D. B. Dingwell, Y. Lavallée, K.-U. Hess, A. Flaws, J. Marti, A. R. L. Nichols, H. A. Gilg, and B. Schillinger Earth and Environmental Sciences, Ludwig-Maximilians-Universität, Theresienstr. 41/III, 80333 München, Germany Earth, Ocean and Ecological Sciences, University of Liverpool, Liverpool, UK Consejo Superior de Investigations Cientificas, Institute of Earth Sciences Jaume Almera, Barcelona, Spain Research and Development Center for Ocean Drilling Science, Japan Agency for Marine Earth Science and Technology (JAMSTEC), 2-15 Natsushima-cho, Yokosuka, Kanagawa 237-0061, Japan Ingenieursfakultät Bau Geo Umwelt, Technische Universität München, München, Germany Forschungsreaktor FRM-II, Technische Universität München, Garching, Germany

In the assessment of volcanic risk, it is often assumed that magma ascending at a slow rate will erupt effusively, whereas magma ascending at fast rate will lead to an explosive eruption. Mechanistically viewed, this assessment is supported by the notion that the viscoelastic nature of magma (i.e., the ability of magma to relax at an applied strain rate), linked via the gradient of flow pressure (related to discharge rate), controls the eruption style. In such an analysis, the physical interactions between the magma and the conduit wall are commonly, to a first order, neglected. Yet, during ascent, magma must force its way through the volcanic edifice/structure, whose presence and form may greatly affect the stress field through which the magma is trying to ascend. Here, we demonstrate that fracturing of the conduit wall via flow pressure releases an elastic shock resulting in fracturing of the viscous magma itself. We find that magma fragmentation occurred at strain rates seven orders of magnitude slower than theoretically anticipated from the applied axial strain rate. Our conclusion, that the discharge rate cannot provide a reliable indication of ascending magma rheology without knowledge of conduit wall stability, has important ramifications for volcanic hazard assessment. New numerical simulations are now needed in order to integrate magma/conduit interaction into eruption models.

This new software package has been purpose built for 3-d tomographic analysis of geomaterials. Its unique datafile structure allows the user to explore datasets of many gigabytes in size without loading times and using very little of the systems resources. This means that research groups can enjoy the benefits of 3-d tomographic analysis without investing in expensive high-end computer hardware. Sub-volumes can be selected from the dataset for advanced processing, rendering and characterisation. A broad range of data processing and morphology tools are available, allowing the user to remove noise and improve the overall image quality. It is possible to overlay and compare multiple datasets using interpolated slicing. As an example, this utility has been used to study the relative attenuation of neutron and X-ray computed tomography, highlighting the presence of water. Tomoview also features powerful and reliable segmentation algorithms to convert the dataset into a database of objects at the press of a button (isolating individual crystals, cracks and pores). These algorithms have been designed to cope with the subtle phase contrast, high image noise and artefacts often present in geomaterial tomography. The database gives each object a unique label and stores information such as its location; volume, mean-attenuation and boundingbox. It can be augmented using ellipsoid fitting to characterise the sizes, shapes and orientations of the objects. This allows us to quantitatively study subtle features such as particle anisotropy and flow dynamics. Complex structures, such as crack networks, cannot be well described by an ellipsoid. For these features, an algorithm has been developed to quantify the 3-d anisotropy based on the edge-gradient. This software package has already been applied to a number of geomaterial studies, several of which will be featured in this presentation.

The application of state-of-the-art detector systems at the Advanced Neutron Tomography and Radiography Experimental System (ANTARES; Munich, Germany) has led to significant improvements in spatial resolution and contrast for geomaterial imaging. Resolutions of ∼16–100 μm are now possible with fields of view of 33–205 mm, a level now comparable with X-ray computed tomography (XCT), for which a micro-XCT at the Institute for Mineralogy, Crystallography and Material Science at the University of Leipzig was used. Fine pixel resolution comes at the cost of image quality and increased exposure time, so that the optimum configuration for each sample must be determined on a case by case basis. Our interdisciplinary approach has yielded an efficient system of data acquisition, processing, and quantification that is well suited for geomaterial imaging. We expect to find applications in a much wider spectrum of geomaterial research, including the formation of natural glasses, the characterization of limited and/or precious samples such as scientific drill cores, and biomineralization studies.

During periods of volcanic unrest, magma is transported to the surface by swarms of dykes, which must overcome the strength of the country rock in order to propagate. Field and theoretical studies have shown that the majority of dykes are arrested long before they actually get to the surface. As such, the importance on understanding dyke propagation through field, theoretical and experimental means is of wide importance in hazard mitigation, both directly due to eruption and also indirectly due to edifice stability. However, to date, there exists a paucity of laboratory information to compliment the field and analytical hypotheses on dyke movement, in particular in terms of how pressurised conduits and dykes interact with the surrounding country rock in order to overcome its tensile strength and hence, propagate to the surface (or otherwise). Previous hydrofracture data has largely concentrated on the pressurization of a central conduit at ambient (room temperature) conditions through the use of a low viscosity pressurisation fluid. Although useful for examining general physical processes, such experiments cannot reveal details due to the temperature of the pressurising fluid (magma) or interaction between melt and the simulated country rock. Conversely, rock mechanics experiments at high temperature have previously been performed at representative temperatures, but without considering a fluid filled conduit; instead relying on mechanical means to fracture the sample, and thus allow the tensile strength and/or critical stress intensity factor (K1c) to be calculated. Therefore, we present a new set of experiments in which a two ‘phase’ (conduit/shell) system is examined under conditions similar to magma ascent, and that subsequently lead to dyking and fracturing due to the overpressurised fluid (conduit). The experiments are conducted in a uniaxial press at temperatures of 828°C, 867°C and 914°C and under strain rates approximating 10-5 s-1. In our experiments we pressurise a crystal-poor granitic melt (obsidian from cougar creek, Yellowstone national park, USA within a cylindrical shell of basalt from Mt. Etna volcano, Italy. Pressurization of the melt, with a known temperature-dependence of viscosity, is used to impose a force on the inner wall of the shell in a manner essentially analogous to dyke pressurisation and movement. As the conduit is incompressible, the force provided by the loading ram can be directly used to calculate an imposed stress and thus conduit pressure. Importantly, to simulate as precisely as possible the cyclical nature of volcanic pressurisation, a strain rate was imposed in a series of strain steps. Eventually, the outer shell fractures under the stress. For each temperature tested, shell failure is confirmed through a peak in acoustic emission energy, at which point the conduit pressure falls to zero. Our results show that a prominent stress relaxation response accompanies each period of decrease in strain rate. temperature controls the rate of stress relaxation. Samples at lower temperatures (and thus higher viscosity) take longer to relax compared to higher temperatures. We observe a noticeable dependence of peak stress upon temperature; conduit pressures at fracture of approximately 44 MPa, 18 MPa and 15 MPa measured for sample temperatures of 828°C, 867°C and 914°C respectively (corresponding to conduit viscosities of 9.5, 9.0 and 8.0 log unit respectively. Finally, we present a simple model for (a) calculating the simple tensile strength based on sample geometry and conduit pressure, and (b), determining conduit properties (viscosity) based on the measured stress relaxation data.

The failure of magma and onset of an explosive eruption is commonly preceded by accelerations in discharge rate and seismic activity. As magma is mechanically forced through the ductile-brittle transition, cracks propagate and coalesce, thereby generating an increasing amount of seismic energy, which can be used to forecast the onset of explosive eruptions. In this study dome lavas from Volcan de Colima (Mexico) are deformed at 930 °C under various stresses (1 to 76 MPa) in a uniaxial press. Crack propagation in the magma is monitored by two acoustic emission sensors and used as a proxy to forecast the time of failure (e.g., Kilburn, J. Volcanol. Geotherm. Res. 2003). We assess the degree of fracture damage and anisotropy of the fracture network in 3-D via high-resolution (30 micron) Neutron Computer Tomography imaging. Tomography images of magmas experimentally deformed in the ductile-brittle transition reveals that at high strain rate, failure occurs rapidly along macroscopic cracks and requires less strain than failure at lower strain rate. The time window to accurately forecast the failure of magma thus diminishes with strain rate.

Forecasting volcanic eruptions is a fundamental objective of volcanology. Given the complexity of magma ascent dynamics and that direct observations are impossible, laboratory experiments provide a promising approach to better understand the processes that lead to and feed explosive eruptions. Fragmentation of porous natural samples in a shock tube apparatus gives insights into the behavior of volcanic rocks during rapid decompression. In this preliminary study, we use acoustic emissions (AE) to monitor the generation of cracks and describe the signal that accompanies the fragmentation of natural volcanic rocks. Rapid decompression experiments were carried out in a fragmentation bomb at room temperature. We used sample sets from Merapi volcano (Indonesia) and Montserrat volcano (West Indies, UK) with open porosities ranging between 20% to 67%. Cylindrical samples were pressurized with Argon gas in an autoclave. To overcome the fragmentation threshold of the specific rocks, applied pressures varied from 5-20 MPa. Subsequent rapid decompression of the samples caused fragmentation. During fragmentation, acoustic emissions were monitored by a two 2- channel- AE system that allows sampling rates of 1-5 MHz/channel. Two piezzoelectric sensors (100-1000 kHz) were attached to the autoclave and simultaneously recorded the micro-seismic events. An enhanced setup with a waveguide connecting the sample to the AE sensors allowed a better quality of the recorded signals. Furthermore, to find the optimum sensor position, various settings were tried during decompression experiments. The recorded AE were analyzed with respect to their distribution, frequency, amplitude, and the energy released during fragmentation. The AE energy parameter takes into account both the number of hits and their sizes. Moreover, changes in size and characteristics of these acoustic emissions with explosion energy, magma state and energy partitioning will be discussed. The results of this study may contribute to better understand volcanic processes, and improve our ability to correctly evaluate the seismic nature of explosive eruptions, necessary to implement forecasting methods.

(1) LMU, Earth and Environmental Sciences, Munchen, Germany (hess@lmu.de), (2) Fachgebiet Tektonik und Gefugekunde, Technische Universitat Munchen, 80333 Munchen, Germany (kruhl@tum.de), (3) Forschungsstelle GIScience, Osterreichische Akademie der Wissenschaften, 5020 Salzburg, Austria (robert.marschallinger@oeaw.ac.at), (4) Forschungsreaktor FRM-II, Technische Universitat Munchen, 85747 Garching, Germany (burkhard.schillinger@frm2.tum.de)