AUTHOR=Rosi Mauro , Acocella Valerio , Cioni Raffaello , Bianco Francesca , Costa Antonio , De Martino Prospero , Giordano Guido , Inguaggiato Salvatore TITLE=Defining the Pre-Eruptive States of Active Volcanoes for Improving Eruption Forecasting JOURNAL=Frontiers in Earth Science VOLUME=10 YEAR=2022 URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2022.795700 DOI=10.3389/feart.2022.795700 ISSN=2296-6463 ABSTRACT=

A crucial feature to manage a volcanic crisis is the ability of volcanologists to promptly detect an impending eruption. This is often affected by significant uncertainty, mainly for the difficulty in interpreting the monitoring signals in terms of the exact timing of a possible eruption. Here we contribute to this problem, focusing on the states of active volcanoes with closed conduit, as deduced from monitoring data. Four main states can be identified. In the quiescence state 1) the monitoring data lie on a baseline, suggesting the lack of shallow magma/fluid movement. The unrest state is highlighted by minor 2) to major 3) variations in the intensity and rate of monitoring data; in both cases, radial ground deformation pattern and non-migrating seismicity imply shallow magma and/or fluid accumulation. The state of impending eruption 4) is characterised by non-radial, asymmetric ground deformation pattern and migrating seismicity, which suggest that magma approaches the surface through a propagating dyke. As early recognition of this distinctive state is crucial for timely eruption forecast, monitoring activity should be aimed at its prompt detection. The application of this rationale to two types of active volcanoes in densely inhabited areas, a restless caldera (Campi Flegrei) and a quiescent stratovolcano (Vesuvio), highlights its feasibility and importance in eruption forecasting. This rationale may foster a general reference framework to be adopted in case of unrest, supporting in interpreting the monitoring data, as well as more effective: 1) operationally-oriented, monitoring system; 2) probabilistic forecast; 3) use of volcanic alert levels.