Tsunamigenic mass flows at Stromboli Volcano- analysis and modelling after the 3rd of July events

Tsunamis generated by large mass flows (described in this case then as tsunamigenic), like large landslides and pyroclastic density currents (PDCs), constitute a serious hazard in volcanic areas. This has been clearly demonstrated by the tsunami triggered by the Anak Krakatau eruption-induced flank collapse in December 2018 in Indonesia, which resulted in hundreds of fatalities. Nevertheless, these phenomena are still poorly understood as they are unusual and complex events. In addition, tsunami monitoring systems are typically placed in areas of high seismic risk rather than along unstable coasts, consequently instabilities that could produce tsunamis are largely unpredictable, difficult to assess and often poorly constrained.
Stromboli is one of the most active volcanoes in the world, and has been extensively monitored and studied in the last few decades. Many tsunamigenic landslides (sub-aerial and/or submarine) have taken place; at least seven occurred in the last 150 years and during the late middle ages, a devastating one has reached the coast of Naples at more than 200 km distance. Because the activity of the volcano has remained similar ever since, larger eruptions, instabilities and tsunamis need to be considered and efforts to mitigate such catastrophes are necessary owing to their potential destructive impact. In particular, the 30 December 2002 a landslide-induced tsunami wave of up to 10 metres caused extensive damage to the coast and has increased awareness around this otherwise largely disregarded hazard. Associated with the 3rd July 2019 eruption, which caused extensive damage and one fatality, at least three mass flows were triggered along the Sciara del Fuoco slope (a highly unstable flank situated on the western side of the volcano), two subaerial PDCs and a submarine landslide. Simultaneously, three buoys in different locations registered the height of a tsunami wave, i.e. 1.5 m, 0.5 m close to the Sciara del Fuoco and 0.2 m near the village of Ginostra. The present project aims to collect data about these mass flows to accurately assess the process that triggered the tsunami. In particular, we will collect samples to characterise the mobilized material in terms of the geotechnical properties and we will build a detailed DTM of both the subaerial and submarine part of the slope. This will allow us to correlate the magnitude of the registered wave with the type of instability, the mobilized volume and the material of the mass flows. Even if the tsunami waves studied here are smaller than the one that could constitute a threat for the population living in this area, they represent an event that, if the present project is funded, could constitute a natural laboratory example that will help us to characterise and better understand the process involved in these phenomena.
In addition, we will use this well studied and characterized event to run accurate back analysis to characterize the material and the behaviour of the tsunamigenic mass flow, i.e. knowing the final and actual characteristics of the deposits and of the tsunami wave, we are able to identify under which type of conditions they have been generated. Starting from these results, we will then be able to run a larger number of numerical simulations of possible instability scenarios. Exploring the different potential types of mass flows, will allow us to identify empirical relationships between their characteristics and the magnitude of the tsunami generated. Assessing these empirical laws constitutes crucial input data to improve the early warning system and to reduce the risk related to these unpredictable but extremely dangerous phenomena.

The Aeolian Islands host a population of more than 15000 permanent residents; this reaches peaks of hundreds of thousands during the high touristic season. Tourism is also the main source of income for the population living in Stromboli where visitors are mainly attracted by the opportunity to see an active volcano from close by. While in general the population is used and prepared to low or moderate explosive activity, little is known about the landslide induced tsunamis and the related risks. In 2002 after a tsunami of this type caused extensive damage, efforts have been made by the scientific community to increase our understanding of these phenomena. Since then early warning systems have been put in place. Larger and more devastating events though have been recorded in the past during the late middle ages and their extent reached the coast of Naples (around 250 km away). Because the activity of the volcano has remained similar ever since, larger eruptions and potential tsunami need to be considered. Therefore, efforts to mitigate such catastrophes are necessary owing to the potential impact on the population in Stromboli, on the surrounding coastal area and to the economy of the Aeolian Island region. The early warning system currently in place at Stromboli is mainly based on the recording of tsunami wave height registered by buoys. Assessing the volumes and type of mass flows that could induce significant tsunami waves will enable us to improve this system. In fact, with a continuous monitoring of the slope, the empirical model derived from this project could contribute in assessing when the probability of the failure of significant volumes (subaerial and submarine) becomes high enough to trigger an early warning. The Italian partners of this project are amongst the main responsible of the monitoring of the activity at Stromboli and they are affiliated to the Department of the Italian Civil Protection. For these reasons, the findings of this project will be directly used for the risk management of the island and will have a direct impact on the monitoring and early warning system. After the necessary modifications and adaptations, the empirical model could be used as starting point in other similar contexts. In addition, as affiliated to the UNESCO Chair on "Prevention and Sustainable Management of Geo-Hydrological Hazard", the University of Florence project partner will disseminate the progress and the results of the present research (see letter of support). This will have a global impact, as it will ultimately help increasing awareness and understanding of these lethal but still poorly constrained phenomena.