M. Lazaridou

Latest aspects of earthquake prediction in Greece based on seismic electric signals, II☆

Abstract Since 1983, continuous monitoring of the electrotelluric field has been carried out using an array of measuring stations located at various sites in Greece. The basic physical properties of the transient changes—seismic electric signals (SES)—in the electrotelluric field that are forerunners of earthquakes were first described six years ago. Since then a large body of data has been collected resulting in new insight into various aspects of the method. The present paper reviews the latest developments in SES-based earthquake prediction and describes the current procedures used to predict the epicenter and magnitude of an impending earthquake. A detailed list of the predictions officially issued in Greece during the past 3 years (January 1, 1987–November 30, 1989) is also given. Public warnings were issued well before the most destructive seismic activity.

Transmission of stress induced electric signals in dielectric media

The conditions under which pressure (stress) variations on solids, containing charged defects, can lead to the emission of transient electric signals, are discussed. The resulting electric field E varies as 1/d3 (where d denotes the distance from the emitting source), in the simple case when the surrounding medium is homogeneous and isotropic. We show that this behavior changes to 1/d when studying the electric field within a cylindrical channel of radius R and infinite length having conductivity appreciably larger than that of the host medium; this holds up to a certain (reduced) distance d/R, which increases versus the conductivity ratio. We also investigate the variation of the electric field, versus the distance, inside a layer of width w and infinite extent having conductivity appreciably larger than that of the host medium; we then find that the electric field decreases as 1/d2, in a wide range of distances up to a certain value of d/w, which is controlled by the conductivity ratio. In both conducti...

Investigation of seismicity after the initiation of a Seismic Electric Signal activity until the main shock

The behavior of seismicity in the area candidate to suffer a main shock is investigated after the observation of the Seismic Electric Signal activity until the impending main shock. This is based on the view that the occurrence of earthquakes is a critical phenomenon to which statistical dynamics may be applied. In the present work, analysing the time series of small earthquakes, the concept of natural time chi was used and the results revealed that the approach to criticality itself can be manifested by the probability density function (PDF) of kappa(1) calculated over an appropriate statistical ensemble. Here, kappa(1) is the variance kappa(1)(=-(2)) resulting from the power spectrum of a function defined as Phi(omega)= summation operator(k=1)(N) p(k) exp(iomegachi(k)), where p(k) is the normalized energy of the k-th small earthquake and omega the natural frequency. This PDF exhibits a maximum at kappa(1) asymptotically equal to 0.070 a few days before the main shock. Examples are presented, referring to the magnitude 6 approximately 7 class earthquakes that occurred in Greece.

Fluctuations, under time reversal, of the natural time and the entropy distinguish similar looking electric signals of different dynamics

We show that the scale dependence of the fluctuations of the natural time itself under time reversal provides a useful tool for the discrimination of seismic electric signals (critical dynamics) from noises emitted from man-made sources, as well as for the determination of the scaling exponent. We present recent data of electric signals detected at the Earth’s surface, which confirm that the value of the entropy in natural time as well as its value under time reversal are smaller than that of the entropy of a “uniform” distribution.

Natural entropy fluctuations discriminate similar-looking electric signals emitted from systems of different dynamics.

Complexity measures are introduced that quantify the change of the natural entropy fluctuations at different length scales in time series emitted from systems operating far from equilibrium. They identify impending sudden cardiac death (SD) by analyzing 15 min electrocardiograms, and comparing to those of truly healthy humans (H). These measures seem to be complementary to the ones suggested recently [Phys. Rev. E 70, 011106 (2004)]] and altogether enable the classification of individuals into three categories: H, heart disease patients, and SD. All the SD individuals, who exhibit critical dynamics, result in a common behavior.

Numerical model of the selectivity effect and the ΔV/L criterion

Numerical solutions of Maxwell equations for a model earth with a reasonably conducting channel indicate that the electric field values are intensified within a certain region only (i.e., above the end of the channel), thus explaining the observed selectivity effect. In this region, the electric field may reach detectable values (5–10mV/km), while the magnetic field still remains low (10−2nT). The results are compatible with those obtained recently by analytical solutions (Varotsos et al. [1998]). Both the numerical and the analytical solutions lead to a natural explanation of the ΔV/L≈const criterion. This criterion, however, should not be applied over an area with strongly inhomogeneous electrical structure.

Official earthquake prediction procedure in Greece

Abstract Since Jan. 19, 1983, the authors have been operating a telemetric network of electrotelluric stations in Greece. The real-time observation of specific pre-seismic variations in the electric field of the earth (seismic electric signals, SES) allows us to predict the time, epicenter and magnitude of an impending earthquake (EQ). Initially the predictions were announced by expediting telegrams prior to the event. After the Ministry of Public Works established the Earthquake Prediction Council (EPC) in September 1985, predictions were dealt with in special sessions of the council prior to the occurrence of an earthquake. It was decided to announce only EQs with an expected magnitude Ms ⩾ 4.8 for epicenters predicted within or near the perimeter surrounding the network of measuring stations, and Ms > 5 for EQs predicted outside the perimeter. This procedure was followed from September 1985 to March 18, 1986 with the following results. 1. (1) Within the network. Only one EQ (with Ms = 4.9) occurred during the period mentioned. The prediction of the magnitude was accurate to within a few tenths of a magnitude-unit, while the epicenter was predetermined to within some tens of kilometers. No predictions were made that were not followed by earthquakes. 2. (2)Outside the network. Three EQs with Ms = 5.4, 5.2, and 5.3 occurred on Nov. 9, Dec. 18, and Dec. 23, 1985, respectively. The first was missed. The magnitudes of the other two were predicted to be Ms = 5.2 and 5.5 respectively. The corresponding epicentral coordinates were forecast with an accuracy of 150 and 160 km respectively. The sessions were discontinued on March 18, 1986. On March 24, 1986, a Ms = 6.1 EQ was predicted and directly announced to the Ministry of Public Works. The event occurred on March 29 with the predicted magnitude and less than 50 km from the predicted epicenter.

Summary of the five principles suggested by Varotsos et al. [1996] and the additional questions raised in this debate

The present paper cannot be considered, either as a rebuttal to any participant, or our overview of the debate. Its publication became necessary due to the fact that various participants raised additional questions, i.e., beyond the points suggested by Varotsos et al. [1996]. We clarify these questions that concern the noise discrimination from our electrical recordings, the recent laboratory experiments which support the emission of electrical precursors, and the question on whether, or not, a retroactive adjustment of the VAN prediction parameters was made, after the period 1987–1989 discussed in this debate. We draw attention to the fact that a continuous 9 year (i.e., 1987–1995) sample of VAN predictions is now available. For the benefit of the reader, the present paper also summarizes the essence of the five Principles suggested by Varotsos et al. [1996] (as a consequence, attention is drawn to a correct definition of the success rate). This essence remains exactly the same as it was initially suggested, because we do not feel, after the debate, that the various contributions cast a sound doubt on the correctness of any of these Principles. The calculations which claim that VAN predictions can be ascribed to chance strongly violate these Principles; the incorrectness of these calculations is beyond any doubt, because they “reject” even an ideal earthquake prediction method. On the other hand, several well founded calculations convince that the VANs success (and alarm) rate is very far beyond chance. The study of this paper is highly recommended to the reader before going through the details of each of our individual Replies.

Identifying sudden cardiac death risk and specifying its occurrence time by analyzing electrocardiograms in natural time

Sudden cardiac death (SCD) is a frequent cause of death and may occur even if the electrocardiogram seems to be similar to that of a healthy individual. A method which not only identifies the risk but also provides an estimate of the time of an impending cardiac arrest is proposed. Analyzing 159 electrocardiograms in natural time, the authors find that the key quantity is the entropy change under time reversal. After it becomes maximum at the scale of 13 heartbeats, ventricular fibrillation starts within ∼3h in 16 out of 18 SCDs. The method also distinguishes congestive heart failure patients from SCD.

Prediction of the 6.6 Grevena-Kozani earthquake of May 13, 1995

Abstract Seismic Electric Signals (SES) were recorded by VAN-group on April 18–19, 1995, at Ioannina station; they resulted in an official prediction that was sent (two weeks before the earthquake occurrence) to the Greek authorities as well as to various International Institutes. The observation of these electrical variations was confirmed by Gruszow et al. (1996); however, they claim that these signals could be attributed to a (non determined) nearby artificial source with huge intensity (IL≈4 × 10 4 Am, for r ≈ 2 km. or 1.6 × 10 5 Am, for r ≈ 4 km). This claim is not valid, because, such an artificial source (cf. horizontal point current dipole) should have produced: (a) electrical field variations having amplitudes two orders of magnitude, larger than the observed ones; this is theoretically shown and experimentally verified and (b) magnetic field variations mainly on the horizontal field, while, in the present case, they have been recorded mainly on the vertical component. Furthermore, we show that the above SES obey the criteria, suggested by Varotsos and Lazaridou (1991), for discriminating SES from noise.