Autopsy of an eruptive phase of Tungurahua volcano (Ecuador) through coupling of seismo-acoustic and SO 2 recordings with ash characteristics

Journal article, 2019

Eruption style and dynamics are controlled by various parameters including magma supply rate, magma viscosity, volatile content, and the permeability of the conduit. Rapid changes of these parameters can significantly modify the hazards associated to the eruption processes and understanding their relationship with multiparametric geophysical monitoring data can greatly improve our forecasting capacities. From 2008 to 2016, volcanic activity at Tungurahua was characterized by eruptive phases separated by episodes of quiescence. These phases displayed great variability of eruptive patterns including Vulcanian and Strombolian explosions, low pyroclastic fountaining, continuous or sporadic ash emissions and passive degassing. We use the comparison between geophysical data (seismic, acoustic and SO 2 emission), recorded by permanent monitoring networks, and the characteristics of the emitted ash to track changes in eruption dynamics during an eruptive phase that lasted from late December 2009 to March 2010. We show that the correlation between the analyzed parameters allows imaging and interpretation of the conditions at the vent. At Tungurahua, these conditions can rapidly change at the time scale of a single eruptive phase, corresponding to various degrees of opening, plugging and permeability of the conduit. Two magma intrusions could be identified during a single eruptive phase showing transitions between violent Strombolian and Vulcanian activity. Changes in the componentry of the analyzed ash samples, together with the geophysical data, nicely highlight these evolutions. Studying these parameters simultaneously provides a unique insight into the physical processes controlling superficial volcanic activity and offers a potential tool for better understanding volcanoes and detecting changes in their activity. The joint interpretation of multiparametric data which we propose is potentially applicable to multiple andesitic volcanoes.