Radio Science

ULF/ELF magnetic field variations from atmosphere induced by seismicity

Authors


Abstract

[1] Local variations of the magnetic field in the ULF-ELF frequency range associated with seismicity are studied with the data of more than 3 a observations at Karimshimo complex observatory (latitude 52.83°N, longitude 158.13°E, Kamchatka, Russia). A wideband emission is found to start about 5 d before an earthquake and last until 5 d after it. Seismic ULF/ELF emission in the frequency range of 4–6 Hz as compared with the seismically quiet background has enhanced Phh/Pdd spectral ratio and reduced standard deviation of ellipse orientation angle and the ellipticity, and it has a more linear polarization. Parameters of this emission are studied for more than 30 individual earthquakes and statistically with the superposed epoch method. The reliability of the earthquake predicting hypothesis is verified, and the favorable parameters for the earthquakes together with those for ELF magnetic field are selected. The following earthquake parameters are favorable for this emission: depths H < 50 km, magnitudes MS > 5.5, and epicenter distances R < 300 km. The changes of natural ULF/ELF emissions during the periods of enhanced seismic activity are interpreted as the result of the excitation of additional ULF/ELF emissions in the seismic zone to the east of the observatory or the redistribution of lightning discharges with their possible concentration near the active crust fault. The earthquake prediction hypothesis is verified for the complex field parameter ΔS and proved to be successful.

1. Introduction

[2] Electromagnetic emissions associated with earthquakes have been intensively studied for several decades [see, e.g., Hayakawa and Molchanov, 2002]. The frequencies of the observed emissions range from 10−3 to 108 Hz. Local high-amplitude ULF signals were registered near the epicenters of the violent MS > 7 earthquakes Loma Prieta [Fraser-Smith et al., 1990], Spitak [Kopytenko et al., 1990; Molchanov et al., 1992] and some others. However, numerous attempts to replicate these results at longer distances or for weaker earthquakes gave rather ambiguous results, probably, due to the high level of natural geomagnetic fluctuations in the ULF range below ≃2 Hz. Indeed, the spectral amplitude here decreases rapidly with frequency and it is 3 orders lower at 1 Hz than at 0.01 Hz. At higher frequencies a level of natural geomagnetic emissions is much lower, but artificial interferences dominate at frequencies above 50 Hz. Thus it is natural to choose the frequency range from few Hz to few tens of Hz to look for weak signals associated with seismicity. In this range the narrowband spectrum maxima near Schumann resonance at about 8, 14, 20, and 24 Hz dominate, and the energy source for them is attributed to the worldwide thunderstorm activity [e.g., Nickolaenko and Hayakawa, 2002]. The influence of the world geomagnetic activity is weak within this frequency range, although regular diurnal and seasonal variations exist [Yatsevich et al., 2005]. The signal azimuth at Schumann Resonance frequencies turns following the activity of World thunderstorm centers [Belyaev et al., 1999]. However, at frequencies below first Schumann resonance local thunderstorm activity dominates and signal azimuth is predominantly determined by local (<500 km) sources [Surkov et al., 2006; Fedorov et al., 2006].

[3] Until now, the majority of publications on seismoelectromagnetics except the above mentioned ULF effect have described the observational results at frequencies above hundred Hz, so-called electromagnetic radiation (EMR) [Gokhberg et al., 1982] (see also corresponding reviews by Hayakawa and Fujinawa [1994], Hayakawa [1999], and Hayakawa and Molchanov [2002]) and there are only few publications on seismoelectromagnetics in the frequency range 1–40 Hz [Ohta et al., 2003; Hayakawa et al., 2005]. These works describe individual cases or give the statistical analysis of rather short time intervals.

[4] In the present paper the data are analyzed of more than 3-a monitoring of the magnetic field fluctuations in the frequency range 1–40 Hz in a seismically active region. We compare different parameters of natural ULF/ELF emissions for the seismically active and the quiet time intervals and select the field parameters, which are more sensitive to seismicity.

2. Measurements and Data Processing

[5] Variations of the magnetic field are measured at the Karimshino observatory (latitude 52.83°N, longitude 158.13°E) since June 2000 until now with the three-component induction magnetometer. Its parameters are given in Table 1.

Table 1. Parameters of Magnetometer
ParameterValue
Frequency band0.003–40 Hz
Noise level0.16 F−1 pT/Hz1/2
Conversion function for f < 4 Hz0.4 F V/nT
Conversion function for f > 4 Hz1.6 V/nT

[6] The sensors for the horizontal components H and D are oriented along the magnetic meridian and transversally to it, and the Z sensor is vertical. The parameters of all three sensors are identical with the deviations less than 3% in absolute values of the conversion function and 2° in phase. These discrepancies are corrected with the help of calibration circuits. To suppress seismic, wind, acoustic and other interference, the sensors are put into a concrete box filled with dry sand. The output signal is digitized at the frequency 160 Hz with the 24-bit data acquisition system (DAS) and written to the DAS hard disk. The data are copied and sent for the further analysis.

[7] Preliminary routine data processing includes substituting interpolated data for short (several points) instrumental peaks and data gaps and excluding of intervals that cannot be corrected, filtration and decimation to the 50 Hz sampling frequency. For the analysis of pulse-like wideband signals, the eighth-order Complex Gauss wavelet is used with the parameters chosen from the trade of time and frequency resolution. The real and the imaginary parts of the wavelet are shown in Figure 1.

Figure 1.

Real and imaginary parts of eighth-order complex Gauss wavelet.

[8] Then, the power spectral densities (PSD) are calculated for the field components: meridional (Phh), azimuthal (Pdd), and vertical (Pzz) together with the cross spectra of the horizontal components Phd. First, the periodograms for 256 points samples (∼5 s) are calculated and then averaged over 30 min intervals. The parameters of polarization ellipse are found by the conventional procedure [Fowler et al., 1967]. We denote by θ the angle between the big axis of the polarization ellipse and the H axis and its tangent is found from

equation image

The angle θ in the interval [−π/2, π/2] is totally determined by (1) and the sign of the numerator of the right-hand side of (1). This angle is connected with the azimuth angle of the source being the electric dipole with the current moment m (m = IL, where I is the source current and L is the length of the dipole). Let the distance r between the receiver and the source satisfy the condition Lr, so that with no reflection from the boundaries of the Earth-ionosphere cavity, we obtain the following magnetic field:

equation image

[9] Here the first term corresponds to the induction (near-zone) field and the second term is the radiation or the far-zone field [e.g., Uman, 1987]. It is easy to find that

equation image

and Phh/Pdd ≈ tan−2(θ) ≈ tan2(α), 2 Re [Phd/(Phh + Pdd)] ≈ sin(2θ) that leads to equation (1). Actually, the sources are distributed in space and the receiving is characterized by a distribution function F(α, ω). If sources concentrate along some prevailing direction αα1, then function F(α, ω) has a maximum here, 〈Phh/Pdd〉 ≈ tan2α1. We conclude that when 〈Phh/Pdd〉 < 1, the emission arrives predominantly from the north (south), and if 〈Phh/Pdd〉 > 1, the sources are located mainly in the east (west). Finally, in the case of the isotropic angular distribution of sources 〈Phh/Pdd〉 = 1. Thus, with the account taken of the specific distribution of the sources in the vicinity of Karimshino, the parameter 〈Phh/Pdd〉 − 1 can be used as a measure of the anisotropy of the source distribution. In addition we compute the parameters describing the ellipticity and the sense of the polarization in terms of the angle β [Fowler et al., 1967]:

equation image

[10] The ellipticity, or ratio of minor axis to major axis, is defined by tan(β) and the sense of the polarization depends on the sign of β, i.e., β > 0 and β < 0 are regarded as the right-hand (RH) and left-hand (LH) polarizations, correspondingly.

[11] To minimize the influence of industrial interference and daytime geomagnetic disturbances, only nighttime intervals (±5 h from local midnight) are used for the analysis. For each local night both the mean value and the standard deviation for each parameter is found. We test different combinations of energetic and polarization parameters to find the most sensitive to seismicity.

[12] Seismic data are taken from the local seismic catalogue (see http://www.emsd.iks.ru) and the local seismic index KS [Molchanov et al., 2003] was calculated for each earthquake. It is proportional to the earthquake seismic energy in the observation point and is computed as follows:

equation image

where R is the distance from the observation point and MS is the earthquake (EQ) magnitude. The geomagnetic activity is estimated with the daily ΣKp index.

[13] To overcome a well-known difficulty in discrimination of time and spatial variations in single-point observations, we select the frequency range (4–6 Hz) almost free from the influence of the global geomagnetic activity dominating at lower frequencies and global thunderstorm activity that prevails at higher frequencies and apply the standard deviation of ellipse (SPE) analysis of seismic and nonseismic intervals with identical seasonal distribution to exclude possible influence of random inhomogeneity in seasonal distribution of earthquakes.

3. Results of the Analysis

3.1. Selection of the Parameter That Is the Most Sensitive to Seismicity

[14] To select the best among the ULF/ELF parameters, the record was studied in the interval about 1.5 month long around the seismic swarm in the middle of March 2003. We choose this interval to study a relation between the seismicity and ULF/ELF field variations because of their remarkable behavior. The first half of the interval is seismically absolutely quiet, and the second one starts with the MS = 5.9 shock on 15 March. This earthquake is the first in the EQ series with of slowly decreasing intensity. The second peak in seismic activity corresponds to MS = 6 EQ registered on 19 March. Epicenters of almost all the earthquakes lie in the sea eastward from the observation point. Seismic and geomagnetic activity and ULF/ELF field parameters are summarized in Figure 2.

Figure 2.

(first panel) KS seismic index, ΣKp index of the global geomagnetic activity; (second and third panels) the spectra of the horizontal components of the magnetic field H and D; and (fourth panel) the spectral ratio Phh/Pdd.

[15] Figure 2 shows that intervals of the enhanced Phh/Pdd ratio start several days prior to an earthquake and last several days after it. A similar but weaker effect is seen in the power spectra of the field components. Such a behavior of spectral parameters may correspond to a source located eastward (westward) from the observation point. Coincidence of the intervals with high seismicity and steep variations of the field parameters makes the assumption plausible about a physical relation between them. Namely, we can assume that an additional local source of ULF/ELF magnetic field fluctuations appears in the epicenter zone at the last stage of the EQ preparation and after the earthquake.

[16] Characteristics of the ULF/ELF geomagnetic field in frequency bands aside of Schumann resonances are shown in Figures 3 and 4 The indexes of seismic (KS) and global geomagnetic activity (Kp) are given in the first panel. Then, from top to bottom, the power spectral density of the horizontal components of the magnetic field, their spectral ratio Phh/Pdd, ellipticity and ellipse orientation are given. Figure 3 shows the results for the frequency band 4–6 Hz, below the fundamental harmonic of Schumann resonance, and Figure 4 gives the same parameters for the frequency band 20–24 Hz, above the third harmonic of Schumann resonance.

Figure 3.

Characteristics of ELF noise in the frequency band F = 4–6 Hz for the same period of observation. From top to bottom: KS seismic index, ΣKp index of the global geomagnetic activity; Phh and Pdd spectral power density; Phh/Pdd spectral ratio; the signal ellipticity; and the orientation of polarization ellipse. In the third and fourth panels, gray lines show current values of parameters, medium black line is a running mean value (24 h window), and upper and lower black lines deviate at σ from the mean value.

Figure 4.

Same as in Figure 3 but for the frequency range of F = 18–22 Hz.

[17] In both frequency bands the spectral power of H component and the ratio Phh/Pdd increase 3–4 d before the first shock of the EQ swarm started on 17 March 2003. Simultaneously, the absolute value of ellipticity and the standard deviation of both ellipticity and ellipse orientation angle decreases.

[18] For the further analysis, we take the nighttime intervals of ±5 h from the local midnight and calculate the mean values and the standard deviations (RMS) of the power spectral densities, of the Phh/Pdd spectral ratio, the ellipticity and ellipse orientation angle in the frequency band 4–6 Hz. Different combinations of spectral and polarization parameters were tested to select one the most sensitive to seismicity. Behavior of different combinations of spectral and polarization parameters is summarized in Figure 5.

Figure 5.

Seismicity, geomagnetic activity, and parameters of the magnetic field averaged over nighttime intervals in the frequency band 4–6 Hz. (first panel) KS seismic index, ΣKp index of the global geomagnetic activity; (second to fifth panels) Phh/Pdd − 1, ΔS = (Phh/Pdd − 1)/rms(tan(β)), ΔSt = (Phh/Pdd − 1)/rms (θ), ΔSβt = (Phh/Pdd − 1)/rms (tan(β))/rms(θ).

[19] It is seen in Figure 5 that sensitivities of all the parameters exceed that of the Phh/Pdd − 1 and are approximately equivalent. However, a lateral extension of a source and its explicit position influence the ellipse orientation rather than the ellipticity. Taking into account that the location and the size of a source can vary within a limited zone, we have chosen the parameter

equation image

which presents the seismic influence better than the other parameters.

[20] The results of comparison of the efficiency of two parameters ΔS and Phh/Pdd − 1 are shown in Figure 6 for two intervals: 24 February 2003 to 6 April 2003 (Figure 6a), and 12 July 2004 to 8 August 2004 (Figure 6b). Both parameters increase with seismicity, but the time correspondence of ΔS enhancements with the groups of earthquakes is really amazing. The parameter Phh/Pdd − 1 demonstrates several peaks at seismically quiet intervals, which correspond to low amplitudes and/or unstable polarization of the signal. The relevant peaks are suppressed in ΔS. Thus enhanced ΔS indicates the appearance of an additional signal with the polarization ellipse oriented along the magnetic meridian and the ellipticity stable at timescales of several hours.

Figure 6a.

(left) Seismicity, (middle) geomagnetic activity and the parameters Phh/Pdd − 1, and (right) ΔS averaged over nighttime intervals in the frequency band 4–6 Hz for the intervals 24 February to 6 April 2003.

Figure 6b.

Same as in Figure 6a but for the intervals 12 July to 8 August 2004.

[21] In a more general case of arbitrary direction to the source Phh and Pdd should be changed by Pnn and Ptt, respectively, where indexes n and t correspond to the directions perpendicular and parallel to the direction from the observation point to the source, respectively. Some details of this procedure are discussed in section 2. The effect is seen in the clearest way in the frequency band 4–6 Hz.

3.2. Statistics and Reliability

[22] A detailed description of spectral and polarization parameters of natural signals at 1–40 Hz is given by Nickolaenko and Hayakawa [2002]. The average parameters of the signal measured at Karimshino agree qualitatively with their results. Diurnal and seasonal variation of spectral power density, ellipticity and ellipse orientation are similar to those described by Nickolaenko and Hayakawa [2002].

[23] The spectral density of the total horizontal power G = 〈Phh〉 + 〈Pdd〉 for the whole period of observation is shown in Figure 7. Indices of local seismic (KS) and global geomagnetic (Kp) activity are shown in Figure 7 (top), ΔS and G are given in Figures 7 (middle) and 7 (bottom), respectively. They are calculated with 2-d averaging over night hours in the frequency band 4–6 Hz. The total horizontal spectral power demonstrates the typical seasonal variation with the maximum at local summer. On the other hand, the seasonal variation is not obvious in the variation of the parameter ΔS, which demonstrates an evident correlation with seismic activity. Five intervals of high seismicity are clearly seen in Figure 7 (top) and each of them corresponds to the interval of the parameter ΔS increased. It is also important that the parameter ΔS is not influenced by the geomagnetic activity.

Figure 7.

(top) Seismic activity, (middle) ΔS and (bottom) total horizontal spectral power for the whole period of observations. Field parameters are calculated in the frequency range 4–6 Hz for local nighttime.

[24] Influence of individual earthquakes on ΔS is illustrated in Figure 8. It is seen in Figure 8 that the earthquakes located to the east from the observatory at distances R < 300 km contribute mostly to the ΔS variation. Because of specific distribution of earthquakes in the observation region, these eastward earthquakes are simultaneously the closest to Karimshino.

Figure 8.

A map of the main shocks in the vicinity of Karimshino (shown with a crossed circle and station code KRM). EQ epicenters are shown with circles. The color (in divisions of gray) corresponds to the maximal ΔS during the last 5 d before an earthquake, and the size is proportional to the magnitude of the earthquake. Only the earthquakes with KS > 1 and depths <50 km are shown.

[25] The analysis of data for individual earthquakes shows that the following earthquake parameters correspond to noticeable changes in the magnetic field polarization and, especially, in ΔS: depths H < 50 km, magnitudes MS > 5.5 (E > 1013 J) and epicenter distances R < 300 km.

[26] Earthquakes with great magnitude are relatively rare. Thus, if only few years are included into analysis, the false correlations can occur due to random nonhomogeneity of the seasonal distribution of earthquakes. Among all the intervals with 21-d duration (a central day ±10 d) two opposite groups of intervals are selected: seismic (with the EQ of MS > 4.5, KS > 2 and H < 70 km occurred during the central day of an interval) and seismically quiet (with no EQs of such parameters throughout the interval). The data coverage for the whole period of observations is good and the months' distribution of EQs is almost homogeneous for the total array. However, nonseismic intervals are concentrated in summer and a deficit of earthquakes exists in summer months. This may result in false interdependences between seismic activity and parameters of ULF/ELF emissions. To estimate a relation between seismic activity and ELF parameters free from seasonal effect, the following technique is used. Tree arrays of the seismic a seismically quiet intervals and general sample with the identical distribution of number of intervals over months are formed. The selection procedure is based on the condition of maximal total number of seismic intervals. Namely, we take all the months with nonzero number of both seismic and nonseismic intervals (Months 1, 3, 6, 8, 9, 12) and average the parameters for seismic intervals with unit weight, while for nonseismic intervals ELF parameters are averaged with the weight equal to ratio of seismic to nonseismic number of intervals. The resulting month distribution of seismic and nonseismic intervals and weights are given in Table 2.

Table 2. Month Distribution of Seismic and Nonseismic Intervalsa
MonthSeismic TotalNonseismic
TotalWeight
  • a

    The number of seismic intervals taken for SPE analysis, total number of nonseismic intervals for each month, and the weight for averaging nonseismic ELF parameters are given.

1380.375
3551
61580.017
84140.286
9431.333
125160.313
All22  

[27] The parameters of the ELF magnetic field are calculated for seismic and nonseismic subarrays and for the general sample and compared. The results for the ellipse orientation angle θ and the parameter characterizing the degree of linear polarization l = (1 − ∣tan(β)∣)/(∣tan(β)∣ + 1) are shown in Figure 9. The latter parameter varies from zero for circular polarization to unity for linear polarization.

Figure 9.

(left) Results of SPE analysis for (top) θ and (bottom) l. For the seismic intervals (S), zero of the time axes corresponds to main shocks. The nonseismic intervals (Q) and the general sample including all the intervals with seismicity not taken into account (T) have the same length and seasonal distribution as seismic intervals. Thin lines around thick Q and S curves show the results for two subsets of corresponding samples. (right) Validity of the hypothesis (top) about θS ≠ θQ and (bottom) lSlQ.

[28] In Figure 9 (top) the results for the angle of ellipse orientation θ are shown. In a general case, average ellipse orientation is controlled by the distribution of sources in rather a wide sector. Appearance of additional sources, concentrated near the seismic zone, results in a small variation of average ellipse orientation for seismic intervals. Actually, the polarization ellipse rotates at 5–7° toward the N-S direction 3–4 d before and after an earthquake.

[29] The degree of linear polarization l for seismic, nonseismic and general samples is given in Figure 9 (bottom). For the time interval from −7 to 5 d it proves to be obviously higher for seismic intervals than for nonseismic ones and for the general sample. Two maxima of l are seen about several days before and after an earthquake.

[30] The observation point is characterized by moderate seismicity and weak earthquakes in the vicinity of Karimshino take place almost every day and nonseismic intervals are relatively rare. Thus the minimum in the beginning of a nonseismic curve can be of the same nature as one at seismic curve at 6–7 d after an EQ, because the majority of nonseismic intervals take place after shock series. Besides, the level of normal thunderstorm activity at Kamchatka is low [e.g., Watt, 1967, p. 486], and the effect of seismicity in ELF parameters is seen in the different results for both seismic and nonseismic intervals from the general sample.

[31] To estimate the reliability of the difference between seismic and nonseismic values of θ and l, the Fisher criterium for samples with different standard deviations is applied [Korn and Korn, 1968]. Figure 9 (right) shows the level of reliability of the hypothesis about the different mean values of seismic and nonseismic values of θ and l. P(θ) exceeds the 80% level at 3–4 and 1 d before an EQ and the picture for P(l) is similar. As the ∼15% of probability to adopt an invalid hypothesis is not negligible, we use an additional test to verify the hypothesis splitting of both samples in two of equal lengths. Thin dashed and dotted curves in Figure 9 (right) show the results for two subsamples. For seismic curves dispersion decreases in 3–4 d vicinity of an EQ and during this interval all seismic and nonseismic curves are clearly separated from each other. It is important, that the maximum of the precursor effect at 3–4 d before an EQ is reproduced in different techniques.

[32] The observed difference between ELF parameters for seismic and nonseismic intervals cannot be an artifact of random seasonal inhomogeneity of EQs and seasonal variation of the ELF parameters. Excluding of the seasonal variations does not eliminate a preseismic effect in θ and l. Thus the SPE analysis confirms the results seen from the individual EQ series. Probable false correlations caused by the existence of the “seasonal variation” in the EQ occurrence rate is also tested and proved to have a little influence on the found dependencies between the seismicity and the parameters of natural ULF/ELF signals. The probability to get this result by random coincidence is about 15% but additional indirect arguments such as reproducibility of results for two subsets of data are in favor of the adopted hypothesis.

[33] Finally, let us estimate the reliability of the effect by using a conventional approach, which was developed for estimation of the seismic precursor efficiency. We reproduce the definitions from Console [2001]:

[34] 1. Target volume Vt is a volume in 3-D space (time and 2 coordinates of the Earth surface) determined by time of observation and geographical area of observation. Each earthquake with preconditioned magnitude threshold or target event is depicted as a point in the volume Vt.

[35] 2. Alarm volume Va is a volume in which an EQ related to that precursor is expected.

[36] 3. If an EQ occurs in the alarm volume, it is called a success (S).

[37] 4. If an EQ occurs outside of alarm volume, it is a failure of predicting.

[38] 5. An alarm that is not associated to any target EQ is called a false alarm.

[39] If NS, NA and NE are the number of success, the number of alarms and the total number of EQs in the target volume then commonly considered parameters in earthquake prediction evaluation are the following:

[40] 1. Success rate = NS/NA is the rate at which precursors are followed by target events in the alarm volume. False alarm rate = 1 − NS/NA is the rate at which precursors are not followed by target events.

[41] 2. Alarm rate NS/NE is the rate at which target events are preceded by precursors.

[42] 3. Failure rate is 1 − NS/NE.

[43] 4. Probability gain PG = [NS/(NAVa)]/[NE/Vt] is the ratio between the rate at which target events occur in the alarm volume and the average rate at which target events occur over the whole target volume.

[44] Generally, a precursor can be considered as reliable if it achieves a PG value greater than one [Console, 2001].

[45] In our case of single-site observation, both the target volume and the alarm volume turn into the time intervals; hence the PG relation is

equation image

To estimate the timescale of the ΔS preseismic variation and its threshold level ΔSth, which maximizes the probability gain, we use a conventional Superposed Epoch (SPE) analysis with the dates of main shocks with KS > 1 taken as centers of time intervals. The analysis of the probability gain in dependence on the ΔSth shows that noticeable changes occur in the interval ±5 d around the EQ date and the optimal threshold value of ΔS is ΔSth = 10. The results of SPE analysis for 16 intervals with KS > 1 central earthquakes and max(ΔS(τ)) > ΔSth, where −15 < τ < 0 and τ is time in days from the EQ day, are shown in Figure 10.

Figure 10.

ΔS variation in the 15-d vicinity of 16 main shocks with KS > 1 and max(ΔS) ≥ ΔSth.

[46] Figure 10 shows that the effect is almost symmetrical on the moment of the first shock and the leading time is about 5 d. We calculate the daily averaged ΔS and use the following rules:

[47] 1. An interval with ΔS exceeding the threshold level ΔSth ≈ 10 is considered an alarm interval.

[48] 2. The duration of the alarm interval is 5 d after the start day (see Figure 9).

[49] 3. The anomalies occurred from the main shock until 5 d after it are considered as associated with the aftershock activity and are excluded from consideration.

[50] Application of these simple rules is evident from Figure 11. The results are summarized in the Table 3.

Figure 11.

Illustration of the EQ prediction technique for the interval from 22 January 2001 to 30 December 2001. (top) Earthquakes with KS > 1 (hexagonal stars), ΣKp daily indices. (bottom) ΔS. Alarm intervals are shown by markers below the horizontal axis, black squares correspond to success, open squares correspond to false alarms, and black circles indicate EQs without precursors. Anomalies just after EQ are excluded.

Table 3. Summary of ΔS Efficiency as an Earthquake Precursor for 3 a
NObservation Period Te, dNENANSSuccess RateAlarm RatePG
13439760.850.666.47
2364151470.500.472.45
3264101330.230.301.22
Total9713434160.470.472.68

4. Discussion

[51] The above analysis has shown that parameters of natural ULF/ELF emissions at Kamchatka are different for seismic and nonseismic intervals. These changes can be explained under the assumption that an additional signal appears several days prior to an This signal has nearly linear polarization and its polarization ellipse is oriented approximately meridionally. The difference between seismic and nonseismic signals is more pronounced in the polarization parameters than in the power spectra. The maximum effect is found at frequencies between Schumann resonance harmonics, especially, below the first Schumann resonance frequency. These features of seismo-related emission indicate on its generation by nearby sources.

[52] The present analysis is, in fact, the first step of a kind of iteration procedure. At this first stage, the existence of any relation between seismicity and parameters of magnetic field is studied. Thus no specific assumptions can be done about properties of this signal. Thus we work with the most common parameters such as power spectral densities and main polarization properties. However, in our specific situation the Phh/Pdd spectral ratio has been effectively used. This is due to the fact that the closest to the observation point seismic zone lies eastward from it. Actually, the same results could be obtained with invariant parameters only (see Figure 9). The parameter ΔS was chosen as the most effective among all the parameters tested. It includes Phh/Pdd ratio in the numerator. Under the present source distribution Phh/Pdd is approximately equal to the ratio Pnn/Ptt, where the index n (t) corresponds to the direction perpendicular (parallel) to the direction from the observation point to the source. At the first stage, when the details of the relation and the physical mechanisms for it are unclear, the difference between east-west direction and the direction from the observation point to the nearest seismic zone can be neglected. The denominator of ΔS increases for the intervals with unstable polarization, and thus enhanced ΔS indicates the existence of a source to the east (west) from Karimshimo, with the polarization stable at timescales of several hours. Nonseismic peaks in Phh/Pdd, cased by signals with low amplitudes and/or unstable polarization, are not seen in ΔS.

[53] The influence of the specific EQ location can be taken into account at the second step. For example, by assuming that position of the source of the seismic emission coincides with the epicenter of the corresponding EQ, the emissions with the necessary polarization can be studied. If this assumption is valid, the method should give higher seismic-to-nonseismic contrast.

[54] It is worth noting, that the rather optimistic results summarized in Table 3 were obtained with all the KS > 1 earthquakes included. At the next step, the direction to each EQ must be taken into account. If this approach improves the results, it is an additional argument in favor of the generation of ULF/ELF signals in the EQ preparation zone.

[55] The other principle limitation of the above analysis is the averaging of signal properties over the 30-min interval. This can lead to losing of short-living pulse-like signals of moderate amplitudes. However, such short-living pulses are often met. Waveforms and wavelet spectra of such preseismic signals are shown in Figure 12.

Figure 12.

Waveforms and wavelet spectra of the signal registered when ΔS is maximal at local nighttime on 13 March 2003. Top (bottom) magnetogram and spectra correspond to H(D) component, respectively. Horizontal axis show time in seconds from 0000:00 LT.

[56] Thus the other possibility to improve the efficiency of detection of the signals related to seismicity is the analysis of separate pulse-like events by using some of the detection techniques adopted to the signal typical waveform as the quasiperiodic wavelet shown in Figure 1. Anyhow, with all the limitations of present data processing, the difference in ULF/ELF parameters for seismic and nonseismic periods seems obvious.

[57] Concerning the generation mechanisms of the supposed additional emission, two candidates are usually discussed: the perturbations of the electric field induced by the preseismic ionized gas release (e.g., radon) and the atmospheric gravity waves (AGW) or the infrasound turbulence excited by sporadic water/gas eruptions or by foreshocks and aftershocks during a time interval about 2 weeks around the EQ time [e.g., Hayakawa and Molchanov, 2002]. However, taking into account the position of the perturbed region over the sea at the offshore distance about 100 km (see Figure 8), the first possibility looks less probable than the second. Using the results of hydrology/geochemistry observations and satellite remote sensing before and after EQs, Mareev et al. [2002] and Molchanov [2004] considered the excitation of the AGW turbulence with the characteristic horizontal wavelength λ ≃ 10 km. This is of the same size as the typical source area on the ground surface. They showed that the AGW energy is translated almost horizontally along the elevation angle ξ that depends on the AGW period T: ξ = arcsin(TB/T). Here TB is Brant-Vaisala period (about 5 min). For the characteristic periods 0.5–1 h the oblique propagation is typical for the AGW energy under the angles 5–7°. It corresponds to the distances 100–200 km at the heights of 10–20 km. The resulting density perturbations are small (1%) and the velocity of the energy propagation is about 1 m/s. It is a rather intriguing question: how such a small but resonant perturbation (λ is of the order of the cloud length) can trigger the observed effect, but similar speculation is beyond the paper scope.

[58] Anyhow, the future study of preseismic ULF/ELF emissions and related physical mechanisms should include the long-term continuous magnetic observations together with monitoring of lightning discharges, the atmospheric electricity, gas content and the ionospheric turbulence.

Acknowledgments

[59] This research is supported in a frame of ISTC project 2990.

Ancillary