Skip to content

Volcanic Hazard Assessment of Paroxysms at Stromboli (Italy) with Uncertainty Quantification: Deposit-derived Pyroclastic Density Currents

Andrea Bevilacqua1 , Augusto Neri1, Alessio Di Roberto1, Federico Di Traglia2, Massimo Pompilio1, Antonella Bertagnini1, Lucas Corna1, Mattia de'Michieli Vitturi1, Zeno Geddo1, Alessandro Tadini1

  • Affiliations:  1Istituto Nazionale di Geofisica e Vulcanologia, Sezione di Pisa, Pisa, Italia; 2Istituto Nazionale di Geofisica e Vulcanologia, Osservatorio Vesuviano, Napoli, Italia 

  • Presentation type: Poster

  • Presentation time: Friday 16:30 - 18:00, Room Poster Hall

  • Poster Board Number: 104

  • Programme No: 6.7.15

  • Theme 6 > Session 7

Abstract

Stromboli volcano (Italy) features persistent explosive activity and occasional paroxysms from the craters located at ≈750 m atop Sciara del Fuoco, a horseshoe-shaped depression on the NW flank, and at ≈2 km from the two villages along the coast. Deposit-derived pyroclastic density currents (PDCs) arise from the gravitational instability of eruptive deposits at elevated temperatures. At Stromboli, these flows can occur outside Sciara del Fuoco during the paroxysms, and pose a significant additional risk to both residents and those ascending volcanic slopes, due to their high mobility. Notable historical examples include PDCs generated during the 1930 and 1944 paroxysms, which affected both inhabited and uninhabited areas. Similar phenomena have been documented on other mafic to intermediate volcanoes around the world, often with devastating consequences. By using Markov chain models, we reproduced clustering in paroxysmal events, and quantified average probabilities at 8.6% and 49% of one or more deposit-derived PDC occurring outside Sciara del Fuoco in the next 10 and 50 years, respectively. We estimated a 90% confidence interval of the PDC probability conditional on a paroxysm, identifying the occurrence rate as one of the key uncertainties of this hazard. This study develops probabilistic hazard maps for Stromboli's deposit-derived PDCs, conditional on paroxysms. Using historical data, field observations, and depth-averaged flow modeling, Monte Carlo simulations were employed to account for uncertainties in source area, thickness, and dissipation parameters. Application to the 1930 event shows that the model can reproduce the main flow observables from field evidence and historical chronicles.