Location and Waveform Classification of Seismicity at Tungurahua Volcano (Ecuador) During the February and April 2014 Eruptive Phases
Abstract:We examined 6 months of seismic recordings collected in 2014 during a temporary experiment carried at Tungurahua. This andesitic stratovolcano has been erupting since 1999 and its activity since late 2008 is characterized by the occurrence of eruptive phases lasting from few weeks to months, separated by quiescence periods. These phases have quite variable temporal evolutions. They include the emission of ash and gases with the occurrence of Strombolian to Vulcanian explosions. We use data from the temporary network and permanent monitoring network which includes 6 broadband and 6 short period stations. The temporary network was installed at the end of October 2013 to improve the density and azimuthal coverage. In 2014, it included 11 broadband stations located up to 4200 m elevation. This network is still running at the time of writing.
We processed the data including the eruptive phases of February and April 2014 which started with major Vulcanian explosions generating plumes up to 13,000 m elevation and pyroclastic flows. We first applied waveform classification techniques to search for characteristic short period repeating events. Results outline the presence of several families mostly grouping explosion quakes (EQs) and Long Period events(LPs). We then focused on the location of the seismo-volcanic sources. Volcano-tectonic events are not frequent at Tungurahua. As suggested by waveform classification most of the seismicity is related to eruptive processes with EQs, LPs and tremors. To locate such signals which are often emergent or steady, we used a method based on the decay of the amplitude as a function of the distance. Amplitudes calculated at the different stations are corrected for site effect using coda site amplification factors. Approximating that seismic amplitudes decay as a function of distance as body waves in a homogeneous medium, we use an inversion technique to locate the source of the events. We used this technique to locate the source of tremor observed during the vent opening phases of February 1st and April 4, the source of explosion quakes and the source of a few episodes of harmonic tremor. At this point all source locations indicate an almost common source at about 1 km below the crater. A downward migration of tremor sources is possibly observed during the February vent opening phase.