Earthquakes are mechanical instabilities in the brittle lithosphere produced by the progressive accumulation of stress in the crust over centuries. The more extended the fractures, called faults, the larger the earthquakes. The magnitude of a seismic event is estimated with the moment magnitude, which is a measure of the energy released by the event on a logarithmic scale. Most earthquakes are small, but larger events are also possible, approximately with a 1:10 ratio compared with the frequency of events with a magnitude of one lower degree. For decades, seismologists have been debating whether precursors exist and, if so, what are the signals providing reliable information about the approach of catastrophic events. In this work recently published in the authoritative journal edited by the American Geophysical Union, the Journal of Geophysical Research – Solid Earth, a team of researchers from Sapienza University of Rome, the National Institute of Geophysics and Volcanology, the National Research Center and the National and Kapodistrian University of Athens suggest an answer this question. To do so, they combined theoretical modeling and statistical analysis. Specifically, they investigated the properties of earthquake clusters occurred in California during the last thirty years before the main events as a function of their magnitudes. Research shows that the so-called foreshocks, small-to-moderate earthquakes which sometimes are thought to forewarn more violent earthquakes, tend to spread over larger areas, are featured by highly variable magnitudes, are more numerous and energetic than swarms (clusters characterized by small maximum magnitudes). Conversely, swarms and foreshocks share the same distribution of duration, intensity, and frequency of events. Results, appropriately supported by statistical tests, suggest that earthquake clusters with several events extended over large regions with high energy entropy, the probability that a minor seismic activity can culminate in a major event is higher than in other conditions. The work goes beyond statistical analysis, attempting to explain the observations. The proposal is that stressed rock volumes progressively become more and more globally unstable and trigger each other seismic activity over more and more extended time intervals and areas, producing clusters of small events. The larger the correlated area, the higher the chances that a run-away earthquake can involve the whole unstable fault system. Therefore, a cascade feedback mechanism acts on stress history release in previous events producing future seismic activity. If this idea were correct, i.e., seismicity is mainly a memory phenomenon, the causal relationship between the series of foreshocks and the main earthquake would disappear. Therefore, according to this point of view, it would not make sense to distinguish between foreshocks and swarms. In support of this hypothesis there is evidence of large earthquakes occurred without being preceded by any seismic anomaly or even in the presence of a decrease in seismic activity, as in the case of the Amatrice earthquake in 2016, as well as the failure of several statistical tests.
https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2023JB027337?af=R