Forecasts and warnings of volcanic eruptions.

Caption: Long-term patterns of unrest before eruption at large calderas.  View full poster.

Most of the world’s active volcanoes are not regularly monitored, especially those that have been in repose for centuries or more. When such volcanoes reawaken, short-term forecasts of eruption are normally based on data from rudimentary monitoring networks installed after the start of unrest. As a result, forecasts are uncertain; uncertainty fosters scepticism about warnings; and scepticism compromises the success of emergency procedures.

We have developed a new physical model to improve the reliability of forecasts. The core idea is elementary: long-quiescent volcanoes must break themselves open to allow magma to erupt. Complexity appears because we cannot see what is happening to molten rock, or magma, underground. We have to rely on signals that monitor how magma disturbs the surrounding rock, such as movement of the ground and local, small earthquakes. Our model explains the behaviour of such signals seen during emergencies that have continued from months to decades. It also helps us to understand the physical reasons why our forecasts contain an element of doubt. This, in turn, helps us to explain to vulnerable communities why we cannot be one hundred percent certain about a forecast – and the explanations encourage the communities to be less sceptical about the uncertainty in warnings.

Some of the challenges for scientific study and the communication of warnings are shown in the posters and accompanying three-minute videos about the continuing unrest at the Campi Flegrei volcano, immediately west of Naples in southern Italy.

More information on our studies can be found published in open access papers:

• Forecasting Volcanic Eruptions: Beyond the Failure Forecast Method. Kilburn, C. Frontiers in Earth Science 2018
• Progressive approach to eruption at Campi Flegrei caldera in southern Italy. Kilburn, C., De Natale, G. & Carlino. Nature Commun 8, 15312 (2017).