From eruptive histories to volcano monitoring: probabilistic eruption forecasting and volcanic hazard assessment at varying temporal and spatial scales

In this chapter, we introduce the reader to the concepts of quantitative eruption forecasting and volcanic hazard assessment. These are two key tasks that modern volcanology increasingly tackles by means of probabilistic approaches, in order to account for the different types and sources of uncertainty that affect both volcanic processes and our knowledge of them. We detail how these two tasks are accomplished at different temporal scales from the short- to the long-term, based on different pieces of information, and using a variety of statistical models and approaches. Such models and approaches are also relevant for volcanic hazard assessments at varying spatial scales, spanning from a local, single-volcano scale to more comprehensive assessments of hazard from multiple volcanoes. We describe how different factors are accounted for in the available literature, in terms of monitoring data, eruption size or style, vent location, and volcanic phenomena (such as tephra fallout, ballistic ejection, gas dispersal, pyroclastic density currents, lava flows and lahars). Future developments in the area may revolve around: (i) tackling data scarcity through creating and maintaining harmonized (global) volcanological databases, combined with the use of analogue volcanoes; (ii) further developing frameworks to test, validate and/or compare physical and statistical models in order to optimize their application to probabilistic volcanic hazard assessment and eruption forecasting; (iii) harmonization of data formats and data structures to enable scaling of hazard and risk assessments across hazardous phenomena and sectors of society.