V21C-0721
Transition between fragmentation and permeable outgassing of low viscosity magmas
Explosive volcanic eruption requires that magma fragments into discrete parcels. Silicic magma can fragment through brittle failure or other processes that depend on the viscoelasticity of the melt. Owing to the low viscosity of basaltic magmas, however, the fragmentation mechanism must be different and will be governed by fluid mechanics alone. We perform a series of decompression experiments on bubbly Newtonian fluids with viscosities and surface tensions similar to those of basaltic magmas. For sufficiently rapid expansion, the bubbly fluid expands continuously, eventually tearing into several pieces. We find that the fragmentation threshold is governed by a critical Reynolds number of ~ 1, indicating that it is the inertia of the expanding fluid that drives the continued expansion and ultimate breakup into discrete parcels. Experiments in which the fluid does not fragment allow us to determine the gas permeability of the bubbly fluid as the bubbles expand. Permeability remains small until the volume fraction of bubbles exceeds about 70%. We scale the results of the laboratory experiments to basaltic eruptions and find that the predicted fragmentation threshold is consistent with the exit velocities that characterize effusive and explosive eruptions. Our experimental results suggest that the mechanism for fragmentation of low viscosity basaltic magma is fundamentally different from that of high viscosity silicic magma, and that magma with low viscosities can fragment easily.
V21C-0722
Magmatic Carbon Dioxide Emissions From Mammoth Mountain, California -- A Decreasing Trend From 1996 to 2007
Mammoth Mountain is a Pleistocene volcano along the crest of the central Sierra Nevada Range and comprises about 20 domes of dacitic to rhyolitic composition extruded between about 110 and 57 ka. The most recent volcanic activity consisted of a few small phreatic eruptions on the north flank of the mountain about 700 yrs ago. Magmatic CO2 emissions from soils on Mammoth Mountain were first detected and measured in 1994 but later, based are 14C in tree rings, were shown to have begun in 1990. The CO2 is released during decompression and cooling of small magma bodies that intrude the crust beneath the mountain. The magmatic gas is thought to have collected in a sealed-, natural-reservoir, 1-4 km deep, which was breached by crustal deformation during an episode of seismic unrest in 1989. Recent measurements of the diffuse emission of magmatic CO2 at Mammoth Mountain show that the total output of CO2 decreased by 80% between 1996 and 2006. CO2 discharged from steam vents in fumarolic areas is minor and was not quantified. Estimates of the total mass of CO2 emitted from soils have been made based on kriging of ground-based, closed-chamber CO2 flux measurements. In five areas where measurements were made in 1996 and 2006, the rate of CO2 emission decreased by 58 to 85% in the individual areas. In the same period the total output of CO2 from the five areas decreased from 495 to 100 tonnes per day. The decrease in CO2 emission is consistent with the depletion of gas stored in a reservoir that filled over a period much longer than the 18 years since the reservoir seal was breached in 1989. Measurements of magmatic gas emissions at volcanoes such as Mammoth Mountain may not always provide reasonable estimates of the volumes of recent shallow degassing intrusions because the source of these emissions may be accumulated gas from older intrusions. Preliminary estimates from data being collected in 2007 indicate that the trend of declining CO2 emissions continues.
V21C-0723
Permeability, anisotropy and tortuosity measurements of pumices using X-ray computed microtomography
X-ray computed microtomography (μCT) has become a widely-applied technique to obtain density maps of heterogeneous media; it allows gathering non-destructively qualitative observations as well as quantitative information on the 3D geometries of multi-phase samples. In this study, we obtained 3D images of different pumice types found in the rhyolitic Kos Plateau Tuff (KPT) deposits (160 ky, South Aegean Arc, Greece), and combined anisotropy and (geometrical) tortuosity measurements of these scans with permeability data to gain insights into the development of pathways through magmatic foam and how it affects syn-eruptive degassing. The rhyolitic KPT pumices are particularly prone to textural analysis because (1) the deposits are non-welded, (2) the high viscosity of the magma helped preserving information on the state of the magmatic foam in the conduit immediately prior to fragmentation (i.e., disruption of magma into pyroclastic fragments) and (3) pumices display variable macroscopic textures including tubular and near-spherical networks of bubbles. The stacks of grey-scale μCT images were cropped and segmented to obtain 3D binary volumes. These volumes were submitted to anisotropy and tortuosity measurements using existing softwares. Results suggest a significantly more convoluted path through the spherical bubble networks than the tubular bubble networks. To complement this geometrical characterisation of pumices, permeability values on the same binary volumes will be acquired using two numerical codes (one is based on a Finite Difference scheme, the other using the Lattice Boltzmann technique).
V21C-0724
Crystal Control on Pore Pathways and Degassing of Mafic Volcanic Rocks
The degree to which magma can degas, and thereby relieve gas pressure prior to an eruption, determines whether an eruption will be effusive or explosive. Potential gas-escape pathways can be imaged by combining measurements of connected porosity, permeability, and electrical conductivity of volcanic rocks with analysis of thin section images. Most work on volcanic pore geometries has focused on crystal-poor silicic magmas. Here we examine crystal-rich mafic magmas to see how pore geometries and pathways are affected by the presence of a crystalline network. We measure connected porosity of one-inch diameter, one-inch long cores using helium pycnometry; permeability is measured using a capillary flow porometer; and resistance to electrical flow is measured by an LCR meter attached to a core saturated in a conductive saline solution. Electrical tortuosity (a proxy for sinuosity and/or roughness of connected pathways) is calculated using the connected porosity of the core and its resistance to electrical flow. Plots of Darcian permeability versus connected porosity for a range of rock types reveal that mafic (basalts and basaltic andesites) and silicic (dacites and rhyolites) samples cluster in separate regions with little overlap. Mafic rocks are generally more permeable and have lower electrical tortuosities than silicic rocks with the same connected porosities. These differences most likely reflect some combination of larger pore apertures and lower pore path sinuosities in mafic samples. In a detailed study of a suite of basaltic andesites from Bezymianny volcano, Kamchatka, Russia, we see that permeability and porosity vary widely with sample type (samples include dense cryptodome, dense to vesicular lava and pycroclastic flows, and highly vesicular airfall deposits) and that electrical tortuosity decreases with increasing connected porosity. Preliminary examination of SEM images shows that all samples have abundant phenocrysts and microlites. Low porosity cryptodome samples have isolated large pores that are deformed and elongated around and between crystals. In contrast, high porosity airfall samples have abundant small vesicles, particularly along crystal margins. We suggest that the presence of crystals facilitates bubble nucleation, and that as crystallinity increases, bubble deformation becomes more pronounced as bubbles are able to elongate and coalesce to form crack-like pathways. This implies that even at low connected porosities, permeable networks are able to form and provide effective pathways for degassing.
V21C-0725
On 3D reconstruction of bubbles in volcanic ash particles
Bubbles in volcanic ash particles are primarily represented by the remnants of films and plateau borders from disrupting foam. Without preservation of complete bubbles, measuring bubble size distributions a challenging task, but one for which we have taken a novel approach. Concavities in ash particles retain a record of bubble sizes in the curvature of their concave surfaces that resulted from bubble fragmentation and quenching during energetic magma eruptions. We have used two methods to measure bubble fragment curvature on the basis of 3D reconstruction of ash particle surfaces. One is based on High Resolution X-Ray Tomography (HRXRT) and the second one is based on stereo images from tilting Scattered Electron Microscopy (SEM). Both methods allow the creation of Digital Elevation Model (DEM) datasets of the ash particle surfaces which in turn are used to identify and measure vertical cross-sectional profiles of the individual bubble fragments ("craters"). Function fit analysis for circular or elliptical functions are applied to each bubble cross sectional profile in two orthogonal directions to reconstruct sizes of the original, complete bubbles. The method allows measurement of submicron (SEM; XUM), micron or larger (HRXRT) bubbles in ash particles. The bubble size distributions so obtained can provide valuable insights regarding magma dynamics and vesiculation that lead to explosive eruptions, as well as the processes of fragmentation in eruption columns. There are no previous systematic information/databases of vesiculation metrics for explosive silicic eruptions, but this new method can be used to produce these and thus provide better insights into prehistoric eruption styles for volcanic hazard assessment.
V21C-0726
Methods of Permeability Calculation for Vesicular Materials
Knowing the permeability of magma is key to understanding the degassing processes of volcanoes. Permeability is controlled by several factors including porosity, specific surface area, bubble connectivity, and pathway tortuosity, which makes calculation often difficult. However, because of the scale-dependence of permeabilities, measurements of permeability only provide correct values when measurements are taken at the appropriate scale of interest, which is not possible in the case of volcanic conduits. Preserved pumice samples give us smaller-scale snapshots of the bubble pathways in the volcano conduit at the time of eruption, which may be upscaled to a representation of the entire conduit. The permeability of this representation may then be calculated using various methods, each explored further here. The simplest of the methods is a Kozeny-Carman approach. In this case, a two-point correlation function is used to determine porosity and specific surface area, which can then be used in an equation to approximate permeability. The weakness of this method is that in vesicular materials, the small aperture radius between bubble connections is not accounted for. A second method, which does account for a smaller aperture radii, is the tube network model described by Jon Blower (Blower, 2001). This method treats gas flow through bubble networks similar to current in electrical circuits, where each aperture provides a certain resistance to flow. For this method, information is needed regarding each aperture in the bubble network. The Lattice-Boltzmann method provides a third way to calculate permeability. It is the most accurate of these methods, however, it typically is computationally expensive. In this presentation, we compare results of permeability determinations from measurements and the three theoretical/numerical approaches discussed above. We then discuss implications of permeabilities determined using these methods with respect to magmatic volatile degassing and related volcanic eruption dynamics.
V21C-0727
Magma Vesiculation and Infrasonic Activity in Open Conduit Volcanoes
At persistently active basaltic volcanoes such as Stromboli, Italy degassing of the magma column can occur in "passive" and "active" conditions. Passive degassing is generally understood as a continuous, non explosive release of gas mainly from the open summit vents and subordinately from the conduit's wall or from fumaroles. In passive degassing generally gas is in equilibrium with atmospheric pressure, while in active degassing the gas approaches the surface at overpressurized conditions. During active degassing (or puffing), the magma column is interested by the bursting of small gas bubbles at the magma free surface and, as a consequence, the active degassing process generates infrasonic signals. We postulated, in this study, that the rate and the amplitude of infrasonic activity is somehow linked to the rate and the volume of the overpressured gas bubbles, which are generated in the magma column. Our hypothesis is that infrasound is controlled by the quantities of gas exsolved in the magma column and then, that a relationship between infrasound and the vesiculation process should exist. In order to achieve this goal, infrasonic records and bubble size distributions of scoria samples from normal explosive activity at Stromboli processed via X ray tomography have been compared. We observed that the cumulative distribution for both data sets follow similar power laws, indicating that both processes are controlled by a scale invariant phenomenon. However the power law is not stable but changes in different scoria clasts, reflecting when gas bubble nucleation is predominant over bubbles coalescence and viceversa. The power law also changes for the infrasonic activity from time to time, suggesting that infrasound may be controlled also by a different gas exsolution within the magma column. Changes in power law distributions are the same for infrasound and scoria indicating that they are linked to the same process acting in the magmatic system. We suggest that monitoring infrasound on an active volcano could represent an alternative way to monitor the vesiculation process of an open conduit system.
V21C-0728
Submarine Strombolian Eruptions Observed at NW Rota-1 Volcano, Mariana Arc
Extraordinary video and hydrophone observations of a submarine explosive eruption were made with a remotely operated vehicle in April 2006 at a depth of 550-560 m on NW Rota-1, a conical, basaltic-andesite submarine volcano in the Mariana arc. The observed eruption evolved from effusive to explosive, while the eruption rate increased from near zero to 10-100 m3/hr. During the peak in activity, cyclic explosive bursts 2-6 minutes long were separated by shorter non-eruptive pauses lasting 10-100 seconds. The size of the ejecta increased with the vigor of the explosions. A portable hydrophone deployed near the vent recorded sounds correlated with the explosive bursts; the highest amplitudes were at frequencies between 10-50 Hz, but extended up to at least 1500 Hz. The acoustic data allow us to quantify the durations, amplitudes, and evolution of the eruptive events over time. The low eruption rate, high gas/lava ratio, and rhythmic eruptive behavior at NW Rota-1 are most consistent with a Strombolian eruptive style. We interpret that the eruption was primarily driven by the venting of magmatic gases, which was also the primary source of the sound recorded during the explosive bursts. The rhythmic nature of the bursts can be explained by partial gas segregation in the conduit and upward migration in a transitional regime between bubbly flow and fully-developed slug flow. The strongest explosive bursts were accompanied by flashes of red glow and oscillating eruption plumes in the vent, apparently caused by magma- seawater interaction and rapid steam formation and condensation. These data are unique because this is the first time submarine explosive eruptions have been witnessed with simultaneous near-field acoustic recordings.
V21C-0729
Melt CO2 Enrichment by Permeable Flow and Resorption
Matrix and melt inclusion glasses commonly have CO2/H2O ratio values that are too high to be explained by standard closed- or open-system degassing. We explore the possibility of increasing CO2/H2O in melts through cycles of decompression, permeability development and gas flow, followed by repressurization and volatile resorption. As a case study we consider the spring and summer 1980 eruptions at Mt. St. Helens. Glasses from the climactic May 18 eruption show the least CO2 enrichment whereas the next eruption, on May 25, as well as several other eruptions, show substantial CO2 enrichment. We suggest that the episodic nature of the eruption sequence could have caused portions of the magma to experience multiple pressure cycles leading to overall increases in the CO2/H2O content of the melt. The proposed cycle begins with magma containing isolated bubbles with gas pressures similar to the surrounding melt. Depressurization of the magma (e.g., by edifice collapse and eruption of more shallow magma) causes sufficient vesiculation for bubbles to touch and form apertures, creating a connected network of gas. As gas escapes, the pressure of the vertically-connected gas falls from the melt pressure (related to the weight of magma above) down towards a gas-static pressure. The resulting discrepancy between the melt and gas pressures will cause 1) volatile diffusion from the melt into the relatively low-pressure bubble phase, and 2) bubble collapse and thus permeability reduction. Once the bubbles become isolated by bubble collapse, gas pressure will return to magmastatic values, driving volatile resorption back into the melt. CO2/H2O enrichment can occur if the gas added, which comes from deeper magma, has a higher CO2/H2O content than the gas lost by flow upwards. To assess the feasibility of increasing CO2/H2O ratios by this process, requires comparison of three timescales: 1) gas escape, 2) bubble collapse by viscous flow of melt and 3) diffusion of volatiles through melt into bubbles. As gas escapes, the gas pressure gradient decreases and gas flux reduces. In contrast, the pressure difference driving volatile diffusion into the bubbles, and bubble collapse (pressure of melt - pressure inside bubble) will initially be small, but will increase as gas escapes.
V21C-0730
Bubble - Crystal Interactions in Magmatic Three-Phase Systems
The influence of crystals on the movement of bubbles through basaltic magmas is poorly understood. We study the interaction of bubbles with a suspension of crystals in a viscous fluid through analog experiments. In our experiments, an air bubble rises through a suspension of plastic beads in a viscous corn syrup - water mixture; we vary bubble volumes, crystal spacings and fluid viscosities. We observe the following change in interaction styles with increasing bubble volume: (1) bubble migration through the crystal network with little bubble deformation, (2) bubble movement through the crystal network with deformation (and sometimes bubble splitting), and (3) displacement of the liquid-crystal mixture by the rising bubble. Interactions change from type (1) to (2) when the bubble is approximately the same size as the crystals forming the network. Transition to type (3) behavior depends on both bubble volume and the thickness of the crystal-liquid layer. In all cases, bubble rise is impeded by the presence of crystals. Preliminary results suggest that impedance is most pronounced for bubbles slightly larger than the crystals (a condition that promotes the maximum bubble deformation). Additionally, very small bubbles may be trapped for long times in the crystal network, suggesting that a shallow reservoir of crystal-rich magma may actually trap rising bubbles from below. These observations provide an alternative interpretation to that of small undeformed bubbles representing late-stage bubble nucleation and large irregularly shaped bubbles forming by coalescence of smaller bubbles (e.g. Lautze and Houghton, 2006). Furthermore, we observe in our experiments that large bubbles can spread out and move laterally underneath a crystal layer. This is not usually considered in models of bubble migration and may explain focusing of gas escape from magma reservoirs and volcanic vents. We apply our experimental results to analysis of bubble populations at Stromboli volcano, Italy, where gases rising from a deep crystal-poor magma reservoir travel through, and entrain, shallow crystal-rich magma. Preliminary results from image analysis on SEM and optical microscope images suggest that the smallest bubbles are most abundant and that their sizes are within the modal size range of the crystals. This can mean that (i) smaller bubbles were initially more abundant or (ii) bigger bubbles have deformed and split into smaller bubbles, which would confirm our conclusion from the analog experiments that interactions change from type (1) to (2) when the bubbles reach the size of the crystals in the network. In addition, the abundance of crystal-size bubbles in these samples suggests relative accumulation, possibly through trapping or extreme impedance, of the bubbles within the crystal-rich layer. Combined with our observation (from the analog experiments) that most bubbles are indeed significantly slowed within the crystal layer, this could suggest that bubble number densities are not direct reflections of bubble nucleation rates.
V21C-0731
Gas slug ascent as a source for very long-period seismicity: a numerical investigation
Very long period (VLP) signals provide a unique insight into the source process of seismic signals recorded on active or restless volcanoes. This is due to the large wavelengths that may allow the structural heterogeneity to be ignored hence source inversions can be calculated. The source process involved in such events is thought to be related to magma transport, gas slug ascent or dike resonance initiated by a fluid driven process. Numerical modelling of gas slug ascent (constrained by results from published laboratory experiments) shows that this process is capable of generating VLP like signals. Inputting the fluid dynamic derived forces into an elastic medium, with ‘volcano' like topography, VLP-like signals are reproduced. A moment tensor inversion is performed on these synthetic VLP signals. The moment tensor inversion produces a Compensated Linear Vector Dipole (CLVD) source mechanism for a pipe structure.
V21C-0732
Vesicle evolution in primary volcaniclastic material from Loihi seamount, Hawaii, with implications for submarine basalt explosivity
The extent and timing of magmatic degassing in ascending basaltic melts has implications for the rheological and physical properties of the magma at the time of eruption, and thus are controls on eruptive style. They are relatively well understood for subaerial eruptions, but have not been rigorously assessed for submarine ones. We present a quantitative textural evaluation of vesicular fragmental basalt collected at ~1140 mbsl on Loihi seamount, Hawaii, which provides insight into the microtextural evolution occurring within individual clasts. Using these clasts, we determine rates of vesicle growth by quantifying the advancement of a seawater-induced cooling front. The work presented here builds toward a semi-quantitative model explaining the origin of primary fragmental submarine deposits that share common features with the products of explosive (Hawaiian and/or strombolian) subaerial eruptions. Representative lapilli-sized Loihi clasts have vesicularities ranging 47-69%, with a well-defined modal vesicularity of 55%. These values are significantly higher than previously reported for lava flows and pillow lavas on submarine Hawaiian basalt of similar composition, despite eruption depths exceeding 1 km. Clasts of modal vesicularity are found to have margin-parallel zones with vesicle and groundmass textures that mature from rim- to-core over several cm. Seawater-quenched rims are dominated by small close-packed sub-spherical vesicles (Na=1.5×103 cm-2, median diameter ~400 μm), hosted in sideromelane glass, and grade through 1 or 2 intermediate zones to texturally mature cores dominated by fewer, large, amoeboid vesicles (Na=1.7×102 cm-2, median diameter ~1650 μm) in a tachylitic groundmass. We infer that the rims represent the texture of the magma at the time of eruption and fragmentation, and hence the starting point from which the more mature textures evolved during cooling. Calculation of conductive cooling rates for basalt clasts in contact with water (+/- steam) indicates that the cores of Loihi clasts took on the order 1 to 10 minutes (~102-103 s) to reach the glass transition temperature (Tg). Below Tg, glass yield strength in the solidified melt inhibited further expansion, migration, or coalescence of vesicles. Vesicle quantification through the margin-parallel zones allows us to assess the rate-limited relative roles of vesicle nucleation, growth, coalescence, and loss, as well as associated rates of groundmass crystallization.