Search

The 15 January 2022 climactic eruption of Hunga volcano, Tonga, produced an explosion in the atmosphere of a size that has not been documented in the modern geophysical record. The event generated a broad range of atmospheric waves observed globally by various ground-based and spaceborne instrumentation networks. Most prominent was the surface-guided Lamb wave (≲0.01 hertz), which we observed propagating for four (plus three antipodal) passages around Earth over 6 days. As measured by the Lamb wave amplitudes, the climactic Hunga explosion was comparable in size to that of the 1883 Krakatau eruption. The Hunga eruption produced remarkable globally detected infrasound (0.01 to 20 hertz), long-range (~10,000 kilometers) audible sound, and ionospheric perturbations. Seismometers worldwide recorded pure seismic and air-to-ground coupled waves. Air-to-sea coupling likely contributed to fast-arriving tsunamis. Here, we highlight exceptional observations of the atmospheric waves.

The mild activity of basaltic volcanoes is punctuated by violent explosive eruptions that occur without obvious precursors. Modelling the source processes of these sudden blasts is challenging. Here, we use two decades of ground deformation (tilt) records from Stromboli volcano to shed light, with unprecedented detail, on the short-term (minute-scale) conduit processes that drive such violent volcanic eruptions. We find that explosive eruptions, with source parameters spanning seven orders of magnitude, all share a common pre-blast ground inflation trend. We explain this exponential inflation using a model in which pressure build-up is caused by the rapid expansion of volatile-rich magma rising from depth into a shallow (<400 m) resident magma conduit. We show that the duration and amplitude of this inflation trend scales with the eruption magnitude, indicating that the explosive dynamics obey the same (scale-invariant) conduit process. This scale-invariance of pre-explosion ground deformation may usher in a new era of short-term eruption forecasting. Here, the authors use 20 years of ground deformation data from Stromboli and correlate this data with eruptive records. They find that duration and amplitude of the inflation trend scales with eruption magnitude, indicating that explosive dynamics obeys the same (scale-invariant) conduit process. [ABSTRACT FROM AUTHOR]

The temporal evolution of effusion rate is the main controlling factor of lava spreading and emplacement conditions. Therefore, it represents the most relevant parameter for characterizing the dynamics of effusive eruptions and thus for assessing the volcanic hazard associated with this type of volcanism. Since the effusion rate curves can provide important insights into the properties of the magma feeding system, several efforts have been performed for their classification and interpretation. Here, a recently published numerical model is employed for studying the effects of magma source and feeding dike properties on the main characteristics (e.g., duration, erupted mass, and effusion rate trend) of small‐volume effusive eruptions, in the absence of syn‐eruptive magma injection from deeper storages. We show that the total erupted mass is mainly controlled by magma reservoir conditions (i.e., dimensions and overpressure) prior to the eruption, whereas conduit processes along with reservoir properties can significantly affect mean effusion rate, and thus, they dramatically influence eruption duration. Simulations reproduce a wide variety of effusion rate trends, whose occurrence is controlled by the complex competition between conduit enlargement and overpressure decrease due to magma withdrawal. These effusion rate curves were classified in four groups, which were associated with the different types described in the literature. Results agree with the traditional explanation of effusion rate curves and provide new insights for interpreting them, highlighting the importance of magma reservoir size, initial overpressure, and initial width of the feeding dike in controlling the nature of the resulting effusion rate curve. Key Points: The effects of magma source and feeding dike properties on the evolution of small‐volume effusive eruptions are studied numericallyTotal erupted mass is mainly controlled by magma reservoir conditions (in particular, dimensions and overpressure) prior to the eruptionDifferent types of effusion rate curves were modeled, classified, and compared with the effusion rate trends defined in the literature [ABSTRACT FROM AUTHOR]

Explosive volcanic eruptions can eject large amounts of ash into the atmosphere, posing a serious threat to populations living near the volcano. The abrupt occurrence of such events requires a rapid response and proper volcanic hazard evaluation. Current monitoring procedures still require human intervention, which often results in significant delays between the occurrence of an eruption and notifications being dispatched. We show how dedicated infrasound array processing can be used to detect and notify the authorities, automatically and in real time, of the onset of explosive eruptions. Conceptually, our method relies on the strong coupling between infrasound and the explosive process, and it is not based on probabilistic considerations but on the ability infrasound has to detect the early stage of the explosive phase. This procedure has been tested for the last 8 years, and it is currently applied to issue early warnings for explosive eruptions at Etna Volcano. We show that the system is able to provide a prealert ~1 hr before the eruption, and it has a 96.6% success rate, with only 1.7% false positive alerts and no false negative alerts. This is, to our knowledge, the first example of an operational early warning system totally based on an unsupervised algorithm that provides automatic notifications of eruptions to a government agency. We show that the same early warning concept might be applicable to arrays at large distances (>500 km), suggesting that infrasound could be successfully used to issue automatic notifications of ongoing eruptions at regional to global scales. Plain Language Summary: Most of the volcanic eruptions are rapidly evolving phenomena, strongly limiting the possibility to prompt activate emergency plans. We still lack the possibility to notify volcanic eruptions automatically, making our society highly exposed to the effects of large explosive eruptions. We present the first early‐warning system based on the acoustic waves generated by volcanic eruptions. Our approach relies on the strong coupling between sound and explosive process, and, as for the earthquake, it is not based on probabilistic consideration but on a quasi‐deterministic approach. In the last 8 years, the system was providing a prealert notification ~1 hr before the eruptive onset with 96.6% rate of success, 1.7% positive false alerts, and no negative false alerts. This is, in our knowledge, the first example of an operational early‐warning system totally based on automatic and unmanned algorithm that provides automatic notification of eruption to government agency automatically and without man supervision. Early warnings can be applied even at regional distances (˃500 km), and notification of ongoing volcanic eruptions would be of valuable importance for aviation safety especially for the many active volcanoes worldwide that are still lacking of geophysical monitoring systems. Key Points: We present the first example of operational early warning for volcanic eruptions based on automatic and unsupervised algorithmThe infrasound array processing detects in real‐time explosive eruptions with a 96.6% of success and no false negative alertsThe early‐warning algorithm automatically delivers prealert notification ~1 hr before the occurrence of the eruptive onset [ABSTRACT FROM AUTHOR]

Abstract: Dome collapse is a dramatic volcanic process, yet the dynamics and evolution of dome collapse still present open questions. Observational data are rare, and this limits our ability to interpret the evolution of this phenomenon in terms of risk assessment. We show how the partial dome collapse of Soufrière Hills Volcano on 2010 evolved in less than 45 min and was characterized by five main different episode of dome failure process. Time and amplitude of seismic and infrasonic records associated with successive pyroclastic density currents show a nearly quadratic temporal trend suggesting a self‐accelerating process increasing in intensity up to the failure limit. Each episode generated gravity waves in the atmosphere, representing the first evidence of internal waves formed due to propagation of density currents in stratified fluids. Finally, we use gravity waves to estimate the total erupted mass and the plume height of the Vulcanian explosions triggered by the decompression induced by the collapse. [ABSTRACT FROM AUTHOR]

Many effusive eruptions are characterized by effusion rates that decay exponentially with time, a trend which is generally ascribed to elastic relaxation of a deep magma chamber. Thermal emissions, detected by satellite during the A.D. 2014-2015 Bárðarbunga-Holuhraun eruption (Iceland), indicate that the volume of the erupted magma and effusion rates followed an overall exponential trend that fits the observed major subsidence of the Bárðarbunga caldera floor. This trend continued until a critical flow rate was reached. Hence, the subsidence slowed down and the eruption rapidly ceased, reflecting the ultimate closure of the magma path. We present a model of inelastic magma withdrawal that very closely reproduces all the observed phenomena and provides new insights into the caldera collapses and the driving pressure of other effusive eruptions. [ABSTRACT FROM AUTHOR]

Mass Flow Rate is one of the most crucial eruption source parameter used to define magnitude of eruption and to quantify the ash dispersal in the atmosphere. However, this parameter is in general difficult to be derived and no valid technique has been developed yet to measure it in real time with sufficient accuracy. Linear acoustics has been applied to infrasonic pressure waves generated by explosive eruptions to indirectly estimate the gas mass erupted and then the mass flow rate. Here, we test on Stromboli volcano (Italy) the performance of such methodology by comparing the acoustic derived results with independent gas mass estimates obtained with UV cameras, and constraining the acoustic source by thermal imagery. We show that different acoustic methods give comparable total gas masses in the 2 to 1425 kg range, which are fully consistent with the gas masses derived by UV cameras and previous direct SO 2 measurements. We show that total erupted gas mass, estimated by infrasound is not simply a function of the initial pressure, but rather the full infrasonic waveform should be considered. Thermal imagery provides evidence that infrasound is generated during the entire gas thrust phase. We provide examples to show how total gas masses derived by infrasonic signals can be affected by large uncertainties if duration of the signal is neglected. Only when duration of infrasound is included, the best correlation (0.8) with UV cameras and the 1:1 direct linear proportionality is obtained. Our results open new perspective for remotely derived gas mass and mass flow rates from acoustic signals. [ABSTRACT FROM AUTHOR]

Abstract: During operational ash-cloud forecasting, prediction of ash concentration and total erupted mass directly depends on the determination of mass eruption rate (MER), which is typically inferred from plume height. Uncertainties for plume heights are large, especially for bent-over plumes in which the ascent dynamics are strongly affected by the surrounding wind field. Here we show how uncertainties can be reduced if MER is derived directly from geophysical observations of source dynamics. The combination of infrasound measurements and thermal camera imagery allows for the infrasonic type of source to be constrained (a dipole in this case) and for the plume exit velocity to be calculated (54–142m/s) based on the acoustic signal recorded during the 2010 Eyjafjallajökull eruption from 4 to 21 May. Exit velocities are converted into MER using additional information on vent diameter (50±10m) and mixture density (5.4±1.1kg/m3), resulting in an average ∼9×105 kg/s MER during the considered period of the eruption. We validate our acoustic-derived MER by using independent measurements of plume heights (Icelandic Meteorological Office radar observations). Acoustically derived MER are converted into plume heights using field-based relationships and a 1D radially averaged buoyant plume theory model using a reconstructed total grain size distribution. We conclude that the use of infrasonic monitoring may lead to important understanding of the plume dynamics and allows for real-time determination of eruption source parameters. This could improve substantially the forecasting of volcano-related hazards, with important implications for civil aviation safety. [Copyright &y& Elsevier]

Abstract: The Stromboli volcano erupted on February 27, 2007, after an intense Strombolian activity lasted about 2 months. The eruption was characterized by a series of rapidly evolving phenomena, like the propagation of an effusive fracture along the crater rim, the opening of lateral effusive vents, an unusually large effusive flux (>10 m3/s), the collapse of the crater system, and a major strombolian explosion. This eruption was monitored by an integrated network of multiparameter instruments: broad-band seismometers, infrasonic array, thermal cameras and bore-hole tiltmeters. All the information collected and processed in real-time allowed to draw a clear picture of the eruption dynamics. Thermal imagery documented in real-time how the effusive fracture propagated at 12:24 GMT of February 27 along the northern rim of the NE crater towards the NE direction. The infrasonic array showed that explosive and degassing activity ceased at 10:32 GMT, a couple of hours before the opening of the effusive fractures. The end of the explosive activity coincided with an intense phase of high-frequency (4–22 Hz) tremor which showed two episodes of harmonic spectrum gliding from 4–7 Hz to ~22 Hz. The end of this phase was characterized at 18:26 GMT by a strong infrasonic signal located by the array in the Sciara del Fuoco and associated with the landslide induced by the opening of a lateral effusive vent at 400 m of elevation. This eruption provided a clear example on how the strategy of integrating in near-real-time geophysical information can pinpoint the transition from explosive to effusive activity of the shallow volcanic system at Stromboli. [Copyright &y& Elsevier]

Abstract: Eruptive plume dynamics is a direct expression of explosive style, where duration and altitude of eruptive clouds and dispersion of erupted products is proportional to the degree of fragmentation. We present an analysis of infrasonic and thermal records for explosions at Villarrica (Chile), Stromboli (Italy), Santiaguito and Fuego (Guatemala) volcanoes. Across these four systems magma composition spans from basaltic to dacitic and explosive activity is typically described as ranging from Strombolian to Vulcanian. We use this analysis to provide a quantitative, geophysically-based description of, and discrimination between, the different explosive styles that characterize the four volcanoes. While infrasound is directly related to the emission of over-pressurized gas, and thus solely reflects the plume emission, both plume emission and ascent are detected thermally. Thus, the two data sets together provide a complete description of the plume dynamics. In particular, while infrasound solely reflects the gas-thrust phase driving plume emission, thermal radiation energy is also affected by buoyancy during plume ascent. Thermal radiation energy estimated for explosions at Stromboli and Villarrica (104–107 J) is lower than that for events at Santiaguito and Fuego (108–109 J), but infrasonic energies overlap. This suggests a greater contribution of buoyancy for eruptive clouds at Santiaguito and Fuego when compared with Stromboli and Villarrica. We further investigated the plume dynamics by comparing infrasonic energy, which reflects gas-thrust (E GT ), with the difference between thermal radiation and infrasonic energies, which mostly reflects buoyancy (E B ). Our data distribution reveals two separate clusters. Explosions at Stromboli and Villarrica share low values of buoyancy, pointing to a gas-thrust dominated emission, efficient coupling of the infrasonic source to the atmosphere, and a Strombolian-type source process to generate a plume rich in coarse fragments. In contrast, explosions at Santiaguito and Fuego share large buoyancy (E B ), and are consistent with Vulcanian-type events in which a large part of infrasonic energy is spent in magma fragmentation leading to ascent of buoyant, ash-rich plumes. We demonstrate that the comparison between thermal radiation and infrasonic energies of explosions allows a quantitative discrimination between, and characterization of, Strombolian- and Vulcanian-type explosions, which leads us to suggest that such an approach may be used as an index to further (and quantitatively) discriminate between eruptive styles. [Copyright &y& Elsevier]

On July 18, 2001, two main eruptive vents opened on the southern flank of Mount Etna volcano (Italy) at ∼2100 m and ∼2550 m a.s.l., respectively. The former vent fed mild strombolian activity and lava flows, while the latter represented the main explosive vent, producing strong phreato-magmatic explosions. Explosions at this latter vent, however, shifted to a strombolian style in the following days, before switching back to phreato-magmatic activity towards the end of the eruption, which ended on August 9, 2001.On August 3, a small seismoacoustic array was deployed close to the eruptive vents. The array was composed of three stations, which recorded seismic and infrasonic waves coming from both of the eruptive vents. A further seismoacoustic station, equipped with a thermal-infrared sensor, was also installed several kilometers north of the first array. Seismic signals relating to the strombolian activity at the 2100-m vent were characterized by a strong decompression at the source. Analysis of the time delays between seismic, infrasonic and infrared event onsets also revealed that ejection velocities during explosions from both vents were subsonic. Time delays between the onset of explosive events apparent in the infrared and infrasound data indicated that the explosion source at the 2550-m vent was located 220–250 m below the crater rim. In comparison, the depth of the seismic source was estimated to be between 230 and 335 m below the rim. This converts to 120–150 and 130–235 m below the preexisting ground surface. In addition, time delays between seismic and infrasonic signals recorded for the lower (2100 m) vent also revealed a seismic source that was no more than a few tens of meters deeper than the fragmentation surface. [Copyright &y& Elsevier]

Among ground-based volcano monitoring techniques, infrasound is the only one capable of detecting explosive eruptions from distances of thousands of kilometers. We show how infrasound array analysis, using acoustic amplitude and detection persistency, allows automatic, near-real-time identification of eruptions of Etna volcano (Italy), for stations at distances greater than 500 km. A semi-empirical attenuation relation is applied to recover the pressure time history at the source using infrasound recorded at global scale (>500 km). An infrasound parameter (IP), defined as the product between the number of detections, filtered for the expected back-azimuth of Etna volcano, and range corrected amplitude, is compared with the explosive activity at Etna volcano that was associated with aviation color code RED warnings. This shows that, during favourable propagation conditions, global arrays are capable of identifying explosive activity of Etna 87% of the period of analysis without negative false alerts. Events are typically not detected during unfavourable propagation conditions, thus resulting in a time variable efficiency of the system. We suggest that infrasound monitoring on a global scale can provide timely input for Volcanic Ash Advisory Centres (VAAC) even when a latency of ~1 hour, due to propagation time, is considered. The results highlight the capability of infrasound for near-real-time volcano monitoring at a regional and global scale. [ABSTRACT FROM AUTHOR]

Pulsatory eruptions are marked by a sequence of explosions which can be separated by time intervals ranging from a few seconds to several hours. The quantification of the periodicities associated with these eruptions is essential not only for the comprehension of the mechanisms controlling explosivity, but also for classification purposes. We focus on the dynamics of pulsatory activity and quantify unsteadiness based on the distribution of the repose time intervals between single explosive events in relation to magma properties and eruptive styles. A broad range of pulsatory eruption styles are considered, including Strombolian, violent Strombolian and Vulcanian explosions. We find a general relationship between the median of the observed repose times in eruptive sequences and the viscosity of magma given by η ≈ 100 ⋅ t median . This relationship applies to the complete range of magma viscosities considered in our study (10 2 to 10 9 Pa s) regardless of the eruption length, eruptive style and associated plume heights, suggesting that viscosity is the main magma property controlling eruption periodicity. Furthermore, the analysis of the explosive sequences in terms of failure time through statistical survival analysis provides further information: dynamics of pulsatory activity can be successfully described in terms of frequency and regularity of the explosions, quantified based on the log-logistic distribution. A linear relationship is identified between the log-logistic parameters, μ and s . This relationship is useful for quantifying differences among eruptive styles from very frequent and regular mafic events (Strombolian activity) to more sporadic and irregular Vulcanian explosions in silicic systems. The time scale controlled by the parameter μ , as a function of the median of the distribution, can be therefore correlated with the viscosity of magmas; while the complexity of the erupting system, including magma rise rate, degassing and fragmentation efficiency, can be also described based on the log-logistic parameter s , which is found to increase from regular mafic systems to highly variable silicic systems. These results suggest that the periodicity of explosions, quantified in terms of the distribution of repose times, can give fundamental information about the system dynamics and change regularly across eruptive styles (i.e., Strombolian to Vulcanian), allowing for direct comparison and quantification of different types of pulsatory activity during these eruptions. [ABSTRACT FROM AUTHOR]

Abstract: We report additional radon data collected at Stromboli during 2002–2007. The whole data set of periodic measurements has been systematically analyzed to retrieve the values of background, threshold and anomaly for all the stations of the network. Maps of radon concentrations in space and time correlate with changes in volcanic activity. Higher radon emissions are essentially concentrated at summit stations and may also affect some stations located onto the summit-eastern sector of the cone (along the N60°E fracture zone), as well as two stations placed at the NE edge of Sciara del Fuoco. Most of these stations are well above their threshold values during effusive eruptions (namely 2002–2003 and 2007) due to the progressive structural adjustment of the volcano edifice coupled with the opening and resetting of the fracture network. A two-year-long timeseries of automatic 222Rn measurements (2005–2007) was analyzed together with local environmental parameters and selected geophysical data (on seismic tremor and infrasonic puffing). The data show good correlation between radon and the latter parameters particularly during periods of high Strombolian activity (March–April 2006), and before the onset of the February 27, 2007 lava effusion (with radon emissions being well above 20,000Bq/m3). Extremely high radon emissions (up to ~1.7×106 Bq/m3) were recorded along the fractures of the SW crater during the first two months of the 2002–2003 effusive eruption. Very high radon emissions (up to ~470,000Bq/m3 for 222Rn, and ~780,000Bq/m3 for 220Rn, respectively) were also detected by the automatic measurements at the summit station during June 20–early July 2007: i.e., just prior to the resuming of the Strombolian activity at summit craters. These data give us the opportunity to estimate the relative depths of the summit fractures during fracturing (being constrained between 200 and 310m). This is in good agreement with the hypocentral depths of the so called “hybrid events” and the source of the very long period seismic signals (VLP) localized at a depth of about 500ma.s.l. below the upper–central part of the Sciara del Fuoco. [Copyright &y& Elsevier]

Abstract: Masses erupted during normal explosions at Stromboli volcano (Italy) are notoriously difficult to measure. We present a method that uses thermal infrared video for cooling bomb fields to obtain the total power emitted by all hot particles emitted during an explosion. A given mass of magma (M) will emit a finite amount of thermal power, defined by M c p (T e −T 0), c p and T e being magma specific heat capacity and temperature, and T 0 being ambient temperature. We use this relation to convert the total power emitted by the bomb field to the mass required to generate that power. To do this we extract power flux curves for the field and integrate this through time to obtain total power (E). This is used to estimate mass (Q) in Q=E/c p (T e −T 0). When applied to individual bombs we obtain masses of between 1 and 9kg per bomb, or a volume of 970 and 6500cm3. These volumes equate to spheres with diameters 12 and 27cm. For the entire bomb field we obtain volumes of 7–28m3. We calculate masses for 32 eruptions and obtain typical bomb masses of between 103 and 104 kg per eruption. In addition, we estimate that between 102 and 103 kg of gas and ash are emitted as part of a mixed plume of bombs, gas and ash. We identify two types of eruption on the basis of the erupted bomb masses and the ratio of the plume''s gas-and-ash component to the bomb component. The first type is bomb-dominated, is characterized by bomb masses of 104 kg and has ash–gas/ bomb ratios of ∼0.02. The second type is ash-and-gas dominated, is characterized by erupted bomb masses of 103 kg and has ash–gas/bomb ratios of around one, and as high as two. There is no correlation between the quantity of bombs and quantity of gas–ash erupted. In addition, while source pressure for each explosion correlates with the quantity of gas and ash erupted, the mass of bombs emitted varies independently of pressure. [Copyright &y& Elsevier]

Abstract: Twelve years of night-time MODIS (Moderate Resolution Imaging Spectroradiometer) observations, has been analysed to detect and quantify the radiative heat power emitted by Stromboli volcano (from March 2000 to September 2011). Using an accurate background subtraction of the MODIS signal at 4μm, we were able to discriminate two main regimes of thermal radiation, related to different levels of volcanic activity. Effusive eruptions (occurred on December 28, 2002 and February 27, 2007) radiated at an average of ~186MW with a frequency of alert detection of 50–95%. Conversely, during the typical strombolian activity, an average of ~9MW is radiated, with a frequency of alert detection of 0–45%. Although during the effusive eruptions the radiative power is basically controlled by the lava discharge rate, our results suggest that during non-effusive periods (strombolian regime) both the intensity and the frequency of MODIS alerts are controlled by the height of the magmatic column feeding the activity at the surface. In particular we found that a radiative power of ~50MW corresponds to a high magma column which is exposed, in the vent area, at the same rate in which the deeper gas-rich magma is typically supplied within the feeding system of Stromboli (~0.3m3 s−1). In this condition the magmatic system approaches steady state regimes. Above this threshold a transition from strombolian to effusive regimes may occur as shown by the detection of ~50MW, 8–10days before the onset of both the last two major flank eruptions. These values were reached after 1–2months of gradual increase of the radiative power which was likely associated the rising of the magma column within the shallowest portion of the conduit. In addition our data suggest that over the years 2000–2011 several cycles of rise and fall of the magma column have occurred, which however did not culminate into an effusive eruption but only into recurrent episodes of sustained spattering or fountaining and summit overflows. These fluctuations has substantially increased in frequency and intensity after the 2007 eruption thus suggesting that this event has perturbed in some way the shallow plumbing system of Stromboli. We stress that the detection of a radiative power higher than 50MW is a clear evidence of a very high magma column, which may prelude the onset of an effusive eruption and/or periods of sustained vent activity. In conclusion, we suggest that systematic analysis of MODIS data can be used to detect variations in the intensity of strombolian activity and may considerably improve volcano surveillance at Stromboli, as well as at other open-system volcanoes. [Copyright &y& Elsevier]

Abstract: We report on the evaluation of in situ measurements of eruption velocities and relative mass flux, collected by Doppler radar, as well as acoustic and infrared data, that were recorded at Stromboli volcano. Doppler radar observations were made alongside thermal, acoustic and seismic measurements, to (1) further investigate the complex waveforms of the SW crater by combining infrasonic, infrared and Doppler radar measurements, (2) establish a relationship between infrared, acoustic and Doppler radar measurements and (3) verify that all instruments observe the same behavior in terms of relative mass eruption rate. We also explore the relationship between kinetic and acoustic energy released during an eruption. Comparing the different methods to each other we are able to show that the mass erupted can be estimated either from the total reflected Doppler radar energy or from infrared observations. However, neither thermal nor reflected energy can provide a value in terms of absolute mass in kg. The erupted masses of different eruptions can only be evaluated relative to each other. Using the combined three data sets we especially focus on the eruption dynamics of the SW crater of Stromboli, namely its fluctuations in eruption strength. These pulses in one eruptive event, so called ’pulsations’ dominate more than 40% of the eruptions of SW crater. Previous models that explain pulsations to be generated by multiple consecutively exploding bubbles are supported by our combined analysis of infrasound, thermal data, particle velocities, and the reflected energy. [Copyright &y& Elsevier]

Abstract: On April 5, 2003, Stromboli volcano (Italy) produced the most violent explosion of the past 50 years. The event was exceptionally well documented thanks to the presence on the island of several scientists and a large number of instruments deployed over the preceding months to monitor the effusive eruption that began in December 2002. Integration of visual documentation, deposit features and geophysical data allowed an accurate reconstruction of the explosive event and its dynamics. The eruption consisted of a 8-min long explosive event which evolved through four phases whose timing was precisely recorded by an infrared thermometer located about 450 m from the summit crater. Phases 2 and 3 lasted 39 and 42 s, respectively. Both had an impulsive character, were responsible for ejecting almost the entire mass of the pyroclastic products. Phases 1 and 4 represented, respectively, a short-lived precursory event and a waning tale. During Phase 2, meter-sized ballistic blocks were launched with velocities of 170 m/s to altitudes of up to 1400 m above the craters. These fell on the volcano flanks and on the village of Ginostra, about 2 km distant from the vent. A vertical jet rose above the craters which developed to feed a convective plume that reached a height of up to 4 km. The calculated mass of the Phase 2 fallout deposit and mass discharge rate were 1.1–1.4×108 kg and 2.8–3.6×106 kg/s, respectively. During Phase 3 a scoria flow deposit, with an estimated volume of 0.9–1.1×104 m3, was erupted from the same vent that fed the ongoing sustained lava flow. The average mass discharge rate for this phase was 2.5–3.1×105 kg/s. Products emitted during Phases 2 and 3 consisted of lithic and fresh magmatic material in similar proportions. The juvenile fraction consisted of a deep-originated, almost aphyric, highly vesicular pumice mingled with a shallow-derived, crystal-rich, moderately vesicular scoria. Similarities with the eruption dynamics of other historical paroxysms at Stromboli makes the April 5, 2003 explosion representative of these highly energetic events that constitute the most hazardous volcanic phenomena at Stromboli volcano. [Copyright &y& Elsevier]