A new picking algorithm based on the variance piecewise constant models
AbstractIn this paper, we propose a novel picking algorithm for the automatic P- and S-waves onset time determination. Our algorithm is based on the variance piecewise constant models of the earthquake waveforms. The effectiveness and robustness of our picking algorithm are tested both on synthetic seismograms and real data. We simulate seismic events with different magnitudes (between 2 and 5) recorded at different epicentral distances (between 10 and 250 km). For the application to real data, we analyse waveforms from the seismic sequence of L’Aquila (Italy), in 2009. The obtained results are compared with those obtained by the application of the classic STA/LTA picking algorithm. Althoug…
Optimization of Low-Cost Monitoring Systems for On-Site Earthquake Early-Warning of Critical Infrastructures
In the last years, monitoring systems based on low-cost and miniaturized sensors (MEMS) revealed as a very successful compromise between the availability of data and their quality. Also applications in the field of seismic and structural monitoring have been constantly increasing in term of number and variety of functions. Among these applications, the implementation of systems for earthquake early warning is a cutting-edge topic, mainly for its relevance for the society as millions of peoples in various regions of the world are exposed to high seismic hazard. This paper introduces the optimization of an already established seismic (and structural) monitoring system, that would make it suit…
An Active Learning Approach for Classifying Explosion Quakes
In this work, an Active Learning approach for improving the classification of passed seismo-volcanic events is proposed. Here we study the specific case of Explosion Quakes from Stromboli Volcano versus other seismo-volcanic events, recorded as seismograms, and the use of Random Forest as a Classification method. In conformity with the active learning paradigm, the approach recalls the human intervention for the annotation of uncertain data. The uncertainty is established by the event probabilities, predicted by a trained random forest classifier. The human intervention consists of editing and relabelling the data into these main three classes: Explosion Quakes, Non-Explosion Quakes or Non-…