Search

Explosive sequences are quite common at basaltic and andesitic volcanoes worldwide. Studies aimed at short-term forecasting are usually based on seismic and ground deformation measurements, which can be used to constrain the source region and quantify the magma volume involved in the eruptive process. However, during single episodes of explosive sequences, integration of camera remote sensing and geophysical data are scant in literature, and the total volume of pyroclastic products is not determined. In this study, we calculate eruption parameters for four powerful lava fountains occurring at the main and oldest Mt. Etna summit crater, Voragine, between 3 and 5 December 2015. These episodes produced impressive eruptive columns and plume clouds, causing lapilli and ash fallout to more than 100 km away. We analyse these paroxysmal events by integrating the images recorded by a network of monitoring cameras and the signals from three high-precision borehole strainmeters. From the camera images we calculated the total erupted volume of fluids (gas plus pyroclastics), inferring amounts from 1.9 × 10 9 m 3 (first event) to 0.86 × 10 9 m 3 (third event). Strain changes recorded during the first and most powerful event were used to constrain the depth of the source. The ratios of strain changes recorded at two stations during the four lava fountains were used to constrain the pyroclastic fraction for each eruptive event. The results revealed that the explosive sequence was characterized by a decreasing trend of erupted pyroclastics with time, going from 41% (first event) to 13% (fourth event) of the total erupted pyroclastic volume. Moreover, the volume ratio fluid/pyroclastic decreased markedly in the fourth and last event. To the best of our knowledge, this is the first time ever that erupted volumes of both fluid and pyroclastics have been estimated for an explosive sequence from a monitoring system using permanent cameras and high precision strainmeters. During future explosive paroxysmal sequences this new approach might help in monitoring their evolution also to understand when/if they are going to finish. Knowledge of the total gas and pyroclastic fractions erupted during each lava fountain episode would improve our understanding of their processes and eruptive behaviour.

The effusion rate is the most important parameter to gather when a volcanic eruption occurs, because it controls the way in which a lava body grows, extends and expands, influencing its dimensional properties. Calculation of lava flow volume from thermal images collected by helicopter surveys has been largely used during the last decade for monitoring subaerial effusive eruptions. However, due to the depths where volcanic activity occurs, monitoring submarine volcanic eruptions is a very difficult task. The 2011–2012 submarine volcanic eruption at El Hierro, Canary Islands, has provided a unique and excellent opportunity to monitor eruptive processes occurring on the seabed. The use of a hand-held thermal camera during daily helicopter flights allowed us to estimate for the first time the daily and total erupted magma volumes from a submarine eruption. The volume of magma emitted during this eruption has been estimated at 300 Mm3, giving an average effusion rate of ~ 25 m3 s− 1. Thermal imagery by helicopter proved to be a fast, inexpensive, safe and reliable technique of monitoring volcanic eruptions when they occur on the shallow seabed.

The 15 March, 2007 Vulcanian paroxysm at Stromboli volcano was recorded by several instruments that allowed description of the eruptive sequence and unraveling the processes in the upper feeding system. Among the devices installed on the island, two borehole strainmeters recorded unique signals not fully explored before. Here we present an analysis of these signals together with the time-lapse images from a monitoring system comprising both infrared and visual cameras. The two strainmeter signals display an initial phase of pressure growth in the feeding system lasting ∼2 min. This is followed by 25 s of low-amplitude oscillations of the two signals, that we interpret as a strong step-like overpressure building up in the uppermost conduit by the gas-rich magma accumulating below a thick pile of rock produced by crater rim collapses. This overpressure caused shaking of the ground, and triggered a number of small landslides of the inner crater rim recorded by the monitoring cameras. When the plug obstructing the crater was removed by the initial Vulcanian blast, the two strainmeter signals showed opposite sign, compatible with a depressurizing source at ∼1.5 km depth, at the junction between the intermediate and shallow feeding system inferred by previous studies. The sudden depressurization accompanying the Vulcanian blast caused an oscillation of the source composed by three cycles of about 20 s each with a decreasing amplitude, as well recorded by the strainmeters. The visible effect of this behavior was the initial Vulcanian blast and a 2–3 km high eruptive column followed by two lava fountainings displaying decreasing intensity and height. To our knowledge, this is the first time that such a behavior was observed on an open conduit volcano.

Volcanic eruptions are commonly preceded, accompanied, and followed by variations of a number of detectable geophysical and geochemical manifestations. Many remote sensing techniques have been applied to tracking anomalies and eruptive precursors, and monitoring ongoing volcanic eruptions, offering obvious advantages over in situ techniques especially during hazardous activity. Whilst spaceborne instruments provide a distinct advantage for collecting data remotely in this regard, they still cannot match the spatial detail or time resolution achievable using portable imagers on the ground or aircraft. Hand-held infrared camera technology has advanced significantly over the last decade, resulting in a proliferation of commercially available instruments, such that volcano observatories are increasingly implementing them in monitoring efforts. Improved thermal surveillance of active volcanoes has not only enhanced hazard assessment but it has contributed substantially to understanding a variety of volcanic processes. Drawing on over a decade of operational volcano surveillance in Italy, we provide here a critical review of the application of infrared thermal cameras to volcano monitoring. Following a summary of key physical principles, instrument capabilities, and the practicalities and methods of data collection, we discuss the types of information that can be retrieved from thermal imagery and what they have contributed to hazard assessment and risk management, and to physical volcanology. With continued developments in thermal imager technology and lower instrument costs, there will be increasing opportunity to gather valuable observations of volcanoes. It is thus timely to review the state of the art and we hope thereby to stimulate further research and innovation in this area.

Basaltic 'a'ā lava flows often demonstrate compound morphology, consisting of many juxtaposed and superposed flow units. Following observations made during the 2001 eruption of Mt. Etna, Sicily, we examine the processes that can result from the superposition of flow units when the underlying units are sufficiently young to have immature crusts and deformable cores. During this eruption, we observed that the emplacement of new surface flow units may reactivate older, underlying units by squeezing the still-hot flow core away from the site of loading. Here, we illustrate three different styles of reactivation that depend on the time elapsed between the emplacement of the two flow units, hence the rheological contrast between them. For relatively long time intervals (2 to 15 days), and consequently significant rheological contrasts, superposition can pressurise the underlying flow unit, leading to crustal rupture and the subsequent extrusion of a small volume of high yield strength lava. Following shorter intervals (1 to 2 days), the increased pressure caused by superposition can result in renewed, slow advance of the underlying immature flow unit front. On timescales of < 1 day, where there is little rheological contrast between the two units, the thin intervening crust can be disrupted during superposition, allowing mixing of the flow cores, large-scale reactivation of both units, and widespread channel drainage. This mechanism may explain the presence of drained channels in flows that are known to have been cooling-limited, contrary to the usual interpretation of drainage as an indicator of volume-limited behaviour. Because the remobilisation of previously stagnant lava can occur swiftly and unexpectedly, it may pose a significant hazard during the emplacement of compound flows. Constant monitoring of flow development to identify areas where superposition is occurring is therefore recommended, as this may allow potentially hazardous rapid drainage events to be forecast. Reactivation processes should also be borne in mind when reconstructing the emplacement of old lava flow fields, as failure to recognise their effects may result in the misinterpretation of features such as drained channels.

The 26 October 2002–28 January 2003 eruption of Mt. Etna volcano was characterised by lava effusion and by an uncommon explosivity along a 1 km-long-eruptive fissure on the southern, upper flank of the volcano. The intense activity promoted rapid growth of cinder cones and several effusive vents. Analysis of thermal images, recorded throughout the eruption, allowed investigation of the distribution of vents along the eruptive fissure, and of the nature of explosive activity. The spatial and temporal distribution of active vents revealed phases of dike intrusion, expansion, geometric stabilization and drainage. These phases were characterised by different styles of explosive activity, with a gradual transition from fire fountaining through transitional phases to mild strombolian activity, and ending with non-explosive lava effusion. Here we interpret the mechanisms of the dike emplacement and the eruptive dynamics, according to changes in the eruptive style, vent morphology and apparent temperature variations at vents, detected through thermal imaging. This is the first time that dike emplacement and eruptive activity have been tracked using a handheld thermal camera and we believe that its use was crucial to gain a detailed understanding of the eruptive event.

The 5 April 2003 paroxysmal explosion at Stromboli volcano was one of the strongest explosive events of the last century. It occurred while the effusive eruption, begun on 28 December 2002 and finished on 22 July 2003, was still on going and the summit craters of the volcano were obstructed. In this paper, we present a reconstruction of the sequence of events based on thermal and visual images collected from helicopter before, during and immediately after the paroxysm. One month before the blast, ash emission and temperature increase at the bottom of the summit craters were observed. An increasing amount of juvenile components in the emitted ash during March suggested that the magma level within the crater was rising accordingly. Hot degassing vents at the bottom of the summit craters were not persistent, and the craters remained almost entirely obstructed by talus accumulation until the paroxysm occurred. Three minutes before the explosion, we recorded a significant increase in temperature inside Crater 1, accompanied by a thicker gas plume. Thirty-two seconds before the blast, reddish ash was emitted from Crater 1. The paroxysm produced a vulcanian explosion that opened the feeder conduit, obstructed for over three months. The blast was accompanied by a shock wave recorded by the INGV seismic network at 07:13:37 GMT. Explosions with hot material started from Crater 1, and after 15 s propagated to Crater 3, about 100 m away. The velocity of ejecta was ∼80 m s − 1 , and increased when the eruptive plumes from both craters merged together during the vulcanian phase. An eruptive column rose 1 km above the top of the volcano, and explosions continued mainly at Crater 3. The paroxysm lasted about 9 min, with bombs up to 4 m wide falling on the village of Ginostra, on the west flank of the island, and destroying two houses. This event signalled the start of the declining phase of the effusive eruption, suggesting that the feeder conduit was returning to its former steady conditions, with open vents and continuous, mild strombolian activity.

The role of sector collapse in the generation of catastrophic volcanigenic tsunami has become well understood only recently, in part because of the problems in the preservation and recognition of tsunami deposits. Tinti et al. [Tinti, S., Bortolucci, E., Romagnoli, C., 2000. Computer simulations of tsunamis due to sector collapse at Stromboli, Italy. J. Volcanol. Geotherm. Res. 96, 103–128] modeled a tsunami produced by the c. 5,000 years BP collapse of the Sciara del Fuoco on the island volcano Stromboli. Although deposits associated with this event are generally lacking on the island, volcaniclastic breccias on the SE side of the island extending to an elevation above 120 m a.s.l. may have been generated by this tsunami. Deposits above 100 m are dominated by coarse breccias comprising disorganized, poorly sorted, nonbedded, angular to subangular lava blocks in a matrix of finer pyroclastic debris. These breccias are interpreted as a water-induced mass flow, possibly a noncohesive debris flow, generated as colluvial material on steep slopes was remobilized by the return flow of the tsunami wave, the run-up of which reached an elevation exceeding 120 m a.s.l. Finer breccias of subrounded to rounded lava blocks cropping out at 15 m a.s.l. are similar to modern high-energy beach deposits and are interpreted as beach material redeposited by the advancing tsunami wave. The location of these deposits matches the predicted location of the maximum tsunami wave amplitude as calculated by modeling studies of Tinti et al. [Tinti, S., Bortolucci, E., Romagnoli, C., 2000. Computer simulations of tsunamis due to sector collapse at Stromboli, Italy. J. Volcanol. Geotherm. Res. 96, 103–128]. Whereas the identification and modeling of paleo-tsunami events is typically based on the observation of the sedimentary deposits of the tsunami run-up, return flow may be equally or more important in controlling patterns of sedimentation.

Effusion rate and degassing data collected at Mt. Etna volcano (Italy) in 2001 show variations occurring on time scales of hours to months. We use both long- and short-term data sets spanning January to August to identify this variation. The long data sets comprise a satellite- and ground-based time series of effusion rates, and the latter include field-based effusion rate and degassing data collected May 29–31. The satellite-derived effusion rates for January through August reveal four volumetric pulses that are characterized by increasing mean effusion rate values and lead up to the 2001 flank eruption. Peak effusion rates during these 23–57 day pulses were 1.2 m3 s−1 in Pulse 1 (1 Jan–4 Mar), 1.1 m3 s−1 in Pulse 2 (5 Mar–21 Apr), 4.2 m3 s−1 in Pulse 3 (24 Apr–18 Jun), 8.8 m3 s−1 in Pulse 4 (23 Jun–16 Jul), and 22.2 m3 s−1 during the flank eruption (17 Jul–9 Aug). Rank-order analysis of the satellite data shows that effusion rate values during the 2001 flank eruption define a statistically different trend than Etna's persistent activity from Jan 1 to Jul 17. Data prior to the flank eruption obey a power-law relationship that may define an effusion rate threshold of ∼3–5 m3 s−1 for Etna's typical persistent activity. Our short-term data coincide with the satellite-derived peak effusion period of Pulse 3. Degassing (at-vent puff frequency) shows a general increase from May 29 to 31, with hour-long variations in both puff frequency and lava flow velocity (effusion rate). We identify five 3–14 h degassing periods that contain 26 shorter (19–126 min-long) oscillations. This variation shows some positive correlation with effusion rate measurements during the same time period. If a relationship between puff frequency and effusion rate is valid, we propose that their short-term variation is the result of changes in the supply rate of magma to the near-vent conduit system. Therefore, these short-term data provide some evidence that the clear weeks- to months-long variation in Etna's effusive activity (January–August 2001) was overprinted by a minutes- to hour-scale oscillation in shallow supply.

We have undertaken detailed observations of the formation of the ‘Laghetto’ cinder cone, a new cone that formed during a 2-week period of intense activity in Piano del Lago, on the upper slopes of Mount Etna in summer 2001. We describe the events leading to the formation of a small graben, the formation of pit craters on the base of the graben, the onset of phreatomagmatic activity, a transition to intense Strombolian activity, and a return to phreatomagmatic activity as the eruption came to an end. We discuss the reasons for these transitions, and describe the morphological development of the cone during these events. Arcuate cracks on the southern part of the cone were related to withdrawal of magma at the end of the eruption. Other slope instabilities that developed during the eruption include the formation of small radial grain flows on the outer flanks of the cone and the collapse into the crater of part of the crater rim. Some of the failure planes we observed were first identified using a FLIR TM 695 thermal infrared camera. This is the first time that infrared thermography has been used to detect instability of volcanic structures. Results obtained during this test case demonstrate that thermal cameras are a very useful tool for studies of volcanic instability.

A number of tumuli formed on the aa-dominated lava fan complex which developed in the medial zone of the 1983 flow-field of Mount Etna during the later stages of the eruption. This complex flow-field formed on shallow sloping ground below a scarp between 1900 and 1700 m asl. A major tube system fed a branching tube network in the fan complex. Numerous tumuli and break-outs of lava formed in the fan. Three main types of tumulus are identified: (1) Focal tumuli, which are formed from the break-up and uplift of ‘old’, thick lava crust and themselves become sustained sites for the distribution of lava both as flows and within distributary tubes. These focal tumuli are significant centres associated with major tubes. (2) Satellite tumuli, which are typically elongate, whale-back shaped features that branch out from focal tumuli. These satellite tumuli were initially lava flows erupted from a focal tumulus. The crust of the flow slowed or came to a halt and the rigid crust became uplifted and fractured, forming a dome-shaped ridge feature. These satellite tumuli continued to be fed from the focal tumulus and became sites of lava emission with numerous break-outs. (3) Distributary tumuli formed on the fan associated with short-lived break-outs from tubes and are relatively simple structures formed from limited effusion of toey lobes and pahoehoe lava. The major tumuli on the fan complex show distinct dilation fractures. The fracture surfaces provide good exposure of the crust and three distinct zones are recognised – an upper zone showing columnar jointing, a middle zone consisting of planar fracture surfaces and a basal zone with distinctive banded planar fracture surfaces showing evidence of both brittle and ductile formation. Using these data a model is proposed for tumulus growth. Field analysis of the fan complex shows how it was fed by a branching tube system, leading to flow thickening, formation of tumuli and numerous ephemeral boccas.

There is a requirement for rapid remote monitoring of ground deformation of volcanoes, especially during eruptions. A long-range reflector-less laser scanning instrument, manufactured by Riegl Laser Measurement Systems, was used on Mount Etna in October 2000, and its potential for monitoring ground surface changes associated with volcanic activity was evaluated. The 3DLM LR1 scanner system can measure distances up to 2 km with an accuracy of ±25 mm in typical conditions. Measurement rates range from 1 point per second at 2 km to 4 points per second at 10 m. The system also records reflective intensity, and this proved useful in distinguishing between different types of volcanic material. Cloud and gas obscured some measurements, but the scanning nature of the instrument, coupled with effective data filtering software, allowed robust digital terrain models to be created.

Cellular Automata provide an alternative approach to standard numerical methods for modelling some complex natural systems, the behaviour of which can be described in terms of local interactions of their constituent parts. SCIARA is a 2-D Cellular Automata model which simulates lava flows. It was tested on, validated by, and improved on several Etnean lava events such as the 1986–1987 eruption and the first and last phase of the 1991–1993 event. With respect to forecasting the surface covered by the lava flows, the best results were acceptable. The model has been used to determine hazard zones in the inhabited areas of Nicolosi, Pedara, S. Alfio and Zafferana (Sicily, Italy). The main goal of the current work in the Etnean area from Nicolosi to Catania has been the verification of the volcanic hazard effects of an eruptive crisis similar to the event that occurred in 1669. The simulation uses the volcanic data of the 1669 eruption with present-day morphology. Catania has been affected by some historical Etnean events, the most famous one being the 1669 eruption, involving 1 km 3 of lava erupted over the course of 120 days. The simulation of ephemeral vents and the use of different histories within the experiments have been crucial in the determination of a new hazard area for Catania. In fact, during the simulation the city was never affected without the introduction of ephemeral vents, proving the fact that lava tubes played a fundamental role in the 1669 Catania lava crisis.

Hydroclastite beds are interbedded with pillow lavas in an overturned section exposed near Acicastello, on the Ionian coast of Sicily. Two distinct hydroclastite facies are recognized in a continuous 15-m section in a cliff face overlying pillow lavas. The first, an hyaloclastite facies, comprises black hyaloclasts and pillow fragments, many of which retain a partial pillow shape and glassy rim, in a matrix of smaller, partially devitrified and palagonitized hyaloclasts. Beds of this facies are predominantly inversely graded, exhibit both matrix and clast support, and commonly contain outsize clasts at or near the bed top. Beds of the hyaloclastite facies beds are interpreted as the deposits of noncohesive submarine debris flows transporting fragments formed primarily by cooling-contraction granulation. The second, a pillow-fragment breccia facies, forms wedge-shaped beds that are interbedded with the hyaloclastite facies. Pillow-fragment breccias comprise non-graded breccias of basalt blocks that formed by spalling of joint blocks from pillow lavas. Beds of this facies are clast-supported and contain a matrix of marly sediment. This facies formed by accumulation of pillow fragments along the sides of pillow-lava ridges, creating wedges of talus which interfinger with the debris-flow deposits of the hyaloclastite facies.

Lava tubes play a pivotal role in the formation of many lava flow fields. A detailed examination of several compound ‘a‘a lava flow fields on Etna confirmed that a complex network of tubes forms at successively higher levels within the flow field, and that tubes generally advance by processes that include flow inflation and tube coalescence. Flow inflation is commonly followed by the formation of major, first-order ephemeral vents which, in turn, form an arterial tube network. Tube coalescence occurs when lava breaks through the roof or wall of an older lava tube; this can result in the unexpected appearance of vents several kilometers downstream. A close examination of underground features allowed us to distinguish between ephemeral vent formation and tube coalescence, both of which are responsible for abrupt changes in level or flow direction of lava within tubes on Etna. Ephemeral vent formation on the surface is frequently recorded underground by a marked increase in size of the tube immediately upstream of these vents. When the lining of an inflated tube has collapsed, ‘a‘a clinker is commonly seen in the roof and walls of the tube, and this is used to infer that inflation has taken place in the distal part of an ‘a‘a lava flow. Tube coalescence is recognised either from the compound shape of tube sections, or from breached levees, lava falls, inclined grooves or other structures on the walls and roof. Our observations confirm the importance of lava tubes in the evolution of extensive pahoehoe and ‘a‘a flow fields on Etna.

Previously undescribed debris-avalanche deposits occur in two locations downslope from the open end of the Valle del Bove. These outcrops comprise unstratified, ungraded deposits of metre-scale lava blocks in a matrix of weathered and fractured lava clasts. The avalanche deposits are unconformably overlain by matrix- to clast-supported conglomerates, representing debris-flow and interbedded fluvial deposits, that constitute most of the Milo Lahar sequence. We present evidence that the Milo Lahar sequence, which crops out just at the exit of the Valle del Bove, formed during the opening and enlargement of this depression. The presence of the avalanche deposits at the base of the Milo Lahar sequence indicates that catastrophic landslides were involved in the formation of the Valle del Bove. The composition of lavas in the debris avalanche deposits is similar to that of most of the Ellittico volcanic sequence exposed along the northern wall of the Valle del Bove. Radiocarbon dates of 8400 and 5300 years BP from the base and top, respectively, of the debris-flow sequence indicate that the Milo Lahars are correlative with the exposed part of the Chiancone deposit. The basal lahars of the Chiancone, which contain lava blocks whose compositions partially overlap that of blocks in the avalanche deposits, may have formed by water concentration in the distal end of the avalanche causing transformation to debris, or alternatively by reworking of the avalanche deposit.

The Chiancone (CH) deposit is a volcaniclastic fan on the eastern flank of Mt. Etna. It crops out from the exit of the depression of the Valle del Bove (350 m a.s.l.) to the sea in an area which covers about 40 km 2 , with a maximum thickness of about 30 m and a slope of 3–4 ° towards the east. The total volume of the deposit is unknown because the base does not crop out, but some geophysical data suggest a maximum thickness of about 300 m, leading to an estimated maximum volume of about 12 km 3 which is comparable to the volume of the Valle del Bove (VDB). Since this horseshoe-shaped valley is considered to be the CH's source region, a study of this deposit may allow an understanding of the mechanisms responsible for the origin of the VDB. On the basis of stratigraphic and sedimentary studies we divided the CH sequence into two main parts. The basal portion (Lithofacies 1) is visible only along the coast and has a width of about 6.5 km, a thickness of 1–4 m, and appears as an indurated horizontal deposit with a flat, locally eroded upper surface. We consider Lithofacies 1 as a huge mud flow which may be associated with an important eruptive event. The upper part (lithofacies 2–5) is mainly composed of fluvial beds (63% of measured thicknesses) that we attribute to events reworking the previously deposited layers. The less abundant (23%) very coarse-grained lenses were caused by fluvial floods. Thin and discontinuous pyroclastic layers are accidentally (3%) interbedded into the sequence. The upper portion of the CH sequence is up to 30 m thick and formed after 7590 ± 130 yr B.P. as established by our new radiocarbon dating. When the chemical data obtained on lava blocks sampled from lithofacies 1–4 are compared with products of older eruptive centres it may be argued that the major direction of transport occurred from west to east and involved erosion of those old eruptive centres which were located in the central-northern part of the VDB. Our data suggest the occurrence of an important eruptive event at least 7590 yr B.P. which may be associated with the deposition of the huge basal mud flow. This event was followed by fluvial reworking and deposition at a rate of at least 4 mm/yr. Extending this deposition rate to the whole thickness of the CH deposit (300 m) would imply a maximum age of 80,000–70,000 yr B.P. However, the deposition rate of the hidden part of the CH deposit was almost certainly much greater than 4 mm/yr implying an age for the onset of the CH sequence significantly less than 80,000–70,000 yr B.P. This very strongly suggests that the Trifoglietto volcano (which is older than 80,000 yr) was not involved in the opening of the VDB and the formation of the CH deposit. It also suggests that the first event of the opening of the VDB was much older than the 5000 yr proposed by some authors. It is possible that the CH was deposited during, or slightly later than, the life-span of the Ellittico volcano (40,000–15,000 yr B.P.).