What can satellites tell us about the link between volcanic inflation and eruption?

The bulge that formed on flank of
Mount St Helens prior to eruption
in May 1980. (Image: United States
Geological Survey).

Ground deformation at volcanoes

In order to assess and monitor the eruption potential of volcanoes worldwide, scientists use an array of observations including seismicity, gas emissions and deformation (motion or changes in the shape) of the ground. In the simplest case, a volcano will inflate before an eruption as the underlying magmatic system pressurises. This is perhaps most memorable in the bulge that formed on the flank of Mount St Helens prior to its eruption in May 1980. Observations of ground deformation not only tell us about escalating eruptive activity, but also shed light on the whole eruptive cycle, from the drainage of magma following an eruption, to the passage and storage of magma in the crust. However, many of the techniques used to monitor ground deformation are limited by their resolution in time (e.g. repeat surveys performed once each summer season) or their spatial resolution (e.g. in-situ equipment recording motion at a single or small network of points).

The role of satellites

Since the early 1990s, satellite data has revolutionised the way in which ground deformation is used as a tool for monitoring and understanding volcanoes. Rather than recording deformation at single points or at widely spaced time intervals, satellite imagery enables us to record ground deformation at millions of data-points, over 100s of km2, with repeat times up to every 12 days. This technology, known as InSAR (Interferometric Synthetic Aperture Radar), works by comparing consecutive satellite images to calculate how much the ground has moved using changes in the phase of the returned radar wave. This technique is particularly useful in hazardous or remote areas, which are inaccessible for ground-based surveys. It is also invaluable in developing countries, which host many of the world’s volcanoes as, in the absence of other equipment, satellite imagery may provide the only indicators of escalating unrest and ultimately, impending eruption.

The European Space Agency satellite
Sentinel-1 to be launched Thursday
3rd April. (Image: European Space
Agency).

We are currently just days away from the long-awaited launch of the European Space Agency Sentinel-1 satellite, and what has been described as a “new era in earth observation”. This satellite is part of the Copernicus programme: the most ambitious Earth observation programme to date. Sentinel-1 will collect data more rapidly and with better global coverage than its predecessor ENVISAT, imaging the entire earth every 6 days for a minimum of 7 years. It is therefore the ideal time to synthesise and reflect upon what we have learnt from the wealth of InSAR data collected by the past generation of InSAR satellites.

A global dataset

A new study, led by the University of Bristol and published in Nature Communications, collates the last 18 years of InSAR data, including observations at over 500 volcanoes, 198 of which have undergone systematic observations of ground deformation. In this study, the authors assess the significance of ground deformation as an indicator of a volcano’s long-term potential to erupt. The results show that many (46%) of deforming volcanoes also erupted, and almost all (94%) non-deforming volcanoes did not erupt. This demonstrates the importance of ground deformation as an indicator of unrest, and also shows that InSAR is an ideal tool to gauge the eruptive state of volcanoes on an individual, and global basis.

Animation demonstrating the use of InSAR to monitor volcanoes in East Africa. (Video: European Space Agency).

Many past systematic studies have targeted volcanoes with long histories of unrest. However, when observations of deformation are made at volcanoes that have not previously been studied, it is much more difficult to gauge the significance of ground deformation and whether or not it indicates an eruption is imminent. This is particularly true in the absence of additional monitoring equipment. This study demonstrates how, in these cases, we can use data from a global dataset to predict how the composition of the magma, the type of volcano, and the tectonic setting might influence the relationship between observed deformation and eruption. For example, the authors show that globally, deformation observed at volcanoes in subduction zone settings has a higher positive predictive value (i.e. is more likely to result in eruption) than deformation observed at volcanoes in extensional rift settings.  This approach of using global observations to inform local predictions, has the potential to be incorporated into hazard assessments

The future

With the launch of new satellites comes a new age of more systematic and regular data acquisitions, enabling more volcanoes to be monitored systematically. This will inevitably reveal new cases of ground deformation at previously unstudied volcanoes. In these cases, where historical records are short or non-existent, the integration of a global set of observations will be extremely helpful in unravelling the link between deformation and eruption.

New technology and improved data quality will allow the scientific community to improve the accuracy and rate at which satellite imagery is processed and used for hazard assessments. This will enable us to add to this global dataset, strengthening conclusions and widening the global effort to better understand the significance of volcanic unrest at individual volcanoes.

“Global link between deformation and volcanic eruption qualified by satellite imagery” (Biggs et al. 2014) is published today in Nature Communications.

Read the official University of Bristol press release A satellite view of volcanoes finds the link between ground deformation and eruption

Amy Parker, is a PhD student in the School of Earth Sciences at the Cabot Institute, University of Bristol. For more information email Amy.Parker@bristol.ac.uk or tweet @amylauraparker.

Amy Parker