Magma source, pre-eruptive dynamics and timescales of major explosions at Stromboli volcano (Italy)
- Authors: Insinga, L.; Voloschina, M.; Marianelli, P.; Bartolomeo, E.; Bertagnini, A.; Métrich, N.; Rotolo, S.G.; Aiuppa, A.; Ripepe, M.; Pistolesi, M.
- Publication year: 2025
- Type: Articolo in rivista
- OA Link: http://hdl.handle.net/10447/688703
Abstract
At Stromboli volcano (Italy) the regular, persistent activity is occasionally (2–4 events/year) interrupted by sudden, short-lived and more energetic major explosions, being intermediate in size between regular Strombolian activity and even more powerful paroxysmal explosions. Despite being frequent and hazardous, the magma source depth, the triggering mechanisms and timescales of such major explosions are still poorly understood. Here, we focus on three major explosions which occurred on 3 May, 8 November and 24 November 2009. We present a dataset of major element composition and dissolved volatile contents in olivine-hosted melt inclusions, embayments and glassy groundmass. We combine them with Fe–Mg diffusion profiles in olivine phenocrysts, and with volcanic gas plume and ground tilt measurements. Olivine phenocrysts display Fo69-72 compositions, with reverse zoning (up to Fo83) in the 24 November and to a minor extent in the 8 November eruption. Glassy groundmass of the November events ranges from the more evolved towards more primitive compositions, whereas the 3 May 2009 glassy groundmass has evolved compositions. Dissolved H2O and CO2 in glassy and bubble-free melt inclusions are low, with CO2 below the detection limit and H2O up to 2.3 wt.%. Volcanic plume measurements indicate the occurrence of these events in phases of high CO2 fluxes, with substantial geochemical changes preceding the onset of the 24 November event. Based on these results, we propose that the 3 May major explosion — the smallest in intensity/magnitude — was likely driven by the accumulation of gas bubbles into the shallower (< 1 km below sea level, b.s.l.) HP reservoir. In contrast, we find the November major explosions to be triggered by the injection of variable amounts of deeper-stored LP magma and gas into the HP reservoir, over hours to weeks before.