Forecasting Eruptions at Volcanoes after Extended Repose

FEVER: Forecasting Eruptions at Volcanoes after Extended Repose.

Summary.

The overarching goals of Project FEVER are to understand how volcanoes reawaken after generations at rest and to devise more reliable methods for forecasting eruptions. Most eruptions from long-quiescent volcanoes occur in Low-to-Middle Income Countries, where they jeopardize some of the world's most vulnerable populations. The threat is under-estimated because long intervals without eruption often result in a volcano being unmonitored and for the memories of previous activity to have faded. As a result, when unrest returns, forecasts of eruption must rely on measurements obtained from hastily-installed monitoring networks. The forecasts are uncertain - a feature that hinders mitigation measures and diminishes the trust of vulnerable communities. A compelling social need therefore exists for reliable forecasts of eruptions at long-quiescent volcanoes, using emergency data obtained after the start of unrest.

Small local earthquakes and ground movement are the most reliable methods for monitoring a reawakening volcano. They measure how a volcano stretches and fractures when it is put under pressure by molten rock, or magma, attempting to reach the surface. It has long been recognised that changes in these signals contain important clues about the approach of an eruption. However, judging when the changes have become critical is heavily influenced by personal experience, so that forecasting is still as much an art as it is a science.

We aim to make eruption forecasts more reliable by taking advantage of the fact that volcanoes seal up their magmatic systems during long intervals of repose. In such cases, a new eruption must be preceded by the volcanic edifice again breaking itself open to allow magma to escape. Rupture occurs under a restricted range of physical conditions, which promotes repeatable patterns of deformation and fracture that can be detected at the surface. We argue that these patterns can be used to determine the stability of a volcano and, because they depend on physical conditions that we can quantify, will allow forecasts to be less subjective than at present and also to be made far enough in advance to be of practical value.

We have supporting evidence that our goals are feasible from a new model that we have developed to describe how rock within and below volcanoes can trigger earthquakes while being stretched. The next steps are to test our methods under controlled conditions during rock-physics experiments in the laboratory. The results will allow us to connect field data to stress in the crust and, from this, to calculate how much more unrest is needed before the crust will break. Once the model has been fully tested, we will transform it into a robust and objective method for forecasting rupture before eruptions.

Our investigations will be performed by an international team built around a long-standing UK collaboration between volcanologists at University College London and experimental rock physicists at the University of Portsmouth. Our partners are the Seismic Research Centre of the University of the West Indies, which is responsible for monitoring volcanoes across the English-speaking Caribbean, and the Vesuvius Observatory, which is the world's oldest volcano observatory and monitors the volcanoes around Naples in southern Italy. Together, we will incorporate our results into existing emergency procedures to force a step change in the reliability of real-time forecasts of the state of volcanoes before eruption.