V12B-01
Segregating Gas from Melt: an Experimental Look at Ostwald Ripening in Rhyolitic Magma
Magmatic degassing is a critical process that controls explosivity of volcanic eruptions, volcano deformation, and long-period seismicity. Degassing is promoted by increasing bubble size, which is attained through 1) decompressional expansion, 2) coalescence, and/or 3) Ostwald Ripening. Although rarely considered, ripening may be an important mechanism of bubble growth and segregation in viscous magmas during sub-volcanic storage. A series of piston-cylinder experiments was conducted to examine the influence of ripening on the segregation of CO2-rich gas from crystal-free rhyolitic magma residing at mid to upper crustal levels. Experiments were run at 400 MPa and 850°C, using a sample from Panum crater (Long Valley, CA) with added ~3 wt% CO2, ~2 wt% H2O, and run times of 1 day, and 1, 2, and 4 weeks. Nucleation of gas bubbles occurred in each experiment; however there was little to no migration or coalescence. Instead, each sample showed a progressive increase in bubble size and decrease in bubble number density due to ripening. Sample vesicularity ranges from ~ 15-20%, number density from ~104-105 mm-3, and bubble size from ~1-150 microns. These quantitative data were obtained through image processing of 2 dimensional images of sample cross sections. 3-dimensional x-ray microtomography images for the samples have been obtained at the Advanced Light Source; Lawrence Berkeley Lab. Current work to process such images will be commented on. Preliminary data are in good agreement with theory for diffusive-controlled ripening, which predicts average radius increases at a rate of t1/3 and number density decreases proportional to t-1. The trend for our data suggests that bubble radius will increase one order of magnitude in 100 years, which, according to Stokes Law, translates to a two order of magnitude increase in buoyant rise velocity. Ripening therefore appears to be an effective mechanism in generating buoyancy-driven migration of bubbles to the top of reservoir, where gases could leak from the magma to generate seismicity and/or feed hydrothermal systems. These data also show that experiments designed to investigate coalescence on bubble size distributions should take into account bubble growth due to ripening.
V12B-02
Numerical Reconstructions of Volcanic Tephra: Using Tomography and Two-Point Correlation Functions to Determine Magma Permeability
Understanding the processes that cause volcanic eruptions to be either effusive or explosive is vital to improve predictability and minimize hazards associated with volcanic eruptions. Explosivity appears to be linked to gas pressure build-up within magma, which is in turn affected by the degree of degassing of magmatic volatiles through permeable bubble networks or fractures in the magma. Magma permeability inside a volcano conduit is typically estimated experimentally by measuring the permeability of small pumice clasts (e.g., Klug and Cashman, Bull. Volcanol., 1996). However, permeability has been shown to be very scale-dependent (e.g., Hyun et al., Water Resor. Res., 2002), leaving substantial uncertainty in magma permeability-dependent calculations, such as magmatic volatile degassing rates. The objective of this study is to up-scale the permeability and microstructure (bubbles and crystals) of volcanic ejecta, and related magma degassing rates, to volcano-conduit scales. This is achieved by creating a numerical reconstruction method using X-ray tomography images of pumice clasts and two-point correlation functions. These numerical reconstructions reproduce the statistics of the spatial relationships of bubbles found in a given pumice clast. Once the bubble network is reconstructed, we are able to determine the porosity, tortuosity, and specific surface area of the bubble networks in the numerical reconstruction. In addition, lattice-Boltzmann simulations can be employed to numerically determine the bubble network's permeability.
V12B-03
Modeling the near-surface expansion of gas slugs in basaltic magma
The ascent and burst of large gas bubbles (slugs) in low-viscosity magmas generates a variety of geophysical signals that can be used to investigate sub-surface processes. Many of these signals are continuously recorded at several volcanoes (e.g. seismic and infrasonic data at Stromboli and Erebus) and, in order to interpret them in terms of slug parameters, models of the fluid flow associated with slug ascent are required. Due to the difficulties involved with two-phase flow modeling in complex conduit geometries we present two initial slug ascent models, a 1D mathematical model and a 3D computational fluid dynamic (CFD), based on ascent within vertical tubes. The performance of the models is assessed by comparing the results with data from laboratory experiments. These models are anticipated to capture the first order processes and represent the foundation from which more complex models can be derived. In order to simulate near-surface expansions appropriate to volcanic scenarios, the laboratory experiments were carried out under reduced surface pressure conditions. Following the observations made, the 1D model defines a constant rise velocity for the base of the slug and calculates the gas expansion during ascent, allowing the slug nose to accelerate through the overlying fluid. The model reproduces the evolution of rapidly expanding gas slugs observed in the experiments well and, at volcano scales, indicates that at-surface overpressures of several bars can result from the ascent processes alone. The 1D model accounts for the viscous and inertial forces resulting from bulk fluid motions but it does not provide details of the flow dynamics. Consequently, pressure distributions and forces exerted on the conduit cannot be realistically represented unless full 3D CFD simulations are carried out. CFD models are much slower to run, but calculate pressures and shear forces exerted throughout the conduit. At volcano scales, the vertical single forces during slug ascent are ~106 N, two orders of magnitude smaller than those associated with very-long- period seismic events at Stromboli. This supports a previous interpretation of these events in which they are generated by gas slugs flowing through changes in conduit geometry, rather than being the direct result of slug eruption processes.
V12B-04
Experimental Studies for Modeling the Explosions of Basaltic Volcanoes
In order to study the processes taking place within feeding systems of basaltic volcanoes during the ascent of magmatic melt to the surface, the Complex Apparatus for Modeling Basaltic Eruptions (CAMBE) has been developed. The purpose of the experiment is to study the processes taking place during the formation and ascent of gas-liquid mixtures within vertical conduits as an analogue to the flow of liquid basaltic magmas within volcano feeding systems. CAMBE has been assembled at the Institute of Volcanology and Seismology RAS (Petropavlovsk-Kamchatsky, Russia). The device is 18 meters high, and consists of two major systems – modeling and recording. For the convenience of discussion of experimental studies, components of the model system are named according to volcanic terminology. The modeling system includes a "chamber zone" (tank for preparing gas-saturated model liquid), a "feeding conduit" (transparent hose), and "crater area" (unit for accepting the supplied model liquid). Processes occurring within the volcano feeding conduit are modeled in this part of the device. The ratio of working section of the hose's inner diameter to its height is about 1:1 000, which is close to actual parameters of volcanic feeding channels. The recording system consists of several devices for dynamic video observations; electronic altimeter, speedometer, video-recording unit, acoustic recording unit, synchronizing device, and shut-down system. For the first time during physical modeling, conditions have been created for the supply of moving model gas- saturated liquid into the conduit, which allowed the study of bubble nucleation, growth and coalescence, as well as the formation and transformation of gas structures, and the kinetic peculiarities of gas phase evolution. The experiments resulted in detecting and describing a new, never before known, mode of gas-liquid two-phase flow in a vertical column – defined here as cluster regime, which is characterized by regular alteration of dense gas bubble clusters separated from each other by the liquid not containing free gas phase. The mechanism of the cluster regime formation is conditioned by the processes of blocking of the hose working section by one big bubble or several smaller ones. It has been demonstrated that liquid, bubble, cluster and slug regime are regularly sequential and present polymorphic modifications of gas-saturated liquids migrating within vertically oriented conduits. The surface (crater area) manifestation of cluster or slug regimes leads to basaltic explosions typical both for Strombolian and Hawaiian eruption types. Analysis of available data on explosions at basaltic volcanoes, given the obtained experimental data on the mechanism of this process, allows rendering the genesis of basaltic explosions from a new viewpoint. Polymorphic gas-hydrodynamic transformations within a vertical volcano feeding conduit result in the development of cluster and slug regimes, accounting for the mechanism of basaltic explosions.
V12B-05
Quantification of the Gas Mass Emitted During a Single Explosion on Stromboli with the SO2 Camera
The recent development of the SO2 camera is a powerful new tool for volcanologists that allows quantitative imaging of SO2 amounts within volcanic plumes. Such SO2 images are information-rich compared to traditional measurements by COSPEC or compact UV spectrometer systems which measure the equivalent of one pixel in an SO2 image. Furthermore, using time-series of SO2 images, we can examine short- lived events on the order of seconds and quantitatively constrain wind speed, the greatest source of error in flux determination. The new capabilities offered by the SO2 camera herald a major step forward for understanding degassing processes and therefore civil defence. On October 3 and 4, 2006, we performed SO2 imaging measurements at Stromboli volcano, Italy. The SO2 camera was placed on the north-eastern flank of the volcano about 2.4 km from the summit craters and recorded SO2 images with a time-interval of 5.0 seconds on Oct. 3 and 3.2 seconds on Oct. 4, respectively. More than 1500 SO2 plume images were collected. Major variations in SO2 flux were observed over a timescale of a few minutes and several peaks were seen. By relating the major increases of the flux with a movie of the SO2 plume images, we can conclude that the main cause of these variations was strombolian explosions. In the afternoon of Oct. 3, we took the SO2 camera to the summit of the volcano. The aim of the measurements was to quantify the amount of SO2 emitted during a single explosion. The SO2 camera was set up 400 m north of the craters and a single UV band-pass filter (Center wavelength: 310 nm, HWHM 10 nm) was used in the measurements. We obtained more than 600 plume images with time interval between frames of ~2 seconds. By analyzing the SO2 images, we could separate several explosive events from the background quiescent degassing. The amount of SO2 emitted by a single explosion was 15 - 40 kg which corresponds to approximate total gas mass of 350 - 900 kg. Assuming that the gas slug originated from ~75 MPa (as demonstrated by recent FTIR measurements on Stromboli) we may calculate the volume of the slug at that pressure. Using the ideal gas law and assuming a spherical bubble in the conduit, at a pressure of 75 MPa and temperature of 1150°, this gas amount will occupy a sphere with radius 0.8 - 1.1 m.
V12B-06
Characterisation of juvenile pyroclasts from the Kos Plateau Tuff (Aegean Arc): insights into the eruptive dynamics of a rhyolitic caldera-forming eruption
Silicic pumices formed during explosive volcanic eruptions are faithful recorders of the state of the magma in the conduit, close to the fragmentation level, as viscosity is generally high enough to limit their post-fragmentation deformation to a minimum. Pumice textures are therefore of great interest to shed light into bubble nucleation, growth and coalescence in silicic magmas ascending in a volcanic conduit, which ultimately controls the style of eruption. In this study, we have characterised the pumices produced by the rhyolitic 161 ka Kos Plateau Tuff (KPT) eruption, Aegean Arc, Greece. Four types of pumices were distinguished macroscopically in the non-welded deposits, and have been characterized using thin section observation, SEM imagery, porosimetry, and permeametry. We show that the types of pumices defined are confirmed by distinct petrophysical characteristics, and the measured differences in porosity and permeability are the result of either conduit processes, differences in crystallinity or magma mixing. We also show that permeability does not depend solely on porosity, as implied by the percolation theory. Size of pore aperture, tortuosity and pathway wall rugosity also play a fundamental role in the flow of gas through the permeable magmatic foam.
V12B-07
Systematic gas flux cycles observed using very high temporal resolution DOAS and UV camera imagery at Santiaguito Volcano, Guatemala.
Cyclic gas production has been observed at Santiaguito volcano, Guatemala, using differential optical absorption spectroscopy (DOAS) and UV imaging systems. Santiaguito explodes regularly, on the order of once an hour, and produces weak ash plumes, that are mostly buoyancy driven, to an altitude of 1-2 km above its dome. Dara from campaigns in 2004-2006 yield striking similarities in the production of SO2 across an explosive cycle. This is only observable due to the instruments' rapid data acquisition of no more than a few seconds per measurement. There is an initial increase in output associated with the explosion (more than 2 kg per second SO2) which quickly returns to ‘background' levels (between 0.1-0.5 kg per second SO2). These background levels vary systematically between explosions and have a sinusoidal pattern. Normalization of the gas flux to a constant rise rate indicates the maximum passive output occurs at approximately 0.4-0.7 cycles which is followed by a significant drop in emission rate before the next explosion. These methods have the potential to provide real predictive capability at Santiaguito and provide insight into shallow processes governing gas production and release in quasi-open systems. Three potential mechanisms exist to explain the length of time the system requires to reset – (i) pistoning associated with unsteady flow, (ii) resealing of cracks and subsequent pressurization and (iii) defluidisation of the porous cap after perturbation. Each has its own merits and issues when explaining the observed gas cyles.
V12B-08
Model for high frequency Strombolian tremor inferred by wavefield decomposition and reconstruction of asymptotic dynamics
We study the volcanic tremor time series recorded by a broadband three component seismic network installed at Stromboli volcano during 1997. By using decomposition methods in both frequency and time domains, we prove that Strombolian tremor can be described as a linear combination of nonlinear signals in time domain. These "components" are similar to those obtained for explosion-quakes, with the only difference being the amplitude enhancement. We characterize each of these nonlinear signals both in terms of their wavefield properties as well as dynamic systems. Moreover, we take into account the complex processes of magma flow and turbulent degassing, looking at time and amplitude modulation of tremor on a suitable scale. The distribution of tremor amplitudes is Gaussian while the inter-times between the maxima in a suitable scale are described by a Poisson clustered process. Starting from these analyses, a first approximate model for volcanic tremor field can be deduced. The recorded signals, i.e. the elastic vibrations at a point, can be described by a nonlinear equation which gives limit cycles (different observed "nonlinear modes"). This equation is governed by a time dependent threshold which represents the variability of bubble flux. We take into account some inelasticity in the medium perturbing the elastic potential with a Gaussian function on a suitable scale. It acts as a radiance function modulating the frequency of the limit cycle. This proposed model is able to reproduce waveform, Fourier spectrum, and phase space dimension of one of the extracted nonlinear wave packets.