distinguish plume top versus umbrella spreading height, and the height of ash versus sulphur dioxide injection. All parameter values provided have been vetted independently by at least two experts. Uncertainties are quantified systematically, including flags to describe the degree of interpretation of the literature required for each estimate. IVESPA also includes a range of additional parameters such as total grain size distribution,eruption style, morphology of the plume (weak versus strong), and mass contribution from pyroclastic density currents, where available. We discuss the future developments and potential applications of IVESPA and make recommendations for reporting ESPs to maximize their usability across different applications. IVESPA covers an unprecedented range of ESPs and can therefore be used to evaluate and develop eruptive column models across a wide range of conditions using a standardized dataset.

International audience; 25 It is commonly accepted that effusive activity emplaces the main emitted magmatic volume in basaltic 26 shield volcanoes. At Piton de La Fournaise (La Réunion Island, France), eruptive activity occurs mostly within the 27 non-populated Enclos Fouqué caldera and generally does not pose any risk to the population. However, historical 28 observations, recent monitoring data and field work on tephra deposits suggest that some eruptions have produced 29 unusual and unexpected explosive phases. A comprehensive sampling of tephra from major historical and recent 30 eruptions on this volcano allowed us to perform systematic componentry, grain size, chemical and morphological 31 analyses in order to characterize the eruptive dynamics involved in these explosive basaltic eruptions. This 32 integrative approach reveals highly variable characteristics of the studied tephra reflecting different fragmentation 33 efficiencies and multiple associated mechanisms. Primary ductile and partially brittle fragmentation of various 34 intensity of juvenile magma emitted during Hawaiian fountaining or mild Strombolian explosions were identified 35 as the most common fragmentation mechanisms, in particular during the 2014-2018 dominantly effusive eruptive 36 sequence. In parallel, we distinguished more efficient short-lived fragmentation events related to (i) plug 37 pressurization and brittle fragmentation enhanced by syn-eruptive crystallization, (ii) magma/water interactions, 38 (iii) rare phreatic explosions and (iv) secondary fragmentation producing fine ash during caldera collapse. We 39 conclude that textural, geochemical and morphological analysis make it possible to identify and characterize the 40 variability in eruptive processes, with the grain size and the componentry of the ash particles being probably the 41 most important parameters to quantify both the efficiency and the nature of the fragmentation. 42 43

Eruptive column models are crucial for managing volcanic crises, forecasting future events, and reconstructing past eruptions. Given their central role in volcanology and the large uncertainties weakening their predictions, the evaluation and improvement of these models is critical. Such evaluation is challenging as it requires independent estimates of the main model inputs (e.g. mass eruption rate) and outputs (e.g. column height). Despite recent efforts to extend datasets of independently estimated eruption source parameters (ESP) (e.g. Mastin 2014, Aubry et al. 2017), there is no standardized, maintained, and community-based ESP database devoted to the evaluation of eruptive column models.Here we present a new ESP database designed to respond to the needs of the plume modelling community, and which will also be valuable to observatories, field volcanologists, and volcanic ash advisory centers. We compiled data for over 130 eruptive events with independent estimates of: i) the mass eruption rate; ii) the height reached by the column; and iii) atmospheric conditions during the eruption. In contrast with previous ESP datasets, we distinguish estimates of column height that relate to different phases (ash and SO2) and parts of the column (plume top or umbrella). We additionally provide the total grain size distribution, uncertainties in eruption parameters, and multiple sources for atmospheric profiles for events where these parameters are available. The database also includes a wealth of additional information which will enable modelers to distinguish between different eruptions when evaluating or calibrating models. This includes the type of eruption, the morphology of the plume (weak/transitional/strong), and the occurrence and mass entrained within pyroclastic density currents.We will apply the new database to revisit empirical scaling relationships between the mass eruption rate and “plume height”. In particular, we will show how such relationships depend on the type of height (e.g. SO2 height vs. ash top height) and eruption (e.g. magmatic vs. phreatomagmatic) considered. We will also discuss the difficulties and limitations of compiling ESP estimates from the literature as well as characterizing fundamentally unsteady volcanic events by a single value for each ESP.

Co-auteur étranger; International audience; Numerical modelling of tephradispersaland depositionis essential forevaluation of volcanichazards. Manymodels consider reasonable physical approximations in order toreducecomputational times, but this may introduce a certain degree of uncertainty in the simulation outputs.The important step of uncertainty quantification is dealt in this paper with respect to a coupled version of a plume model (PLUME-MoM) and a tephra dispersal model (HYSPLIT). The performances of this model are evaluated through simulationsoffourpasteruptions of different magnitudesand styles from three Andean volcanoes, and the uncertainty is quantified by evaluating the differences between modeled and observed data of plume height (at different time steps above the vent)as well asmass loading and grain sizeat givenstratigraphic sections.Different meteorological datasets were alsotestedand hada sensible influence on the model outputs. Other results highlight that the modeltendsto underestimate plume heights whileoverestimatingmass loading values, especially for higher magnitude eruptions.Moreover, the advective part of HYSPLIT seems to work more efficiently than the diffusive part.Finally, though the coupledPLUME-MoM/HYSPLIT model generally is less efficient inreproducing deposit grain sizes, we propose itmay be used for hazard maps production for higher magnitude eruptions (sub-Plinian or Plinian)for what concern mass loading.

International audience; Interactions between clasts in pyroclastic density currents (PDCs) generate volcanic ash that can be dispersed to the atmosphere in co-PDC plumes, and due to its small size, is far-travelled. We designed a series of experiments to determine the effects of pyroclast vesicularity and crystal content on the efficiency and type of ash generated by abrasion. Two different pyroclastic materials were used: (1) basaltic-andesite pyroclasts from Fuego volcano (Guatemala) with ~ 26–46% vesicularity and high groundmass crystallinity and (2) tephri-phonolite Avellino pumice (Vesuvius, Italy) with ~ 55–75% vesicularity and low groundmass crystallinity.When milled, both clast types produced bimodal grain size distributions with fine ash modes between 4 and 5φ (32–63 μm). Although the vesicular Avellino pumice typically generated more ash than the denser Fuego pyroclasts, the ash-generating potential of a single pyroclast was independent of density, and instead governed by heterogeneous crystal and vesicle textures. One consequence of these heterogeneities was to cause the vesicular Avellino clasts to split in addition to abrading, which further enhanced abrasion efficiency. The matrix characteristics also affected ash shape and componentry, which will influence the elutriation and transport properties of ash in the atmosphere. The experimental abrasion successfully replicated some of the characteristics of natural co-PDC ash samples, as shown by similarities in the Adherence Factor, which measures the proportion of attached matrix on phenocrysts, of both the experimentally generated ash and natural co-PDC ash samples. Our results support previous studies, which have shown that abrasion is an effective mechanism for generating fine ash that is similar in size (~ 5φ; 30 μm) to that found in co-PDC deposits. We further show that both the abundance and nature (shape, density, components, size distribution) of those ash particles are strongly controlled by the matrix properties of the abraded pyroclasts.

Volcanic ash clouds are common, often unpredictable, phenomena generated during explosive eruptions. Mainly composed of very fine ash particles, they can be transported in the atmosphere at great distances from the source, having detrimental socio-economic impacts. However, proximal settling processes controlling the proportion (ε) of the very fine ash fraction distally transported in the atmosphere are still poorly understood. Yet, for the past two decades, some operational meteorological agencies have used a default value of ε = 5% as input for forecast models of atmospheric ash cloud concentration. Here we show from combined satellite and field data of sustained eruptions that ε actually varies by two orders of magnitude with respect to the mass eruption rate. Unexpectedly, we demonstrate that the most intense eruptions are in fact the least efficient (with ε = 0.1%) in transporting very fine ash through the atmosphere. This implies that the amount of very fine ash distally transported in the atmosphere is up to 50 times lower than previously anticipated. We explain this finding by the efficiency of collective particle settling in ash-rich clouds which enhance early and en masse fallout of very fine ash. This suggests that proximal sedimentation during powerful eruptions is more controlled by the concentration of ash than by the grain size. This has major consequences for decision-makers in charge of air traffic safety regulation, as well as for the understanding of proximal settling processes. Finally, we propose a new statistical model for predicting the source mass eruption rate with an unprecedentedly low level of uncertainty.

How and how much the mass of juvenile magma is split between vent-derived tephra, PDC deposits and lavas (i.e. mass partition) is related to eruption dynamics and style. Estimating such mass partitioning budgets may reveal important for hazard evaluation purposes. We calculated the volume of each product emplaced during the August 2006 paroxysmal eruption of Tungurahua volcano (Ecuador) and converted it into masses using high-resolution grainsize, componentry and density data. This dataset is one of the first complete descriptions of mass partitioning associated with a VEI 3 andesitic event. The scoria fall deposit, near-vent agglutinate and lava flow include 28, 16 and 12 wt. % of the erupted juvenile mass, respectively. Much (44 wt. %) of the juvenile material fed Pyroclastic Density Currents (i.e dense flows, dilute surges and co-PDC plumes), highlighting that tephra fall deposits do not depict adequately the size and fragmentation processes of moderate PDC-forming event. The main parameters controlling the mass partitioning are the type of magmatic fragmentation, conditions of magma ascent, and crater area topography. Comparisons of our dataset with other PDC-forming eruptions of different style and magma composition suggest that moderate andesitic eruptions are more prone to produce PDCs, in proportions, than any other eruption type. This finding may be explained by the relatively low magmatic fragmentation efficiency of moderate andesitic eruptions. These mass partitioning data reveal important trends that may be critical for hazard assessment, notably at frequently active andesitic edifices. This article is protected by copyright. All rights reserved.

On March 19, 2008 a small explosive event accompanied the opening of a 35-m-wide vent (Overlook vent) on the southeast wall of Halema‘uma‘u Crater in Kīlauea Caldera, initiating an eruptive period that extends to the time of writing. The peak of activity, in 2008, consisted of alternating background open-system outgassing and spattering punctuated by sudden, short-lived weak explosions, triggered by collapses of the walls of the vent and conduit. Near-daily sampling of the tephra from this open system, along with exceptionally detailed observations, allow us to study the dynamics of the activity during two eruptive sequences in late 2008. Each sequence includes background activity preceding and following one or more explosions in September and October 2008 respectively. Componentry analyses were performed for daily samples to characterise the diversity of the ejecta. Nine categories of pyroclasts were identified in all the samples, including wall-rock fragments. The six categories of juvenile clasts can be grouped in three classes based on vesicularity: (1) poorly, (2) uniformly highly to extremely, and (3) heterogeneously highly vesicular. The wall-rock and juvenile clasts show dissimilar grainsize distributions, reflecting different fragmentation mechanisms. The wall-rock particles formed by failure of the vent and conduit walls above the magma free surface and were then passively entrained in the eruptive plume. The juvenile componentry reveals consistent contrasts in degassing and fragmentation processes before, during and after the explosive events. We infer a crude ‘layering’ developed in the shallow melt, in terms of both rheology and bubble and volatile contents, beneath a convecting free surface during background activity. A tens-of-centimetres thick viscoelastic surface layer was effectively outgassed and relatively cool, while at depths of less than 100 m, the melt remained slightly supersaturated in volatiles and actively vesiculating. Decoupled metre-sized bubbles rising through the column burst through the free surface frequently, ejecting fragments of the outgassed upper layer. When the surface was abruptly perturbed by the rock-falls, existing mm-sized bubbles expanded, leading to the acceleration of adjacent melt upward and consecutive explosions, while renewed nucleation created a minor population of 10-micron-sized bubbles. After each explosive event in September–October 2008, this layering was re-established but with decreasing vigour, suggesting that the magma batch as a whole was becoming progressively depleted in dissolved volatiles.

Crystal-rich intermediate magmas are subjected to both primary and secondary fragmentation processes, each of which may produce texturally distinct tephra. Of particular interest for volcanic hazards is the extent to which each process contributes ash to volcanic plumes. One way to address this question is by fragmenting pyroclasts under controlled conditions. We fragmented pumice samples from Soufriere Hills Volcano (SHV), Montserrat, by three methods: rapid decompression in a shock tube-like apparatus, impact by a falling piston, and milling in a ball mill. Grain size distributions of the products reveal that all three mechanisms produce fractal breakage patterns, and that the fractal dimension increases from a minimum of ~ 2.1 for decompression fragmentation (primary fragmentation) to a maximum of ~ 2.7 by repeated impact (secondary fragmentation). To assess the details of the fragmentation process, we quantified the shape, texture and components of constituent ash particles. Ash shape analysis shows that the axial ratio increases during milling and that particle convexity increases with repeated impacts. We also quantify the extent to which the matrix is separated from the crystals, which shows that secondary processes efficiently remove adhering matrix from crystals, particularly during milling (abrasion). Furthermore, measurements of crystal size distributions before (using x-ray computed tomography) and after (by componentry of individual grain size classes) decompression-driven fragmentation show not only that crystals influence particular size fractions across the total grain size distribution, but also that free crystals are smaller in the fragmented material than in the original pumice clast. Taken together, our results confirm previous work showing both the control of initial texture on the primary fragmentation process and the contributions of secondary processes to ash formation. Critically, however, our extension of previous analyses to characterisation of shape, texture and componentry provides new analytical tools that can be used to assess contributions of secondary processes to ash deposits of uncertain or mixed origin. We illustrate this application with examples from SHV deposits.

(IF 2,217); International audience; This contribution aims at understanding how magmatism, hydrothermal alteration, and deformation may have interacted to localize a detachment (a low-angle normal fault) in a basaltic volcano. Piton des Neiges, an inactive volcano of La Réunion Island, has been deeply cut by erosion, allowing its inner structure to be investigated. The deepest unit observed in the edifice is a kilometer-scale plutonic complex, the top of which is intruded by multiple sills. This zone of repeated sill intrusions has been interpreted as a detachment because it displays evidence of hydrothermal alteration in the greenschist facies linked to a brittle-ductile shear deformation. Deformation begins with cataclasis and is followed by mylonitization and chlorite crystallization, then by hydrofracturing and pumpellyite crystallization. Subsequent and post-deformation calcite crystallization occurs in voids such as fractures and vacuoles. Aluminium substitutions in chlorite suggest that the syn-deformation hydrothermal alteration did not exceed 250 °C and peaked in the deformation zone. Comparison of bulk-rock major element analyses of fresh, altered and deformed rocks shows that the zone of sill intrusion and deformation localized increased concentrations of P and K otherwise depleted in the footwall and hangingwall rocks, suggesting that the detachment acted as a trap for fluids.In contradiction with proposed models of volcano spreading, it is apparent that the portion of Piton des Neiges accessible to observation did not deform by creep of a large hydrothermal system or a plutonic complex below its solidus. Instead, the interface between the already cooled plutonic complex and the host rock acted as a brittle failure zone and was repeatedly intruded by magma injections. This localized heat source promoted hydrothermal alteration and low temperature creep in and around the discontinuity. The same process of magmatism-related weakening might occur on active volcanoes; it may, for instance, explain the slow post-eruptive deformation of the Eastern flank of Piton de La Fournaise (the active volcano of La Réunion) observed since the April 2007 eruption.

A co-pyroclastic density current (co-PDC) plume forms as a mixture of fine-grained (

International audience; The 18 May 1980 eruption of Mount St. Helens started with a lateral blast that fed a pyroclastic surge, which then uplifted to form a co-blast plume. Thirty minutes later, Plinian activity started at the vent and fluctuated in intensity for ~9 h. The resulting fallout deposit, documented to > 600 km from vent, presents some striking features: (1) displacement of the overall deposit to the north of the vent, (2) a secondary thickness and mass maximum at ~300 km from vent, (3) a total grain size distribution dominated by fine ash (62 wt % of the deposit < 63 μm), and (4) individual grain size distributions that vary dramatically in the crosswind direction from strongly bimodal in the south to skewed unimodal in the north. Results from a new deconvolution of the individual grain size distributions show that they are a combination of a coarse subpopulation that decreases in size with distance from vent and a constant fine subpopulation with a mean of ∼15 μm. Relative proportions of each subpopulation vary asymmetrically in the crosswind directions, with the fine subpopulation preponderant toward the north and the coarse one dominating the south of the deposit, both reach their absolute maxima in mass on the deposit axis. Componentry analyses of selected samples show that blast-derived material is greatly enriched toward the north of the deposit. These results indicate that the co-blast plume dispersed fine-grained material over great distances and dominated the fine subpopulation. Comparison with reanalysis data of atmospheric wind fields and satellite images of the spreading ash cloud suggests contrasting ash transport and depositional processes for the (early) co-blast plume and the (later) vent-derived Plinian plumes. The co-blast plume is displaced to the north; it had a high overshoot height, and eastward dispersion via strong winds low in the stratosphere (~10-15 km). The Plinian plumes were lower and dispersed most of the material to the southeast as the direction of high-velocity winds shifted just before the late climactic Plinian eruptive phase. Fine ash (fine subpopulation) was deposited continuously throughout the deposit, with an increase of sedimentation rate ~300 km from the vent where there is a secondary maximum in the deposit mass and thickness. Fine ash probably settled by a combination of enhanced sedimentation mechanisms, including not only aggregation but also gravitational convective instabilities of cloud base, hydrometeor formation and destruction, and entrainment of small particles by larger ones. Finally, we show that half of the deposit (by mass) in the medial area was deposited by the co-blast plume, and that a significant proportion of the Mount St. Helens fallout deposit is nonjuvenile, which has implications for the magmatic budget of this eruption.

Long-lasting andesitic eruptions sometimes include strong short-lived explosive events, which can pose significant hazards in populated regions. The origin and dynamics of such violent eruptions remain poorly known and may involve a combination of different factors. Tungurahua volcano, Ecuador, reawakens in 1999 and is an example of such an open-vent system that experienced a strong and deadly andesitic pyroclastic flow-forming event in August 2006. Inspection of the deposits suggested that the event could have been triggered by magma mixing (coexistence of both silicic pumices and andesitic scoria in the tephra), magma-water interaction (presence of xenolithic clasts) or deep andesitic magma reinjection (based on mineral chemistry). Here we investigate these options by performing a high-resolution mass budget analysis of the scoria fall deposit. This is achieved by analysing componentry compositions and their mass distribution pattern in the layer, which allow us to document and integrate exponential and power laws mass decay rates over wide areas. The results yield a total mass for the tephra layer of similar to 2 x 10(10) kg. The pumice mass fraction is far too small (

International audience; The violent August 16-17, 2006 Tungurahua eruption in Ecuador witnessed the emplacement of numerous scoria flows and the deposition of a widespread tephra layer west of the volcano. We assess the size of the eruption by determining a bulk tephra volume in the range 42-57 × 106 m3, which supports a Volcanic Explosivity Index 3 event, consistent with calculated column height of 16-18 km above the vent and making it the strongest eruptive phase since the volcano's magmatic reactivation in 1999. Isopachs west of the volcano are sub-bilobate in shape, while sieve and laser diffraction grain-size analyses of tephra samples reveal strongly bimodal distributions. Based on a new grain-size deconvolution algorithm and extended sampling area, we propose here a mechanism to account for the bimodal grain-size distribution. The deconvolution procedure allows us to identify two particle subpopulations in the deposit with distinct characteristics that indicate dissimilar transport-depositional processes. The log-normal coarse-grained subpopulation is typical of particles transported downwind by the main volcanic plume. The positively skewed, fine-grained subpopulation in the tephra fall layer shares close similarities with the elutriated co-pyroclastic flow ash cloud layers preserved on top of the scoria flow deposits. The area with the higher fine particle content in the tephra layer coincides with the downwind prolongation of the pyroclastic flow deposits. These results indicate that the bimodal distribution of grain size in the Tungurahua fall deposit results from synchronous deposition of lapilli from the main plume and fine ash elutriated from scoria flows emplaced on the western flank of the volcano. Our study also reveals that inappropriate grain-size data processing may produce misleading determination of eruptive type.

A general expression to describe particle density distribution in tephra fall deposits is essential to improve fallout tephra mass determination and numerical modelling of tephra dispersion. To obtain particle density distributions in tephra fall deposits, we performed high-resolution componentry and particle density analyses on samples from the 2006 subplinian eruption of Tungurahua volcano in Ecuador. Six componentry classes, including pumice and scoria, have been identified in our sample collection. We determined the class of 300 clasts in each 0.5ϕ fractions from −4.5ϕ to 3.5ϕ and carried out water pycnometry density measurements on selected size fractions. Results indicate that the mean particle density increases with ϕ up to a plateau of ~2.6 g/cm3 for clasts finer than 1.5ϕ. The density of scoria and pumice increases between −3 and 1ϕ, while dense particle density is sub-constant with grainsize. We show that the mean particle density μ of the vesicular fractions is a function of grainsize i (ϕ scale) given by a sigmoidal law: $$ \mu (i)={{{K+\beta }} \left/ {{\left( {1+\alpha {e^{-ri }}} \right)}} \right.} $$ , where K, β, α and r are constants. These sigmoidal distributions can be used to determine accurately the load of each componentry class and should be applicable to many tephra deposits and for modelling purposes.

International audience; Ash particle terminal settling velocity is an important parameter to measure in order to constrain the internal dynamics and dispersion of volcanic ash plumes and clouds that emplace ash fall deposits from which source eruption conditions are often inferred. Whereas the total Particle Size Distribution (PSD) is the main parameter to constrain terminal velocities, many studies have empirically highlighted the need to consider shape descriptors such as the sphericity to refine ash settling velocity as a function of size. During radar remote sensing measurements of weak volcanic plumes erupted from Stromboli volcano in 2015, an optical disdrometer was used to measure the size and settling velocities of falling ash particles over time, while six ash fallout samples were collected at different distances from the vent. We focus on the implications of the physical parameters of ash for settling velocity measurements and modeling. Two-dimensional sizes and shapes are automatically characterized for a large number of ash particles using an optical morpho-grainsizer MORPHOLOGI G3. Manually sieved ash samples show sorted, relatively coarse PSDs spanning a few microns to 2000 μm with modal values between 180 and 355 μm. Although negligible in mass, a population of fine particles below 100 μm form a distinct PSD with a mode around 5–20 μm. All size distributions are offset compared to the indicated sieve limits. Accordingly, we use the diagonal of the upper mesh sizes as the upper sieve limit. Morphologically, particles show decreasing average form factors with increasing circle-equivalent diameter, the latter being equal to 0.92 times the average size between the length and intermediate axes of ash particles. Average particle densities measured by water pycnometry are 2755 ± 50 kg m−3 and increase slightly from 2645 to 2811 kg m−3 with decreasing particle size. The measured settling velocities under laboratory conditions with no wind

Volcanic ash clouds are common, often unpredictable, phenomena generated during explosive eruptions. Mainly composed of very fine ash particles, they can be transported in the atmosphere at great distances from the source, having detrimental socio-economic impacts. However, proximal settling processes controlling the proportion (ε) of the very fine ash fraction distally transported in the atmosphere are still poorly understood. Yet, for the past two decades, some operational meteorological agencies have used a default value of ε = 5% as input for forecast models of atmospheric ash cloud concentration. Here we show from combined satellite and field data of sustained eruptions that ε actually varies by two orders of magnitude with respect to the mass eruption rate. Unexpectedly, we demonstrate that the most intense eruptions are in fact the least efficient (with ε = 0.1%) in transporting very fine ash through the atmosphere. This implies that the amount of very fine ash distally transported in the atmosphere is up to 50 times lower than previously anticipated. We explain this finding by the efficiency of collective particle settling in ash-rich clouds which enhance early and en masse fallout of very fine ash. This suggests that proximal sedimentation during powerful eruptions is more controlled by the concentration of ash than by the grain size. This has major consequences for decision-makers in charge of air traffic safety regulation, as well as for the understanding of proximal settling processes. Finally, we propose a new statistical model for predicting the source mass eruption rate with an unprecedentedly low level of uncertainty.