Discriminating seismic events of the Llaima volcano (Chile) based on spectrogram cross-correlations

Abstract The surveillance of active volcanoes around the world has become a critical security issue for many countries, requiring a continuous monitoring of seismic signals. By analyzing such signals, we intend to understand volcanic activities (e.g. explosions, eruptions and depressurization) and take decisions to reduce the effects and damages to the economy and society of nearby regions. Active volcanoes constantly produce signals, which may then be characterized as data streams, making impractical the presence of specialists to monitor and label every activity. To overcome this drawback, we present a new and straightforward approach to discriminate volcano seismic signals using spectrogram cross-correlations (SPCC) in conjunction with the K-Nearest Neighbors algorithm, a supervised machine learning strategy. Experiments were performed on signals collected from the Llaima volcano (Chile), and results confirmed the ability of our approach to discriminate events with a sensitivity over 95 % for three out of the five classes considered.