Tracking volcanic explosions using Shannon entropy at Volcán de Colima

Abstract

The main objective of this work is to show that Shannon Entropy (SE) calculated on continuous seismic signals can be used in a volcanic eruption monitoring system. We analysed three years of volcanic activity of Volcán de Colima, México, recorded between January 2015 and May 2017. This period includes two large explosions, with pyroclastic and lava flows, and intense activity of less energetic explosion, culminating with a period of quiescence. In order to confirm the success of our results, we used images of the Visual Monitoring system of Colima Volcano Observatory. Another of the objectives of this work is to show how the decrease in SE values can be used to track minor explosive activity, helping Machine Learning algorithms to work more efficiently in the complex problem of distinguishing the explosion signals in the seismograms. We show that the two big eruptions selected were forecasted successfully (6 and 2 days respectively) using the decay of SE. We conclude that SE could be used as a complementary tool in seismic volcano monitoring, showing its successful behaviour prior to energetic eruptions, giving time enough to alert the population and prepare for the consequences of an imminent and well predicted moment of the eruption.

Universal machine learning approach to volcanic eruption forecasting using seismic features

“…The study of parameters like Kurtosis and Frequency Index reflects the type of activity present in the volcano; kurtosis characterizes the presence of spectral peaks in temporal series, so impulsive high frequency signals like volcanic tectonic events will induce changes in the Kurtosis values (Cortés et al, 2015); on the other hand, the frequency index indicates changes in the spectral tendency of the signal, so energetic tremor or a swarm of long period events, which are both low frequency signals, will produce a displacement of the Energy content to the lower frequency bands, reflected in the trend of the Frequency Index (Bueno et al, 2019;Rey-Devesa et al, 2023a). In addition, Shannon Entropy is a measure of the uncertainty, or the amount of information, of a dataset, which provides a quantitative value of the predictability of the system; Shannon Entropy decreases whenever the volcanic seismic signals are homogeneous; thus, the changes of a volcano selforganizing prior to an eruption are reflected in a decreasing trend of the temporal evolution of the Shannon Entropy to minimum values (Shannon, 1948;Delgado-Bonal and Marshak, 2019;Rey-Devesa et al, 2023b). With these seismic features, we built a temporal matrix that involves data from several volcanoes; the data correspond to both eruptive and non-eruptive phases, involving seismicity associated to the noise prior of the eruption, the preeruptive activity, and the explosions.…”

“…Besides these limitations, signal processing techniques have enabled the development of volcanic early warning tools, demonstrating its capability to detect significant changes in volcanic and seismic activity (Rey-Devesa et al, 2023b;Ardid et al, 2022;Caudron et al, 2020;Dempsey et al, 2020). This allows better hazard evaluation policy and protection for the population living in…”

Entropy of the Canonical Occupancy (Macro) State in the Quantum Measurement Theory

“…The concept of physical entropy is controversial since the times of Boltzmann and Gibbs . The subsequent excerpt from an interview with Shannon, whose crucial contribution to the comprehension of entropy is now widely acknowledged not only in physics [1] but also in diverse fields such as medicine, seismology, and finance (refer to [2][3][4] for recent applications of Shannon entropy in these domains), is available in [5]: "My greatest concern was what to call it. I thought of calling it 'information,' but the word was overly used, so I decided to call it 'uncertainty.'…”

Machine Learning Analysis of Seismograms Reveals a Continuous Plumbing System Evolution Beneath the Klyuchevskoy Volcano in Kamchatka, Russia

“…Due to advancements in information processing and the introduction of continuous recordings, some studies have shown that continuous seismograms contain relevant information beyond classical event catalogs. The entropy of the seismic noise and the seismic background level seems to vary significantly prior to eruptions (Glynn & Konstantinou, 2016; Ichihara et al., 2023; Rey‐Devesa et al., 2023). Machine learning strategies, in particular unsupervised learning, provide a promising approach for automatically analyzing large amounts of continuous seismograms and inferring such patterns without requiring predefined labels (e.g., Holtzman et al., 2018; Jenkins et al., 2021; Köhler et al., 2010; Ren et al., 2020; Seydoux et al., 2020; Steinmann, Seydoux, Beaucé, & Campillo, 2022; Steinmann, Seydoux, & Campillo, 2022; Zali et al., 2023).…”