Stochastic analysis of geo-electric field singularities as seismically correlated candidates

Abstract

The study of the Earth's electromagnetic field prior to the occurrence of strong seismic events has repeatedly revealed cases where transient electric potential anomalies, often deemed as possible earthquake precursors, were observed on electromagnetic field recordings. In an attempt to understand the nature of such signals, several models have been proposed based upon the exhibited characteristics of the observed anomalies, often supported by different mathematical models simulating possible generation mechanisms. This paper discusses a candidate Electric Earthquake Precursor (EEP) signal, accompanying the Kythira <i>M_w</i>=6.9 earthquake in Greece (occurred on 8 January 2006). Neuro-Fuzzy along with stochastic models are currently incorporated for the modelling and analysis of the recorded Earth's electric field. The results of the study indicate that the Neuro-Fuzzy model treats the observed possible EEP signal as an external additive component to the recorded Earth's electric field, while the stochastic TARMA models accurately represent the recorded electric signals in both the time and the frequency domains. The complementary findings of both methodologies might potentially contribute to the future development of a more accurate and generalized framework for the efficient recognition and characterization of possible EEP's.