Alain Burgisser

Redox evolution of a degassing magma rising to the surface

Volatiles carried by magmas, either dissolved or exsolved, have a fundamental effect on a variety of geological phenomena, such as magma dynamics and the composition of the Earth’s atmosphere. In particular, the redox state of volcanic gases emanating at the Earth’s surface is widely believed to mirror that of the magma source, and is thought to have exerted a first-order control on the secular evolution of atmospheric oxygen. Oxygen fugacity (fO2) estimated from lava or related gas chemistry, however, may vary by as much as one log unit, and the reason for such differences remains obscure. Here we use a coupled chemical–physical model of conduit flow to show that the redox state evolution of an ascending magma, and thus of its coexisting gas phase, is strongly dependent on both the composition and the amount of gas in the reservoir. Magmas with no sulphur show a systematic fO2 increase during ascent, by as much as 2 log units. Magmas with sulphur show also a change of redox state during ascent, but the direction of change depends on the initial fO2 in the reservoir. Our calculations closely reproduce the H2S/SO2 ratios of volcanic gases observed at convergent settings, yet the difference between fO2 in the reservoir and that at the exit of the volcanic conduit may be as much as 1.5 log units. Thus, the redox state of erupted magmas is not necessarily a good proxy of the redox state of the gases they emit. Our findings may require re-evaluation of models aimed at quantifying the role of magmatic volatiles in geological processes.

A rapid mechanism to remobilize and homogenize highly crystalline magma bodies

The largest products of magmatic activity on Earth, the great bodies of granite and their corresponding large eruptions, have a dual nature: homogeneity at the large scale and spatial and temporal heterogeneity at the small scale. This duality calls for a mechanism that selectively removes the large-scale heterogeneities associated with the incremental assembly of these magmatic systems and yet occurs rapidly despite crystal-rich, viscous conditions seemingly resistant to mixing. Here we show that a simple dynamic template can unify a wide range of apparently contradictory observations from both large plutonic bodies and volcanic systems by a mechanism of rapid remobilization (unzipping) of highly viscous crystal-rich mushes. We demonstrate that this remobilization can lead to rapid overturn and produce the observed juxtaposition of magmatic materials with very disparate ages and complex chemical zoning. What distinguishes our model is the recognition that the process has two stages. Initially, a stiff mushy magma is reheated from below, producing a reduction in crystallinity that leads to the growth of a subjacent buoyant mobile layer. When the thickening mobile layer becomes sufficiently buoyant, it penetrates the overlying viscous mushy magma. This second stage rapidly exports homogenized material from the lower mobile layer to the top of the system, and leads to partial overturn within the viscous mush itself as an additional mechanism of mixing. Model outputs illustrate that unzipping can rapidly produce large amounts of mobile magma available for eruption. The agreement between calculated and observed unzipping rates for historical eruptions at Pinatubo and at Montserrat demonstrates the general applicability of the model. This mechanism furthers our understanding of both the formation of periodically homogenized plutons (crust building) and of ignimbrites by large eruptions.

Reconciling Pyroclastic Flow and Surge: the Multiphase Physics of Pyroclastic Density Currents.

Abstract Two end-member types of pyroclastic density current are commonly recognized: pyroclastic surges are dilute currents in which particles are carried in turbulent suspension and pyroclastic flows are highly concentrated flows. We provide scaling relations that unify these end-members and derive a segregation mechanism into basal concentrated flow and overriding dilute cloud based on the Stokes number ( S T ), the stability factor ( Σ T ) and the dense–dilute condition ( D D ). We recognize five types of particle behaviors within a fluid eddy as a function of S T and Σ T : (1) particles sediment from the eddy, (2) particles are preferentially settled out during the downward motion of the eddy, but can be carried during its upward motion, (3) particles concentrate on the periphery of the eddy, (4) particles settling can be delayed or ‘fast-tracked’ as a function of the eddy spatial distribution, and (5) particles remain homogeneously distributed within the eddy. We extend these concepts to a fully turbulent flow by using a prototype of kinetic energy distribution within a full eddy spectrum and demonstrate that the presence of different particle sizes leads to the density stratification of the current. This stratification may favor particle interactions in the basal part of the flow and D D determines whether the flow is dense or dilute. Using only intrinsic characteristics of the current, our model explains the discontinuous features between pyroclastic flows and surges while conserving the concept of a continuous spectrum of density currents.

Synchrotron X-ray microtomography and lattice Boltzmann simulations of gas flow through volcanic pumices

To illustrate the advances made in permeability calculations combining X-ray microtomography and lattice Boltzmann method simulations, a sample suite of different types of pumices was investigated. Large three-dimensional images at high spatial resolution were collected at three different synchrotron facilities (Elettra, SLS, and ESRF). Single phase gas flow simulations were done on computer clusters with a highly parallelized lattice Boltzmann code, named Palabos. Permeability measurements obtained by gas flow simulation were compared to lab measurements of pumices produced by the Kos Plateau Tuff eruption to validate the method. New permeability data for pumices from other silicic volcanic deposits is presented, and an empirical model for permeability is tested using geometrical and topological data, i.e., tortuosity, specific surface area, and total and connected porosity.

Mechanisms of bubble coalescence in silicic magmas

Bubble coalescence is an important process that strongly affects magmatic degassing. Without coalescence, bubbles remain isolated from one another in the melt, severely limiting gas release. Despite this fact, very little has been done to identify coalescence mechanisms from textures of magmatic rocks or to quantify the dynamics of bubble coalescence in melts. In this paper, we present a systematic study of bubble-coalescence mechanisms and dynamics in natural and experimentally produced bubbly rhyolite magma. We have used a combination of natural observations aided by high-resolution X-ray computed tomography, petrological experiments, and physical models to identify different types of bubble–bubble interaction that lead to coalescence on the timescales of magma ascent and eruption. Our observations and calculations suggest that bubbles most efficiently coalesce when inter-bubble melt walls thin by stretching rather than by melt drainage from between converging bubble walls. Orders of magnitude are more rapid than melt drainage, bubble wall stretching produces walls thin enough that inter-bubble pressure gradients may cause the melt wall to dimple, further enhancing coalescence. To put these results into volcanogical context, we have identified magma ascent conditions where each coalescence mechanism should act, and discuss the physical conditions for preserving coalescence structures in natural pumice. The timescales we propose could improve volcanic eruption models, which currently do not account for bubble coalescence. Although we do not address the effect of shear strain on bubble coalescence, the processes discussed here may operate in several different eruption regimes, including vesiculation of lava domes, post-fragmentation frothing of vulcanian bombs, and bubbling of pyroclasts in conduits.

Porosity redistribution enhanced by strain localization in crystal-rich magmas

Magma degassing, characterized by changes in permeability and porosity distribution, has a crucial control on the style of eruption. During ascent, magma might develop large porosities and crystallize while it is subjected to shear. Shear, in turn, enhances complex fabrics that result from the reorganization of the different phases (crystals, gas, melt). Such fabrics have not yet been evaluated experimentally on a three-phase system. We performed torsion experiments on a synthetic crystal-rich hydrous magma at subsolidus conditions with 11 vol% porosity to establish a link between strain partitioning and porosity redistribution. Crystals induce non-Newtonian deformation, resulting in localization of the shear strain. Three-dimensional microtomography and two-dimensional scanning electron microscope imaging show gas accumulation in local microstructures caused by shear-induced crystal fabric. Our data show that strain localization is a mechanism that could enable magma degassing at very low vesicularity.

Simulating the behavior of volatiles belonging to the C–O–H–S system in silicate melts under magmatic conditions with the software D-Compress

Modeling magmatic degassing, or how the volatile distribution between gas and melt changes at pressure varies, is a complex task that involves a large number of thermodynamical relationships and that requires dedicated software. This article presents the software D-Compress, which computes the gas and melt volatile composition of five element sets in magmatic systems (O-H, S-O-H, C-S-O-H, C-S-O-H-Fe, and C-O-H). It has been calibrated so as to simulate the volatiles coexisting with three common types of silicate melts (basalt, phonolite, and rhyolite). Operational temperatures depend on melt composition and range from 790 to 1400°C. A specificity of D-Compress is the calculation of volatile composition as pressure varies along a (de)compression path between atmospheric and 3000 bars. This software was prepared so as to maximize versatility by proposing different sets of input parameters. In particular, whenever new solubility laws on specific melt compositions are available, the model parameters can be easily tuned to run the code on that composition. Parameter gaps were minimized by including sets of chemical species for which calibration data were available over a wide range of pressure, temperature, and melt composition. A brief description of the model rationale is followed by the presentation of the software capabilities. Examples of use are then presented with outputs comparisons between D-Compress and other currently available thermodynamical models. The compiled software and the source code are available as electronic supplementary materials.

Backward tracking of gas chemistry measurements at Erebus volcano

Erebus volcano in Antarctica offers an exceptional opportunity to probe the dynamics of degassing - its behavior is characterized by an active lava lake through which sporadic Strombolian eruptions occur. Here, we develop a framework for interpreting contrasting degassing signatures measured at high temporal resolution, which integrates physical scenarios of gas/melt separation into a thermodynamic model that includes new volatile solubility data for Erebus phonolite. In this widely applicable framework, the measured gas compositions are backtracked from surface to depth according to physical templates involving various degrees of separation of gas and melt during ascent. Overall, explosive signatures can be explained by large bubbles (gas slugs) rising slowly in equilibrium from at least 20 bars but at most a few hundred bars in a magmatic column closer to the stagnant end-member than the convecting end-member. The span of explosive signatures can be due to various departure depths and/or slug acceleration below a few tens of bars. Results also reveal that explosive gases last equilibrated at temperatures up to 300 degrees C colder than the lake due to rapid gas expansion just prior to bursting. This picture (individual rise of gas and melt batches from a single, potentially very shallow phonolitic source) offers an alternative to the conclusions of previous work based on a similar data set at Erebus, according to which differences between quiescent and explosive gas signatures are due to the decompression of two deep, volatile-saturated sources that mixed to various degrees (phonolite at 1-3 kbar and basanite at 5-8 kbar).

Experimental constrains on shear‐induced crystal breakage in magmas

Crystal breakage occurs along margins of conduit walls and basal zones of lava flows. It is usually interpreted as flow-related textures developed at large finite strains and strains rates. We have investigated the grain size and shape distributions in an experimentally deformed crystal-melt suspension in order to constrain the temperature T, the strain γ and the strain rate γr ranges of the crystal breakage process. The starting crystal-melt suspension is composed of a haplogranitic melt with 54 vol% alumina crystals. Torsion experiments were performed in a gas medium Paterson apparatus at 300 MPa confining pressure and subsolidus temperatures. Crystal size distribution and aspect ratio of alumina grains were measured on polished sections normal to the shear direction, i.e. from the centre to the rim of the deformed cylinders. A first minor occurrence of crystal breakage is evidenced in all experiments and low strains. It is related to intense stress localisation at some grain contacts in the initially connected solid framework. A second intense and penetrative crystal breakage process is observed for T≤ 550°C and γr > 6.2x10-4 s- 1. The evolution of the size distribution as a function of finite strain and the reduced aspect ratios of preserved largest crystals in intensely strained zones support that breakage occurs by abrasion of the larger crystals. This abrasion can be attributed to the partial stress propagation over both the melt and partially isolated crystals under visco-elastic conditions. Mechanical data show a transition from slight shear softening at low strain rates and highest temperatures to strain hardening for experiments that produced penetrative crystal breakage. The crystal-melt suspension exhibits a shear thinning behaviour with a stress exponent larger than 2.06 over the explored strain rate and temperature domain for the experiments without intensive crystal breakage. Our results are applicable to the interpretation of the crystal breakage often observed at the base of lava flows, in domes, and near conduit walls. This experimental reproduction of a process observed in nature is important because the controls of stress-induced breakage we quantified are also key parameters governing magma transport.

Time scales of crystal mixing in magma mushes

Magma mixing is widely recognized as a means of producing compositional diversity and preconditioning magmas for eruption. However, the processes and associated time scales that produce the commonly observed expressions of magma mixing are poorly understood, especially under crystal-rich conditions. Here we introduce and exemplify a parameterized method to predict the characteristic mixing time of crystals in a crystal-rich magma mush that is subject to open-system reintrusion events. Our approach includes novel numerical simulations that resolve multiphase particle-fluid interactions. It also quantifies the crystal mixing by calculating both the local and system-wide progressive loss of the spatial correlation of individual crystals throughout the mixing region. Both inertial and viscous time scales for bulk mixing are introduced. Estimated mixing times are compared to natural examples and the time for basaltic mush systems to become well mixed can be on the order of 10 days.