A Bayesian Approach to Infer Volcanic System Parameters, Timing, and Size of Strombolian Events From a Single Tilt Station

Abstract

AbstractPersistently active volcanoes are characterized by frequent eruptions, in which volatiles dissolved in magma play an important role in controlling the explosivity. Inverting techniques on geodetic data sets have been used to retrieve information about key controlling parameters of these eruptions. However, up to date, several data sets are combined to obtain reliable estimates of the physical parameters using a physical model, hindering the possibility to provide forecasting tools for time and magnitude of eruptions at volcanoes with limited monitoring network. In this work, we propose an approach to extract valuable information out of limited data sets through inverting techniques dealing with limited number of sensors, but high frequency of events. Our method exploits time series of tilt signals recorded by a single station to estimate, by mean of the Bayesian statistics and a physics‐based model, the range of the controlling parameters. The method was developed and tested on a synthetic volcanic system before being applied on data from Semeru volcano (Indonesia). Finally, we tested the possibility to forecast explosion magnitude and timing using data recorded by a single tilt station. Results show that data from a limited network or even a single tilt station is sufficient to estimate the controlling parameters. The information obtained is shown to be useful for estimating the time and magnitude of future events, which can enhance the monitoring systems of those volcanoes characterized by frequent, potentially dangerous events.