Data uncertainty management in volcanic hazard assessment: review and examples

Abstract

The availability of new volcanological data employed as input parameters in numerical models have allowed scientist over the past years to produce more accurate hazard maps. However, it is also true that these latter products, which are fundamental for decision-makers when dealing with long-term planning or with the management of emergency situations, might be affected by a certain degree of uncertainty. This latter one depends on many factors, and it is important to quantify the two main sources of uncertainty of a volcanic system: i) the first one related both to the incomplete knowledge of the system under investigation and to errors in data acquisition (epistemic uncertainty); ii) the second one related to the intrinsic physical variability of the system (aleatoric uncertainty). If a volcanic hazard map might be provided to civil protection authorities and decision-makers with a quantification of these sources of uncertainty, the resultant emergency planning might be undertaken with a better awareness. This presentation is therefore aimed at briefly reviewing the state of the art of the distinction between different types of uncertainty applied to volcanic hazard assessment: after a theoretical introduction, two examples from two Italian high-risk volcanoes (Vesuvio and Campi Flegrei) will illustrate how data uncertainty has been managed in specific cases related to volcanic hazard assessment.