The effect of total grain-size distribution on the dynamics of turbulent volcanic plumes

- We present a model of volcanic plume including actual grain-size distribution (GSD).
- We validate the model against a specially assembled data set of Plinian eruptions.
- We show that sedimentation and GSD (fragmentation) strongly affect plume dynamics.
- The model greatly improves the prediction of eruptive mass flux from column height.
- The model provides better estimates of ash injection rate into the atmosphere.

The impact of explosive volcanic plumes on climate and on air traffic strongly depends on the concentration and grain-size distribution (GSD) of pyroclastic fragments injected into the atmosphere. Accurate and robust modelling of the evolution of GSD during pyroclast transport from the vent to the ash cloud is therefore crucial for the assessment of major volcanic hazards. Analysis of field deposits from various recent Plinian eruptions shows that their total GSD is well described by a power law, as expected from the physics of magma fragmentation, with an exponent (D) ranging from 3.0 to 3.9. By incorporating these measured GSD into the initial conditions of a steady-state 1D model of explosive eruption columns, we show that they have a first-order impact on the dynamical behaviour of explosive eruption columns. Starting from an initial value of D, the model tracks the evolution of GSD in the column and calculates the dynamical consequences of particle sedimentation. The maximum height reached by the column, one of the first-order results relevant to aircraft safety, changes by 30% for mass fluxes of 107kgs−1 or larger, and by 45-85% for mass fluxes between 105 and 107kgs−1, depending on exponent D. We compare our predictions to a specially assembled set of geologic field data and remote sensing observations from 10 Plinian eruptions for which maximum column height and mass flux are known independently. The incorporation of realistic power-law GSD in the model greatly improves the predictions, which opens new perspectives for estimation of ash load and GSD in volcanic clouds from near real-time measurements available from satellite payloads. Our results also contribute to the improvement of volcanic source term characterization that is required input for meteorological dispersion models.

71