At 5.42 on 8 October 2026 the network of Italy’s National Institute of Geophysics and Volcanology, INGV, recorded a local magnitude 4.7 earthquake in the southern Tyrrhenian Sea, north of the Aeolian Islands. The epicentre was at sea. The hypocentre lay 372 kilometres down. On the map the point is some forty kilometres from the summit of Marsili, the largest submarine volcano in Europe.
The volcano is not at that depth. The Ionian plate is, sinking beneath the Tyrrhenian. Seismic tomography shows it as a body about 250 kilometres thick, dipping 70 degrees to the north-west and reaching beyond 400 kilometres. The deep earthquakes of the Tyrrhenian start inside this slab. They mark the plate’s movement, and they are the distant engine of the volcanoes in the area. The sinking plate releases fluids that in turn melt the mantle above it, and magma rises.
A quake that deep reaches the surface damped and spread over a wide area. It is not a sign of awakening. A volcano that stirs shakes the crust beneath it, a few kilometres down, with small, frequent tremors.

Marsili does. In July 2006 an INGV ocean-bottom seismometer with a hydrophone was placed on the volcano’s flat summit, 790 metres down, and stayed on the seafloor for nine days. Antonino D’Alessandro and colleagues counted more than a thousand signals. Only 19 were tectonic earthquakes, and one of those came from Java. The rest belonged to the volcano. Eight hundred and seventeen were small volcano-tectonic shocks in swarms, almost all in the first seven days. In the last two the instrument recorded continuous tremor and bursts of high-frequency tremor, like water boiling in cracks heated by magma. The authors read it as intense hydrothermal and volcanic activity. The instrument was funded under an agreement between Italy’s Civil Protection Department and INGV, within a geothermal exploration programme promoted by a private company, PRAMA.

Marsili is about sixty kilometres long and twenty wide. It rises more than 3,000 metres from the floor of its 3,500-metre-deep basin, and the summit lies about 500 metres below sea level. Almost all of it grew in the last million years. It is not extinct. A sediment core taken in 2010 at 943 metres, on the flank of the summit, holds the ash of at least four explosive eruptions between 7,200 and 2,100 years ago. Giuseppe Ventura’s group at INGV and the National Research Council showed this in 2015. Beneath, magma sits at several levels. A 2024 study from the University of Parma places the basalts at the boundary between crust and mantle, at the base of an oceanic crust 10 to 12 kilometres thick, with smaller, shallower reservoirs under the crest. Hot water vents at the summit, and the fluids alter the rock.
That is where the risk lies. Marsili’s crest is narrow, the upper flanks are steeper than 30 degrees, and the summit rocks are weakened by hydrothermal activity. In 2021 Glauco Gallotti, Stefano Tinti and colleagues at the University of Bologna and the Institute of Marine Sciences of the National Research Council simulated five landslides on the flanks. The possible causes, they write, are eruptions, perhaps together with seismic activity.

Depth decides everything. A 2.8 cubic kilometre slide starting between 2,400 and 3,200 metres brings a few centimetres of water to the coast. One of similar size, 2.7 cubic kilometres, starting from 800 metres, gives waves of 3 to 4 metres in southern Campania and 2 to 3 metres in Calabria and Sicily. The worst case is a 17.6 cubic kilometre collapse between the summit and the eastern flank. It would bring at least 5 metres to Campania, Basilicata and Calabria, more than 20 in Basilicata, 5 to 10 in Sicily and 2 to 4 in Sardinia. The Aeolian Islands would be reached in about ten minutes, Calabria and Sicily in twenty, Sardinia in three quarters of an hour.

The authors add three warnings. The worst case is not a hazard estimate but an example of what a large collapse could do. There is no recurrence rate, because the data are not enough for a probabilistic calculation. Past landslides are inferred from the shape of the relief, and only the smallest has left recognised deposits at the foot of the volcano.
There is one last observation, and it concerns the land, not the sea. Italy has no warning system for tsunamis caused by landslides. Risk perception on the Tyrrhenian coasts is low, the authors write, and in summer those coasts are heavily populated.
Marsili’s landslides can be read in the shape of the seafloor. None of them has a date yet. The summit lies five hundred metres below the sea, with steep flanks and soft rock, and it gives off hot water.

Highest wave on the worst-hit coasts in five simulated landslides on the flanks of Marsili: for each case, the landslide volume and the detachment depth. Slides (a), (b) and (c) reconstruct past events from the shape of the seafloor, (d) and (e) are hypothetical. Case (e) is the worst case and the authors warn that it should not be read as a hazard estimate. Chart produced by ScienceOnline from data in Gallotti G. et al., Bulletin of Volcanology 2021, doi 10.1007/s00445-020-01425-0 (article licensed CC BY 4.0). Original newsroom chart.

References
Gallotti G., Zaniboni F., Pagnoni G., Romagnoli C., Gamberi F., Marani M., Tinti S., «Tsunamis from prospected mass failure on the Marsili submarine volcano flanks and hints for tsunami hazard evaluation», Bulletin of Volcanology, 83, 2, 2021, published online 8 December 2020, open access under CC BY 4.0. doi: 10.1007/s00445-020-01425-0
D’Alessandro A., D’Anna G., Luzio D., Mangano G., «The INGV’s new OBS/H: Analysis of the signals recorded at the Marsili submarine volcano», Journal of Volcanology and Geothermal Research, 183, 17–29, 2009. doi: 10.1016/j.jvolgeores.2009.02.008
Tamburrino S., Vallefuoco M., Ventura G., Insinga D.D., Sprovieri M., Tiepolo M., Passaro S., «The proximal marine record of the Marsili Seamount in the last 7 ka (Southern Tyrrhenian Sea, Italy): Implications for the active processes in the Tyrrhenian Sea back-arc», Global and Planetary Change, 2015. doi: 10.1016/j.gloplacha.2015.07.005
Colle F., Masotta M., Costa S., Giacomoni P.P., Trua T., Marani M., «A micro-scale insight into a back-arc trans-crustal plumbing system: The case of Marsili volcano, Southern Tyrrhenian Sea», Lithos, 482–483, 107675, 2024, open access under CC BY-NC-ND. doi: 10.1016/j.lithos.2024.107675
Trua T., Serri G., Marani M., Renzulli A., Gamberi F., «Volcanological and petrological evolution of Marsili Seamount (southern Tyrrhenian Sea)», Journal of Volcanology and Geothermal Research, 114, 441–464, 2002.
Monecke T., Petersen S., Augustin N., Hannington M., «Seafloor hydrothermal systems and associated mineral deposits of the Tyrrhenian Sea», Memorie Descrittive della Carta Geologica d’Italia, 104, 41–74, 2019.
Istituto Nazionale di Geofisica e Vulcanologia, Terremoto Tirreno Meridionale [Mare], ML 4.7, 8 October 2026, 05:42:50 Italian time, evento 47468602, terremoti.ingv.it.

*Board Member, SRSN (Roman Society of Natural Science)
Past Editor-in-Chief Italian Journal of Dermosurgery