Many volcanic eruptions are shortly preceded by new magma injection into a pre-existing, shallow (< 10 km) magma chamber, causing convection and mixing between the incoming and resident magmas. These processes may trigger dyke propagation and further magma rise, inducing long-term (days to months) volcano deformation, seismic swarms, gravity anomalies, and changes in the composition of volcanic plumes and fumaroles, eventually culminating in an eruption. Although new magma injection in shallow magma chambers is a potentially hazardous event, its occurrence is still not systematically detected and recognized. Here we present the results of numerical simulations of magma chamber replenishment by buoyant magma of deeper origin, and the associated gravity changes, seismicity, and ground deformation. Synthetic gravity changes and ground deformation patterns are then inverted with classical methods, to check their capability to detect the source of signals. The results show that the invaded shallow chamber may be not revealed by inversion of ground deformation, as a consequence of non-homogeneous pressure changes resulting into substantial deviations from usual simplifying assumptions when inverting the data. While ground deformation patterns and volcanic seismicity tend to illuminate the deeper regions of the magmatic system, gravity changes are controlled by the shallow system where gas expansion dominates. These results suggest that i) classic simplifications in data inversion techniques may be largely inadequate for magmatic systems, and ii) more robust inversions require joint use of a variety of data including gravity changes.

4-5. Open issues and missing links in volcanology and volcano physics
Micol Todesco* (micol.todesco@bo.ingv.it), Eleonora Rivalta, Lucia Zaccarelli, Laura Sandri, Francesca Bianco, Chiara P. Montagna

On the unrest and eruptive behaviour of large calderas: the Campi Flegrei case

P. Papale, G. Chiodini, M. Todesco

Large calderas are the site of the most devastating eruptions occurred on Earth; they often display substantial unrest dynamics that puzzle volcanologists, and in some cases like the Campi Flegrei case, trouble them as well as the society for the enormous risks associated to their eruptions. Calderas display sequences of signals that would almost certainly prelude to an eruption if observed at central volcanoes; nonetheless, volcanic eruptions may not follow, while they may happen with definitely much weaker signals preceding them, as for the Rabaul eruption in 1994. Although largely debated, the origin of this controversial behaviour is still unclear. The caldera structure favours the development of large geothermal circulation, that is often invoked as an important controlling factor for the observed geophysical and geochemical changes. At least at Campi Flegrei, the structural setting of the caldera appears to have repeatedly favoured emplacement of small magma bodies at shallow (< 4 km) depth, creating a network of interconnected reservoirs capable to exchange mass and heat. The different efficiency of interconnections likely controlled the scale of the eruptions, resulting therefore in limited role of the most shallow batch on the eruption impact, and complicating the forecasts. Essentially, although our knowledge of caldera systems has evolved substantially, our understanding is still limited, contributing to increase their associated risk.

1. Geophysical modelling
2. magma dynamics