ST-HASSET for volcanic hazard assessment: A Python tool for evaluating the evolution of unrest indicators

Abstract

Short-term hazard assessment is an important part of the volcanic management cycle, above all at the onset of an episode of volcanic agitation (unrest). For this reason, one of the main tasks of modern volcanology is to use monitoring data to identify and analyse precursory signals and so determine where and when an eruption might occur. This work follows from Sobradelo and Martí [Short-term volcanic hazard assessment through Bayesian inference: retrospective application to the Pinatubo 1991 volcanic crisis. Journal of Volcanology and Geothermal Research 290, 111, 2015] who defined the principle for a new methodology for conducting short-term hazard assessment in unrest volcanoes. Using the same case study, the eruption on Pinatubo (15 June 1991), this work introduces a new free Python tool, ST-HASSET, for implementing Sobradelo and Martí (2015) methodology in the time evolution of unrest indicators in the volcanic short-term hazard assessment. Moreover, this tool is designed for complementing long-term hazard assessment with continuous monitoring data when the volcano goes into unrest. It is based on Bayesian inference and transforms different pre-eruptive monitoring parameters into a common probabilistic scale for comparison among unrest episodes from the same volcano or from similar ones. This allows identifying common pre-eruptive behaviours and patterns. ST-HASSET is especially designed to assist experts and decision makers as a crisis unfolds, and allows detecting sudden changes in the activity of a volcano. Therefore, it makes an important contribution to the analysis and interpretation of relevant data for understanding the evolution of volcanic unrest.