University of Munich (LMU)

Products of magma fragmentation can pose a severe threat to health, infrastructure, environment, and aviation. Systematic evaluation of the mechanisms and the consequences of volcanic fragmentation is very difficult as the adjacent processes cannot be observed directly and their deposits undergo transport-related sorting. However, enhanced knowledge is required for hazard assessment and risk mitigation. Laboratory experiments on natural samples allow the precise characterization of the generated pyroclasts and open the possibility for substantial advances in the quantification of fragmentation processes. They hold the promise of precise characterization and quantification of fragmentation efficiency and its dependence on changing material properties and the physical conditions at fragmentation. We performed a series of rapid decompression experiments on three sets of natural samples from Unzen volcano, Japan. The analysis comprised grain-size analysis and surface area measurements. The grain-size analysis is performed by dry sieving for particles larger than 250 Am and wet laser refraction for smaller particles. For all three sets of samples, the grain-size of the most abundant fraction decreases and the weight fraction of newly generated ash particles (up to 40 wt.%) increases with experimental pressure/potential energy for fragmentation. This energy can be estimated from the volume of the gas fraction and the applied pressure. The surface area was determined through Argon adsorption. The fragmentation efficiency is described by the degree of fineparticle generation. Results show that the fragmentation efficiency and the generated surface correlate positively with the applied energy.

For an improvement in the quality of conduit flow and dome-related explosive eruption models, knowledge of the preeruption or precollapse density of the rocks involved is necessary. As close investigation is impossible during eruption, the best substitute comes from quantitative investigation of the eruption deposits. The porosity of volcanic rocks is of primary importance for the eruptive behaviour and, accordingly, a key-parameter for realistic models of dome stability and conduit flow. Fortunately, this physical property may be accurately determined via density measurements. We developed a robust, battery-powered device for rapid and reliable density measurements of dry rock samples in the field. The density of the samples (sealed in plastic bags at 250 mbar) is determined using the Archimedean principle. We have tested the device on the deposits of the 1990–1995 eruption of Unzen volcano, Japan. Short setup and operation times allow up to 60 measurements per day under fieldwork conditions. The rapid accumulation of correspondingly large data sets has allowed us to acquire the first statistically significant data set of clast density distribution in block-and-ash flow deposits. More than 1100 samples with a total weight of 2.2 tons were measured. The data set demonstrates that the deposits of the last eruptive episode at Unzen display a bimodal density distribution, with peaks at 2.0F0.1 and 2.3F0.1 g/cm3, corresponding to open porosity values of 20 and 8 vol.%, respectively. We use this data set to link the results of laboratory-based fragmentation experiments to field studies at recently active lava domes.

In response to rapid decompression, porous magma may fragment explosively. This occurs when the melt can no longer withstand forces exerted upon it due to the overpressure in included bubbles. This occurs at a critical pressure difference between the bubbles and the surrounding magma. In this study we have investigated this pressure threshold necessary for the fragmentation of magma. Here we present the first comprehensive, high temperature experimental quantification of the fragmentation threshold of volcanic rocks varying widely in porosity, permeability, crystallinity, and chemical composition. We exposed samples to increasing pressure differentials in a high temperature shock tube apparatus until fragmentation was initiated. Experimentally, we define the fragmentation threshold as the minimum pressure differential that leads to complete fragmentation of the pressurized porous rock sample. Our results show that the fragmentation threshold is strongly dependent on porosity; high porosity samples fragment at lower pressure differentials than low porosity samples. The fragmentation threshold is inversely proportional to the porosity. Of the other factors, permeability likely affects the fragmentation threshold at high porosity values, whereas chemical composition, crystallinity and bubble size distribution appear to have minor effects. The relationship for fragmentation threshold presented here can be used to predict the minimum pressure differential necessary for the initiation or cessation of the explosive fragmentation of porous magma.

To visualize the behavior of erupting magma in volcanic conduits, we performed shock tube experiments on the ductile–brittle response of a viscoelastic medium to diffusion-driven bubble expansion. A sample of shear-thinning magma analogue is saturated by gas Ar under high pressure. On rapid decompression, Ar supersaturation causes bubbles to nucleate, grow, and coalesce in the sample, forcing it to expand, flow, and fracture. Experimental variables include saturation pressure and duration, and shape and lubrication of the flow path. Bubble growth in the experiments controls both flow and fracturing, and is consistent with physical models of magma vesiculation. Two types of fractures are observed: i) sharp fractures along the uppermost rim of the sample, and ii) fractures pervasively diffused throughout the sample. Rim fractures open when shear stress accumulates and strain rate is highest at the margin of the flow (a process already inferred from observations and models to occur in magma). Pervasive fractures originate when wall-friction retards expansion of the sample, causing pressure to build-up in the bubbles. When bubble pressure overcomes wall-friction and the tensile strength of the porous sample, fractures open with a range of morphologies. Both types of fracture open normally to flow direction, and both may heal as the flow proceeds. These experiments also illustrate how the development of pervasive fractures allows exsolving gas to escape from the sample before the generation of a permeable network via other processes, e.g., bubble coalescence. This is an observation that potentially impact the degassing of magma and the transition between explosive and effusive eruptions.