No abstract

Settling-driven gravitational instabilities observed at the base of volcanic ash clouds have the potential to play a substantial role in volcanic ash sedimentation. They originate from a narrow, gravitationally unstable region called a Particle Boundary Layer (PBL) that forms at the lower cloud-atmosphere interface and generates downward-moving ash fingers that enhance the ash sedimentation rate. We use scaled laboratory experiments in combination with particle imaging and Planar Laser Induced Fluorescence (PLIF) techniques to investigate the effect of particle concentration on PBL and finger formation. Results show that, as particles settle across an initial density interface and are incorporated within the dense underlying fluid, the PBL grows below the interface as a narrow region of small excess density. This detaches upon reaching a critical thickness, that scales with (ν2/g′)1/3, where ν is the kinematic viscosity and g′ is the reduced gravity of the PBL, leading to the formation of fingers. During this process, the fluid above and below the interface remains poorly mixed, with only small quantities of the upper fluid phase being injected through fingers. In addition, our measurements confirm previous findings over a wider set of initial conditions that show that both the number of fingers and their velocity increase with particle concentration. We also quantify how the vertical particle mass flux below the particle suspension evolves with time and with the particle concentration. Finally, we identify a dimensionless number that depends on the measurable cloud mass-loading and thickness, which can be used to assess the potential for settling-driven gravitational instabilities to form. Our results suggest that fingers from volcanic clouds characterised by high ash concentrations not only are more likely to develop, but they are also expected to form more quickly and propagate at higher velocities than fingers associated with ash-poor clouds.

Aggregation of volcanic ash is known to significantly impact sedimentation from volcanic plumes. The study of particle aggregates during tephra fallout is crucial to increase our understanding of both ash aggregation and sedimentation. In this work, we describe key features of ash aggregates and ash sedimentation associated with eleven Vulcanian explosions at Sakurajima Volcano (Japan) based on state-of-the-art sampling techniques. We identified five types of aggregates of both Particle Cluster (PC) and Accretionary Pellet (AP) categories. In particular, we found that PCs and the first and third type of APs can coexist within the same eruption in rainy conditions. We also found that the aerodynamic properties of aggregates (e.g., terminal velocity and density) depend on their type. In addition, grainsize analysis revealed that characteristics of the grainsize distributions (GSDs) of tephra samples correlate with the typology of the aggregates identified. In fact, bimodal GSDs correlate with the presence of cored clusters (PC3) and liquid pellets (AP3), while unimodal GSDs correlate either with the occurrence of ash clusters (PC1) or with the large particles (coarse ash) coated by fine ash (PC2).

No abstract

Field observations and laboratory experiments have shown that ash sedimentation can be significantly affected by collective settling mechanisms that promote premature ash deposition, with important implications for dispersal and associated impacts. Among these mechanisms, settling-driven gravitational instabilities result from the formation of a gravitationally-unstable particle boundary layer (PBL) that grows between volcanic ash clouds and the underlying atmosphere. The PBL destabilises once it reaches a critical thickness characterised by a dimensionless Grashof number, triggering the formation of rapid, downward-moving ash fingers that remain poorly characterised. We simulate this process by coupling a Lattice Boltzmann model, which solves the Navier-Stokes equations for the fluid phase, with a Weighted Essentially Non Oscillatory (WENO) finite difference scheme which solves the advection-diffusion-settling equation describing particle transport. Since the physical problem is advection dominated, the use of the WENO scheme reduces numerical diffusivity and ensures accurate tracking of the temporal evolution of the interface between the layers. We have validated the new model by showing that the simulated early-time growth rate of the instability is in very good agreement with that predicted by linear stability analysis, whilst the modelled late-stage behaviour also successfully reproduces quantitative results from published laboratory experiments. The results show that the model is capable of reproducing both the growth of the unstable PBL and the non-linear dependence of the fingers’ vertical velocity on both the initial particle concentration and the particle diameter. Our validated model is used to expand the parameter space explored experimentally and provides key insights into field studies. Our simulations reveal that the critical Grashof number for the instability is about ten times larger than expected by analogy with thermal convection. Moreover, as in the experiments, we found that instabilities do not develop above a given particle threshold. Finally, we quantify the evolution of the mass of particles deposited at the base of the numerical domain and demonstrate that the accumulation rate increases with time, while it is expected to be constant if particles settle individually. This suggests that real-time measurements of sedimentation rate from volcanic clouds may be able to distinguish finger sedimentation from individual particle settling.

<p>The determination of Eruptive Source Parameters (ESPs) is crucial especially for very active volcanoes whose eruptive intensity can vary significantly. In this aim, new strategies are being developed to determine in near real time the total erupted mass (TEM), total grain-size distribution (TGSD) and plume height from ground sampling and remote sensing methods. Since 2011, Etna volcano has produced about 100 paroxysmal episodes characterized by the emission of fountain-fed tephra plumes whose heights reached up to 15 km (above sea level). In this work, we present multi-sensor strategies based on data acquired by the complementary set of remote sensing systems available at Etna. In fact, multi-sensor strategies may help to refine and assess the uncertainty of ESP estimates made by individual sensors, which can present various limitations such as narrow field of views (e.g., visible imagery) and/or low temporal resolution (e.g., satellite-based infrared). First, we show how the combination between tephra-fallout deposit and satellite-based estimates, along with numerical modelling, can help to refine estimates of TEM and TGSD, especially for weak explosive eruption such as the 29 August 2011 paroxysm. We use the model TEPHRA2 and compute synthetic data of ground accumulation to successfully fill significant sampling gaps in the tephra-fallout deposits. Moreover, we find that the Rosin-Rammler equation can be used to inform on missing part of the TGSD, including the tail of very fine ash also detected by satellite-based platforms. Additionally, we compare all estimates of Mass Eruption Rates, Plume height and grain-size distributions made by all available methods including Doppler radar detection, visible and infrared imagery, infrasound arrays, gravimetric signals and tephra-fallout deposit sampling. Accordingly, based on each sensor limitation and capacities, we obtain new constraints on ESP estimates acquired during several paroxysms between 2011-2013 and February 2021. We also bring new insights into the differences and complementarities that exist between the available remote sensing methods, especially in the case of future eruptive events at Mount Etna.</p>

Field observations and laboratory experiments have shown that ash sedimentation can be significantly affected by collective settling mechanisms that promote premature ash deposition, with important implications for associated impacts. Among these mechanisms, settling-driven gravitational instabilities result from the formation of a gravitationally-unstable particle boundary layer (PBL) that grows between volcanic ash clouds and the underlying atmosphere. The PBL destabilises once it reaches a critical thickness characterised by a dimensionless Grashof number, triggering the formation of rapid, downward-moving ash fingers that remain poorly characterised. We simulate this process by coupling a Lattice Boltzmann model, which solves the Navier-Stokes equations for the fluid phase, with a Weighted Essentially Non Oscillatory (WENO) finite difference scheme which solves the advection-diffusion-settling equation describing particle transport. Since the physical problem is advection dominated, the use of the WENO scheme reduces numerical diffusivity and ensures accurate tracking of the temporal evolution of the interface between the layers. We have validated the new model by showing that the simulated early-time growth rate of the instability is in very good agreement with that predicted by linear stability analysis, whilst the modelled late-stage behaviour also successfully reproduces quantitative results from published laboratory experiments. The results show that the model is capable of reproducing both the growth of the unstable PBL and the non-linear dependence of the fingers' vertical velocity on both the initial particle concentration and the particle diameter. Our validated model is used to expand the parameter space explored experimentally and provides key insights into field studies. Our simulations reveal that the critical Grashof number for the instability is about ten times larger than expected by analogy with thermal convection. Moreover, as in the experiments, we found that instabilities do not develop above a given particle threshold. Finally, we quantify the evolution of the mass of particles deposited at the base of the numerical domain and demonstrate that the accumulation rate increases with time, while it is expected to be constant if particles settle individually. This suggests that real-time measurements of sedimentation rate from volcanic clouds may be able to distinguish finger sedimentation from individual particle settling.