The dynamics of the 2001 Etna eruption as seen by full moment tensor analysis

The Mt. Etna eruption of July 2001 was announced by severe seismic activity and by the opening of a 7-km-long zone of densely distributed fractures. The large amount of data collected gave a unique opportunity to study the magma migration process and to infer the position and geometry of the uprising dyke. Results from multidisciplinary approaches suggest that the observed phenomenology was the result of the rapid intrusion of a vertical dyke, oriented roughly N–S and located a few km south of the summit region. To add new constraints to the dynamics of the eruption process, in this study we determine the full seismic moment tensors of 61 earthquakes, selected among those occurring between July 12 and July 18 (Md ≥ 2.2), located in a depth ranging from 1 km above sea level (a.s.l.) to 3 km below sea level (b.s.l.). At the beginning of the seismic swarm, the dominant component of the seismic source tensor is double-couple percentage (around 65 per cent on average) statistical significant at 95 per cent confidence level and in the following hours the non-double-couple components increase at the expenses of the double-couple. Such observations are related well with the system of fractures formed just before the eruption, whereas the increasing non-double-couple components can be explained as the response of the confining rocks to the rising magma and degassing processes. The type of focal mechanisms retrieved are predominantly of normal fault type (44 per cent), strike slip (30 per cent) and thrust mechanisms (9 per cent), and outline a scenario that concurs with the stress regime induced by a dyke injection. The space–time analysis of seismic source locations and source moment tensors (1) confirms the evidence of a vertical dyke emplacement that fed the 2001 lateral eruption and (2) adds new insights to support the hypothesis of the injection of a second aborted dyke, 2 km SE from the fractures zone.