Procedura near real-time per la valutazione dell’hazard da eruzioni laterali all’Etna (Sicilia, Italia)

Abstract

Mt. Etna is a composite stratovolcano located along the Ionian coast of the Sicily. The frequent flank eruptions (mostly along the NE, S and W rift zones) lead to a high volcanic hazard that is getting worse by the intense urbanization on the flanks of the volcano. In the framework of the project PON SIGMA (Integrated Cloud-Sensor System for Advanced Multirisk Management), we developed a completely automatized computer-assisted analysis that evaluate in near real-time the areas prone to the highest vent opening probability and the corresponding hazard related to effusive eruptions. The Bayesian Event Tree_Eruption Forecasting (BET_EF) algorithm is firstly used to evaluate a long-term hazard map based on the last 4.0 ka flank effusive activity. The analysis and inversion of real time monitoring data, such as seismic location and tremor sources, allowed us to evaluate the short-term Probability Density Function (PDF). A further application of the BET_EF algorithm provided a short-term scenario for lava flow simulations. The short-term scenario and the application of a specific algorithm allowed us to estimate the hazard related to the expected effusive eruption. The final intrusion of the May 2008 – July 2009 flank eruption at Mt. Etna was selected as test case. The future vent forecast highlighted the area with higher probability. It is worth noting that a good accordance is evident if the highest probability area was compared with the real vent location. A series of 200 lava flow simulations was run to assess the areas prone to the highest lava invasion hazard. Finally, the density of the simulated flows was evaluated showing that the highest values are in accordance with the area actually covered by the 2008-2009 lava field.