V31E-0702
Turbulent dynamics and pyroclastic flow generation during the Mount St. Helens May 18th, 1980 eruption
Eruption behavior at Mount St. Helens changed greatly over the course of 18 May: a buoyant, Plinian column dominated the morning phases of eruption, whereas during the early afternoon, the column partially collapsed, such that a Plinian column and non-buoyant pyroclastic flows were simultaneously erupted. Changes in the plume's turbulent flow dynamics, pyroclastic fall and flow deposit grain size distributions (GSDs), and character of the plume reflect this evolution in eruption dynamics. Optical flow velocimetry of video of the plume immediately above the crater rim indicates the sizes of the largest structures in the plume decreased from a range of ~300 to >1000 m during the morning to 150-200 m during the afternoon. These measurements agree with visual inspection of photographs showing eddy size decreasing from a range of 200 to >500 m (average 300 m) in the morning to a range of 150-350 m (average 250) in the afternoon. During this same time interval, the rotation speed of eddies (as measured by the rms values of the 2D velocity field) increased by a factor of 1.6. Furthermore, the appearance of the column changed through the course of the eruption. In the morning, the column was characterized by discontinuous, large eddies frequently depositing "curtains" of pyroclasts, and an indentation was present on the column's southern margin. In contrast, the column margins were completely covered by smaller eddies and no curtains of sedimenting pyroclasts during the afternoon. Given that during the morning most mass erupted as buoyant plumes, we have estimated total eruptive GSDs from fall deposit GSDs using known mass fluxes and plume sedimentation models. Accounting for changes in buoyant mass flux and depositional axis, the afternoon Plinian fall deposits are 0.5 to 1 phi units coarser than models predict if the bulk, buoyant GSD remained the same. Although the majority of pyroclastic flows were emplaced to the north of the crater during the afternoon, smaller pyroclastic flows were emplaced over the rim of the crater. The GSDs of these "over-the-rim" flow deposits are significantly finer grained than those to the north, despite their shorter run out. Increased total mass flux caused the change from dominantly buoyant to dominantly non-buoyant behavior observed between the morning and afternoon of 18 May. This increase was likely the product of increases in at least two components of mass flux: column density and mean velocity. Shorter length scales of turbulence observed in the afternoon reflect shorter Kelvin-Helmholtz type instability wavelengths resulting from eruption of a denser jet into the atmosphere. Increases in jet velocity are indicated by the increase in turbulence intensity. Coarsening in bulk, buoyant plume GSDs and different groups of flow GSDs suggest either less efficient fragmentation during the afternoon, thus a comparatively coarse-skewed GSD erupted from the vent, or size segregation of pyroclasts between buoyant and non-buoyant parts of the column.
V31E-0703
Eruptive and Transportation Processes During Caldera-Forming Eruptions of Sete Cidades Volcano, São Miguel, Azores
Sete Cidades volcano forms the Western part of the island of São Miguel, Azores, which is hosting three active trachytic central volcanoes (Sete Cidades, Fogo, Furnas). Volcanic activity in the archipelago exhibits a strong tectonic control and on São Miguel, the NW-SE trending basaltic Terceira Rift is intersecting the central volcanoes. All three have erupted since the settlement of the island in the 15{th} century. The Eastern part of the island is considered extinct. The oldest dated subaerial rocks of Sete Cidades exhibit an age of 210 ka. Morphology of the present summit caldera (5 km diameter, up to 350 m deep), stratigraphy, and distribution of the deposits suggest a multiple-stage evolution and at least three caldera-forming eruptions (CFE) are assumed to have occurred. 14C-dating revealed ages of 36, 29, and 16 ka, respectively, for the most recent ones. Today, the average slope angle is 12° and the maximum distance of the coastline from the caldera rim approx. 5 km. Assuming a comparable situation at the time of the CFE, a large portion of the eruptive products has probably not been deposited on land. After a pause of several thousand years, eruptive activity resumed approx. 5 ka ago and started filling the caldera. As deposits of minor thickness and distribution can be found between the deposits of the CFE, it is unclear whether the caldera formation is completely finished. Climatic factors (e.g. precipitation, air humidity) have affected the deposits by erosion, weathering, and possibly significant reworking and caused dense vegetation on all flanks of the volcano. Still, it was possible to establish distribution and thickness of the deposits of the CFE and constrain differences in eruptive behaviour and transport/emplacement mechanisms. They are composed of air-fall deposits and pyroclastic density currents but show significant differences amongst them: (1) Degree of pre- and syn-eruptive magma-magma interaction and syn-eruptive magma-water interaction. (2) Ratio of juvenile/lithic content and basaltic/trachytic magma. (3) Degree of vesiculation and crystal content of the juvenile material. (4) Percentage of air-fall deposits within the deposits of a single CFE and the timing of their deposition. (5) Distribution of air-fall deposits. (6) Degree of welding. The results highlight the bandwidth of possible eruptive scenarios at this trachytic central volcano cut by an active rift. Based on the study of these eruptions, volcanic hazard maps can be produced that are essential for adequate risk assessment.
V31E-0704
Experimental and Numerical Investigation of the Boundary Conditions of Over-Water Pyroclastic Flows
In over-water pyroclastic flows the ability of particles to sink through the water interface influences the speed of the flow and the distance it travels. In order to understand how the water interface affects flow mobility, we performed lab measurements to characterize the dynamics of particles at the water interface and then used these experimental results in numerical simulations. To quantify the dynamics of particles that impact the water surface we perform a series of 150 experiments of pumice-water collisions. We measure the fraction of particles that bounce and the energy lost following the collision using high-speed photography. We vary the collision angle, impact speed, particle mass and shape. We find that the restitution coefficient (the fractional decrease of speed) is insensitive to the initial speed and particle mass, but does depend on the collision angle. The amount of energy lost during the collision is proportional to the amount of time that the particle spends in contact with the water during the collision. We fit empirical models for the fraction of particles that sink and the fraction of energy lost during the collision to the experimental data. Similar experiments are currently being conducted for collisions with a bed of pumice particles. In order to determine how far and how fast flows will travel we use the results of continuum numerical simulations, similar to those in Dufek and Bergantz, 2007 (Journal of theoretical and computational fluid dynamics). We integrate probability distributions for sinking and energy lost over the velocity distribution at the water interface to determine the sink terms in the mass, momentum and energy equations. Flows that have what we think are realistic boundary conditions travel less than ~ 10 % farther and faster than a completely leaky end-member due to the high rate of energy dissipation; they have however ~ 50-60 % less runout distance than equivalent overland flows.
V31E-0705
The Rheology of Phlegrean Field Magmas
In order to quantify the effect of crystals on the rheology of Monte Nuovo magma, a series of torsional and compressional deformation-experiments have been performed in an internally heated Paterson-type apparatus at the ETH (Zurich). The porosity present in the scoriaceous samples has previously been removed, annealing the samples at 300 MPa confining P and 1073 K. Unconfined uniaxial compression experiments were also performed at the Geological Sciences Dep. of the University of Roma Tre; the EOS department at the UBC, Vancouver and the LMU, Munich. The three presses are complementary in nature, with different maximum loads (10, 44 and 300 KN, respectively) and displacement rates (from 10-1 to 10-6 cm/s, for the first two presses and 100 to 10-8 cm/s for the large press in Munich). Results from multiphase rheology of bubble-free (Paterson) and vesicle-bearing (uniaxial presses) samples show that viscosity of crystal-bearing samples is affected by the strain rate, showing clear non-Newtonian behaviour (shear-thinning) already at low deformation-rates. Within the regime of applied strain rate no yield stress can be observed. In addition to the multiphase rheological investigation, we have also investigated the liquid and, where possible, the liquid+crystal rheology of remelted products of Agnano Monte Spina (AMS), Monte Nuovo (MNV) and Fondo Riccio (FR) . Viscosities of pure liquids and liquid+crystals mixtures within the range 101-105 Pa s and T between 1100 and 1500° C were measured by concentric cylinder. Low-T viscosity was measured using the micropenetration technique. Isothermal holds measurements on MNV and AMS samples did not show any significant effect on the viscosity due to sample crystallization. However, viscosity measurement on FR samples at variable rates of strain suggest that non-Newtonian rheology occurs even at low crystal contents. In addition to evaluating the onset of non-Newtonian rheology, isothermal holds viscosity measurements can also define a time-temperature-viscosity window over which crystallization occurs. Results indicate that crystallization is dominantly governed by T (favouring) rather than viscosity (inhibiting). Below 1150° C a very rapid viscosity increase occurs due to a high crystallization rate.
V31E-0706
Dynamics of Experimental Rapid Shear Flows
Pyroclastic flows are dense mass flows of hot (< ~ 700° C) particles and gas generated by volcanic eruptions. Their fluidal behavior is attributed to high interstitial gas pore pressures and associated fluidization effects. We carried out experiments on the dynamics of laboratory-scale flows of fluidized volcanic ash at 170° C, which is hot enough to render negligible effects of humidity-derived interparticle cohesion. The flows were generated in a 3.5-m-long, horizontal lock-exchange flume built of aluminum and pyrex. The ash was expanded by fluidization to 0 to 42 vol % above loose packing, then released down the flume as thin (< 10 cm), but fast-moving (1.5-2.3 m/s) shear flows. Since the floor of the flume was impermeable, the ash defluidized progressively until motion ceased. A key feature of the flows was that deposition occurred by progressive sediment aggradation at the flow base. Particle hindered settling velocities in the flows were indistinguishable within error from those measured in (static) 1-D bed-collapse tests at given values of expansion. In order to investigate the flow dynamics in more detail, we filmed a number of experiments using a high-speed (1000 frames per second) video camera. By repeating each experiment several times, we were able to observe the sedimentation behavior both as function of time and with distance down the flume in the short-lived and highly unsteady flows. This was possible because the experiments were reproducible. Using specially designed image analysis software, we measured the sediment aggradation rates and the temporal and spatial evolution of velocity fields and velocity gradients in the flows. Video analysis shows a distinct deposition behavior between expanded flows and non expanded flows. Flow duration was governed by the ratio of two times: tsett, a characteristic time of hindered settling and tgrav, a characteristic time of gravitational acceleration. Analyses confirme that this ratio increases with increasing initial expansion.
V31E-0707
Ash plume dynamics at Santiaguito volcano, Guatemala, from thermal and high resolution video
The regular (1-2 per hour) explosion plumes at Santiaguito volcano provide a valuable opportunity to track ash plume dynamics continuously from vent levels to neutral buoyancy heights, permitting a rare opportunity to analyze virtually the entire rise lifespan of an eruption plume. In January and June, 2007, we imaged 184 plumes at Santiaguito at frame rates of 7.5-30 fps, using both a thermal (FLIR) camera and a conventional camcorder situated at the Santiaguito observatory (6.5 km distance) and a second thermal (Variocam) camera at Santa Maria summit (2.5 km distance). The plumes we analyzed had near-vent velocities of 5-15 m/s, resulting in neutral buoyancy heights of approx. 0.5 to 2 km. Initial analyses of plume motion and dimensions suggest that bulk densities of the plume attain near-ambient values within a few hundred meters height, suggesting rapid and efficient air entrainment in initial stages of the plume rise. In a qualitative sense, the thermal data show daily variations in January in the character of the eruption plumes, ranging from robust, impulsive plumes with a well- developed plume front to more diffuse, emergent behaviors. We are currently investigating how these trends in velocity and character relate to concomitant infrasound and seismicity. It appears as though the relatively low intensity infrasound pressures associated with explosions (< few Pa at 1 km) reflects relatively low inertia at the vent and a plume that is largely buoyancy-driven.
V31E-0708
Effect of Particle Size on Tephra Sample Drying
The analysis of excavated tephra and other pyroclastic material often includes the removal of environmental moisture that has migrated into the individual clasts subsequent to their deposition. Elimination of the moisture content of a sample allows researchers to gather quantitative data concerning the mass, density, and particle size of the ejecta at the time of the eruptive event. Progress in the modeling of past eruptions using stratigraphic evidence has increased interest in this type of data. Many of the formulae used in the modeling of ash fall from volcanic events include variables for the density and particle size (Carey and Sparks 1986) of the ejected material. Hence, the first step of data processing in the laboratory is drying of the tephra and proceeding to a particle size analysis. While standards for the removal of moisture from soil and rock samples exist for use in the geotechnical and construction fields (ASTM D 2216-98), no methodological guidelines have been created for use in the field of volcanology. The method described in ASTM D 2216-98, in which samples are baked at a standard temperature of 110° C until they reach a constant mass, is similar to that used by researchers studying tephra. However, lack of access to the exact apparatus specified by ASTM D 2216-98 in many volcanology and geophysics labs, as well as the fact that it is not specifically concerned with pyroclastic material, may prohibit acceptance of this standard by researchers. This paper provides information that may assist in the construction of a future standard of drying volcanic ash fall deposits. We tested whether the location of moisture within the bulk tephra sample is affected by particle size and what effect the unequal distribution of moisture between different sized particles has on drying times. The pyroclastic material used for this study was taken in the summer of 2007 from the Inyo Crater area of California. The tephra was deposited by a series of Plinian and sub-Plinian eruptions approximately 550 years ago (Miller 1985). It is composed of mostly pumice, with smaller amounts of lithics and glass. Samples were sieved in their naturally humid stage and immediately dried as individual particle size dependent groups to a constant mass in the laboratory at 150 degrees centigrade. The weight of the groups was taken every 30 minutes during drying. As a control, each of the dry groups was weighed and moisture was added until it increased the sample's mass by 10 percent. The samples were then re-dried with measurements again being taken at 30-minute intervals. Preliminary results indicate that large pumice particles contain a disproportionate amount of the deposits moisture.
V31E-0709
Revising Methods for Moisture Removal in Tephra Bulk Samples
The current standard for removing water content from tephra seems inefficiently long. Tephra, depending on composition, can be porous and requires less time than other rocks and soil to dry. The Standard Test Method for Laboratory Determination of Water (ASTM D 2216-98) content of soil and rock by mass is not accurate specific to tephra bulk samples. A study was conducted to provide valuable information on drying dynamics to aid a possible new guideline specific to this type of material. The pyroclastic material used for this study was taken in the summer of 2006 from the Inyo Crater area of California. The tephra was deposited by a series of Plinian and sub-Plinian eruptions approximately 550 years ago (Miller 1985). Tephra bulk samples from two eruptions, Obsidian Flow and Deadman Creek near Mammoth, California were tested in the lab. Both samples were the result of rhyolitic eruptions composed of pumice, obsidian, and other lithics. Particle size ranges from .01 mm to 4.0 mm. It was found after testing the existing standard (24h) that tephra bulk samples require much less time to dry in an oven than suggested. Maximum drying time in the oven was 6 hours, and as little as 5 hours for smaller samples. However, the drying time depends on the amount of water content the samples retained before being placed in an oven and the mass of the sample. It seemed that soaking the samples for a control experiment over night were more comparable to the samples that are freshly taken in the field. Samples that were just wetted with a defined amount of water and stirred before drying dried up in less than one hour. After experimenting in the lab with drying tephra samples, it was concluded that the current standard was not accurate for tephra bulk samples. They required much less time than were stated, only 5-6 hours compared to 16-24 hours. Samples were baked at 150 degrees centigrade until all water content was evaporated, as the mass of the samples was measured every 30 minutes and compared to pre-drying humidified mass of the tephra we started with.
V31E-0710
Calm Before the Storm? Immediate Identification of Volcanic Eruption Intensity: Promising Test of a New Monitoring System at the Active Volcano Popocatépetl, Mexico
Experiments by the Physikalisch Vulkanologisches Labor (PVL) in Wuerzburg, Germany, have shown that the intensity of violent volcanic eruptions, occurring when magma undergoes brittle fragmentation, is mirrored within brief electrical charges that can be detected on a short timescale (ms). Laboratory studies and certain explosion experiments offer the opportunity to calibrate the energy release of volcanic eruptions. Based on these results, a new high-precision, low-cost, real-time surveillance system is developed and tested at the active volcano of Popocat\´{e}petl, Mexico. This volcano, situated about 60 km southeast of Mexico City, offers excellent testing conditions, erupting regularly and intensively and violent eruptions are expected in the near future. The system, which detects short-term electrostatic field gradients (dc voltage against local ground), mainly consists of an antenna and a specially-designed amplifier. Depending on eruption intensity, as little as two or three eruptions will provide a sufficient amount of data. Amount, size, and shape of erupted particles give important indications about the physical fragmentation process which formed the pyroclasts, and hence about the type and intensity of the eruption. The evaluation and analysis of the samples collected at the volcano after each documented eruption will be carried out at the PVL. This physics lab, with a specially-designed experimental setup, allows controlled explosion experiments wherein rock from lava or bombs - related to the sampled pyroclasts - will be melted and subsequently brought to explosion. The energy released during these laboratory experiments will be calibrated to Popocat\´{e}petl using the ejecta volume of the observed eruptions, allowing a correlation of the actual energy release to the registered electrical field data. The aims of the project are: (1) quantification of individual magma properties of Popocat\´{e}petl (2) on-line measurement of mechanical energy release and mass flux and (3) immediate risk assessment of ongoing volcanic eruptions using permanent sensors located on volcanoes with pyroclastic eruptions.
V31E-0711
Large Scale Failures on Volcanoes of Kurile Islands: the First Data
Investigations of air and space images of volcanoes of the Kurile arc, supplemented by observations from a
vessel as well as by on-land field work on several islands, have allowed us to identify 23 active volcanoes with
well-preserved horseshoe-shaped scars formed by large-scale edifice failures. Breaches of most of the scars
(14 cases) range from 0.5 to 2 km wide, indicating moderate failure volumes – around 1 km
V31E-0712
Current State of an Intelligent System to Aid in Tephra Layer Correlation
We are developing a computer based intelligent system to correlate tephra layers by using the lithologic, mineralogic, and geochemical characteristics of field samples, to aid geologists in interpreting eruption patterns of volcanic chains and fields. The intelligent system is used to define groups of tephra source vents by utilizing geochemical data, and to correlate tephra layers based on lithostratigraphic characteristics. Understanding the eruption history of a volcano from stratigraphic studies is important for forecasting future eruptive behavior and hazards. In volcanic chains and fields with a complex eruptive history and no central vent, determining the spatio- temporal eruption patterns is difficult. Sedimentologic and chemical variability, and sparse sampling often result in relatively large variances and imprecision in the dataset. Lithostratigraphic and geochemical interpretation also depends on ones' level of expertise and can be subjective. The processing of lithostratigraphic features is conducted by a hybrid classifier, composed of supervised artificial neural networks (ANNs) combined within the framework of the Dempster-Shafer theory of evidence. Since lithostratigraphic features vary with distance from source, hypothetical vent locations are determined by using expert domain knowledge and geostatistical methods. Geochemical data are processed by a suit of fuzzy k- means classifiers. Each fuzzy k-means classifier assigns observations to multiple clusters with various degrees, called membership coefficients. The assignment minimizes a function of the total distance between the centers of clusters and the individual geochemical data patterns weighed by the membership coefficients. Improved clustering results of geochemical data are achieved by the fusion of individual clustering results with an evidential combination method. Lithostratigraphic data from individual tephra beds of the North Mono eruption sequence are used to test the effectiveness of the intelligent system for tephra layer correlation. Geochemical data from tephra bedsets of the Mono and Inyo Craters, CA, are used to test the effectiveness of the intelligent system for eruption sequence correlation. The intelligent system aids correlation by showing matches and disparities between data patterns from different outcrops that may have been overlooked in initial interpretations. Initial results show that the lithostratigraphic classifier is able to accurately differentiate known layers 76% of the time. Output from the lithostratigraphic classifier can furthermore be plotted directly as isopleth maps that can aid in rapid recognition of tephra layers as well as determination of eruption characteristics, e.g. eruption volume, plume height, etc. The intelligent system produces a useful recognition result, while dealing with the uncertainty from sparse data and the imprecise description of layer characteristics.
V31E-0713
Preliminary SAGE Simulations of Volcanic Jets Into a Stratified Atmosphere
The SAGE (SAIC Adaptive Grid Eulerian) code employs adaptive mesh refinement in solving Eulerian equations of complex fluid flow desirable for simulation of volcanic eruptions. The goal of modeling volcanic eruptions is to better develop a code's predictive capabilities in order to understand the dynamics that govern the overall behavior of real eruption columns. To achieve this goal, we focus on the dynamics of underexpended jets, one of the fundamental physical processes important to explosive eruptions. Previous simulations of laboratory jets modeled in cylindrical coordinates were benchmarked with simulations in CFDLib (Los Alamos National Laboratory), which solves the full Navier-Stokes equations (includes viscous stress tensor), and showed close agreement, indicating that adaptive mesh refinement used in SAGE may offset the need for explicit calculation of viscous dissipation.We compare gas density contours of these previous simulations with the same initial conditions in cylindrical and Cartesian geometries to laboratory experiments to determine both the validity of the model and the robustness of the code. The SAGE results in both geometries are within several percent of the experiments for position and density of the incident (intercepting) and reflected shocks, slip lines, shear layers, and Mach disk. To expand our study into a volcanic regime, we simulate large-scale jets in a stratified atmosphere to establish the code's ability to model a sustained jet into a stable atmosphere.
V31E-0714
A Computational Thermodynamic Model for Explosive Volcanism
The initial state of the system and the thermodynamic constraints under which the calc-alkaline Erciyes Volcano in Central Turkey evolved has been modeled using the MELTS algorithm. Results of MELTS simulations show that basaltic andesite magma with 3 wt% water and fO2=QFM+3 at 1 GPa pressure evolved by closed system fractional crystallization to produce 63 Km3 of dacitic and rhyolitic Plinian fall and pumice deposits (Sen et al., 2003). In order to determine the initial state of the system simulations of fractional crystallization of the most primitive basaltic andesite magma were carried out in the pressure range of 0.0001 to 1 GPa, fO2 range of QFM-3 to QFM+3 and at H20 ranging from zero to 4 wt%. The phase equilibria calculations that gave the best fit to natural rock compositions are P= 0.1 GPa, fO2=QFM+3 and H2O=3 wt%. Using these system parameters "parental" basaltic andesite magma was allowed to fractionate under closed system conditions starting at the liquidus temperature of 1107 C. MELTS calculations showed that water began to separate from the melt at 1027 C and the volume faction of water increased from 0.0049 at 1027 C to 0.69 at 767 C. Below this temperature the volume fraction of water increased to 5.4 and both physical and chemical conditions of the magma changed drastically, marking the explosive eruption of the volcano.
V31E-0715
Study New Pregress On Volcanic Phreatomagmatic Eruption
As an essential and important type of volcanic eruption on earth, phreatomagmatic eruption is characterized by groundwater-related explosive eruption and subsequent base surge deposit and maar lakes. Base surge deposit and maar lakes are widely distributed all over the world, and also in the Northeast China and the southern China. Study of phreatomagmatic eruption maybe dated back to 1921, and in the following over 80 years, many works have been done on phreatomagmatic eruption, using various of methods of volcanic geology, petrology, sedimentology, physical volcanology and digital modeling, to discuss its origin and mechanism. In this paper, we focus on the geological feature of the base surge deposit and dynamic mechanism of the phreatomagmatic eruption. When ascending basaltic magma meets with ground ( surface ) water, violent explosion would occur, this action was called phreatomagmatic eruption. The main product of this kind of eruption are maars and base surge. As to the base surge, it has long been treated as sedimentary tuff by mistake. Usually, base surge is distributed around maar, different from the distribution of sedimentary tuff. Typical phenomena of base surge caused by phreatomagmatic eruption can be observed through the detail field work, such as large-scale and low-angle cross-bedding, slaty-bedding, current-bedding and distal facies accretionary lapilli. In order to explain the dynamic mechanism of phreatomagmatic eruption thoroughly, we propose a simple model in this paper in light of the elasticity theory. Some conclusions can be drawn as follows: the larger the radius of maar, the larger the explosive wallop needed for the formation of maar is; provided that the radius of maar and depth of explosive point are limited, then the larger the area of contact surface between magma and groundwater, the stronger the explosive energy will be; if the explosive energy and area of explosive point are restricted, the larger the radius of maar, the greater the depth of explosive point can be inferred; when the explosive energy and radius of maar are qualified, the depth of explosive point decreases with increasing of the area of contact surface between magma and groundwater. As for the maximum stress, undoubtedly it should occur on the surface of the overlying formation.
V31E-0716
Changes in Explosive Eruptive Styles From Morne Trois Pitons, Dominica, Lesser Antilles
The island of Dominica, which is located in the central part of the Lesser Antilles island arc, has witnessed three large-volume Plinian eruptions within the last 100,000 years. One of these eruptions issued from the Morne Trois Pitons caldera and produced the widespread (>62 km2) Layou ignimbrite deposits. The Layou ignimbrite is exposed in two pyroclastic fans that extend NE and NW from the Morne Trois Pitons caldera (now infilled by a Pelean dome complex). Field observations of the NW fan show that the ignimbrite sequence can be subdivided into an unlithified and a lithified faces. The former occurs in distal exposures along the margins of the ignimbrite, whereas the latter is dominant in deposits that occupy central valley-fill locations. In its most distal exposure, the Layou ignimbrite overlies a thick sequence of fluviatile conglomerates. Here the basal unit is an approximately 18 cm thick bed of pumiceous lapilli fall. Lithic clast sizes in this basal unit are consistent with an eruption column height of circa 20-25 km. The basal fall deposit is overlain by 10 cm of ash fall that contains abundant accretionary lapilli (up to 8 mm in diameter), suggesting magma-water interaction and a phreatoplinian style of eruption. SEM imaging of these accretionary lapilli reveal two lapilli types: 1) an aggregate variety comprised entirely of ash (heterogeneous lapilli) and 2) an agglomerated variety composed of ash rimming a core of pumice (homogeneous lapilli). The heterogeneous accretionary lapilli contain concentric stratification of fine and coarse ash laminae suggesting pyroclastic cycling within the moist phreatoplinian column. Overlying these basal fall deposits are at least 3 unlithified pyroclastic flow units with a maximum thickness of over 20 m. Within the Layou Valley west of the caldera the flow units increase in thickness and reach a maximum of 180 m in the upper Layou Valley. The Layou sequence records transitions from plinian fall to phreatoplinian fall to ignimbrite during a single explosive event from Morne Trois Piton caldera. These transitions are likely related to changes in column buoyancy and degree of magma-water interaction.
V31E-0717
Multiple Eruptive Phases and Deposits of a Monogenetic Volcano: Tabernacle Hill Volcano, Utah, USA
Tabernacle Hill volcano, located near the eastern edge of North America's great basin, is one of a group of monogenetic small-volume (0.47 km3) basaltic volcanoes forming a long-lived (~ 1 Ma) north-south trending alignment in Utah's Black Rock Desert. Initial phreatomagmatic eruptions at Tabernacle Hill are reported to have begun 14,320 ± 90 years ago. The initial eruptive phase produced a tuff cone approximately 80 m high (1,511 m a.s.l.) and 1.5 km in diameter with distinct bedding layers. Recent mapping and sampling of Tabernacle Hill's lava and tuff cone deposits has been aimed at better constraining the sequence of events, physical volcanology, rheology, and geochemistry of this eruption. Blocks located on the rim of the tuff cone of mid-crustal and near-surface origin were mapped and analyzed to yield preliminary minimum muzzle velocities of 70-100 m/s. After the initial phreatomagmatic explosions, the eruption style transitioned to a more effusive phase that partially filled the tuff cone with a semi-steady state lava lake 200 m wide and 15 m deep. Eventually, the tuff cone was breached by the impinging lava resulting in large portions of the cone rafting on top of the lava flows away from the vent. Eruption onto the Lake Bonneville lake bed allowed the Tabernacle Hill lava flows to flow radially from the tuff cone and cover an area of 18.1 km2, producing a very uniform high aspect ratio (100:1) flow field. Tabernacle Hill lava flows are pâhoehoe flows with many large phenocrysts of olivine and plagioclase (>1 cm) and have an average thickness of 26.3 m. Subsequent eruptive phases cycled several times between effusive and explosive, producing scoria cones and more lava flows, culminating in an almost complete drainage of the lava lake through large lava tubes and drainback.
V31E-0718
The Dynamics of Explosions at Santiagutio Volcano, Guatemala
Dynamic processes and explosive activity at dacitic volcanoes that are erupting viscous magma is still a matter of intense research, not the least because such volcanoes produce violent explosions. Here we present Doppler radar observations recorded between Jan.\ 8 and Jan.\ 13 2007 during a multidisciplinary experiment at Santiaguito volcano, Guatemala. The experiment comprised velocities and density of the erupted material (Doppler radar, Univ.\ Hamburg), infrasound and high speed video data (Univ.\ of New Hampshire), seismic recordings (Univ. of New Hampshire and Univ.\ of North Carolina), and temperatures of the dome and eruption column (infrared camera, Univ.\ de Colima). The instruments, operating on a direct line of sight (video, thermal imager, and Doppler radar), were installed on the top of Santa Maria volcano, and were pointed down (inclination 27°) towards the active Caliente lava dome 1200 m below and 2600 m distant. During the observational period we recorded a total of 142 explosions from the dome with the Doppler radar. Here we focus on the Doppler radar data to characterize the different types of explosions. The radar transmits 50mW of power at a frequency of 24GHz enabling us to see particles down to a size of mm even at low concentrations. The data are processed such that we get information on the amount of energy reflected by the particles in the radar beam as a function of the velocity. Because the approximate field of view of the radar beam is 1.6 deg (70m width at the target distance) and covers only a portion of the 200 m wide dome, the orientation of the beam was changed several times during the experiment to observe different portions of the dome exhibiting different eruptive processes. Many of the pyroclastic explosions appeared to emanate from concentric ring shaped fractures comprising an inner and/or an outer ring on the dome. While focused on the center of the dome, we find that the explosions start with very low velocities (\<4m/s), indicating a slow uplift proceeding the main part of an explosion by up to 25s, and before activity can be detected by eye. This slow movement appears to be corroborated by examining high resolution video images of the dome just prior to pyroclastic emissions. Significantly higher velocities (~60m/s) mark the onset of the main phase of the explosion. Most of the explosions are very gas rich, and carry only small amounts of ash. This is deduced from the relatively low reflected energy when targeting the rising eruption clouds above the dome. Another type of explosion is observed when pointing the radar at the outer ring of the dome. Here we find that some of the explosions show two explosions following each other within 30 to 50s. Similar eruption characteristics were observed with the other observation methods, where especially high speed video footage proved as a valuable complement to the radar data.
V31E-0719
The 2006 Eruption of Fourpeaked Volcano, Katmai National Park, Alaska
On September 17, 2006 Fourpeaked Mountain, a glacier-clad stratovolcano with no known Holocene activity, produced a plume of steam, ash, and SO2, which rose to 6000 m above sea level. The plume was observed by eye-witnesses, seen on weather radar, and also appeared in SO2-sensitive satellite imagery. Concurrent with the plume, regional seismic stations recorded a swarm of volcano-tectonic earthquakes while an atmospheric infrasound signal was recorded 800 km away. Air and ground reconnaissance revealed a linear series of vigorously steaming vents in the summit glacier, stretching about 1 km down the north flank of the volcano. Debris flows, emanating from disrupted glacial ice, extended many kilometers downslope and contained hydrothermally altered clasts and clay minerals mixed with ice; the largest clasts approached 5 meters in diameter. Airborne gas measurements indicated SO2 emission rates of thousands of tonnes per day. Ash collected from the glacier surface contained crystal fragments and dense rock particles. Neither the ash nor the debris flows show evidence of juvenile material; most clasts appear to consist of hydrothermally altered material likely derived from rock similar to outcrops exposed near the summit. The eruption response over subsequent weeks and months included regular visual and gas observations and the installation of a basic seismic network and web camera. The ensuing weeks saw continued disruption of the summit glacier, day-long seismic swarms, and the continuation of notable SO2 levels. As of mid-2007, SO2 levels had dropped by more than 90% and seismic swarms have been replaced by scattered earthquake activity. Steam plumes continue, but have diminished in vigor dramatically over the last year. The entire staff of the Alaska Volcano Observatory, as well as other collaborators, contributed to this abstract and the Fourpeaked eruption response. The Alaska Volcano Observatory is a cooperative program of the U.S. Geological Survey, the University of Alaska Fairbanks Geophysical Institute, and the Alaska Division of Geological and Geophysical Surveys.