Stromboli, Italy

Mount Stromboli is referred to as the“Lighthouse of the Mediterranean”due to the amazing continuity of itseruptions. Stromboli is probably themost active volcano on Earth;exhibiting a nearly continuouseruption for the past 1,700 years(Rosi et al., 2000) and possibly aslong as 5,000 years. The Strombolistratovolcano started developing 100ka on 18 km thick crust (Pasqare etal., 1993). Stromboli is presentlycharacterized by twin peaks and avery symmetrical and conical morphology. The summit is 918 meters above sea level (2500 meters abovethe ocean floor); there are three craters with distinctive explosive behavior and histories. The three craters are structurally aligned NE-SW and span a distance of 300 meters; the crater terrace is at an elevation of 750 m asl., below the Pizzo sopra de la Fossa summit. A horse-shoe shaped depression on the flank, Sciara del Fuoco (Road of Fire), is the result of several collapses of the northwest flank over thepast 13,000 years (Tibaldi, 2000). There are two villages on the island, Ginostra on the west and Stromboli on the northeast coast. Two kms to the northeast, Stombolicchio, a volcanic neck, rises from the sea.. Strombolicchio is a much older cone that has subsequently been eroded down to sea level. The San Vincenzo Observatory monitors the volcano (geophysical and geochemical). The figure below from Ripere and others (2008) is a general structural map of the island.

Stromboli is the type location for the“Strombolian” eruption, a distinctivestyle of eruption that consists ofrelatively small gas releases whicheject molten lava fragments in afountain above the vent. In the caseof Stromboli, these gas explosionsoccur relatively frequently, typically,1 to 10 events per hour. These mild,intermittent explosions are caused bylarge gas slugs that move through themagma column and expand rapidlyat the surface due to decompression.These events involve degassed, high-porphyritic basaltic magma from ashallow, relatively cool, conduitsystem (Bertagnini et al., 2003;Blackburn et al., 1976). Theexplosions are classified as ash-rich(long-lasting at 10-20 s) or fragment-rich (short at 5-10 s), depending onthe size of the material ejected. TheSW crater tends to have ash-rich

events and the NE crater tends to be fragment-rich (Ripepe et al., 1993). This is what we observed on ourtwo visits to the volcano.

The physical model for the Stombolian eruptions is supported by geophysical and geochemical evidence. Infrasound, thermal and seismic data indicate that gas expansion occurs at a depth of 260 meters 2 to 20 seconds prior to an explosion. Explosions occur at a typicalrate of 13 per hour with gas jet velocities of 10 to 130 m/s.Studies also found that there is a persistent bursting of smallgas bubbles; this “puffing” releases more gas than “normal”Strombolian explosions. Explosions and puffing exhibitshort-term variability that may be due to changes in the gassupply to the shallow magma chamber (Ripepe et al., 2008).Puffing is an important characteristic of Stromboli eruptionsand there is a positive correlation between explosions andpuffing; they both increase in magnitude and frequency toaccommodate additional gas. These “minor” explosions tapporpyritic, degassed magma from the upper magma chamber. Once or twice a year there are “major” explosions that ejectmeter-size bombs several hundred meters from the vent andshower the slopes of Stromboli with ash and lapilli. Theseevents are usually singular explosions and give no warning.The typical product is a dark, low-vesicular scoria with up to50 millimeter-size plagioclase, pyroxene and olivinephenocrysts. Allard and others (1994) calculated gasemissions of 6,000 to 12,000 t/d or H2O, CO2, SO2, HCl andHF.

Larger eruptions and lava flows are infrequent and violent eruptions are rare. Larger-intensity events are referred to as “paroxysmal” events (Barberi et al., 1993) and are due to violent emission of deeper, volatile-rich, low-porphyritic basalt (Rosi et al., 2013). Some have combined the “major” explosions with paroxysmal eruptions, referring to them as “small” paroxysmals (Metrich et al., 2005; Landi et al., 2008). In general, “small” paroxysmals will endanger those at the summit and paroxysmals may endanger those in Stormboli Town. The lava associated with these events is nearly aphytic golden pumice and crystal-rich scoria similar to those produced in normal events. Despite their textural differences, the composition of the soria and pumice are HK-basalts and shoshonitic. Incandescent scoria and bombs as well as meter-size ballistic lithic blocks can be lobbed several kilometers from the vent (Bertagnini et al., 2003). Paroxysmals last from several hours to a few days and occurred in 1930 and 1944.

What makes Stromboli so unusual is that it maintains an open conduit and, generally, does not experience pressurization of the magma nor plumbing. This open conduit system exhibits efficient conduit convection; gas-rich magma ascends, degasses, and descends in the conduit at a rate of 300 to 1300 kg/s (Stevenson and Blake, 1998). Geophysical and thermal data indicate that all three craters are fed by a common, shallow magma source approximately 250 meters deep (Ripepe et al., 2005). Gas expansion occurs at this depth a few seconds prior to each explosion.

The absence of pressurization means that there is very little ground deformation at the scale generally seen for active volcanoes. The Strombolian system may experience local inflation or deflation due to magma convection or uprising (Chaussard et al., 2013; Di Traglia et al., 2014). Ground deformation has been observed at Stromboli associated with dike intrusion and prior to the opening of new eruptive fractures (Casagli et al., 2009; Aloisi et al., 2008).

Stromboli has a history of collapse; two phases have been identified: an older phase associated with pyroclastic eruptions that resulted in the flank collapse toward the southeast and involved three concentriccalderas; and a younger phase that included two flank failures, one toward the northwest associated with lava effusions and the collapse that created Sciara del Fuoco (Pasquare et al., 1993). Large flank collapse is probably the greatest hazard associated with this volcano (with the possible exception of pyroclastics hitting people visiting the crater).

In the short-term (last century) the primary hazard has been pyroclastic flows and volcanic bombs. A large eruption occurred in 1919 that resulted in four fatalities and the destruction of a dozen homes; the damage was caused by large volcanic bombs. The largest eruption in the past century occurred in 1930. This eruption produced voluminous ash and rare pyroclastic flows that killed three people. In March of 1993, three craters on Stromboli were actively erupting. The most recent large eruption occurred in 2002-03 and it closed the island to nonresidents. This eruption produced a lava flow down Sciara del Fuoco to the sea. This eruption also triggered at least two landslides in the Sciara that created tsunamis that damaged the village of Stromboli (largest wave run-up was 10 meters). Volcanic bombs, once again, damaged several homes on the island. Another large eruption occurred in March, 2007. It is interesting to note that the town of Stromboli has installed tsunami evacuation signs and established meeting points; none of the other Aeolian Islands have followed suit.

We were led up the volcano by Lorenzo Russo (Magmatrek Guides). With headlamps, hiking shoes, water and helmet, we set off from Chiesea San Vincenzo in the town of Stromboli at 5:15 pm. We had a full moon (honeymoon) and it was Friday the 13th. It was nearly a three hour climb to the summit at 918 meters, the first 500 meters was through shrubs, once above that elevation the volcano is barren. In the early 1900's, the island of Stromboli supported a greater population (20,000) and was cultivated with extensive vineyards. The agriculture is gone, as are the people (emigrated to Australia) but the abandonedterraces remain on the lower flanks. Despite being evening, it was hot and humid (sunset was not until 8:30). The route up and down are different; on the way up the path follows a basalt flow that provides good traction and the path down traverses lapilli and ash deposits (with bombs). This “soft” trail makes the decent with headlamps much safer. You must be accompanied by a guide over 400 meters and individuals are only allowed to stay one hour at the summit. We reached the summit at 8:00 passing constructed “bomb” shelters and monitoring stations. Explosions could be heard during the entire hike upthe volcano.

There were two active craters, one to NE and another to the SW (the central vent was not erupting). The vents were approximately 200 meters away from the observation point and 200 meters lower. The vent tothe NW exploded three times while we were there (about every 20 minutes) and was much more energetic; incandescent bombs that flew well over 200 meters dropped around the vent and on the flank ofthe observation ridge. Since the wind was blowing to the east, fine black ash, generated by the explosion,

Photos of explosions from the SW crater (ash cloud) and NW crater (lava and ash).

rained on us. The SW vent was much more active, erupting approximately every 5 minutes. The explosions from the SW crater ranged from infrequent loud gas releases that produced a dark (ash) cloud and SO2 emissions to incandescent lava fountains. The lava was much more defined after sunset.

By all indications, Stromboli was unusually active. Lorenzo, who has been a guide for 13 years on Stromboli, was concerned about the “unusual” behavior and was very nervous about the bombs landing just below the observation ridge and indicated he had never experienced similar ash fall. He was most concerned with the activity of the NW vent. After spending 60 minutes on the ridge, we packed up and descended on a trail that dropped off the ridge to the east. We watched the full moon rise as we hiked down Stromboli and were back in town by 11:30.

Stromboli is probably one of the most monitored volcanoes on Earth. A seismic network, established in 2003, consists of 21 broadband seismometers distributed across the volcano. Although major surface inflation/deflation is rare, weak deformations have been recorded by Ground-Based Interferometric Shynthetic Aperature Radar (GBIoSAR). Di Traglia and others (2014) found a correlation between ground deformation tremor amplitude and cumulative aplitudes of very long period signals associated with explosions. In addition to seismic, geodetic, and geochemical monitors, researches at the Istituto Nazionale de Geofisica e Vulcanologia, sezione di Catania (INGV-CT), formerly Istituto Internazionale diVulcanologia (IIV) of Catania, employ visible surveillance as well. Four permanent video cameras were installed in April, 2003, after a paroxysmal event, and are monitored in Catania. The cameras are situatedon the Pizzo sopra de la Fossa (infrared and visible) and the northern edge of Sciara de Fuoco (thermal and visible). The cameras are used to establish the size of explosions and identify the associated tephra for chemical analysis. The average Frequency of Explosions (avFE) is published on the INGV website (www.ct.ingv.it).

The composition of the lava tends to bepotassium-rich. As indicated by the SiO2 vsK2O classification diagram, the magma tendsto be shoshonitic. In this diagram, the plotteddata are from glass erupted during the 2005paroxysmal event, the dashed field is from2002-07 strombolian activity, and the solidfield is from paroxysmal eruptions from2002-07 (Andronico et al., 2008).

The unique “open conduit” system ofStromboli has rendered it relativelypredictable and less hazardous than othervolcanoes. In addition, the continuity ofactivity has provided an ideal setting forestablishing an extensive monitoring programand carrying out volcanic research. More isknown about the volcano than any other inthe world but, since Stromboli is unusual, it isn't clear how much of the research done here can be appliedto other volcanoes. In addition, since violent eruptions (and flank failures) are rare, it is easy to become complacent and ignore the potential hazards associated with this volcano. That said, Stromboli is a fantastic place to observe “red rock”, an exciting experience for geologists and non-geologists. In the 40 years that I have been a “geologist” I have only observed “red rock” one other time and that was a lava flow on Pacaya Volcano in Guatamala.

Eruption frequency for Stromboli from INGV. Note the elevated frequency for the week of June 12.

References

Allard, P., J. Carbonelle, N. Me´trich, H. Loyer, and P. Zettwoog, 1994, Sulphur output and magma degassing budget of Stromboli volcano: Nature, v. 368, p. 326–330.

Aloisi, M., A. Bonforte, M. Mattia, and G. Puglisi, 2008, Ground deformations related to the effusive eruptions of Stromboli: The 2002–2003 case, in The Stromboli Volcano: An Integrated Study of the 2002–2003 Eruption: Geophys. Monogr. Ser., v. 182, edited by S. Calvari et al., p. 247–257, AGU, Washington, D. C.

Andronico, D., Corsaro, R. A., Cristaldi, A., and Polacci, M., 2008, Characterizing high energy explosive eruptions at Stromboli volcano using multidisciplinary data: An example from the 9 January 2005 explosion: Jour of Volcanol and Geophys Res., v. 176, p. 541-550.

Barberi, F., M. Rosi, and A. Sodi, 1993, Volcanic hazard assessment at Stromboli based on review of historical data: Acta Vulcanol., v. 3, p. 173–187.

Bertagnini, A., N. Métrich, P. Landi, and M. Rosi, 2003, Stromboli Volcano (Aeolian Archipelago, Italy): An open window on the deep-feeding system of a steady state basaltic volcano: J. Geophys. Res., v. 108, p. 2336.

Blackburn, E. A., L. Wilson, and R. S. J. Sparks (1976), Mechanisms and dynamics of Strombolian activity, J. Geol. Soc. London, v. 132, p. 429–440.

Casagli, N., A. Tibaldi, A. Merri, C. Del Ventisette, T. Apuani, L. Guerri, J. Fortuny-Guasch, and D. Tarchi, 2009, Deformation of Stromboli Volcano (Italy) during the 2007 eruption revealed by radar interferometry, numerical modelling and structural geological field data: J. Volcanol. Geotherm. Res., v. 182(3), p. 182–200.

Chaussard, E., F. Amelung, and Y. Aoki, 2013, Characterization of open and closed volcanic systems in Indonesia and Mexico using InSAR time series: J. Geophys. Res. Solid Earth, v. 118, p. 3957–3969.

Delle Donne, D., E. Marchetti, M. Ripepe, G. Ulivieri, and G. Lacanna, 2006, Monitoring explosive volcanic activity using thermal images, Stromboli volcano, Italy: Eos Trans. AGU, v. 79, p. 1795.

Di Traglia, F., Cauchie, L., Casagli, N., and Saccorotti, G., 2014, Decrypting geophysical signals at Stomboli Volcano (Italy): Integration of seismic and ground-based InSAR displacement data: Geophys. Res. Lett., v. 41, p. 2753-2761.

Hort, M., R. Seyfried, and M. Voge, 2003, Radar Doppler velocimetry of volcanic eruptions: Theoretical considerations and quantitative documentation of changes in eruptive behaviour at Stromboli volcano, Italy: Geophys. J. Int., v. 154, p. 515–532.

Landi, P., Métrich, N., Bertagnini, A., Rosi, M., 2008, Recycling and “re-hydratation” of degassed magma inducing transient dissolution/crystallization events at Stromboli (Italy): J. Volcanol. Geotherm. Res., v. 174, p. 325–336.

Métrich, N., Bertagnini, A., Landi, P., Rosi, M., 2005, Triggering mechanism at the origin of paroxysms at Stromboli(Aeolian Arcipelago, Italy): The 5 April 2003 eruption: Geophys. Res. Lett., v. 32, p. 10305.

Pasquare, G., Francalanci, L., Garduno, V. H., and Tibaldi, A., 1993, Structure and geologic evolution of the Stromboli volcano, Aeolian Islands, Italy: Acta Vulcanologica, v. 3, p. 79-90.

Patrick, M. R., 2005, Strombolian eruption dynamics from thermal (FLIR) video imagery: Ph.D. dissertation, 281 pp., Department of Geology and Geophysics, Univ. of Hawaii.

Patrick, M. R., A. J. L. Harris, M. Ripepe, J. Dehn, D. A. Rothery, and S. Calvari, 2007, Strombolian explosive styles and source conditions: Insights from thermal (FLIR) video: Bull. Volcanol., v. 69, p. 769–784.

Ripepe, M., M. Rossi, and G. Saccorotti, 1993, Image processing of explosive activity at Stromboli: J. Volcanol. Geotherm. Res., v. 54, p. 335–351.

Ripepe, M., A. J. L. Harris, and E. Marchetti, 2005, Coupled thermal oscillations in explosive activity at different craters of Stromboli volcano: Geophys. Res. Lett., v. 32.

Ripepe, M., Delle Donne, D., Harris, A., Marchetti, E., and Ulivieri, G., 2008, Dynamics of Strombolian Activity: inThe Stromboli Volcano: An Integrated Study of the 2002–2003 Eruption: Geophys. Monogr. Ser., v. 182, edited by S. Calvari et al., p. 247–257, AGU, Washington, D. C.

Rosi, M., A. Bertagnini, and P. Landi, 2000, Onset of the persistent activity at Stromboli Volcano (Italy): Bull. Vulcanol., v. 62, p. 294–300.

Rosi, M., M. Pistolesi, A. Bertagnini, P. Landi,M. Pompilio, and A. Di Roberto, 2013, Stromboli Volcano, Aeolian Islands (Italy): Present eruptive activity and hazards, Geol. Soc. London Mem., v. 37(1), p. 473–490.

Tibaldi, A., 2000, Multiple sector collapses at Stromboli volcano, Italy: How they work: Bull. Volcanol., v. 63, p. 112–125.