Information Modeling to Assess Eruptive Behavior and Possible Threats on Mt. Etna, Italy

One of the best-studied volcanoes of the world, Mt. Etna in Sicily repeatedly exhibits eruptive scenarios that depart from the behavior considered typical for this volcano. Episodes of intense explosive activity, pyroclastic density currents, dome growth, cone collapse, and phreatomagmatic explosions pose a variety of previously underestimated threats to human lives in the summit area of the volcano. However, retrospective analysis of these events shows that they were likely caused by the same very sets of premises and starting conditions as "normal" effusive eruptions, yet combined in an unexpected, probably unique, way. Physical modeling tells us what may happen in terms of physical parameters but not what events we will actually see on a volcano. Bayesian modeling of volcanoes can unite physical parameters and observed events but, unlike physics, it lacks strictness of terms used to describe the events and, hence, may fail to provide a reasonably impartial, complete and self-consistent set of possible scenarios to be expected. Therefore, a tool is needed to process the observational knowledge as strictly as physical matters are treated by mathematics to provide an appropriate event-based framework for assessment of natural hazards during volcanic eruptions. This task requires a modeling not of the volcano itself but of our knowledge of it, and therefore falls into the field of informatis, knowledge engineering, and artificial intelligence. We involved an approach of artificial intelligence developed specially for the needs of geoscience, the method of event bush. Scenarios inferred from event bush fit the observed ones and allow one to foresee other low-probable events that may occur at the volcano. Application of the event bush provides a more impartial vision of volcanic phenomena and may serve as an intermediary between physical modeling, the expert knowledge and numerical assessment, e.g., by means of Bayesian belief networks.