The Prediction of Two Large Earthquakes in Greece

Abstract

The VAN experimental method of shortterm earthquake prediction (named after the initials of three Greek physicists, Panayiotis Varotsos, Kessar Alexopoulos, and Konstantine Nomicos) has been used to monitor preseismic electric signals since the 1980s [see Varotsos, 2005]. From observed telluric current signals, called seismic electric signals (SES), the epicentral area, magnitude, and occurrence time of an impending earthquake are estimated. SES are interpreted as having been emitted when the focal region in which the earthquake in question could occur has entered the critical regime (i.e., a stage close to the rupture). The VAN method recently reached the stage of possibly enabling the narrowing of the time window of earthquake prediction to the order of a few days. This narrowing is made possible by the use of a new method called “natural time analysis.” This analysis has been developed to identify the time when a dynamic system (i.e., a system evolving with time) exhibits behavior similar to a phase change [ Varotsos et al., 2008, and references therein]. On the hypothesis that the main shock earthquake is a critical phenomenon, when SES activity is observed, natural time analysis is conducted on the seismicities of small earthquakes in the suspected future epicentral area solely by considering their order of occurrence and the energy emitted by each of them. The term natural time analysis stems from the disregard of the conventional time of the earthquakes' occurrence. It has been found that such an analysis enables the identification of the time of the main shock usually within a few days before it occurs (see P. Varotsos et al., Seismic electric signals and 1/f “noise” in natural time, at http://arxiv.org/abs/ 0711.3766).