This project has received funding from the European Union's Horizon 2020 research and innovation programme under grant agreement No 858092

Europe flag
Eruzione Etna


Modern volcanology develops through contributions from multiple disciplines such as geology, physics, chemistry, mathematics, engineering, informatics, etc. Volcanoes are monitored and investigated using highly engineered networks of telemetered multi-parametric sensors providing information on on-going sub-aerial and underground dynamics. This science is enabled by sophisticated computer software for data inversion and direct numerical simulations, as well as by high-tech laboratory analyses and experiments aimed at reproducing and testing, at small scale, processes and hypotheses in volcano dynamics.

Two major opportunities have recently emerged that offer potential for unprecedented understanding of volcanic systems:

Krafla is a restless caldera in Iceland (last eruptions in 1975-1984) currently exploited for geothermal energy production by the National Power Company of Iceland Landsvirkjun (Partner LV in IMPROVE). Whilst drilling in 2009 in search of a non-conventional energy source represented by supercritical fluids (the Iceland Deep Drilling Project, IDDP), LV serendipitously encountered rhyolitic magma at 2.1 km depth. That was unexpected, as the results of available inversion of electromagnetic soundings, as well as other available geophysical and geochemical data, had suggested the absence of magma in the uppermost 4 km beneath the caldera floor. Retrospective analysis showed that company drilling had hit magma at similar depths on at least two other occasions over an area of 3.5 km2.

The Krafla case demonstrates that current state-of-the-art geophysical imaging may miss identifying shallow magmatic bodies; a fact with enormously adverse implications for volcanic hazard and risk analyses worldwide. At quiescent or restless volcanoes close to urbanized areas (Vesuvius and Campi Flegrei in Italy, Auckland Volcanic Field in New Zealand, Mount Rainier in US, Unzen in Japan, and many others), the presence of magma at shallow levels would significantly change the expected pre-eruptive sequences and eruptive activities, and consequently, the alert level systems and preparatory plans established by Civil Defence authorities.

Recent advances in geophysical imaging, including joint inversion techniques merging geological and geophysical investigation, novel approaches like virtual source noise-correlation and extraction of coherent body waves from the noise field, seismic interferometry with dense seismic arrays, etc., yield high resolution and are highly promising in order to establish standard practices to detect the magma-rock interface. The Krafla case offers the ideal conditions for such a fundamental advance, as it provides a unique situation where the location of a shallow magma body has been directly observed, therefore allowing ground-truth tests of advanced geophysical imaging methods and techniques, and development of new ones. Krafla is also the site of intense geothermal circulation, that must be accounted for in order to deal with geophysical and geochemical measurements, data interpretation, and inversions. Characterization of geothermal fluids and geothermal circulation modelling has also progressed significantly during last years, as a response to both industrial and volcanic hazard objectives. Characterizing and modelling geothermal circulation at Krafla is necessarily, and desirably, one objective of IMPROVE: it is necessary because geothermal circulation affects geophysical and geochemical observation, and desirable because the establishment of close and effective science-industry relationships is an objective of IMPROVE. The above discussion leads to:


IMPROVE will attack the present-day major limitation in underground imaging of volcanoes, by concentrating the research of 9 ESRs and their scientific tutorship teams in a coordinated, multi-disciplinary effort aimed at testing current-day models, and developing new ones, at Krafla, the unique place where ground-truth testing can be conducted thanks to direct knowledge of the magma-rock interface.

Ground displacement at volcanoes occurs over time scales from the sub-seconds of seismic shaking to the days, months and years of volcano edifice dynamics. Historical progress has separately addressed seismicity and geodesy at the high and low frequency extremes, respectively. Remarkably, these two fields of investigation have largely ignored the intermediate periods from tens/hundreds of seconds to hours, mostly as a consequence of instrumental limitations. Recent expanded use of broad-band seismometers, as well as new data from dilatometers, strain meters, and tiltmeters, can identify rarely reported signals with periods extending to tens or hundreds of seconds. Such measurements demonstrate Ultra-Long-Period (ULP) ground oscillations, the interpretation of which does not benefit yet from the same established background knowledge as for displacement dynamics in the seismic and quasi-static domains. Numerical simulations of magma dynamics allowed us to discover that ULP ground displacements can be due to deep magma convection. Close correspondence between simulated and measured (through dilatometers) ULP ground displacement dynamics at Campi Flegrei nicely support that view.

The above arguments and findings show that i) current methods to measure ground displacements at volcanoes urgently need to systematically fill the gap between seismic and quasi-static domains, and ii) ULP ground displacements located in intermediate frequency range may contain new and extremely precious information on the location of magma as well as on on-going deep magma dynamics, that we are not yet accessing.

Bridging the gap between seismic and quasi-static ground displacement at active volcanoes is likely to open the path to the next level of volcano monitoring and time-series analysis, and of understanding of the deep volcano dynamics, with comparable social benefits in the form of more reliable knowledge and forecasts. Etna in Sicily is the ideal place for addressing such challenges, as it is i) one of the most active volcanoes in the world, continuously offering observations and measurements on deep and shallow magma and volcano dynamics, and ii) one of the best known and most monitored volcanoes in the world, with decades of high-quality time series from a multi-parametric measurement network in principle covering any frequency from milliseconds to decades. This leads to:


IMPROVE will fully explore the ULP ground displacement range and fill the gap between seismic and quasi-static ground displacement dynamics at active volcanoes, by concentrating the research of 6 ESRs and their scientific tutorship teams in a coordinated, multi-disciplinary effort taking advantage of the unique, ideal conditions at Etna in terms of activity, knowledge and data.