Abstract
Most of the current ash transport and dispersion models neglect particlefluid (twoway) and particlefluid plus particleparticle (fourway) reciprocal interactions during particle fallout from volcanic plumes. These interactions, a function of particle concentration in the plume, could play an important role, explaining, for example, discrepancies between observed and modelled ash deposits. Aiming at a more accurate prediction of volcanic ash dispersal and sedimentation, the settling of ash particles at particle volume fractions (ϕ_{p}) ranging 10^{−7}10^{−3} was performed in laboratory experiments and reproduced by numerical simulations that take into account first the twoway and then the fourway coupling. Results show that the velocity of particles settling together can exceed the velocity of particles settling individually by up to 4 times for ϕ_{p} ~ 10^{−3}. Comparisons between experimental and simulation results reveal that, during the sedimentation process, the settling velocity is largely enhanced by particlefluid interactions but partly hindered by particleparticle interactions with increasing ϕ_{p}. Combining the experimental and numerical results, we provide an empirical model allowing correction of the settling velocity of particles of any size, density, and shape, as a function of ϕ_{p}. These corrections will impact volcanic plume modelling results as well as remote sensing retrieval techniques for plume parameters.
Introduction
Volcanic ash is injected in the atmosphere during explosive eruptions and, dispersed by wind and eventually deposited on the ground, may cause health, social, and economic disruption during and after an eruption^{1}. Where, when, and how ash from an eruption may impact human activities is currently forecast by using numerical simulations of ash dispersal. Current simulations incorporate wind advection, atmospheric diffusion, particle aggregation, and simplified sedimentation models that consider the terminal velocity of particles as if settling individually^{2,3}. To overcome this last simplification, here we describe joint experimental and numerical simulations investigating the effect of particle volume fraction on the settling velocity of volcanic ash particles in suspensions.
Complexities of the volcanic cloudatmosphere system and large variations in the size, density, shape and concentration of erupted particles result in large variations in ash settling dynamics and corresponding settling velocity. The terminal settling velocity of a single volcanic ash particle is usually derived analytically and/or experimentally, and previous studies focussed primarily on the effect of particle size, shape, and density, and of atmospheric properties on this parameter^{4,5,6,7,8,9,10,11,12}. This approach considers the settling of an individual particle in a still fluid, ignoring the presence of neighbouring particles. Similarly, the effect of particle volume fraction (ϕ_{p}) on the settling velocity of individual (i.e., not aggregated) particles is not included in most numerical models of ash dispersal^{13,14,15,16,17,18,19,20} which are mostly concerned with the fate of volcanic ash in the dilute, mediumdistal regions of the plume. These models assume particles are only affected by the drag due to the local velocity of the carrier flow (oneway coupling), ignoring the effect of particle motion on the flow itself (twoway coupling) and interparticle collisions (fourway coupling). This approach is justified as long as ϕ_{p} is less than 10^{–6}, below which negligible alteration to the structure of turbulence by particles occurs^{21,22}. With increasing ϕ_{p} in the fluid, however, twoway and fourway coupling effects are expected to play an increasingly important role on the settling velocity of particles^{23,24,25,26,27,28}. Within a few hundreds of kilometres from the vent, ϕ_{p} in the plume can be as high as 10^{−5}, and up to 10^{−3} in more proximal regions^{29,30,31}. The observations and modelling of localized regions of instability, including particlerich ‘fingers’ or ‘scalloped umbrella’^{32,33}, suggest that the dynamics of particle sedimentation is strongly governed by several factors mostly related to particle concentration^{34}.
Only very recently has the effect of other volcanic particles perturbing the surrounding fluidparticle system been addressed by numerical simulations of the eruption plume^{35,36}, but has never been corroborated experimentally. If not calibrated against experimental data on volcanic particles, numerical models may not capture this effect properly, resulting in incorrect estimates of ash depletion rate (or residence time) from the plume. The only experimental studies of gasparticle mixtures linking settling velocity to ϕ_{p} are found within the engineering literature and involve industrial droplets and particles at the ten micron scale^{26,28}. These studies reported an increase in the mean particle settling velocity in the mixture compared to the terminal velocity of individual particles in still fluid, explaining this outcome in terms of the preferential sweeping of the particles downstream within local vortical structures^{26,28}.
With the objective of filling the lack of experimental studies on natural volcanic particles, we have performed a systematic experimental study on the effect of ϕ_{p} on the settling velocity of volcanic ash particles in suspensions 10^{−7} < ϕ_{p} < 10^{−3}. We also performed numerical simulations that reproduced the experiments using both a twoway and a fourway coupling approach, pointing to: i) identify the key physical processes in place during the experiment, and ii) test the ability of the code to capture the coupling dynamics at the laboratory scale. Finally, we derived an empirical relation that links the main particle characteristics (particle size, density and partly shape) and settling velocity to ϕ_{p}.
Methods
Experiments
Starting material
Two volcanic ash samples differing in composition, density, shape and size distributions were used in our experiments: i) dense basaltic ash from Etna (ETB), sieved in the diameter classes 125 μm < d < 500 μm; and ii) pumiceous phonolitic ash from Laacher See (LSP) sieved in the class 500 μm <d <1000 μm. Shape and vesicularity of the starting material were characterized on a population of approximately 50 clasts by Scanning Electron Microscope (JEOL JSM6500F) images analysis (Fig. 1). For each particle, binary images generated by thresholding the total projected area and the area fraction occupied by voids, were analysed using the ImageJ analysis toolbox, obtaining the following parameters: i) projected area, ii) perimeter, iii) major and iv) minor axis, v) void fraction. The shape parameters F^{4} and Ψ^{10} (definition in Supplementary Methods S1), calculated assuming the particles as prolate spheroids, show mean values of 0.67 and 0.74 and 0.63 and 0.71 for ETB and LSP, respectively (Table 1). Vesicularity α (i.e. the area fraction occupied by voids) is always negligible (α < 0.1) for ETB samples, with 56% of the analysed population showing α < 0.01. Vesicularity varies significantly for LSP, spanning from 0.3 to 0.9 with ca. 60% of the analysed samples showing α > 0.7 (Fig. 1).
Density of the solid fraction (ρ_{s}, including closed pores) was measured using a helium pycnometer (Micrometrics AccuPyc II 1340), obtaining 2925 ± 32 kg/m^{3} and 2136 ± 41 kg/m^{3} for ETB and LSP, respectively. The density of the analysed particles was computed as ρ_{p} = ρ_{S}(1−α), assuming that measured 2D vesicularity is representative of the actual 3D vesicularity of the bulk particle. Mean density values of the analysed ETB and LSP samples are 2846 ± 101 kg/m^{3} and 764 ± 298 kg/m^{3}, respectively, equal to vesicularity of approximately 3 and 65%. The average density of about 3 × 10^{4} LSP particles was also measured by an independent method^{37}, obtaining a value of 831 ± 60 kg/m^{3}.
Experimental setup
The experiments consisted of the release of a vertical flow of particles in a box, open at the top and bottom, with sizes 7.6 m (height) × 1.8 m (width) × 1.2 m (depth), at ambient pressure (1016 ± 2 hPa), temperature (27 ± 1 °C) and relative humidity (54 ± 3%) conditions (Fig. 2). For both ETB and LSP, different experimental runs were conducted by varying the mass discharge rate at the flow source from a maximum of 1.14 × 10^{−1} kg/s (run f1) to a minimum of 2 × 10^{−4} kg/s (run f5) using parcels of the same starting material (i.e., same grain size, shape, and density distributions). The resulting experimental durations ranged from a maximum of 176 s to a minimum of 10 s. A highspeed camera (Optronis CR600 × 2) equipped with a 135 mm lens was used to acquire images of the particles (at 2000 Hz and 114 μm pixel resolution) in a small region, or control volume, at the centre of the stream of settling particles. This control volume of size 68.4 mm × 31.9 mm × 5 mm was defined by the camera field of view multiplied by the depth of field and was located 5.1 m below the release system.
Image processing & data analysis
A custom designed routine (PyTV^{38}) was used to track framebyframe each particle in the imaged area of the flow. For each point of the track, the routine provided the xy position and the projected area of the tracked particle (see Supplementary Video S3). Due to particle interaction during the settling (see below), particle volume fraction varied during each experimental run, as observed in similar literature experiments^{26}. Thus, for each experimental run, a 50 frameslong (i.e. 0.025s) bin was sampled every 5s starting from the appearance of the first particles to the end of the experiment. Each bin contained a variable number of tracks ranging ca. 2 to 9 × 10^{2}. The routine tracked ca. 1.1 × 10^{4} particles (sum of all runs), from which the following, trackaveraged parameters were obtained: i) equivalent particle diameter d_{e}, taken as the diameter of an equivalent sphere with the same projected area; ii) settling (vertical) velocity v_{p}. For each bin, we also calculated the particle volume fraction ϕ_{p} as the sum of the volumes of all particles in one frame (calculated as V = 4/3π(d_{e}/2)^{3}) divided by the control volume. The obtained ϕ_{p} is a lower bound, considering that we confidently tracked about 90% of the particles within the control volume. In addition, the frame by frame position and area of 20–50 particles per run (using ImageJ Software, with MTrakJ plugin extension) was obtained manually to check consistency of the parameters obtained with the tracking routine.
Results are compared with the models of Kunii and Levenspiel^{5}, Wilson and Huang^{4}, and Dellino et al.^{10} for the settling velocity of (both volcanic and nonvolcanic) individual particles in still fluids (described in Supplementary Methods S1), computed using d_{e}, mean density, and shape factor values of starting material particles as described above.
Numerical Simulations
An EulerianLagrangian approach was adopted to numerically simulate the settling of particles in our ETB experiments, by using the MPPICFoam OpenFoam solver (based on the MultiPhase ParticleinCell method), which includes the effect of the particle volume fraction on the continuous phase. A detailed description of the modelling approach used and the set of underlying equations can be found in the Supplementary Methods S1. The solver has been used to simulate the release and the fall of particles in a domain closed on the sides and open at the bottom with the same size of the experimental box. For each simulation, about 10^{6} to 10^{7} solid spherical particles are released from the top of the domain from a circular vent with a fixed diameter (0.019 m) and initial null velocity. Three different mass discharges have been investigated: i) a total mass of 1.1 kg for 10 s (1.1 × 10^{−1} kg/s), ii) a total mass of 1.24 kg for 40 s (3.1 × 10^{−2} kg/s), iii) a total mass of 0.0352 kg for 176 s (2.0 × 10^{−4} kg/s), reproducing the experimentally generated maximum (f1), intermediate (f3) and minimum (f5) mass discharge rate, respectively. Particles density is assumed constant (2600 kg/m^{3}) and the diameter is described by a truncated normal distribution with mean value of 300 μm, variance 100 μm and minimum and maximum diameter fixed at 100 and 400 μm respectively, similar to the ETB case. For each mass discharge rate, two types of simulations were performed: one with a 2way coupling (S2W), where the reciprocal drag of the interacting air and particles is considered, and one with a fourway coupling (S4W), where also the collisions between particles are accounted for. Similarly to the experiments, the mean particle volume fraction, velocity, and diameter data of the simulated particles were sampled every 0.2 s within a control volume located 5 m below the release source (Table 1).
Results
The settling velocity (v_{p}) of particles measured in both the ETB and LSP experiments spans between 1 and 6 m/s, with similar modal values around 2.5–3 m/s, and a rather platykurtic Gaussian distribution for the narrow range of particle sizes used. Automatic and manually tracked particle velocities plot in the same range (Fig. 3, Table 1). As expected, v_{p} increases with particle size (d_{e}). For any particle size, the vast majority of the measured velocities are higher than those predicted from models in the literature, which only approximate well the lowest measured velocities. For any particle size, runs with higher discharge rate tend to display higher velocities for the ETB sample (see Supplementary Video S2), while this trend is not evident for LSP.
The numerical simulations of ETB (Supplementary Video S4–S5) match well the range of experimentally observed velocities. For the minimum discharge rate (f5), a good correlation with the literature models of Kunii and Levenspiel^{5} and Wilson and Huang^{4} is also observed, while the correlation is worse when using the model of Dellino et al.^{10}, developed for larger and more vesicular clasts and already known to overestimate the settling velocity of particles smaller than ~500 μm^{2,6}. For the intermediate (f3) and maximum (f1) discharge rates, velocities from the numerical simulations are always higher than those predicted by previous literature models considering the settling of individual particles. With respect to the twoway simulations, at the intermediate (f3) and maximum discharge rates (f1) the settling velocity distributions for the fourway simulations display lower modal values and partial overlapping, thus more closely resembling the experimental results.
More than discharge rate, local particle volume fraction ϕ_{p} is the main factor enhancing particle settling velocity v_{p} (Fig. 4). A power law can fit the dependence of these two parameters in both experimental and numerical cases, with vp increasing by a factor of 3 when ϕ_{p} increases from 2 × 10^{−7} to 2 × 10^{−3}. With respect to the experimental, irregular particles, the numerical, spherical ones display – as expected – higher velocities for any volume fraction (despite their slightly lower density), higher for the twoway case than for the fourway one, which better parallels the experimental data. Moreover, the numerical simulations results display a good match between the discharge rate used for the different runs and the corresponding particle volume fraction, while this match is less evident for the ETB runs and not observed in the LSP ones, outlining the greater variability of the natural starting material with respect to the numerical model. Note that, both in the ETB and LSP cases, the experimentallyderived settling velocities for the lowermost particle volume fractions match well the settling velocity predicted by literature models for individual particles of the same size.
Our final goal is to obtain an empirical fit allowing one to calculate the settling velocity of ash particles of known size, density and shape as a function of particle volume fraction. To this goal we followed an approach similar to that of Haider and Levenspiel^{12}. First we nondimensionalized the settling velocity and particle size to compare the experimental (ETB and LSP) and the fourway numerical results (which, compared to the twoway ones, are more realistic and approximate better the experimental results) as:
where μ_{c} and ρ_{c} are the density and viscosity of the carrier fluid (air) at ambient conditions, respectively, and g is gravitational acceleration. All data are then grouped for different ϕ_{p} intervals and each group is fitted with an empirical expression similar (but not identical) to that used by Haider and Levenspiel^{12} for the settling velocity of spherical and nonspherical particles (Fig. 5)
Given that the use of dimensionless parameters accounts for particle size and density, the coefficients a and b of the fit curves are, in our case, function of particle shape and volume fraction only. The goodness of fit given by Equation (3), expressed by the Root Mean Square Error (RMSE), is quite high for the numerical case (RMSE of 0.07–0.62) and more variable for the experimental one (RMSE of 0.60–1.85). We found that fit error is minimized for b = 1 and b = 1.2 for the spherical particles (numerical case) and the irregular ones (experimental case), respectively. These values are also those that fit better the runs with the lowest particle volume fraction, i.e., those more similar to the case of particles settling individually. Holding the above b values constant, we found that a decreases as a function of volume fraction following a power law (Fig. 6), with a constant coefficient ~2.9 and an exponent almost equal to −0.1b so that Equation (3) can be rewritten as
Equation (4) directly links the dimensionless settling velocity and size of particles to the particle volume fraction, with two distinct (but similar) coefficients for spherical and irregular particles, respectively.
Discussion and Conclusions
Our experiments show that ashsized natural volcanic particles settle at greater absolute velocity when the particle volume fraction ϕ_{p} in the surrounding medium increases. The commonly used terminal settling velocity of an individual particle in a still fluid represents the lowest bound for ϕ_{p} → 0. The rate of settling velocity increase with increasing particle volume fraction is similar for both Etna and Laacher See samples, despite their different particle shape and density.
Particle volume fraction has two impacts on settling velocity. First, the relative motion of an individual particle with respect to the surrounding fluid (twoway coupling) induces turbulence along its trajectory, causing a neighbouring particle to be preferentially swept to the downward side of the eddies, with a resultant net force that accelerates downwards both particles and air^{21}. Second, as the particle volume fraction becomes higher, in addition to the twoway coupling between the particles and fluid, particleparticle collisions (fourway coupling) take place increasingly. Such collisions, enhanced by differential settling velocity, may have a horizontal component and do not necessarily result in a downward acceleration. The numerical simulations allow discriminating the effects of twoway and fourway coupling on our experimental results. The S2W case shows the highest settling velocity for any particle volume fraction and is the one that deviates most from the individual particle case. As expected, twoway coupling strongly enhances settling velocity. In the S4W case, this enhancing effect is attenuated by particle collisions, which randomly deviate trajectories and dissipate downward kinetic energy. As a result, at the same particle volume fraction, settling velocity is lower and with a broader distribution with respect to the S2W case. Fourway effects are usually assumed^{24,25,26} to become important at ϕ_{p} > 10^{−3}, but in our numerical cases, possibly as a result of the relatively broad initial size distribution of particles used, these effects seem to became apparent even at lower values of ϕ_{p} (Fig. 4). For the lowermost ϕ_{p} values, the good match between the experimental velocity values and those predicted by literature models^{4,5} for particles settling individually suggests that at ϕ_{p} ~ 10^{−7} twoway effects are still negligible, in agreement with literature observations^{24,26}. At any ϕ_{p}, the settling velocity distribution of the S4W case replicates the experimental case better than the S2W case (Fig. 3), thus confirming the importance of fourway interactions at the conditions proper for the experimental case. Nonsphericity, surface roughness, and density variations of natural particles account for the larger scatter of the experimental case with respect to the S4W one. The densityinduced deviations are accounted for in our dimensionless analysis. Thus, only shape differences cause the deviations (visible in Fig. 5 and quantified in Fig. 6 and through the b coefficient) between the ideal, spherical particles and the natural, irregular ones.
For a given particle volume fraction, Equation (4) allows adjusting dimensionless velocity and thus the settling velocity of particles of known size and density. In addition, the two values of the b coefficient allow adding, to a limited extent, information on the particle shape. Equation (4) is valid only within our experimental boundaries, i.e., 10^{−7} < ϕ_{p} < 10^{−3} and 7 < d^{*} < 50, corresponding to 10 < Re < 100, typical for the intermediate conditions (between laminar and turbulent) that characterize the settling of coarse ash from volcanic plumes^{11}.
Our results stem from observations at a millimetrescale, and their applicability on a larger scale could be questionable. However, temporal fluctuations in our observation window, resulting from experimental and intrinsic effects, partly compensate this spatial limitation by exposing different flow conditions to our analysis. The larger scale validity of our experimental results is corroborated by both the numerical results, which are free from undesired experimental artefacts, and by previous experimental results that shows how velocity/volume fraction relationships are consistent in different regions of the flow^{26}. Reciprocally, the convergence of results supports the applicability of our numerical methods to capture two and fourway coupling dynamics. Future research will investigate the role of other factors (e.g., grain size distribution, presence of ash aggregates, ambient temperature, pressure and humidity) on our results.
To our knowledge, our experiments are the first to investigate fourway coupling using natural volcanic particles. Our results show a significant deviation from ideal, spherical particles. The range of particle size and volume fraction used in the models and experiments match well those expected, for instance, inside recent eruption plumes from Mt. Etna (Italy) within 20 km from the eruptive vent^{30,39,40,41}. In such a case, an ash accumulation rate on the ground of about one centimeter per hour would translate, assuming particles settling at 1 m/s, to a ϕ_{p} of about 4 × 10^{−6}, with fourway coupling effects starting to be important. Observed accumulation rates of one millimeter per hour^{42} would translate to a ϕ_{p} of about 10^{−7}. However, these ϕ_{p} values apply in the volume of air immediately above the ground, while at the base or inside the plume ϕ_{p} is expected to be higher, advection and dispersal forces diluting particles during settling (as seen, on a smaller scale, in our case, see Supplementary Video S5).
Fourway coupling may play an important role in controlling particle sedimentation in different parts of the volcanic plume. Inside the plume itself, in order to sustain particles in suspension, the upward component of turbulence must balance not only the theoretical, quiescentfluid terminal velocity of individual particles but also the enhanced velocity due to particle interactions. Below the plume, regions of higher particle volume fractions are known to develop, e.g., particlerich ‘fingers’ or ‘scalloped umbrella’^{32,33}, and the settling rate from such regions will change according to the local particle volume fraction. In all these cases, our results show the actual velocity of the particles settling at ϕ_{p} ~ 10^{−3} may reach up to a factor of 3–4 higher than that of individual particles settling in a still fluid. Such an increase in particle settling rate would result in earlier deposition of particles of a given size (and consequent depletion of such particles from the plume), affecting the areal and grain size distributions of eruption deposits. Inside such regions, fourway coupling will also impact particle aggregation and disaggregation processes^{43}, and possibly remove finer particles from the plume because of their sweeping by the downdraft of larger ones. Our observations may be also extended to the study of pyroclastic density currents, where the settling velocity of particles is often used to infer the physical properties of the flow^{44}. In such currents, ϕ_{p} is commonly in the range 10^{−5}–10^{−2}, relevant for fourway coupling. Equation 4 may be a rapid alternative to numerical modelling when accounting for fourway coupling in estimating properties of the currents.
Several plume measurement techniques, including Doppler Radar stations and disdrometers, may use inversion algorithms to determine the grain size of settling particles in atmosphere from the measured velocity^{45,46}. Such algorithms are based on stillfluid, individual particle theoretical models that neglect the effect of particle volume fraction. In the presence of such effects particle size would be overestimated.
Most numerical models of ash dispersal adopt a oneway coupling approach^{13,14,15,16,17,18,19,20}, an assumption justified in the dilute regions of the plume where ϕ_{p} is less than 10^{−6}^{31}, in agreement with our above statements. However, depending on plume height, ϕ_{p} may be 10^{−5}–10^{−6} in regions within a few hundreds of kilometres and up to 10^{−3} more proximally^{29,30,31}. This range of volume fractions is covered by our experiments, whose results detail for the first time the role of twoway and fourway coupling effects for volcanic particle mixtures. Twoway coupling is being recently incorporated in numerical simulations of volcanic plumes, but it requires new code development and the simultaneous solution of the equations describing the atmospheric fields and those describing the ash particles trajectories, with a big impact on the computational cost^{35,36} and the real time applicability during a crisis. Equation (4) could provide an empirical alternative also accounting for fourway effects, with a limited application range as for particle size but easily incorporated into preexisting codes.
Additional Information
How to cite this article: Del Bello, E. et al. Effect of particle volume fraction on the settling velocity of volcanic ash particles: insights from joint experimental and numerical simulations. Sci. Rep. 7, 39620; doi: 10.1038/srep39620 (2017).
Publisher's note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
Rymer, H. Volcanic Hazards in The Encyclopedia of Volcanoes (eds Sigurdsson, H., Houghton, B., McNutt, S., Rymer, H. & Stix, J. ) 895–896 (Elsevier, 2015).
Folch, A. A review of tephra transport and dispersal models: Evolution, current status, and future perspectives. J. Volcanol. Geotherm. Res. 235–236, 96–115 (2012).
Bonadonna, C. & Costa, A. Tephra Dispersal and Sedimentation in The Encyclopedia of Volcanoes (eds Sigurdsson, H., Houghton, B., McNutt, S., Rymer, H. & Stix, J. ) 587–597 (Elsevier, 2015).
Wilson, L. & Huang, T. C. The influence of shape on the atmospheric settling velocity of volcanic ash particles. Earth Planet. Sci. Lett. 44, 311–324 (1979).
Kunii, D. & Levenspiel, O. Fluidization engineering. (Elsevier, 2013).
Alfano, F., Bonadonna, C., Delmelle, P. & Costantini, L. Insights on tephra settling velocity from morphological observations. J. Volcanol. Geotherm. Res. 208, 86–98 (2011).
Coltelli, M., Miraglia, L. & Scollo, S. Characterization of shape and terminal velocity of tephra particles erupted during the 2002 eruption of Etna volcano, Italy. Bull. Volcanol. 70, 1103–1112 (2008).
Ganser, G. H. A rational approach to drag prediction of spherical and nonspherical particles. Powder Technol. 77, 143–152 (1993).
Riley, C. M., Rose, W. I. & Bluth, G. J. S. Quantitative shape measurements of distal volcanic ash. J. Geophys. Res. 108, 1–15 (2003).
Dellino, P. et al. The analysis of the influence of pumice shape on its terminal velocity. Geophys. Res. Lett. 32, 1–4 (2005).
Bonadonna, C., Ernst, G. G. J. & Sparks, R. S. J. Thickness variations and volume estimates of tephra fall deposits: the importance of particle Reynolds number. J. Volcanol. Geotherm. Res. 81, 173–187 (1998).
Haider, A. & Levenspiel, O. Drag Coefficient and Terminal Velocity of Spherical and Nonspherical Particles. Powder Technol. 58, 63–70 (1989).
Pfeiffer, T., Costa, A. & Macedonio, G. A model for the numerical simulation of tephra fall deposits. J. Volcanol. Geotherm. Res. 140, 273–294 (2005).
Costa, A., Macedonio, G. & Folch, A. A threedimensional Eulerian model for transport and deposition of volcanic ashes. Earth and Planetary Science Letters 241 (2006).
Folch, A., Costa, A. & Macedonio, G. FALL3D: A computational model for transport and deposition of volcanic ash. Comput. Geosci. 35, 1334–1342 (2009).
Costa, A., Folch, A. & Macedonio, G. A model for wet aggregation of ash particles in volcanic plumes and clouds: 1. Theoretical formulation. J. Geophys. Res. Solid Earth 115, 1–16 (2010).
De’Michieli Vitturi, M., Neri, A., Esposti Ongaro, T., Lo Savio, S. & Boschi, E. Lagrangian modeling of large volcanic particles: Application to Vulcanian explosions. J. Geophys. Res. Solid Earth 115 (2010).
De’Michieli Vitturi, M., Neri, A. & Barsotti, S. PLUMEMoM 1.0: A new integral model of volcanic plumes based on the method of moments. Geosci. Model Dev. 8, 2447–2463 (2015).
Costa, A. et al. Results of the eruptive column model intercomparison study. J. Volcanol. Geotherm. Res. 326, 2–25 (2015).
Macedonio, G., Costa, A., Scollo, S. & Neri, A. Effects of eruption source parameter variation and meteorological dataset on tephra fallout hazard assessment: example from Vesuvius (Italy). J. Appl. Volcanol. 5, 5 (2016).
Elghobashi, S. & Trusdell, G. C. On the twoway interaction between homogeneous turbulence and dispersed solid particles. I. Turbulence modification. Phys. Fluids A Fluid Dyn. 5, 1790–1801 (1993).
Elghobashi, S. Particleladen turbulent flows: direct simulation and closure models. Appl. Sci. Res. 48, 301–314 (1991).
Maxey, M. R. The gravitational settling of aerosol particles in homogeneous turbulence and random flow fields. J. Fluid Mech. 174, 441 (1987).
Wang, L.P. & Maxey, M. R. Settling velocity and concentration distribution of heavy particles in homogeneous isotropic turbulence. J. Fluid Mech. 256, 27–68 (1993).
Fede, P. & Simonin, O. Effect of particleparticle collisions on the spatial distribution of inertial particles suspended in homogeneous isotropic turbulent flows. In Notes on Numerical Fluid Mechanics and Multidisciplinary Design 110, 119–125 (2010).
Aliseda, A., Cartellier, A., Hainaux, F. & Lasheras, J. C. Effect of preferential concentration on the settling velocity of heavy particles in homogeneous isotropic turbulence. J. Fluid Mech. 468, 77–105 (2002).
Bosse, T., Kleiser, L. & Meiburg, E. Small particles in homogeneous turbulence: Settling velocity enhancement by twoway coupling. Phys. Fluids 18 (2006).
Yang, T. S. & Shy, S. S. Twoway interaction between solid particles and homogeneous air turbulence: particle settling rate and turbulence modification measurements. J. Fluid Mech. 526, 171–216 (2005).
Girault, F., Carazzo, G., Tait, S. & Kaminski, E. Combined effects of total grainsize distribution and crosswind on the rise of eruptive volcanic columns. J. Volcanol. Geotherm. Res. 326, 103–113 (2015).
Corradini, S. et al. A multisensor approach for volcanic ash cloud retrieval and eruption characterization: The 23 November 2013 Etna lava fountain. Remote Sens. 8 (2016).
Bursik, M. I., Sparks, R. S. J., Carey, S. N. & Gilbert, J. N. The concentration of ash in volcanic plumes, inferred from dispersal data. US Geol. Surv. Bull. 2047, 19–29 (1994).
Manzella, I., Bonadonna, C., Phillips, J. C. & Monnard, H. The role of gravitational instabilities in deposition of volcanic ash. Geology 43, 211–214 (2015).
Chakraborty, P., Gioia, G. & Kieffer, S. Volcan Reventador’s unusual umbrella. Geophys. Res. Lett. 33, 1–5 (2006).
Carazzo, G. & Jellinek, M. A. A new view of the dynamics, stability and longevity of volcanic clouds. Earth Planet. Sci. Lett. 325–326, 39–51 (2012).
Cerminara, M., Esposti Ongaro, T. & Berselli, L. C. ASHEE1.0: a compressible, equilibrium–Eulerian model for volcanic ash plumes. Geosci. Model Dev 9, 697–730 (2016).
Van Eaton, A. R. et al. Hail formation triggers rapid ash aggregation in volcanic plumes. Nat. Commun. 6, 7860 (2015).
Taddeucci, J. & Palladino, D. M. Particle sizedensity relationships in pyroclastic deposits: Inferences for emplacement processes. Bull. Volcanol. 64, 273–284 (2002).
Gaudin, D., Moroni, M., Taddeucci, J., Scarlato, P. & Shindler, L. Pyroclast Tracking Velocimetry: A particle tracking velocimetrybased tool for the study of Strombolian explosive eruptions. J. Geophys. Res. Solid Earth 119, 5369–5383 (2014).
Andronico, D. et al. Eruption dynamics and tephra dispersal from the 24 November 2006 paroxysm at SouthEast Crater, Mt Etna, Italy. J. Volcanol. Geotherm. Res. 274, 78–91 (2014).
Andronico, D., Scollo, S., Cristaldi, A. & Lo Castro, M. D. Representivity of incompletely sampled fall deposits in estimating eruption source parameters: a test using the 12–13 January 2011 lava fountain deposit from Mt. Etna volcano, Italy. Bull. Volcanol. 76 (2014).
Pardini, F., Spanu, A., De Michieli Vitturi, M., Salvetti, M. V. & Neri, A. Grain size distribution uncertainty quantification in volcanic ash dispersal and deposition from weak plumes. J. Geophys. Res. B Solid Earth 121, 538–557 (2016).
Barsotti, S. et al. Quantitative assessment of volcanic ash hazards for health and infrastructure at Mt. Etna (Italy) by numerical simulation. J. Volcanol. Geotherm. Res. 192, 85–96 (2010).
Del Bello, E. et al. Experimental investigation of the aggregationdisaggregation of colliding volcanic ash particles in turbulent, lowhumidity suspensions. Geophys. Res. Lett. 42, 1068–1075 (2015).
Dufek, J. The Fluid Mechanics of Pyroclastic Density Currents. Annu. Rev. Fluid Mech. 48, 459–485 (2016).
Bonadonna, C. et al. Tephra sedimentation during the 2010 Eyjafjallajökull eruption (Iceland) from deposit, radar, and satellite observations. J. Geophys. Res. Solid Earth 116 (2011).
Gouhier, M. & Donnadieu, F. Mass estimations of ejecta from Strombolian explosions by inversion of Doppler radar measurements. J. Geophys. Res. Solid Earth 113, 1–17 (2008).
Acknowledgements
This project was supported by the EUfunded FP7 project MEDiterranean SUpersite Volcanoes (MEDSUV), grant agreement 308665. Part of the work was also undertaken under the agreement INGVDPC CPV. The LSP particles used in this study were kindly offered by ROTEC GmbH & Co. KG (MühlheimKärlich, Germany). Measurements and characterization of the particles were performed at INGV HPHT lab of experimental Volcanology and Geophysics. Alix VuDuc provided invaluable lab support. U.K. and J.T. acknowledge the support by the Marie Curie Initial Training Network ‘VERTIGO’, funded through the European Seventh Framework Programme (FP7 2007–2013) under Grant Agreement number 607905. We are grateful to Larry Mastin and an anonymous reviewer for their precious comments that substantially improved the paper.
Author information
Authors and Affiliations
Contributions
E.D. and J.T. conceived the experiment, E.D., J.T., P.S., U.K., D.A., S.S. T.R., conducted the experiments, D.A., S.S., U.K. collected and prepared the ash samples used in the experiments, M.d.M.V. performed the numerical simulations and described the simulation methods. E.D., J.T. and M.d.M.V. analysed the results and drafted the manuscript. All authors discussed the results and reviewed the manuscript.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing financial interests.
Rights and permissions
This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
About this article
Cite this article
Del Bello, E., Taddeucci, J., de’ Michieli Vitturi, M. et al. Effect of particle volume fraction on the settling velocity of volcanic ash particles: insights from joint experimental and numerical simulations. Sci Rep 7, 39620 (2017). https://doi.org/10.1038/srep39620
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/srep39620
Further reading

Numerical and experimental research on the degradation mechanism in a dustfree crusher using DEMCFD method
Computational Particle Mechanics (2022)

Sediment waves and the gravitational stability of volcanic jets
Bulletin of Volcanology (2021)

Acoustic probing of the particle concentration in turbulent granular suspensions in air
Scientific Reports (2020)

Low efficiency of large volcanic eruptions in transporting very fine ash into the atmosphere
Scientific Reports (2019)

Dynamics of shallow hydrothermal eruptions: new insights from Vulcano’s Breccia di Commenda eruption
Bulletin of Volcanology (2018)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.