Home > Research > Publications & Outputs > Cyclicity in slug-driven basaltic eruptions: in...

Associated organisational unit

Electronic data

  • Cyclicity in slug-driven basaltic eruptions: insights from large-scale analogue experiments

    Final published version, 40.2 KB, PDF document

View graph of relations

Cyclicity in slug-driven basaltic eruptions: insights from large-scale analogue experiments

Research output: Contribution to conference - Without ISBN/ISSN Abstract

Published

Standard

Cyclicity in slug-driven basaltic eruptions: insights from large-scale analogue experiments. / Llewellin, Ed; Del Bello, Elisabetta; Lane, Stephen et al.
2013. Abstract from EGU General Assembly 2013, Vienna, Austria.

Research output: Contribution to conference - Without ISBN/ISSN Abstract

Harvard

Llewellin, E, Del Bello, E, Lane, S, Capponi, A, Mathias, S & Taddeucci, J 2013, 'Cyclicity in slug-driven basaltic eruptions: insights from large-scale analogue experiments', EGU General Assembly 2013, Vienna, Austria, 7/04/13 - 12/04/13.

APA

Llewellin, E., Del Bello, E., Lane, S., Capponi, A., Mathias, S., & Taddeucci, J. (2013). Cyclicity in slug-driven basaltic eruptions: insights from large-scale analogue experiments. Abstract from EGU General Assembly 2013, Vienna, Austria.

Vancouver

Llewellin E, Del Bello E, Lane S, Capponi A, Mathias S, Taddeucci J. Cyclicity in slug-driven basaltic eruptions: insights from large-scale analogue experiments. 2013. Abstract from EGU General Assembly 2013, Vienna, Austria.

Author

Llewellin, Ed ; Del Bello, Elisabetta ; Lane, Stephen et al. / Cyclicity in slug-driven basaltic eruptions: insights from large-scale analogue experiments. Abstract from EGU General Assembly 2013, Vienna, Austria.

Bibtex

@conference{3db08a0c32a64a7088929404c5acb414,
title = "Cyclicity in slug-driven basaltic eruptions: insights from large-scale analogue experiments",
abstract = "Basaltic eruptions commonly exhibit cyclical or pulsatory behaviour. Strombolian eruptions are typically discrete and rhythmic, with return interval of minutes-to-hours; lava fountains may fluctuate over similar, or shorter, timescales. In both cases the cyclicity results from the separation of gas from the magma, and its localization into discrete gas slugs or gas-rich packets. We report analogue experiments which model the fluid dynamics of slug-driven basaltic eruptions. Experiments were conducted in liquid-filled vertical pipes at a range of scales,from 0.02 to 0.2 m in diameter, and 2 to 13 m in height, allowing us to investigate Reynolds numbers 10􀀀3 < Re <105, encompassing the natural range for volcanoes. The dynamics of both discrete gas slugs (Taylor bubbles) and continuous sluggy flow were quantified. A significant novelty of this study is that we explore the role played by the boundary conditions at the top and bottom of the conduit, which may be either closed (zero flux) or held atconstant pressure. This setup allows us to mimic plugged or open vent, and the influence of a magma chamber. Our study combines the direct observation of in-conduit fluid dynamic processes with measured pressure variations inside the conduit.Our results demonstrate that, when discrete gas slugs are injected, plugging the vent has a strong influence on the development of overpressure in the system, and on the potential for the system to manifest cyclic behaviour.When gas is injected continuously, with constant pressure boundaries at the top and bottom of the conduit, the system spontaneously self-organizes into rhythmic sluggy flow when the injection rate exceeds a critical value. In both cases we find that the capacity of the system to sustain well-formed, discrete slugs depends strongly on the Reynolds number of the flow. Well-formed Taylor bubbles – which have a smooth hemispherical cap, occupy the width of the conduit, and ascend relatively slowly – only form when the liquid column is stagnant, or is flowingin the laminar regime. Slugs that rise through turbulent liquid are poorly-formed and, at the extreme, ascend as highly-turbulent gas-rich packets; these occupy less of the conduit width and therefore ascend more rapidly than true Taylor bubbles. A major factor influencing the turbulence in the liquid is the separation between slugs; when slugs are closely spaced, the wake of one slug tends to disrupt the next. At the onset of eruption, therefore, the tendency is for poorly-formed, fast-moving slugs to catch up with the well-formed, slow-moving lead slug, increasing the potential for an impulsive, explosive onset. During sustained eruption, lower gas fluxes lead to lower-frequency, higher amplitude fluctuations in eruption rates, whilst higher gas fluxes lead to higher-frequency, lower amplitude fluctuations. The spectrum of natural activity, from discrete Strombolian eruptions to sustained, pulsatory lava fountaining can, therefore, be characterized as a change in the behaviour of the separated gas phase in the conduit.",
author = "Ed Llewellin and {Del Bello}, Elisabetta and Stephen Lane and Antonio Capponi and Simon Mathias and Jacopo Taddeucci",
year = "2013",
language = "English",
note = "EGU General Assembly 2013 ; Conference date: 07-04-2013 Through 12-04-2013",

}

RIS

TY - CONF

T1 - Cyclicity in slug-driven basaltic eruptions: insights from large-scale analogue experiments

AU - Llewellin, Ed

AU - Del Bello, Elisabetta

AU - Lane, Stephen

AU - Capponi, Antonio

AU - Mathias, Simon

AU - Taddeucci, Jacopo

PY - 2013

Y1 - 2013

N2 - Basaltic eruptions commonly exhibit cyclical or pulsatory behaviour. Strombolian eruptions are typically discrete and rhythmic, with return interval of minutes-to-hours; lava fountains may fluctuate over similar, or shorter, timescales. In both cases the cyclicity results from the separation of gas from the magma, and its localization into discrete gas slugs or gas-rich packets. We report analogue experiments which model the fluid dynamics of slug-driven basaltic eruptions. Experiments were conducted in liquid-filled vertical pipes at a range of scales,from 0.02 to 0.2 m in diameter, and 2 to 13 m in height, allowing us to investigate Reynolds numbers 10􀀀3 < Re <105, encompassing the natural range for volcanoes. The dynamics of both discrete gas slugs (Taylor bubbles) and continuous sluggy flow were quantified. A significant novelty of this study is that we explore the role played by the boundary conditions at the top and bottom of the conduit, which may be either closed (zero flux) or held atconstant pressure. This setup allows us to mimic plugged or open vent, and the influence of a magma chamber. Our study combines the direct observation of in-conduit fluid dynamic processes with measured pressure variations inside the conduit.Our results demonstrate that, when discrete gas slugs are injected, plugging the vent has a strong influence on the development of overpressure in the system, and on the potential for the system to manifest cyclic behaviour.When gas is injected continuously, with constant pressure boundaries at the top and bottom of the conduit, the system spontaneously self-organizes into rhythmic sluggy flow when the injection rate exceeds a critical value. In both cases we find that the capacity of the system to sustain well-formed, discrete slugs depends strongly on the Reynolds number of the flow. Well-formed Taylor bubbles – which have a smooth hemispherical cap, occupy the width of the conduit, and ascend relatively slowly – only form when the liquid column is stagnant, or is flowingin the laminar regime. Slugs that rise through turbulent liquid are poorly-formed and, at the extreme, ascend as highly-turbulent gas-rich packets; these occupy less of the conduit width and therefore ascend more rapidly than true Taylor bubbles. A major factor influencing the turbulence in the liquid is the separation between slugs; when slugs are closely spaced, the wake of one slug tends to disrupt the next. At the onset of eruption, therefore, the tendency is for poorly-formed, fast-moving slugs to catch up with the well-formed, slow-moving lead slug, increasing the potential for an impulsive, explosive onset. During sustained eruption, lower gas fluxes lead to lower-frequency, higher amplitude fluctuations in eruption rates, whilst higher gas fluxes lead to higher-frequency, lower amplitude fluctuations. The spectrum of natural activity, from discrete Strombolian eruptions to sustained, pulsatory lava fountaining can, therefore, be characterized as a change in the behaviour of the separated gas phase in the conduit.

AB - Basaltic eruptions commonly exhibit cyclical or pulsatory behaviour. Strombolian eruptions are typically discrete and rhythmic, with return interval of minutes-to-hours; lava fountains may fluctuate over similar, or shorter, timescales. In both cases the cyclicity results from the separation of gas from the magma, and its localization into discrete gas slugs or gas-rich packets. We report analogue experiments which model the fluid dynamics of slug-driven basaltic eruptions. Experiments were conducted in liquid-filled vertical pipes at a range of scales,from 0.02 to 0.2 m in diameter, and 2 to 13 m in height, allowing us to investigate Reynolds numbers 10􀀀3 < Re <105, encompassing the natural range for volcanoes. The dynamics of both discrete gas slugs (Taylor bubbles) and continuous sluggy flow were quantified. A significant novelty of this study is that we explore the role played by the boundary conditions at the top and bottom of the conduit, which may be either closed (zero flux) or held atconstant pressure. This setup allows us to mimic plugged or open vent, and the influence of a magma chamber. Our study combines the direct observation of in-conduit fluid dynamic processes with measured pressure variations inside the conduit.Our results demonstrate that, when discrete gas slugs are injected, plugging the vent has a strong influence on the development of overpressure in the system, and on the potential for the system to manifest cyclic behaviour.When gas is injected continuously, with constant pressure boundaries at the top and bottom of the conduit, the system spontaneously self-organizes into rhythmic sluggy flow when the injection rate exceeds a critical value. In both cases we find that the capacity of the system to sustain well-formed, discrete slugs depends strongly on the Reynolds number of the flow. Well-formed Taylor bubbles – which have a smooth hemispherical cap, occupy the width of the conduit, and ascend relatively slowly – only form when the liquid column is stagnant, or is flowingin the laminar regime. Slugs that rise through turbulent liquid are poorly-formed and, at the extreme, ascend as highly-turbulent gas-rich packets; these occupy less of the conduit width and therefore ascend more rapidly than true Taylor bubbles. A major factor influencing the turbulence in the liquid is the separation between slugs; when slugs are closely spaced, the wake of one slug tends to disrupt the next. At the onset of eruption, therefore, the tendency is for poorly-formed, fast-moving slugs to catch up with the well-formed, slow-moving lead slug, increasing the potential for an impulsive, explosive onset. During sustained eruption, lower gas fluxes lead to lower-frequency, higher amplitude fluctuations in eruption rates, whilst higher gas fluxes lead to higher-frequency, lower amplitude fluctuations. The spectrum of natural activity, from discrete Strombolian eruptions to sustained, pulsatory lava fountaining can, therefore, be characterized as a change in the behaviour of the separated gas phase in the conduit.

M3 - Abstract

T2 - EGU General Assembly 2013

Y2 - 7 April 2013 through 12 April 2013

ER -