Classroom demonstrations of volcanic processes are typically performed with low-temperature materials designed to exhibit behaviors conceptually analogous to magma. However, there are advantages to finding methods by which real magma can be created and used with audiences and students. Here, we describe a method by which magma can be made at home or in a classroom using a standard household microwave. We have tested a particular demonstration of how volatiles such as H2O exsolve from natural obsidian to form pumice and ash, a direct replica of the volcanic phenomenon that drives eruptions on Earth and other planets. To demonstrate this, we used three obsidian samples—from Mono Craters (U.S.A.), Arteni (Armenia), and Hrafntinnuhryggur (Iceland)—within an off-the-shelf ceramics "microwave kiln" in both a 700 and 1000 W microwave. We found that millimetric obsidian chips will vesiculate to form pumice-like textures at low microwave power or low initial H2O content and will fragment safely within the kiln at high microwave power or high initial H2O content. This can be used to supplement existing well-known demonstrations using low-temperature analogs for magma. The demonstration results in real "products" that can be examined just as a geologist would examine natural specimens.

Rhyolites that erupted between the Otowi and Tshirege members of the Bandelier Tuff, known as the Valle Toledo Member, were investigated in the Jemez Mountains volcanic field to infer changes in eruptive rate and flux between two caldera-forming eruptions. Our analysis combines high-precision Ar-40/Ar-39 single-crystal laser-fusion dating of sanidine, matrix glass compositions and mineralogy of pumice, depositional textures, and volumetric estimates of erupted rhyolites to present a revised Valle Toledo Member eruptive chronology that integrates our observations with those of previous studies. With the improved temporal resolution of our high-precision eruption ages (median 2 sigma uncertainty of +/- 2.9 ka), the updated Valle Toledo Member eruption chronology consists of at least 42 temporally or mineralogically distinct eruptions that we group into four periods of time based on eruption rate. Within the first 8.1 +/- 5.1 kyr (similar to 1605-1597 ka) that followed the Otowi caldera-forming eruption at 1605.4 +/- 2.3 ka, six eruptions are recognized. This suggests an average recurrence interval on the order of 1.4 +/- 0.9 kyr. Twenty-seven eruptions occurred during the next 194.4 +/- 5.1 kyr (from similar to 1597-1403 ka) and the average recurrence interval increased to 7.2 +/- 0.2 kyr. Following this second period of slower eruption rate, a previously unrecognized eruption hiatus of up to 162.4 +/- 3.1 kyr occurred from 1402.9 +/- 2.3 ka to 1240.5 +/- 2.1 ka. Resumption of volcanic activity is characterized by a series of at least nine volcanic eruptions during the next 8.6 +/- 2.5 kyr, culminating in the eruption of the 400 km(3) Tshirege Member of the Bandelier Tuff and formation of Valles caldera at 1231.9 +/- 1.3 ka. This last phase of pre-caldera activity has the shortest observed average recurrence interval (1.0 +/- 0.3 kyr) and greatest eruptive flux (>= 0.4 +/- 0.2 km(3)/kyr) that occurred between the two caldera-forming eruptions. We interpret the shifts in eruption rate and flux following the 162.4 +/- 3.1 kyr eruption hiatus, in addition to mineralogical changes present in post-hiatus rhyolites, as indicators that the Bandelier system was trending towards another major eruption. The magmatic system may have grown gradually throughout the ca. 162 kyr period of volcanic repose; however, the heightened eruption rate in the ca. 10 kyr before caldera collapse is consistent with relatively rapid growth of the magmatic system before the Tshirege event as previously proposed by other studies. Furthermore, the new dataset is consistent with prior geochronology studies of the region that show the four largest explosive eruptions in the Jemez field were all proceeded by very slow eruptive rates or hiatuses. This demonstrates that sequences of heightened volcanic activity can initiate on rapid geological timescales from states of volcanic quiescence in the Bandelier system.

Pyroclastic density currents (PDCs) are density-stratified along their vertical axis, with the near-bed portion being denser than the upper portion, resulting from particle settling and ambient air entrainment at current margins. Whereas vertical density stratification likely influences mixing, sedimentation, and buoyancy of PDCs, many depth-averaged models of PDC dynamics assume currents are well-mixed. We investigated this discrepancy by performing sub-aqueous laboratory experiments and conducted complementary numerical simulations to interrogate current dynamics at finer scales. Currents with small temperature difference with the ambient fluid become density-stratified during propagation. The dynamics of such currents resemble two-phase flows, in which particles move freely and particle concentration becomes stratified, but fluid density remains constant. Currents with large temperature difference with the ambient fluid, however, do not develop density stratification during propagation, due to current dynamics becoming dominated by the fluid phase and the lessening importance of particles. Currents that develop density stratification do not lift off from the bed within the domain of the setup, whereas poorly stratified currents do lift off, forming a rising plume. Strong density stratification within currents inhibits turbulence production, preventing entrained ambient fluid on current edges from mixing into current interiors. Poorly stratified currents are highly turbulent, have vigorous internal mixing, thereby achieving lift-off. The strongly stratified currents are analogous to PDCs that result from eruption column collapse, maintaining fast velocity, low internal mixing, and high temperature over long distances. The poorly stratified currents are analogous to dilute ash-cloud surges that develop atop basal avalanches, having short runout distances. Pyroclastic density currents are incredibly destructive volcanic flows made up of hot rock and gases, and they pose a major threat to human populations in the vicinity of active volcanoes. Evidence from PDC deposits suggests that the top of such flows is less dense than the bottom, but mathematical models of such flows often inadequately assume they have unvarying density. We performed laboratory experiments and computer simulations to investigate how the difference in density between the top and bottom of such flows affects how far they travel. We find that hotter currents are less likely than colder currents to develop a density difference between the top and bottom portions. Currents that do not develop a density difference do develop more energetic mixing, which in turn makes them lift-off from the ground and rise vertically as a plume in the shape of a mushroom thermal. These experimental currents are like PDCs that result from the collapse of vertical eruption columns. Currents that do develop a density difference do not energetically mix and do not lift-off into a plume. These experimental currents are like dilute PDCs that result from the collapse of lava domes. Laboratory experiments and numerical simulations were used to study the effects of density stratification in pyroclastic density currents The temperature difference between currents and ambient fluid affects the development of density stratification during propagation Turbulence production depends on density stratification, impacting current buoyancy evolution and has implications for natural currents

Compositionally zoned crystals can record changing melt composition and trace element partitioning behavior during magmatic differentiation. Diffusive reequilibration between compositionally distinct zones in crystals can also produce compositional gradients. Here, we compare the length scales of concentration gradients for different elements in clinopyroxene that originate from the Tshirege Tuff and late Valle Toledo Member rhyolites of the Bandelier magmatic system in New Mexico to determine what petrogenetic information is recorded in the zonation. Within these rhyolites there are unzoned ferrohedenbergite crystals, as well as less common normally-zoned clinopyroxene with ferrohedenbergite rims and ferroaugite cores. Compared to the ferroaugite cores, the ferrohedenbergite rims are enriched in Dy and Yb, but depleted in Co, Ti, Sc, Ce, and Nd. The length scales for fast and slow diffusing elements for most gradients measured are indistinguishable, which argues that the gradients emerged predominantly from changing magmatic composition during crystallization, with diffusion having little to no role in establishing the concentration gradients. Fractional crystallization of the phases present in the rhyolites fails to reproduce all trace-element zonation that occur in the clinopyroxene, however, indicating a more complex origin. Based on the compositional similarity of the ferroaugite cores with pyroxene from rhyolites that erupted ≥ 165 kyr earlier, we interpret the ferroaugite cores as antecrysts scavenged from crystal-rich mush during magmatic rejuvenation. The magmatic rejuvenation that remobilized the parent mush of the ferroaugite antecrysts was likely initiated near the end of a > 100 kyr eruption hiatus that preceded the final runup to the catastrophic Tshirege eruption.

We conducted a set of high-temperature decompression experiments to constrain the mechanisms of heterogeneous bubble nucleation in high-silica rhyolitic melt that contained 4.6-4.8 wt% H2O. The melt was seeded with two different size fractions of magnetite crystals: 1-2 mu m crystals and large crystals of 32-135 mu m (long axis). The number density of bubbles (BND) that nucleated on the small crystals was found to increase from 10(6.5) to 10(8.7) cm(-3) as H2O increasingly supersaturated (Delta P) in the melt from 3 to 23 MPa. At Delta P >23 MPs, however, the number of bubbles nucleated equals the number of small magnetite and no more nucleated with increased Delta P. At the same conditions, the number of bubbles that nucleated on the large crystals increases, from <1 bubble per crystal at Delta P = 3 MPa to 14 +/- 4 bubbles per crystal at 58 MPa. We thus find that Delta P has a significant influence on the mechanisms of heterogenous nucleation, but the observed increases in BND are much greater than would be predicted solely from the increase in Delta P. The discrepancy can be reconciled if there are different sites on the crystals that become activated at greater Delta P, leading to greater numbers of bubbles nucleating. The cumulative BND nucleated on small crystals, however, is capped by the number of crystals present. The BND values generated at Delta P >23 MPa in our experiments overlap with those found in similar to 80 % of naturally occurring pumice. Assuming our experiments are representative of natural pumice, this suggests that explosively erupted magmas either become significantly volatile supersaturated before heterogeneously nucleating bubbles, or that the number of nucleation sites in natural magmas greatly exceed 10(9) cm(-3).

Edentulism-lack of teeth-is a derived condition among salientians (total-group frogs and toads) that has arisen independently at least 22 times within the crown group, Anura. Despite this frequency, edentulism (in part or complete) is seldom documented in the salientian fossil record, and thus its evolutionary history remains obscure. A similar to 90 Ma gap presently exists between the edentulous salientians Notobatrachus reigi (late Toarcian; Argentina) and the five known species from the latest Cretaceous of Argentina, South Africa, Canada, and the U.S.A. Here we report a new instance of edentulism in an Early Cretaceous salientian, Ostrombatrachos nodos gen. et sp. nov., based on a maxilla from an early Albian age (similar to 111 Ma) locality in the Cloverly Formation of Wyoming, U.S.A. Although incomplete, the holotype maxilla exhibits a distinctive suite of features (i.e., edentulous; pit-and-ridge external ornament; shallow groove across anteriormost portion of margo orbitalis and continuing anteroventrally down lateral surface of pars facialis to interrupt the external ornament; medially expanded orbital flange; moderately broad, deep lamina horizontalis; and weakly developed processus pterygoideus) that is unknown in any other salientian. This new taxon represents the oldest occurrence of edentulous maxillae in a Laurasian salientian and helps fill the long temporal gap in edentulism among salientians. O. nodos gen. et sp. nov. does not appear to be closely related to other edentulous Mesozoic salientians, suggesting it represents yet another independent evolution of this feature.

Volcanic eruptions are driven by bubbles that form when volatile species exsolve from magma. The conditions under which bubbles form depend mainly on magma composition, volatile concentration, presence of crystals, and magma decompression rate. These are all predicated on the mechanism by which volatiles exsolve from the melt to form bubbles. We critically review the known or inferred mechanisms of bubble formation in magmas: homogeneous nucleation, heterogeneous nucleation on crystal surfaces, and spontaneous phase separation (spinodal decomposition). We propose a general approach for calculating bubble nucleation rates as the sum of the contributions from homogeneous and heterogeneous nucleation, suggesting that nucleation may not be limited to a single mechanism prior to eruption. We identify three major challenges in which further experimental, analytical, and theoretical work is required to permit the development of a general model for bubble formation under natural eruption conditions. ▪ We review the mechanisms of bubble formation in magma and summarize the conditions under which the various mechanisms are understood to operate. ▪ Bubble formation mechanisms may evolve throughout magma ascent as conditions change such that bubbles may form simultaneously and sequentially via more than one mechanism. ▪ Contributions from both homogeneous nucleation and heterogeneous nucleation on multiphase crystal phases can be captured via a single equation. ▪ Future work should focus on constraining macroscopic surface tension, characterizing the microphysics, and developing a general framework for modeling bubble formation, via all mechanisms, over natural magma ascent pathways.

We present an experimental investigation of surface tension–driven sintering and associated densification of glassy rhyolitic ash and crystals under shallow volcanic conduit conditions. Rhyolitic glass (< 45 μm) and quartz were run in suites of hydrothermal experiments for 30 min to 9 h. Fluid pressure was isobaric ( P_H_2O=40 MPa ) for all runs, and temperature was held constant at a value between 675 and 850 °C. Three size populations of quartz were used: 45–76 µm (fine), 90–125 µm (medium), and 250–500 µm (coarse). All samples evolved from loose, cohesion-less particles to a friable, agglutinated framework of glass with an interconnected network of pores of ≥ 15 vol.

Obsidian pyroclasts are common in deposits from silicic sub-Plinian eruptions and can record pre- and syn-eruptive processes in the volcanic conduit. Previous work focusing on dissolved volatiles and vesicle textures has been useful in extracting timescales of sintering, diffusion, and vesicle relaxation recorded by obsidian pyroclasts. Here we focus on microlite crystals (<100 µm in size) to augment previous work because they form at different rates than vesicles or than rates of volatile degassing. Hence, they have the potential to disclose additional information about processes occurring in an explosive conduit. We examine microlites in 72 samples from tephra deposits of the 1340 C.E. North Mono eruption, California, U.S.A., and complement these measurements with hydrothermal experiments at 800°C, 10–50 MPa, and durations from 1 to 7 h. Three observations of the natural obsidians further elucidate their formation. First, microlite number densities (MND) increased as the eruption progressed. Second, multiple microlite morphologies occur for feldspars (blocky, swallowtail, tabular, cluster, skeletal) and pyroxenes (individual rods or clusters of acicular crystals) in each obsidian, regardless of any other characteristic. Third, microlite orientations correlate with the dominant morphology of vesicles, being generally well aligned in samples with ellipsoid vesicles, generally poorly aligned in samples with spherical vesicles, and either unaligned or aligned into planes in samples with distorted vesicles. In hydrothermal experiments, MND increase with time, microlites display only one morphology, and microlites are randomly oriented at any given pressure or temperature. When compared to natural obsidians, our experiments suggest most of the microlites could have grown in ≤∼7 h. The variety of microlite morphologies and orientations argue for repeated in-conduit fragmentation and sintering, consistent with the idea that each individual obsidian pyroclast is the product of ash sintering at multiple depths in the conduit prior to finally being erupted. During most of the eruption, obsidian pyroclasts were extracted from many depths in the conduit, preserving an array of volatile contents and microlite textures. Near the end of the explosive phase, however, higher MND record longer periods of stalling while dissolved volatile contents record vapor-melt equilibration at shallow depths in the conduit.

In this study, we present new evidence for changes in magma storage conditions that preceded the 1232 ka caldera-forming eruption of the Bandelier magmatic system in the Jemez Mountains Volcanic Field. Using high precision Ar-40/Ar-39 sanidine dating we determine that at least eight rhyolites erupted within 8.6 +/- 3.4 kyr of the-400 km(3) eruption that formed Valles caldera. Some of those rhyolites contain fayalite with or without biotite, others contain only biotite. An eruption of fayalite-bearing rhyolite at 1240.5 +/- 2.1 ka ended an eruption hiatus of at least 100 kyr. Following that first post-hiatus episode of volcanism, at least four more eruptions of fayalitebearing rhyolite and three eruptions of biotite-bearing rhyolite occurred prior to the caldera-forming eruption. We use phase equilibrium experiments and geothermobarometry to infer the storage conditions and processes that led to these differing crystal cargos and ultimately generated-400 km3 of predominantly fayalite rhyolite ignimbrite (Tshirege Member of the Bandelier Tuff). We find that biotite-bearing rhyolites were stored at 695-750(degrees)C, 75-160 MPa, and at an oxygen fugacity more oxidizing than the quartz-fayalitemagnetite (QFM) buffer reaction. Fayalite-bearing rhyolites were similarly stored at 695-745(degrees)C and 70-190 MPa, but at more reducing conditions (fO(2)<= QFM). We suggest that the reduced, fayalite-bearing rhyolite was most likely produced via interaction of crystal poor rhyolitic magma with a reducing, potentially Cl-bearing, and H2O-rich supercritical fluid phase. This fluid flux event was a key component of the substantial magmatic rejuvenation that enabled the mobilization of-400 km(3 )of mostly fayalite-bearing rhyolite prior to not only the Tshirege event, but the older Otowi event as well.

How to build a legacy of scientific leadership: the HR formula PROF. JULIA HAMMER, PHD1, LESLIE BAKER2, JENNI BARCLAY3, MICHAEL R. CARROLL4, MICHELLE COOMBS5, ELIZABETH COTTRELL6, NICHOLAS J DYGERT7, LINDA ELKINS-TANTON8, EMILY FIRST9, JAMES GARDNER10, DAVID GOLDSBY11, JAMES GREENWOOD12, MARIE JOHNSON13, MIKE KRAWCZYNSKI14, CHARLES MANDEVILLE15, MOLLY MCCANTA16, MICHELLE E. MINITTI17, WILLIAM NELSON18, TABB PRISSEL19, DINA VENEZKY20, CATHERINE WEITZ21 AND DIANE WOODRUFF22 1University of Hawaiʻi 2University of Idaho 3University of East Anglia 4Camerino University 5U.S. Geological Survey 6National Museum of Natural History, Smithsonian Institution 7University of Tennessee, Knoxville 8Arizona State University 9Cornell University 10University of Texas at Austin 11University of Pennsylvania 12Wesleyan University 13Cal State Fullerton 14Washington University in St. Louis 15US Geological Survey 16University of Tennessee at Knoxville 17Framework, Silver Spring 18University of Hawaii at Manoa 19NASA 20Smith College 21Planetary Science Institute 22Anadarko Petroleum Company Presenting Author: jhammer@hawaii.edu

One model for formation of obsidian pyroclasts suggests that they form through sintering of ash particles on volcanic conduit walls, which are subsequently torn out and entrained in the gas-particle dispersion out of the erupting vent.Here, we investigate microlite abundances and textures in obsidian pyroclasts in order to determine the time required to produce adequate numbers of microlites, and hence the pyroclasts themselves.We measured microlite number densities (MNDs) and microlite and vesicle orientations in obsidian pyroclasts in tephra deposits from the 1340 A.D. North Mono eruption.MNDs increase with decreasing dissolved H2O concentrations.Also, microlite spatial orientations become less aligned and differ more from vesicle orientations with decreasing dissolved H2O concentrations.MNDs increase from the second layer (P2) through the final layer (P10).To investigate timescales required to replicate MNDs in the North Mono obsidian, we performed time, temperature and pressurecontrolled experiments with rhyolitic glass from the same eruption.MNDs in our experiments initially increase with decreasing pressure (50-35 MPa), then decrease as pressure decreases further(35-10 MPa).MNDs in obsidian from layers P2-P10 were replicated in ˜7 hours or less.Based on these observations we propose a model where during the initial phase of the North Mono eruption most obsidian formed close to the magmatic fragmentation depth, equilibrated for short time periods (< 7 hours) and were then erupted out of the volcanic vent.These obsidian clasts have lower MNDs than subsequent phases, and microlites are well aligned with each other and with vesicles, reflecting their short residence time in the conduit, higher dissolved H2O contents, and lower viscosities.During later phases of the North Mono eruption obsidian formed at various depths in the conduit, equilibrating for longer periods of time ([?] ˜7 hours) before being erupted out of the vent or sintering together with other clasts and equilibrating at shallower depths before being erupted.These obsidian clasts have higher MNDs than earlier phases of the eruption, and microlites are not well aligned with each other or with vesicles, reflecting their variable residence times in the volcanic vent, lower dissolved H2O contents, and higher viscosities.

An incomplete salamander dentary (AMNH FARB 22965) described herein from the upper Maastrichtian Lance Formation, Wyoming, USA, exhibits a puzzling suite of features. Four features—a prominent bony trough extending anteriorly and curving upwards along the lingual surface of the ramus, lack of an obvious Meckelian fossa or groove, an apparent gap in the tooth row, and a symphysial-like first tooth—are likely anomalies. However, the remaining features are interpreted as normal structures and suggest that AMNH FARB 22965 represents a new genus and species of batrachosauroidid, an extinct family of neotenic salamanders that were prominent components of Cretaceous to Neogene freshwater and floodplain paleocommunities in North America and Europe. The new taxon differs from other batrachosauroidids in a unique suite of dentary and dental features, most notably in having a lingual bony flange paralleling the posterior two-thirds of the dentary tooth row, a prominent and robust coronoid process bearing a grooved anterior face, and the anterior portion of the corpus dentalis behind the symphysis is broadly expanded ventrolingually. The presence of a third batrachosauroidid taxon in the Lance Formation was unexpected, considering that the formation has been well sampled and that its two previously recognized batrachosauroidids, namely Opisthotriton kayi and Prodesmodon copei, are known by abundant isolated bones, including dozens of dentaries, from numerous localities in the unit and elsewhere in the North American Western Interior. Known by a unique dentary from the Bushy Tailed Blowout locality, the taxon represented by AMNH FARB 22965 evidently was uncommon within the Lance Formation paleoenvironment.

Large silicic eruptions can be preceded by small eruptions of different styles and volumes. The Holocene Llao Rock, Cleetwood, and climactic eruptions of Mt. Mazama, OR, USA, were sourced from the same magma and followed this pattern. The Llao Rock and Cleetwood eruptions are both relatively small, have pyroclasts with microlites and a wide range of vesicularities, and each consisted of an explosive phase followed by an effusive phase. The climactic eruption had no effusive phase and created highly vesicular pyroclasts with no microlites. We analyzed microlite crystallization using phase equilibria and decompression experiments. Comparing the results to the pyroclasts from the Llao Rock and Cleetwood eruptions, we find that the differences between the small and climactic eruptions are likely caused by different magma ascent dynamics. Our experiments show that plagioclase and pyroxene microlites crystallize only during decompressions that are most likely too slow to result in explosive eruptions. The Llao Rock magma likely stalled at shallow depths before continuing fast ascent, which allowed for microlite crystallization that might have also caused explosive eruptions. The Cleetwood magma likely took two separate ascent paths, a majority fraction that ascended quickly from high pressure without stalling and a minority fraction that stalled at shallow depths before continuing ascent along with the majority fraction. These ascent dynamics of the Llao Rock and Cleetwood magmas led to the creation of obsidian pyroclasts from sidewall sintering of fragmented majority-fraction ash. The climactic magma did not stall at shallow depths and instead ascended from depth quickly to the surface, creating the conditions necessary for caldera collapse.

Silicic volcanic eruptions range in style from gently effusive to highly explosive, and may switch style unpredictably during a single eruption. Direct observations of subaerial rhyolitic eruptions (Chaiten 2008, Cord ' on Caulle 2011-2012, Chile) challenged long-standing paradigms of explosive and effusive eruptive styles and led to the formulation of new models of hybrid activity. However, the processes that govern such hybrid explosive-effusive activity remain poorly understood. Here, we bring together observations of the well-studied 2011-2012 Cord ' on Caulle eruption with new textural and petrologic data on erupted products, and video and still imagery of the eruption. We infer that all of the activity - explosive, effusive, and hybrid - was fed by explosive fragmentation at depth, and that effusive behaviour arose from sticking and sintering, in the shallow vent region, of the clastic products of deeper, cryptic fragmentation. We use a scaling approach to determine that there is sufficient time available, during emplacement, for diffusive pyroclast degassing and sintering to produce a degassed plug that occludes the shallow conduit, feeding clastogenic, apparently effusive, lava-like deposits. Based on evidence from Cord ' on Caulle, and from other similar eruptions, we further argue that hybrid explosive-effusive activity is driven by episodic gas-fracking of the occluding lava plug, fed by the underlying pressurized ash- and pyroclast-laden region. The presence of a pressurized pocket of ash-laden gas within the conduit provides a mechanism for generation of harmonic tremor, and for syn-eruptive laccolith intrusion, both of which were features of the Cord ' on Caulle eruption. We conclude that the cryptic fragmentation models is more consistent with available evidence than the prevailing model for effusion of silicic lava that assume coherent nonfragmental rise of magma from depth to the surface without wholesale explosive fragmentation.

Bubble nucleation is the first step in magma degassing. The kinetics of bubble nucleation during eruptive magma ascent determines the rate of magma degassing, which in turn, has a significant control on the explosivity of eruptions. Here, we examine if bubble nucleation in rhyolite can be continuous over the inferred timescale of magma travel time from reservoir to the surface in explosive silicic eruptions. We performed homogeneous bubble nucleation experiments using cylinders cored from natural rhyolite. Samples were hydrated at an initial pressure and temperature, then decompressed to a final pressure, and held at that pressure for various amounts of time before being quenched. Within a given suite samples had the same temperature as well as initial and final pressures, but were held for different annealing times. The resulting bubble number density of samples within each suite increases by up to 2 orders of magnitude as annealing time increases, indicating that nucleation continued during annealing. We estimate the nucleation rate and duration from measured bubble number densities and show that nucleation in the majority of samples was still ongoing at the time of quenching. Our findings confirm theoretical predictions that bubble nucleation can be a continuous process during eruptive magma ascent in explosive eruptions.

The effects of pyroclastic density currents (PDCs) can be devastating, so understanding their internal dynamics and evolution is important for hazard assessment. We use damaged trees located around Mount St. Helens (USA) as proxy for the dynamic pressure ( P dyn ) of the PDC erupted on 18 May 1980. We recorded the location, distribution, and foliage preservation of damaged trees within the medial and distal parts of the devastated forest. Sub-meter resolution aerial photographs from a month after the eruption allow distinction between standing trees that retained foliage from those that were stripped. Heights of standing trees were estimated from the measured lengths of their shadows. The number of standing trees was counted within defined areas along the propagation paths of PDCs. From the measured tree heights, we estimated tree toppling stresses from P dyn . Overall, P dyn of the PDC head within the medial to distal portions of the blowdown zone ranged from 10 to 35 kPa. P dyn likely waned with distance, as shown by the increased number of standing trees in the outer parts of the devastated area. In addition, we find clusters of standing trees on the lee sides of some hills. We propose that these clusters survived because they were primarily impacted by lower dynamic pressures extant within the PDC body, with foliage retention or stripping as a function of the P dyn evolution in the PDC body. We estimate that P dyn of the body was less than the estimated maximum P dyn of the PDC head by 12 ± 4 kPa.

The controls on the style of silicic eruptions – hazardously explosive, more gently effusive, or hybrid explosive-effusive – are poorly constrained. Current models invoke escape of gas through a connected foam, or through fractures, as the primary mechanism for the transition from explosive to effusive eruption. We propose a new model, in which hybrid and effusive silicic eruptions are typically explosive at depth, but the clastic products of this 'cryptic fragmentation' sinter and weld in the conduit to produce coherent lava at the surface. Drawing on numerous case studies of natural textures within eruptive products and dissected conduits, we show that effusive silicic eruptions are best interpreted as being the welded, squeezed-out remnant of ongoing or recent subsurface explosive behaviour. We demonstrate that effusively erupted lavas have microtextures diagnostic of a welding/sintering genesis and are comparable with those found in rheomorphic and welded ignimbrites. All eruptive products share pore network geometries and associated mechanical and hydraulic property-porosity relationships that are consistent with models for sintered materials. We conclude that silicic lava is generally clastogenic, and that, after it is sinter-assembled, it may undergo gas-driven fracturing that produces lava plug-cutting tuffisites (closed fractures filled with sintered particles), and sintered pyroclasts (from ash- to bomb-sized). At some sites (e.g. Volcán Chaitén 2008), the first material to be extruded from the vent is a pyroclastic rubble similar texturally to the volcanic bombs from the same site. We propose therefore that the shallow conduit is filled with pyroclastic and lithic rubble; a volume that variably compacts over time to produce a plug of densified lava. Envisaging the shallow conduit as a compacting rubble pile instead of a coherent magma-filled pipe or crack leads us to posit that the explosive-effusive transition is a blurred behavioural switch controlled by the competition between material supply at the underlying fragmentation front, and shallow particle capture, welding and lava production above. This framework has broad top-down implications for geochemical and geophysical predictions of shallow silicic volcanism, which we will explore in this presentation.