Fragmentation & explosive eruptions

Class 7 - Blowin’ in the wind

Sébastien Biass

sebastien.biasse@unige.ch

Group of physical volcanology and geological risk - 1st floor, office 103

March 30, 2026

From Class 7

Conduit drivers & processes

Overpressure → reservoir failure → ascent → dike propagation → conduit formation

Key processes

• Rates of bubble nucleation
• Mechanisms of bubble growth
• Conditions of gas escape by either bubble rise or flow through permeable networks
• Degassing-induced crystallization

• Decompression → growth
• Growth → coalescence
• Bubbles mobility → f(viscosity)
  • Trapped? Development of permeability?
• If trapped → acceleration?

Ascent rates

Gaz mobility

Bubbles (B) vs magma (M) ascent rates in decoupled (left) and coupled (right) systems.

Conduit properties

Evolution of key conduit properties for decoupled (blue) and coupled (orange) systems

Importance of:

• Ascent rate
• Decompression rate
• Gas mobility

Halemaumau

Intuitively, what is the difference?

• Gas coupling?
• Size/colour of ejected fragments?
• Appearance → fluidal/solid?

Eyja

Magmatic fragmentation

Decompression rate

End members of explosive mechanisms from bottom-up (→ rapid acceleration) to top-down (→ rapid decompression). Cashman and Scheu (2015)

Fields of eruptive processes mapped on a SiO₂ vs decompression rate. Cashman and Scheu (2015)

Fragmentation

Conduit:

• Decompression
  • Bubble expansion
  • No permeable network (high \(\eta\))
  • Acceleration

Fragmentation when:

• Deformation (strain rate) becomes faster than structural relaxation time of the melt
  • Elastic response
  • Brittle failure

Evolution of key conduit properties for decoupled (blue) and coupled (orange) systems

Illustration of the glass transition. Cashman and Scheu (2015)

Top: - important role of crystallization and degassing during magma ascent and eruption - 1. Hot mafic, 2 Cold silicic

TGSD: - High fragmentation intensity produces more fine particles → problem for the aviation

Fragmentation end members

Ductile fragmentation

→ Magma stretches, thins and tears apart

Brittle fragmentation

→ Brittle fracture when strain rate becomes too large

Total grain-size distribution

• Proportion of mass of the deposit contained within a size fraction
• More intense fragmentation (→ Plinian) produce fine particles

Total grain size distributions for different eruption styles. Cashman and Scheu (2015)

	bombs / blocks	lapilli	coarse ash	fine ash
Diameter	> 64 mm	64 mm – 2 mm	2 mm – 63 µm	< 63 µm
Residence time	≈ sec	≈ min	≈ hours to few days	several days
Travel distance	proximal <10 km	medial <50 km	distal <100 km	very distal <1000 km

Component

Juvenile fragments

• Fragmented magma
• Generally hot

Lithic fragments

• Wall rock fragments
• Usually cold

Gas

• Hot

Volcanic plumes

Plume regions

Umbrella region

• Neutral buoyancy
  • Wind advection

Convective region

• Buoyancy-driven

Gas-thrust region

• Momentum-driven

Conduit fragmentation

Fragmentation → upward acceleration

Gas-thrust region

• Momentum-driven
• \(\rho_{plume} > \rho_{atm}\)

• Deceleration
  • Drag, gravity
• Expansion
  • Rapid expansion to reach pressure equilibrium
• Mixing → turbulences
  • Density equilibrium
  • Thermal equilibrium

Sakurajima

Question 1

What if the plume remains denser than the atmosphere?

Column collapse

• Generates pyroclastic density currents (PDC)
• Lateral deposition system

Question 2

What could allow the plume to continue rising?

Convective region

• Buoyancy-driven
• \(\rho_{plume} < \rho_{atm}\)

• Decrease of the plume’s bulk density
  • Entrainment and heating of air
• If mixture becomes lighter than ambient air
  • Superbuoyancy
  • Secondary acceleration
• Otherwise: Gravitational collapse

Morpho

Eyjafjallajökull 2011

Ruapehu 1996

Cordon Caulle 2011

→ What controls plume morphology?

→ Eruptive vs atmospheric parameters?

Plume morphology

Degruyter and Bonadonna (2013)

• Radial (\(\alpha\)) vs wind (\(\beta\)) entrainment
• Mass eruption rate [kg s^-1]
  • Function of plume height [km] H⁴
• Wind velocity

\[\Pi = \frac{\overline{N}H}{1.8\overline{v}} \left(\frac{\alpha}{\beta}\right)^2\]

• N: Buoyancy frequency

Umbrella region

• Lateral wind advection
• \(\rho_{plume} = \rho_{atm}\)

• Momentum overshoot
• Gravity current
• Spreading at level of neutral buyancy

Umbrella cloud

Hunga Tonga–Hunga Haʻapai , January 15 2022

Source: https://appliedsciences.nasa.gov/media/4618

Pyroclastic density currents

PDC Generation

Lack of buoyancy → pyroclastic density currents (PDC)

Chazot et al. (2017)

Problem

• Multiple triggering mechanisms
• Multiple emplacement mechanisms
• Which can happen simultaneously!!!

= Very complex phenomenon

Sarychev

On June 12, 2009, a fortuitous orbit of the International Space Station (ISS) made it possible for an astronaut on board to capture Sarychev Volcano in the early stages of eruption. The volcano is located on the northwestern end of Matua Island, which is part of the Kuril Islands, a chain of 56 islands northeast of Japan.

Source: https://svs.gsfc.nasa.gov/11446

MSH

Source: https://www.youtube.com/watch?v=rD-RldBQx7U

Unzen

Emplacement mechanisms

• Complex mixtures of gases and solids
  • 200°C to 700°C
  • 50 to 700 km/h
  • Runout: 10’s to 100’s km
  • Affected by topography
• 2 end-members:
  • pyroclastic surges
    
  • pyroclastic flows

Kelfoun and Gueugneau (2022)

“It is . . . unrealistic to hope that all flowage deposits can be conveniently classed into either category and that a single set of transport and depositional processes would apply to each case.”

Pyroclastic deposits

Pyroclastic deposits: Size

Size

• Size class(es)

Source

Source: J. Alean

Source: USGS

Category	Diameter
Fine ash	< 0.063 mm
Coarse ash	0.063 – 2 mm
Fine lapilli	2 – 16 mm
Coarse lapilli	16 – 64 mm
Blocks/Bombs	> 64 mm

Pyroclastic deposits: Sorting/componentry

Sorting/Componentry

• Variability of size
  • Well sorted
  • Poorly sorted
• Componentry
  • Juvenile, lithic, crystals
  • Pristine vs altered
  • Fluidal vs angular
  • Shiny vs mat

Lockwood and Hazlett (2010)

Pyroclastic deposits: Texture

Texture

• Massive vs layered
• Class supported vs matrix supported

Vertical structures

• Grading

Sedimentary structures

• Plane parallel
• Cross bedding
• Erosional contacts

Lockwood and Hazlett (2010)

Flow vs fall 1

Flow vs fall 2

Tephra fallout

Wind advection → ± aerodynamic filter

• Coarse particles
  • Fall close to the vent
  • Have short atmospheric residence time
• Fine particles
  • Are carried away by the wind
  • Have long atmospheric residence time

Deposits are

• Well-sorted
• Massive, clast-supported
• Plane parallel, draping the topography

Sakurajima - Takishita et al. (2024)

Shinmoedake - USGS

Layer 3, Cotopaxi volcano

Tephra fallout: Isopach

• Isopach = contours of equal thickness
• Sedimentation regime linked to particle size
• Consequence → exponential decrease of deposit thickness
  • Deposit volume

Tephra fallout: Isopleth

• Isopleth = contours the maximum distance reached by a particle of a given size
• Isopleth size/geometry reflect plume height and wind
  • Width: Plume height
  • Length Wind speed

• Consequence:
  • Isopleth geometry → Plume height

Reading tephra fallout deposits

Pyroclastic flow deposits

• Deposits
  • Massive, poorly-sorted
  • Follow topography lows
• Block and ash flows → dome collapse
  • Dome fragments, angular
  • Short runout

• Pumice flows → instable/collapsing Plinian column
  • Pumice rich, usually rounded
  • Ignimbrite → welding

Tarawera. Source

Changbaishan (China/DPRK)

Pyroclastic surge deposits

• Deposits
  • Wave form/crossbed facies
  • Erosional surfaces
  • Evolve with distance from the vent
  • Poorly-sorted
  • Can overcome topography

Surge deposit around Vesuvius. Lockwood and Hazlett (2010))

Comparison of pyroclastic deposits

Relationship to topography

Emplacement dynamics

Take home messages

Summary

Fragmentation

Fragmentation is the process of tearing apart magma into pyroclasts. When the deformation rate in the conduit is sufficiently low and/or the magma sufficiently fluid, ductile fragmentation stretches, thins and tears appart magma. This is what happens in Hawaiian fountains. In contrast, if the decompression rate and/or magma viscosity are sufficiently high, the rate of deformation in the conduit caused by bubble growth might exceed the structural relaxation time of the magma, in which case a brittle fragmentation occurs.

Driver of fragmentation

Two end-members mechanism can explain explosive eruptions. On the one hand, rapid acceleration represents a bottom-up mechanism usually caused by a rapid volatile exsolution in the magma reservoir and is associated with heterogeneous bubble nucleation. On the other hand, rapid decompression is a top down process caused by a sudden pressure drop (e.g., dome or flank collapse), which triggers homogeneous bubble nucleation. As always, reality is never one or the other, and an eruption might fluctuate between these two extremes.

Summary

Above the vent

Above the fragmentation layer, the eruptive mixture continues to accelerate in the conduit, and is usually composed of juvenile and lithic fragments within hot gas. At the vent outlet, the mixture possesses a vertical momentum (gas thrust region) will decelerate abruptly due to the atmosphere. In order to equilibrate density and temperature, turbulent mixing will develop.

 

Plume stability

If the plume remains at a higher density than the atmosphere, it will collapse into a pyroclastic density current. But at the right conditions, the plume might become lighter than the atmosphere (→superbuoyancy) and accelerate upwards. Depending on eruptive (→ mass eruption rate, exit velocity, …) and atmospheric properties (e.g., wind speed), the plume can develop into a strong plume - in which case an umbrella cloud will form - or intermediate to weak, bent-over plumes.

Summary

Tephra fall deposits

Fallout deposits are wind-advected and, therefore, “aerodynamically-filtered” by the atmosphere. The consequence is that i) the thinning rates of the deposits can be constrained and used to compute the deposit volume and ii) the geometry of isopleth can help calculating the plume height. Fallout deposits are well sorted, massive, and drape the topography.

Pyroclastic density currents (PDC)

PDC are lateral, gravity-driven mechanisms resulting in the complex emplacement of mixture of hot gas and particles. Several phenomena can cause PDCs, and their deposition mechanisms can be classified in two end members. Pyroclastic flows represent the dense, basal, laminar portion of the flow usually constrained in topographic lows and producing massive, poorly sorted deposits (e.g., block and ash flows, pumice flows). Pyroclastic surges represent the upper, turbulent and dilute part of the PDCs, which can overcome topographic obstacles and usually results in poorly sorted deposits that show sedimentary aggradating structures (e.g., dunes, cross bedding).

References

Bibliography

Cashman KV, Scheu B (2015) Magmatic Fragmentation. In: Sigurdsson H, Houghton B, McNutt S, et al. (eds) The Encyclopedia of volcanoes, 2nd edition. Academic Press, San Diego, pp 459–471

Chazot G, Lénat J-F, Maury RC, et al (2017) Volcanologie. De Boeck Supérieur, Louvain-la-Neuve Paris

Degruyter W, Bonadonna C (2013) Impact of wind on the condition for column collapse of volcanic plumes. Earth and Planetary Science Letters 377–378:218–226

Kelfoun K, Gueugneau V (2022) A Unifying Model for Pyroclastic Surge Genesis and Pyroclastic Flow Fluidization. Geophysical Research Letters 49:e2021GL096517. https://doi.org/10.1029/2021GL096517

Lockwood J, Hazlett RW (2010) Volcanoes: Global Perspectives. Wiley

Takishita K, Poulidis A-P, Iguchi M (2024) Tephra segregation profiles based on disdrometer observations and tephra dispersal modeling: Vulcanian eruptions of Sakurajima volcano, Japan. Earth, Planets and Space 76:29. https://doi.org/10.1186/s40623-023-01952-y