Sensitivity analysis of source parameters for gravitational mass-flow generated tsunamis at Stromboli
Matteo Trolese1 , Matteo Cerminara1, Tomaso Esposti Ongaro1, Mattia de\' Michieli Vitturi1, Alessandro Tadini1, Juan Francisco Rodriguez1,2
Affiliations: 1Istituto Nazionale di Geofisica e Vulcanologia, Sezione di Pisa, Pisa, Italy 2Universidad de Malaga, Malaga, Spain
Presentation type: Poster
Presentation time: Thursday 16:30 - 18:30, Room Poster Hall
Poster Board Number: 151
Programme No: 2.2.15
Abstract
Tsunamis generated by granular flows represent significant hazards to coastal communities, particularly near volcanic islands with steep slopes and persistent activity. Stromboli volcano exemplifies this, with recurrent gravitational activity causing tsunamis over the past century. Here, we investigate the sensitivity of tsunami generation and propagation to key gravitational mass-flow parameters using the Multilayer-HySEA model. We explore variations in granular flow initial position (subaerial versus submarine), volume, density, friction angles, and water-landslide coupling. Results indicate distinct behaviors based on flow position for volumes from 3 to 30 × 106 m3. Submarine flows show a linear relationship between volume and maximum sea surface elevation, with the slope of the fit predicting 0.3 ± 0.3 m per 106 m3 near Stromboli village. Subaerial flows show a logarithmic trend, with maximum elevations per 106 m3 decreasing as volume increases, with typical values of 0.7 ± 0.7 m per 106 m3 near the village. Sobol's analysis shows that initial flow position drives variability in maximum elevation (∼60%), followed by volume (∼20%) and friction angles (∼10%). Separating subaerial and submarine cases reveals differences and localized trends. Tsunamis generated by subaerial flows are primarily influenced by volume (∼60%), whereas submarine flows show balanced sensitivity between volume (∼37%) and position (∼35%). In both scenarios, friction angles and water-landslide coupling contribute to variability, although to a lesser extent. Our study supports prior findings at Stromboli and provides insights for reconstructing tsunami source parameters from sea-level measurements, enabling rapid impact forecasting for the island.