Seismogenic Field in the Ionosphere before Two Powerful Earthquakes: Possible Magnitude and Observed Ionospheric Effects (Case Study)

Abstract

A retrospective analysis of complex geophysical data around the time of the two most powerful earthquakes that occurred in Alaska and had magnitudes M = 8.2 (29 July 2021) and M = 9.2 (28 March 1964), respectively, is carried out. The aim of the research is to assess the maximum possible magnitude of the electric field of a seismogenic nature that penetrated the ionosphere/plasmasphere, which could cause the ionospheric effects observed experimentally. Theoretical calculations have shown that under the geophysical conditions that existed before these earthquakes (favorable for the penetration of the seismogenic field into the ionosphere), the maximum value of a quasi-static electric seismogenic field in the ionosphere, perpendicular to geomagnetic field lines (tens of hours/units of days before the earthquake) for earthquakes with magnitudes M = 8–9 could reach 1–2 mV/m. Such values are sufficient for the formation of a plasmaspheric ULF-ELF-VLF-duct, which is formed in the vicinity of the geomagnetic field-line passing through the epicenter of the earthquake under the influence of a seismogenic electric field that penetrated into the ionosphere/plasmasphere. This leads to an anomalous amplification of the captured ULF-ELF-VLF waves, ULF (DC-16 Hz), ELF (6 Hz–2.2 kHz), VLF (1.8–20 kHz), not only above the epicenter of the future earthquake, but also at the point magnetically conjugated with the epicenter of the earthquake, testifying to the formation of such a duct, stretched along the geomagnetic field from one hemisphere to another, and formed on closed L-shells shortly before the earthquake. This result is confirmed by the measurements of the mission of the CSES satellite (China-Seismo-Electromagnetic Satellite) for the 29 July 2021 earthquake with magnitude M = 8.2.

1. Introduction

The maximum concentration of electrons in the ionospheric F2-layer (NmF2) experiences numerous variations and perturbations [1,2,3]. Regardless the fact that in ionospheric physics the solar activity is considered as the main source of the ionosphere variability, in recent years the strong effects on the ionosphere from below such as volcanic eruptions and dust storms were established [4,5]. Among the coupling processes from below, the strong earthquakes are also considered as perturbation sources [6]. The impact of earthquakes on the F-region of the ionosphere has been documented and reflected in many publications [7,8,9,10,11,12,13,14,15]. At the same time, the monograph [6] gives a large number of references to numerous observations (conducted before 2004) of the ionospheric perturbations preceding earthquakes (IPPE), and later examples of the IPPE [16], which could also be ionospheric precursors of earthquakes (IPE) associated with the processes of preparation of earthquakes in the Earth’s crust. Since then, hundreds of articles have been devoted to this topic, which have been published up to the present time [17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35]. However, all possible physical mechanisms that could be responsible for the perturbations in the ionosphere before earthquakes are still far from being understood. The scientific literature discusses several physical mechanisms that lead to changes in the ionosphere before earthquakes (see relevant links in the work [36]). One of the possible options is the anomalous quasi-static electric field, which is observed before an earthquake as a perturbation of the vertical component of the electric field of the atmosphere near the ground surface (E_z) in earthquake preparation zones. The magnitude of the pre-earthquake perturbation of the vertical component of the quasi-static field at the ground surface ΔE_z has the order of hundreds of V/m [37,38,39,40,41,42]. The characteristic spatial scale of the perturbation region ΔE_z (centered at the epicenter of the earthquake) can exceed hundreds [39] and thousands [42] of km. There are known examples of registration of strong electric fields in natural conditions even without earthquakes but with presence of air ionization [43]. It was demonstrated that this effect is provided by the formation of the large-scale volumetric electric charge due to the charge separation in the area of near ground electrode effect as a result of air ionization by radon [44]. Two very important details concerning the surface layer ionization by radon are described in [45]: (a) at night-time, due to a sharp temperature inversion near the Earth’s surface, a thin layer is formed, where the concentration of radon is sufficient for a negative space charge to appear at an altitude of 0.5–1 m due to inversion; (b) as the turbulent diffusion or electric field increases, the space charge becomes positive. Both of them are exactly what we regularly observe in our studies of the ionospheric precursors [16]. The nocturnal ionospheric positive anomalies observed a few days before earthquakes is explained by the radon concentration increase during the night-time temperature inversion [46] anomaly; both positive and negative anomalies were regularly observed from the very beginning of the seismo-ionospheric studies [6], so this fact is very important to establish the background mechanism of seismo-ionospheric anomaly generation. The difficulties associated with this mechanism include the need for a fairly long existence of an uncompensated charge in the near-surface layer of the Earth.

The second possible underlying mechanism for generating a quasi-static electric field before earthquakes was proposed in [47]. It has been suggested that due to activation of positive holes in igneous rocks by pre-earthquake stress the electric current brings these holes to the ground surface and creates the large-scale positive charge on the surface which creates the quasi-static electric field in the direction opposite to the natural vertical atmospheric fair-weather electric field. This possible mechanism has several disadvantages: not all seismically active regions have the igneous rocks; it does not have local time-dependence characteristics for the ionospheric precursors, and it creates only the positive electric field on the ground surface which does not permit one to explain the ionospheric anomalies of opposite polarity.

A seismogenic electric field that has penetrated into the ionosphere can affect ionospheric plasma [6,21,22,25,30,36,48,49,50]. In addition to electron concentration anomalies, perturbations were detected in the amplitudes of ultra-low-frequency (ULF), extremely-low-frequency (ELF) and very-low-frequency (VLF) electromagnetic fields in the ionosphere at altitudes from 500 to 1000 km in different satellite missions, Intercosmos-19 [51], Cosmos 1809 [52] and DEMETER (Detection of Electro-Magnetic Emissions Transmitted from Earthquake Regions) [53,54], before a series of earthquakes over their epicentral zones. The most intriguing fact is that sometimes the electromagnetic perturbations were registered not only over the seismically active regions but also over the magnetically conjugated regions [52]. The first interpretation of the conjugated effect was proposed in [55]. ELF emission at a frequency of 15 Hz was registered via DEMETER satellite [54] with amplitudes of 1.5–16 mV/m, preferably less than 10 mV/m. Recent measurements on the CSES (China-Seismo-Electromagnetic Satellite) also revealed the conjugated effect registering abnormally high magnitudes of ULF wave amplitudes above the epicenter of a future earthquake with magnitude M = 8.2 (29 July 2021 in the south of the Alaska Peninsula) and over the magnetically conjugated region as follows from the work [56].

As was demonstrated in [55], the IPPE and electromagnetic emissions in conjugated regions are parts of the seismo-ionospheric coupling chain of the processes initiated by the seismogenic electric field on the ground surface. The magnitude of the quasi-static electric field of seismogenic origin, which penetrates into the ionosphere within the earthquake preparation zone, sets the magnitude of the electrodynamic drift in the ionosphere and thereby determines the characteristic size and lifetime of plasma inhomogeneities in the ionosphere/magnetosphere associated with an impending earthquake. The purpose of this work is to estimate the possible maximum values of the electric field of seismogenic origin in the ionosphere (perpendicular to geomagnetic lines of force), before two powerful earthquakes.

Based on experimental results, the seismogenic anomaly of the quasi-static vertical electric field at the ground surface ΔE_z may exist for hours [39,40,41,42], so that the penetration of such a field into the ionosphere can cause a redistribution of ionospheric plasma over the earthquake preparation area, if its magnitude is significant enough. The characteristic spatial scale of the earthquake preparation zone on the ground surface is defined by the radial distance from the epicenter and depends on its magnitude M. In the scientific literature, there are various (empirical and theoretical) estimates of this scale (in km) given in the publications [39,57,58] (hereinafter, we will use the minimum estimate of this scale obtained in [57]), calculated using the elastic deformation on the ground surface produced by ellipsoid inclusion in the crust and its comparison with spatial distribution of the short-term precursors; the maximum distances from the epicenter correspond to the Earth’s crust deformation of order 10⁻⁸. This value can be expressed by the formula R_D = 10^0.43M. It is worth mentioning that the spatial distribution of the pre-earthquake radon activity according to the extended review of [59] follows the law [57].

In this study, we develop the model of the quasi-static electric field penetration into the ionosphere to obtain the effect in the ionosphere which conforms with the experimental measurements for two strong and mega earthquakes in the Alaska area which is also supported by the CSES measurements of electron concentration and electromagnetic emission, topside sounding results from the Alouette satellite and ground-based measurements of the vertical ionospheric sounding and GPS TEC ground-based receivers.

The study of seismo-ionospheric relationships is important both from the point of view of fundamental physics and having in mind the applied aspect of strong earthquake prediction.

2. Experimental Setup

Satellite studies of earthquake-related effects are usually accompanied by limitations of different origins: the proper orbit configuration in relation to the earthquake time and position, availability of ground-based supporting measurements, and availability of earthquakes themselves. For example, there were only six earthquakes in Alaska from 1900 with magnitude higher than 8, and the time of not all of them coincided with the existence of the proper satellite in orbit. Fortunately, for the present work we were able to pick up two Alaska earthquakes with satellite and ground-based data. The first of them with a magnitude of M = 8.2 (hereinafter referred to as EQ-1), occurred on 29 July 2021 in the south of the Alaska Peninsula near the city of Perryville. The second (hereafter EQ-2), which was the third strongest earthquake in human history with a magnitude M = 9.2, also happened in Alaska, but much earlier, on 28 March 1964 near the city of Anchorage (the so-called Good Friday earthquake). For the first earthquake, the satellite measurements were provided by the Chinese CSES (China-Seismo-Electromagnetic Satellite). CSES has a sun-synchronous orbit (02 AM–02 PM), orbital altitude 500 km, inclination ~97.32° and orbit period 94.6 min.

The Alouette-1 satellite had the high inclination orbit 80.5° which created different LT orbit conditions with the drift in the local time nearly 7 min per day. The orbit altitude of Alouette-1 was nearly 1000 km (996–1032 km), orbit period 105.5 min.

Ground-based support of the ionosphere control for the first earthquake period was exercised with GPS TEC in the form of the IONEX index provided by the International GNSS Service (IGS), while for the second earthquake we used the data of the ground-based ionosonde situated in the city of Anchorage 120 km northwest from the epicenter of the earthquake.

Emissions were registered in the night-time part of the CSES orbit (near 02 AM). An electric field detector (EFD) measures the spatial electric field in the frequency band from DC to 3.5 MHz, providing data in the four detection frequency bands: ULF (DC-16 Hz), ELF (6 Hz–2.2 kHz), VLF (1.8–20 kHz).

3. Estimation of the Electric Field in the Ionosphere before the M8.2 29 July 2021 Earthquake (EQ-1) near Perryville, Alaska, USA

The crustal earthquake with a magnitude of M = 8.2 (the depth of the hypocenter is 35 km) occurred at a distance of ~99 km southeast of the city of Perryville, Alaska, USA (south of the Alaska Peninsula). It happened on the border of the subduction zone between the Pacific and North American tectonic plates on 29 July 2021 at 06:15:49 UT (or 28 July 2021 at 21:15:49 LT). The geographical coordinates of the epicenter were as follows: φ_e = 55.36° N, λ_e = 157.89° W (magnetic inclination angle I ≅ 68° and sin I ≅ 0.93). From 25 July 2021 to 31 July 2021 the flux of solar radio emission at a wavelength of 10.7 cm varied within the range 74-82 SFU (1 Solar Flux Unit ≡ 10⁻²² W⋅m⁻²⋅Hz⁻¹), and its average monthly value nearly 81 SFU, i.e., solar activity was low.

An estimate of the radius of the earthquake preparation zone [57] gives R_D = 10^0.43M = 10^0.43×8.2 = R₀ ≅ 3300 km. Measurements of the vertical component of the electric field on the ground surface within the area of preparation of this earthquake were not carried out. However, from the GIM TEC variations over the epicenter of the earthquake (Figure 1) we can conclude that some perturbation started to develop within the time interval of 12–13 UT (1–2 h LT) 27.07.2021 (see purple arrow in Figure 1) in extremely low geomagnetic conditions which lasted almost 5 days. From 00 h UT on 23 July 2021 to 19 h UT 27 July 2021, the geophysical index magnitude was Kp ≤ 1₊, D_ST ≥ −10 nT, −3 nT ≤ IMF Bz ≤ 3 nT according to the Goddard Space Flight Center, NASA, USA (https://omniweb.gsfc.nasa.gov/form/dx1.html, accessed on 10 February 2023).

In the upper part of Figure 1, one can see the variations in the GIM TEC over the earthquake epicenter (red line), running median (green line) and upper (UB) and lower (LB) bounds (grey lines). The lower part of the figure (shadowed black and red columns) shows the deviation in the GIM TEC from the running median where the black color is negative deviation and the red color is positive deviation [56]. The GIM TEC data are provided by the IGS service (https://cddis.gsfc.nasa.gov/pub/gps/products/ionex accessed on 15 March 2023) in the IONEX format which is a matrix whose elements are TEC values multiplied by 10. Matrix resolution is 2.5° in latitude and 5° in longitude. The TEC values over the epicenter are obtained via the linear interpolation between the closest points of the IONEX table matrix.

As we do not have the direct measurements of the electric field on the ground surface, we can only speculate on its possible magnitude. This speculation is based on the measurements of seismogenic electric field for similar cases of strong earthquakes [37,38,39,40,41,42]. Assuming the existence on the ground surface of the vertical seismogenic electric field anomaly ΔE_z associated with the arisen TEC anomaly which started 1.75 days before the shock, we can estimate the possible magnitude of a horizontal (perpendicular to geomagnetic lines of force) seismogenic electric field in the ionosphere E_r^(ion) using the model proposed below.

3.1. Conductivity Model

As shown in [29], in order to calculate the penetration of a quasi-static seismogenic field into the ionosphere, it is necessary to set the altitude profile of conductivity from the Earth’s surface to the lower ionosphere (0–90 km) and the value Σ_P^av = (Σ_P + Σ_P^*)/2. Here, Σ_P is the Pedersen integral conductivity for the geographical coordinates of the epicenter of the earthquake for the corresponding date and time UT, and Σ_P^* is the Pedersen integral conductivity for the magnetically conjugated point (the details of the calculations can be found in [29]). In the calculations performed below, it is assumed that the integral Pedersen conductivity is determined via the formula ${\mathsf{\Sigma }}_P={\int }_{z_{up}}^{z_{\mathrm{\infty }}}{\sigma }_{p}dz$, where z_up = 90 km and z_∞ = 1000 km. According to estimates [36], σ_p(z) becomes a negligibly small quantity and makes no significant contribution to the integral conductivity Σ_P above the level z = 600 km.

For this geographical point and the specified time interval 27 July 2021, the values of the corresponding parameters (as an average between 12 and 13 UT) of conductivity are σ₀(z = 90 km) = 4.64910⁻⁵·S/m, σ₁(z = 90 km) = 6.90210⁻⁸·S/m, Σ_P^av = 0.16 S. These required input parameters were calculated using empirical models of the IRI-2016 ionosphere (https://ccmc.gsfc.nasa.gov/modelweb/models/iri2016_vitmo.php accessed on 15 March 2023), and neutral atmosphere NRLMSIS-00 (https://modelweb.gsfc.nasa.gov/atmos/nrlmsise00.html accessed on 15 March 2023).

Figure 2 shows this altitude conductivity model, where in the lower part of the profile (first zone, [0, z₁ = 20] km) with dashed and dash-dotted lines the lower (LB) and upper (UB) boundaries of the profile indicate the scatter of conductivity in the fair-weather conditions according to [60]. In the second [z₁ = 20; z₂ = 60] and the third [z₂ = 60; z₃ = 70] zones, the vertical conductivity profile used further is close to the model provided in [61]. In the fourth [z₃ = 70; z₄ = 90], anisotropic zone, σ₀(z) and σ₁(z) ≡ σ_p(z) are extrapolated exponentially from their values at z₃ = 70 km to their values at z₄ = 90 km (according to the geographical coordinates of the epicenter of the considered earthquake and the observation moment of the presumed seismogenic anomaly ΔE_z at the ground surface preceding the ionospheric anomaly).

3.2. Method for Calculating the Seismogenic Electric Field in the Ionosphere E_r^(ion)

For calculations of the seismogenic electrostatic field (SEP) penetration into the ionosphere, the formalism proposed in [62] is used in the approximation of vertical direction of the geomagnetic field lines (magnetic inclination I = 90°) used previously in the study of a thundercloud electrostatic field penetration into the ionosphere, and then developed in studies [36]. Due to the anisotropy of conductivity associated with the presence of the geomagnetic field, if at the ground surface level there is only a vertical component of the electric field (E_z) from the axial-symmetric localized source, then (in the cylindrical coordinate system) at the level of the ionosphere only the field component perpendicular to the geomagnetic lines of force (E_r) will be significant.

It follows from [28] (see, in particular, Formula (34) of that publication) that, when the magnetic field inclination angle is taken into account, the magnitude of the E_r component in the ionosphere will attenuate by the factor sin(I) as the field source moves on the Earth’s surface from higher latitudes to the equator, i.e., E_r calculated in the approximation of vertical geomagnetic field lines (I = 90°) must be multiplied by sin(I) to obtain its correct value.

Under steady state conditions, the governing equations are

∇·J = 0, (1)

J = σE,(2)

E = −∇Φ,(3)

where J is the electric current density, σ is the electrical conductivity tensor, E and Φ are the electrostatic field and potential, respectively. Neglecting the Earth’s curvature, and using cylindrical coordinates (r, ϕ, z) centered at a forthcoming earthquake epicenter and with the z axis pointing vertically upward, we represent the seismogenic perturbation in the vertical atmospheric field near the Earth’s surface by the spatial distribution, in the form obtained in [29]:

ΔE_z ≡ E_z = E₀ exp{−([1 + (r/R₀)²]^1/2−1)/(R_E/R₀)}/[1 + (r/R₀)²]^1/2, (4)

where E₀ and R₀ are, respectively, the maximum perturbation value of ΔE_z in the future epicenter of the impending earthquake and the radius of the earthquake preparation zone on the ground surface, and R_E determines the scale of field decrease as it moves away from the epicenter. If one assumes that the geomagnetic field B is vertical, and the electrical conductivity tensor depends only on z, the following equation for the electrostatic potential Φ can be obtained from (1), (2) and (3)

(5)${\partial }^2\mathsf{\Phi }/{\partial }^{2}z+(1/r)\partial \mathsf{\Phi }/\partial r+(1/{\sigma }_p)\partial ({\sigma }_0\partial \mathsf{\Phi }/\partial z)/\partial z=0,$

where σ_p is the Pedersen conductivity, and σ₀ is the specific conductivity. At altitudes below 70 km, the conductivity is isotropic (σ₀ = σ_p) since the geomagnetic field does not affect drifts of charged particles. Above 70 km, the presence of the geomagnetic field results in the anisotropy of the conductivity (σ₀ ≠ σ_p). Equation (5) can be solved analytically if the conductivities σ₀ and σ_p depend exponentially on altitude.

The electric field components are given by

E_r = −∂Φ/∂r, (6)

E_z = −∂Φ/∂z. (7)

Since we assume that the geomagnetic field B is vertical, E_r is perpendicular to B, while E_z is parallel to B.

Above 90 km, the geomagnetic field-aligned conductivity σ₀ is sufficiently high, it is much larger than the transverse conductivity σ_p, so the geomagnetic field lines of force are nearly equipotential lines for the case of perpendicular electrostatic fields with scale sizes of more than a few tens of kilometers. This makes it possible to consider the ionospheric region from ~90 to ~600 km as a thin conducting layer with a geomagnetic field line-integrated Pedersen conductivity Σ_P (note that the local conductivity σ_p is negligible above 600 km). Thus, the continuity equation of the electric current can be written at z = 90 km in the following form:

σ₀·E_z = ∇_⊥·(2Σ_P^av E_⊥), (8)

where ∇_⊥ denotes the two-dimensional gradient operator B (in a plane perpendicular to B); the factor 2 before Σ_P^av accounts for a contribution of the Pedersen conductivity of the magnetically conjugate ionosphere. Equation (10) is explicitly expressed as

σ₀·∂Φ/∂z = 2Σ_P^av [∂²Φ/∂r² + (1/r)·∂Φ/∂r]. (9)

Relations (4) and (9) represent the lower and upper boundary conditions, respectively, to evaluate the electrostatic potential Φ.

3.3. Discussion of the Quasi-Static Seismogenic Field near the Earth’s Surface Source Model

As mentioned above, the distribution of the perturbation of the vertical component of the electric field ΔE_z on the ground surface as a function of the epicentral distance r is taken in the form (4):

ΔE_z ≡ E_z = E₀ exp{−([1 + (r/R₀)²]^1/2−1)/(R_E/R₀)}/[1 + (r/R₀)²]^1/2, (10)

This distribution fits well with long-term measurements of seismogenic anomalies ΔE_z before earthquakes in their epicentral regions, according to studies presented in [39]. From study [29], it follows that the value of E₀ ≅ 1 kV/m at the ground surface, according to the measurements of [39], is achieved already before earthquakes with M within the range 4.1–4.7, and before an earthquake with M = 6.0, the value of E₀ is ≅ 1.3–2.4 kV/m, while the value of the ratio R_E/R₀ lies within the interval [0.6–1.75]. Moreover, the study [42] described a seismogenic anomaly ΔE_z on the ground surface, which appeared ~14 h before the earthquake of 12 June 2010 in the area of the Nicobar Islands (India) with a magnitude of M = 7.5 according to the updated data of the USGS catalog (https://earthquake.usgs.gov/earthquakes/browse/significant.php?year=2010 accessed on 15 March 2023), which existed for ~41 min. It had an amplitude of ΔE_z = 1385 V/m at an epicentral distance R_e = 1763 km. An estimate of the value of E₀ for this earthquake (at its epicenter) made in [29] gives E₀ ≅ 2.9–3.9 kV/m. In Figure 3, a solid line shows the normalized distribution of E_z/E₀ on the ground surface depending on r for the case we are considering, when R₀ = R_E = R_D ≅ 3350 km according to Formula (4). The dash-dotted line shows the corresponding distribution of the seismogenic field on the ground surface, which was used [36] according to the formula ΔE_z = E₀ exp[−(r/R₀)² ln10], derived from general considerations, for the same values of R₀.

It follows from the figure that at a linear distance r = R₀, the formula for the source of the seismogenic field, derived from general considerations and used in [36] by ≅ 4.68 times underestimates the value of the field compared to Formula (4). This fact is of significant importance in calculating the magnitude of the electric field penetrating from the Earth’s surface to the heights of the ionosphere.

3.4. The Expected Pattern of Distribution of the Seismogenic Field E_r^(ion) Penetrated into the Ionosphere Depending on the Epicentral Distance before the EQ-1 Earthquake

The calculation of the distribution of the horizontal (perpendicular to geomagnetic field lines) component of the electric field in the ionosphere E_r^(ion) is performed in accordance with the formalism presented in detail in [29] in the cylindrical coordinate system (the base of the vertical z axis placed at the epicenter of the earthquake on the ground surface) and the input parameters described above. The results of the calculations for I = 90 are illustrated in Figure 4, which shows the ratio E_r^(ion)/E₀ as a function of the epicentral distance r for low (LB, dash-dotted line), medium (AV, solid line) and high (UB, dashed line) conductivity values σ₀ (z = 0) at the ground surface according to the model [60].

From the figure, it can be seen that the ratio E_r^(ion)/E₀ reaches a maximum value at a distance r_max ≅ 5100 km and is equal to 1.92 × 10⁻⁷ (LB), 3.85 × 10⁻⁷ (AV) and 7.7 × 10⁻⁷ (UB), respectively.

Let us now estimate the maximum value of the expected values of E_r^(ion)max in the ionosphere due to a possible seismogenic anomaly of the quasi-static electric field ΔE_z on the ground surface, which may take place ~42 h or 1.75 days before EQ-1. Let us choose the lower boundaries of values E₀ as reference points. For earthquakes with magnitudes M within the range 4.1–7. 5 mentioned above in Section 3.3, we obtain 1 kV/m, 1.3 kV/m and 2.9 kV/m, respectively. Then the average value of ⟨E₀⟩ ≅ 1.7 kV/m, and for average (AV) conductivity values σ₀ (z = 0) at the ground surface, taking into account the factor sin(I) ≅ sin (68°) ≅ 0.93, we obtain that E_r^(ion)max = sin (68°) × ⟨E₀⟩ × 3.85 × 10⁻⁷ × 1000 [mV/m] = 0.93 × 1700 V/m × 3.85 × 10⁻⁷ × 1000 ≅ 0.6 mV/m AV ( ≅ 0.3 mV/m and ≅ 1.2 mV/m for LB and UB, respectively).

Figure 5 shows the latitudinal distribution of the module velocity electrodynamic drift of plasma V = |V_E×B| = |E × B/B²| (m/s) at the geographical longitude of the epicenter of the earthquake EQ-1 (λ_e = 157.89° W ≅ 202° E) at altitude h = 300 km for |E_⊥| ≅ E_r^(ion) = 1 mV/m (solid line) and 0.6 mV/m (dashed line). Here, E_⊥ is perpendicular to the geomagnetic field line component of the electric field (for calculations, we use the International Geomagnetic Reference Field (IGRF-13) model (https://ccmc.gsfc.nasa.gov/modelweb/models/igrf_vitmo.php accessed on 15 March 2023)).

From the figure, it is inferred that at all latitudes the magnitudes of the velocity V for E_r^(ion) = 0.6 mV/m exceed 11 m/s, which is sufficient to create significant effects in the ionosphere.

3.5. Discussion of the Results Obtained for the Earthquake EQ-1 with the Involvement of Additional Data of the CSES Mission (China-Seismo-Electromagnetic Satellite)

Let us sum up some results. The above calculations show that before the EQ-1 earthquake in the night-time ionospheric conditions, 1.75 days before the earthquake, under the low geomagnetic and solar activity, an electric field of a seismogenic nature could be observed. The maximum magnitude of the electric field component perpendicular to magnetic field lines is |E_⊥| max and is of order 0.3–1.2 mV/m while the mean maximum value |E_⊥|_max ≅ 0.6 mV/m at the epicentral distance r_max ≅ 5100 km.

Let us consider now the data of the ULF-band electromagnetic field measurements (from DC to 15 Hz) onboard the CSES mission especially for data taken during the flyby times and trajectories of this satellite passing over the epicentral zone of the EQ-1 earthquake and area magnetically conjugated to it [56]. The authors claim that “An abnormal ULF wave appeared 10 and 2 days before the mainshock”, i.e., anomalous ULF-band electromagnetic emissions were observed on 27 July 2021 simultaneously with anomalies in TEC, the starting moment of their development was laid into the basis for our calculations of the seismogenic electric field penetration into the ionosphere.

Looking carefully at Figure 6 it can be seen that the abnormally high amplitudes of ULF-waves of the specified observation range are observed within latitudinal range ≅± 3–3.5° from the latitudinal location of the EQ-1 earthquake epicenter and at the part of the satellite trajectory magnetically conjugated to the epicenter 2 days before the EQ-1 earthquake. The effect of increasing amplitude in ELF-(10–30 Hz), SLF-(30–300 Hz) and ULF-(300 Hz–3 kHz), VLF-(3–30 kHz) bands over the epicentral zones of subsequent earthquakes is known from much earlier experimental works [63,64]. The very first description of simultaneous observation of VLF emissions and ionospheric critical frequency foF2 anomalies in magnetically conjugated regions registered onboard the Intercosmos-19 satellite before the Irpinia M6.3 earthquake in Italy was published in [51]. The physical interpretation of such effects is given in [6,55,65,66]. Very briefly, it consists in the fact that the amplification of ELF and VLF emissions occur as a result of their capture in the plasmaspheric ELF-VLF-duct stretched along the geomagnetic field (from one hemisphere to the other), which is formed in the vicinity of a geomagnetic field line passing through the epicenter of an impending earthquake under the influence of a seismogenic electric field that penetrated the ionosphere/plasmasphere. Thus, the presence of amplitudes of ULF-ELF-VLF waves at a magnetically conjugated point with the epicenter of the earthquake clearly indicates the formation of such ducts emerged on closed L-shells (for the epicenter of the EQ-1 earthquake, located at fairly high geographical latitudes; however, L_e(EQ-1) ≅ 2.9–3.0). The fact established by the CSES satellite data is an experimental confirmation of the assumption put forward at the beginning of Section 3 that on 27 July 2021 in the time interval of 12–13 h UT (for ~42 h or 1.75 days before the earthquake) in the epicentral zone of the EQ-1 earthquake there was a seismogenic electrical anomaly of the vertical electric field at the ground surface ΔE_z. The characteristics of a “seismogenic” duct (under specific conditions indicated in [61]) are described in detail, and at altitudes of ~2000 km, the size of such a “duct” in the zonal direction (west–east) can be ~2000 km, whereas in the meridional direction (pole–equator) it can extend for ~600 km and is centered on a geomagnetic field line passing through the epicenter of the earthquake. The maximum heterogeneity of the electron concentration relative to its central point in the perpendicular cross-section of the magnetic power tube is approximately 15% for conditions of low solar activity.

4. Estimation of the Electric Field in the Ionosphere before the M9.2 28 March 1964 Earthquake (EQ-2) near Anchorage, Alaska, USA

A catastrophic earthquake EQ-2 with magnitude M = 9.2 according to updated on 07 November 2016 data (Prince William Sound Earthquake or The Great Alaska Earthquake or The Good Friday Earthquake, hypocenter depth 25 km) occurred on 28 March 1964 at 03:36:16 h UT (or 27 March 1964 at 17:36:16 h LT) near Anchorage, Alaska, USA, located about 120 km northwest of the epicenter. The geographical coordinates of the epicenter are φ_e = 60.91° N, λ_e = 147.34° W (magnetic inclination angle I ≅ 74° and sin I ≅ 0.96). From 27 March 1964 to 28 March 1964, the flux of solar radiation at the wavelength 10.7 cm was within the range 74–78 SFU and solar activity was very low. The estimate of the minimum radius of the earthquake preparation zone according to Dobrovolsky [57] gives R_D = 10^0.43M = 10^0.43×9.2 = R₀ ≅ 9000 km.

It should be noted here that the investigation of the ionospheric perturbations preceding earthquakes (IPPE) apparently began with a pioneering work [67] devoted to the study of the ionospheric effects associated with this particular earthquake. In this work, in particular, it was shown that at the Boulder ISVS, distanced in the arc of the great circle at ~3700 km from the epicenter of the earthquake, well-defined disturbances in the ionosphere were clearly recorded. They were observed as simultaneously existing inhomogeneities in the regions E (sporadic layer Es) and F (F-scattering) of the ionosphere for about two hours before the shock. Note here that the sequence of planetary Kp-indices on the day preceding the earthquake and on the day when the earthquake occurred was as follows: Kp (27.03.1964) = {1₊; 0₀; 2₀; 0₊; 0₊; 1_-; 0₊; 1₀} and, respectively, Kp (28.03.1964) = {0₀; 0₀; 0₊; 0₀; 0₀; 0₊; 0₀; 0₀}, i.e., the geomagnetic situation was very quiet. A more detailed analysis of the variations of one of the most important, regularly measured parameters at the ISVS, the critical frequency of F2 layer of the ionosphere (foF2), a few hours before this earthquake was carried out much later, in [68] with the data of eight ISVSs. At the same time, it was found that at least a few hours before the earthquake against a quiet background of heliogeomagnetic activity (two days before the earthquake, the average daily number of sunspots was about 15, and the average value of the AE-index was ~30 nT) specific regional disturbances appeared in the ionosphere both near the epicenter of the anticipated earthquake (~100 km) and at a considerable distance from it (~1000–1500 km), apparently being its precursors. Figure 7 shows the geographical location of all the ISVSs concerning which data are used to further investigate the situation related to the EQ-2 earthquake (the epicenter of the earthquake is marked with an asterisk and almost coincides with the position of the ISVS Anchorage, and the Adak and Boulder ISVS are marked with black squares).

Figure 7 clearly shows the large characteristic spatial scales of physical processes occurring in the zone of preparation of the catastrophic earthquake EQ-2 with a magnitude of M = 9.2, which is due to its enormous power.

Careful study of Figure 8 allows one to find that in the time interval 11–12 h UT (03–04 h LT, night) on 27.03.1964 (it is ~15.5–16.5 h before the earthquake) abnormally low values of foF2 are observed on the ISVS Anchorage compared to the median.

It can be seen from the figure that at the Adak ISVS a few hours before sunrise until the end of the local night the critical frequency exceeds the upper limit foF2_med + 0.75IQR (see Figure 8a, shaded area); regardless, it is significantly distant from the epicenter of the earthquake by ~2090 km along the great circle arc. Nevertheless, it is still inside the zone of preparation of this earthquake. Note that the local time in Anchorage is “ahead” of local time in Adak by ~2 h (LT_Adak ≅ LT_Ancorage −2 h), and Anchorage is located higher in latitude by ~9°.

Contrary to the Adak ISVS, at Boulder ISVS at the same time interval the negative deviation of foF2 is observed beyond the lower boundary of foF2_med −0.75IQR (see Figure 8b). Recall here that the Adak ISVS is distanced from the epicenter of the EQ-2 earthquake by ~2090 km, and the Boulder ISVS by ~3700 km along the arc of a large circle, but the maximum deviation from the median in the specified time interval for both ISVSs is ≈20%.

To explain the difference in the critical frequency deviation sign, we should consider the longitudinal difference between the ionospheric station locations. Adak is located ~26.5 degrees westward from Anchorage, while Boulder is ~44.6 degrees eastward from Anchorage. Taking into account the differences in the local time −2 h for Adak and +3 h for Boulder in relation to Anchorage, and established in [70] dependence of the ionospheric precursors on the local time this difference looks quite natural.

Finalizing the demonstration of the experimental data for the case of the Good Friday earthquake (EQ-2), we have an exceptional lack of data concerning the topside ionosphere monitoring from onboard the Alouette-1 satellite which was in orbit and passed over the EQ-2 preparation zone during day-time hours local time [69]. The synthetic differential maps (composed from several satellite passes) of the critical frequency foF2 are shown in the left panel of Figure 9. They were collected for the time interval when the positive anomalies were observed on the vertical sounding ionograms of the Adak ISVS (Figure 8a). The most striking fact was that for the EQ-1 of 2021 a similar positive anomaly was registered as well on the GIM differential TEC MAP on 27.07.2021 (Figure 9, right panel); the description of the procedure for obtaining such a card is detailed in work [71]. This permits us to be more confident that the observed anomalies are formed under the action of the same physical mechanism: anomalous electrostatic electric field penetrating into ionosphere/plasmasphere. This is also confirmed by estimation of this field for the EQ-2 provided in Section 4.3.

Suppose here (as well as earlier in Section 2) that at 11–12 h UT (03–04 h LT, night near the epicentral point close to the ISVS Anchorage) on 27 March 1964, i.e., 15.5–16.5 h before the earthquake, there was a seismogenic anomaly of the vertical electric field ΔE_z at the ground surface. Let us then make the same assessment of a possible seismogenic electric field in the ionosphere according to the scheme described above in Section 2, omitting the details.

4.1. Conductivity Model

Figure 10 demonstrates the conductivity model used to calculate the magnitude of the seismogenic electric field in the ionosphere (similar to Figure 2) for the time of 11:00 h UT on 27 March 1964 and the geographical point of the epicenter of the EQ-2 earthquake (the Pedersen integral conductivity Σ_P^* ≅ 0.08 S is also calculated for the magnetically conjugated point). For the specified geodetic points and the specified time of 27.03.1964, the values of the corresponding conductivity parameters σ₀(z = 90 km) = 4.38 × 10⁻⁵ S/m, σ₁(z = 90 km) = 5.83 × 10⁻⁸ S/m, Σ_P≅ 0.07 S and Σ _P^av = 0.075 S, taking into account that Σ_P^av = (Σ_P + Σ_P^*)/2. Note here that, despite the higher latitude of the epicenter of the EQ-2 earthquake compared to the EQ-1 earthquake (ϕ_e{EQ-2} = 60.91° N > ϕ_e {EQ-1} = 55.36° N), it remains within the range of closed L-shells, because L_e(EQ-2) ≅ 4.35 (the position of plasmapause is usually about L = 5).

4.2. Model of the Source of a Quasi-Static Seismogenic Electric Field in the Atmosphere near the Ground Surface

The distribution of the perturbation of the vertical component of the electric field ΔE_z on the surface as a function of the epicentral distance r (assuming the cylindrical symmetry of the field source relative to the vertical axis passing through the epicenter of the earthquake) is taken (as in Section 3.3) in accordance with Formula (4). In this case, R₀ = R_D(M = 9.2) ≅ 9000 km is the radius of the earthquake preparation zone on the ground surface. The value of R_E (the scale of field decrease in direction from the epicenter) will be determined on the basis of its possible lower boundary (see Section 3.2), based on the ratio R_E /R₀ = 0.6, i.e., R_E (EQ-2) ≅ 5400 km.

4.3. Expected Pattern of Distribution of the Seismogenic field E_r^(ion) Penetrated into the Ionosphere Depending on the Epicentral Distance before the EQ-2 Earthquake

The results of calculations of the horizontal (perpendicular to geomagnetic field lines) component of the seismogenic electric field in the ionosphere E_r^(ion) for I = 90° illustrates Figure 11. The ratio of E_r^(ion)/E₀ (similarly to Figure 3) is provided as dependence on the epicentral distance r for low (LB, dashed line), medium (AV, solid line) and high (UB, dashed line) conductivity values σ₀ (z = 0) at the ground surface according to the model [60].

From the figure, it can be seen that the ratio E_r^(ion)/E₀ reaches a maximum value at a distance r_max ≅ 10,700 km and is equal to 9.15 × 10⁻⁷ (LB), 1.83 × 10⁻⁶ (AV) and 3.66 × 10⁻⁶ (UB), respectively.

Let us now estimate the maximum value of the expected values of E_r^(ion)max in the ionosphere due to a possible seismogenic anomaly of the quasi-static electric field ΔE_z on the ground surface, which may have occurred ~16 h before the earthquake EQ-2. Let us choose as a reference the lowest limit of the values of E₀, for earthquakes with magnitudes M from 4.1 to 7. 5 referred to in Section 3.3 above, i.e., E₀^min = 1 kV/m. Then for the average (AV) values of conductivity σ₀(z = 0) at the ground surface, taking into account the factor sin I ≅ sin 74° ≅ 0.96, we get that E_r^(ion)max = sin 74° × E₀^min× 1.8 3 × 10⁻⁶ × 1000 [mV/m] = 0.96 × 1000 V/m × 1.83 × 10⁻⁶ (AV) × 1000 ≅ 1.76 mV/m (≅ 0. 88 mV/m and ≅ 3. 51 mV/m for LB and UB, respectively).

Thus, for an EQ-2 earthquake in the ionosphere, |E_⊥| _max∈ [0.88, 3.51] mV/m with an average value of ≅ 1.76 mV/m at an epicentral distance of ≅ 10,700 km. It turns out that at low and medium conductivity values σ₀(z = 0) at the ground surface according to the [55] maximum value of |E_⊥|_max ≈ 1–2 mV/m, which is due to the enormous power of this catastrophic earthquake (which had a very large spatial dimension of the preparation zone) and the very favorable geophysical conditions for the penetration of the seismogenic field (deep night and very low solar activity) into the ionosphere.

5. Discussion and Conclusions

The presented model is only part (but a very important part) of the Lithosphere–Atmosphere–Ionosphere Coupling (LAIC) model [72]. As it is presented in the cited paper, the model considers the chain of several coupling interfaces: lithosphere–atmosphere (crust deformation controlled by b-parameter and connection with the radon emanation), geochemical interface considering the physics of generation of the thermal anomalies in the atmosphere, and the electromagnetic interface where the near-ground modification of the global electric circuit is mapped at the ionosphere and the magnetosphere. Among the three proposed options of electromagnetic interface, the seismogenic electric field penetration into the ionosphere is described in the present publication. This approach does not use the fantastic values of external fields to obtain the effect in the ionosphere which is criticized in [28], but also does not fall into the pessimism advocated by the authors of [28], claiming that electric field penetration in the ionosphere is impossible. We try to conform the theoretical estimations with the results of real experimental measurements of the pre-earthquake electric field in the ionosphere [73] where the experimentally obtained magnitude of the electric field in the ionosphere exactly coincides with our estimations: 0.91 mV/m.

From another side, we made maximum efforts to obtain as many experimental measurements as possible to formulate the initial and boundary conditions and to consider the pre-earthquake effects in the ionosphere obtained via four different techniques of ionosphere monitoring.

The results of the model calculations show that for an earthquake EQ-1 (M = 8.2), the maximum value of a possible quasi-static seismogenic field |E_⊥|_max(EQ-1) in the nocturnal ionosphere 1.75 days before the earthquake against the background of low geomagnetic and solar activity lies in the range 0.3–1.2 mV/m with an average maximum value of |E_⊥|_max ≅ 0.6 mV/m at the epicentral distance r_max ≅ 5100 km.

Similarly, for an earthquake of catastrophic strength EQ-2 (M = 9.2), the maximum value of a possible quasi-static seismogenic field |E_⊥|_max (EQ-2) in the night ionosphere ~16 h before the shock against the background of low geomagnetic and solar activity lies in the range of 0.88–3.51 mV/m with an average maximum value of |E_⊥|_max ≅ 1.76 mV/m at the epicentral distance r_max ≅ 10,700 km.

Summarizing, we can say that for the magnitude of earthquakes M = 8–9 under favorable geophysical conditions (night, low solar activity), the maximum value of the seismogenic quasi-static electric field in the ionosphere |E_⊥|_max tens of hours/units of the day before the earthquake, according to our model estimates, can be 1–2 mV/m. The similarity of the ionospheric effects for both earthquakes was demonstrated via differential ionospheric maps of the critical frequency (for EQ-2) and GIM GPS (for EQ-1).

It should be noted here that the case study we conducted before two powerful earthquakes relies primarily on the study of ionospheric data. However, for the EQ-1 earthquake, the involvement of additional data from the CSES mission made it possible to clearly establish the presence of a geomagnetic field stretched along (from one hemisphere to another) the plasmaspheric ULF-ELF-VLF-duct, which was formed in the vicinity of a geomagnetic field line passing through the epicenter of this earthquake under the influence of a seismogenic electric field that penetrated the ionosphere/plasmasphere. As a result, an anomalous increase of ULF-ELF-VLF wave amplitude was recorded not only over the epicenter of the future earthquake, but also at a point magnetically conjugated with the epicenter position, clearly indicating the formation of such a ductus formed on closed L-shells about two days before the earthquake (27 July 2021).

Conclusions

• Based on the analysis of ionospheric data obtained before the powerful earthquakes EQ-1 (M = 8.2) and EQ-2 (M = 9.2), and model representations of the source of the seismogenic quasi-static electric field on the ground surface obtained in the work [Khegai, 2020], estimates of the maximum value of a possible quasi-static seismogenic field perpendicular to geomagnetic lines (|E_⊥|_max) penetrating into the night-time ionosphere from the earth’s surface (with favorable geophysical conditions - low solar and geomagnetic activity) were made.

• It turned out that under favorable geophysical conditions, the value of |E_⊥| _max (tens of hours/units of the day before the earthquake) for earthquakes with magnitudes M = 8–9 can be 1–2 mV/m. Such values are sufficient to form the plasma-bearing ULF- ELF- VLF-duct, emerging in the vicinity of the geomagnetic field line passing through the epicenter of the earthquake under the influence of a seismogenic electric field that penetrates into the ionosphere/plasmasphere. This leads to an anomalous increase in the ULF-ELF-VLF wave amplitudes captured in the duct, not only over the epicenter of a future earthquake, but also at the point magnetically conjugated to the epicenter of the earthquake, indicating the formation of such a duct, stretching along the geomagnetic field from one hemisphere to another, and emerging at closed L-shells shortly before the earthquake.

• The key parameters determining the magnitude of the seismogenic field in the ionosphere are as follows: E₀—the magnitude of the seismogenic anomaly at the epicenter of its future earthquake before the earthquake; R₀—the radius of the earthquake preparation zone on the ground surface; R_E—the scale of field decreases as it moves away from the epicenter; σ₀(z = 0 km) is the value of conductivity at the ground surface; σ₀(z = 90 km) is the value of conductivity at an altitude of 90 km; σ₁ (z = 90 km) is the value of the local conductivity of Pedersen at an altitude of 90 km; and Σ _P^av = (Σ_P + Σ_P^* )/2 is the average value of Pedersen’s integral conductivity for two hemispheres. Here, Σ_P is the Pedersen integral conductivity for the geographical coordinates of the epicenter of the earthquake for the corresponding date and time UT, and Σ_P^* is the Pedersen integral conductivity for the magnetically conjugate point.

Author Contributions

V.H. developed the idea of manuscript, has provided the model development and calculations, and experimental data of vertical sounding for the Good Friday earthquake, Z.Z. provided and processed the data from the CSES satellite and calculation of the GIM TEC data, S.P. contributed to the methodology, provided the data of Alouette topside sounding, draft preparation and final corection. All authors have read and agreed to the published version of the manuscript.

Institutional Review Board Statement

Not applicable.

Informed Consent Statement

Not applicable.

Data Availability Statement

Data are available upon request to the corresponding author.

Acknowledgments

The authors express their gratitude to NOAA’s National Geophysical Data Center (NGDC) USA; NASA’s Goddard Space Flight Center, USA; UK Solar System Data Centre (UKSSDC), England; World Data Center for Geomagnetism, Kyoto, Japan, whose solar, ionospheric and magnetic data were used in this work, as well as the United States Geological Survey’s (USGS) Earthquake Hazards Program, USA for providing access to earthquake data.

Conflicts of Interest

The authors declare no conflict of interest.

Footnotes

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Figures

Figure 1. GIM TEC over the epicenter of EQ-1. A purple arrow from below marks the beginning of the development of disturbances in TEC, 1.75 days before the earthquake.

Figure 2. Model piecewise continuous vertical profiles of nocturnal conductions σ0 and σ1 from 0 to 90 km at the point of geographical coordinates of the epicenter of the earthquake EQ-1{29 July 2021 with M = 8.2} for 01–02 LT 27 July 2021. The numbers of the corresponding line segments indicate the characteristic scale (in km) of changes in exponentially increasing conductivity, and dashed horizontal lines mark the boundaries of high-altitude zones. The dashed and dash-dot lines indicate the lower (LB; σ0(z = 0) = 0.5 × 10–13 S/m) and upper (UB; σ0(z = 0) = 2.0 × 10–13 S/m) boundaries of the conductivity scatter boundary in the fair-weather region in the first zone according to [60], respectively. Solid line (AV; σ0(z = 0) = 1.0 × 10–13 S/m) shows the average conductivity distribution by altitude. The average value of the Pedersen integral conductivity for the two hemispheres ΣPav = 0.16 S is also shown in the figure.

Figure 3. The normalized distribution of Ez/E0 on the ground surface of the seismogenic field source depending on the epicentral distance r (solid line) when R0 = RE = RD ≅ 3350 km according to Formula (4) and according to the formula ΔEz = E0 exp[−(r/R0)2 ln10] for the same value of R0 used in [34] and obtained from general considerations. The arrow at the abscissa axis notes a characteristic scale of R0 = 3350 km.

Figure 4. The ratio of Er(ion)/E0 depending on the epicentral distance r for lower bound (LB, dash-dotted line), average (AV, solid line) and upper bound (UB, dashed line) conductivity values σ0(z = 0) at the ground surface according to the model [60]. A continuous vertical line with an arrow at the abscissa axis marks the distance rmax ≅ 5100 km at which the maximum value of the ratio Er(ion)/E0 is reached.

Figure 5. Latitudinal distribution of the electrodynamic plasma drift velocity module V = |VE × B| = |E × B/B2| at the geographical longitude of the epicenter of the earthquake EQ-1 at altitude h = 300 km for values of | E⊥| ≅ Er(ion) = 1 mV/m (solid line) and 0.6 mV/m (dashed line).

Figure 6. Yellow “squares” indicate abnormally high values of the amplitudes of ULF waves above the epicenter and at a magnetically conjugated point on the trajectory of the CSES satellite (sun-synchronous orbit, orbital height of 500 km, inclination of ~97.32°, period of 94.6 min) on 27 July 2021 two days before the ~EQ-1 earthquake. It is also visible that on 19 July 2021 the corresponding anomaly at the magnetically conjugated point is weakly expressed.

Figure 7. The geographical location of the ISVS Adak and Boulder and the position of the epicenter of the earthquake EQ-2. The epicenter of the earthquake is marked with an asterisk and almost coincides with the position of the ISVS Anchorage; other ISVSs are marked with black squares.

Figure 8. Critical frequency variations for the time interval 27–28 March 1964. (a) Adak (thick line) and Anchorage (open circles and dots); (b) Boulder (thick line) and Anchorage (open circles and dots). Thin solid lines show the corresponding monthly medians foF2med, and the dashed ones show the boundaries foF2med ± 0.75∙IQR (here IQR is the interquartile range). The thin long vertical lines with arrows indicate the moment of the EQ-2 earthquake, and the vertical dashed line with an arrow ending marks time 11:00 h UT on 27 March 1964, when a significant decrease in the current foF2 values starts on the records of Anchorage ISVS at local night hours (marked with ellipses on both panels). Solid dark thick bars under the abscissa axes mark local time intervals from 18 to 6 h LT (night-time). Breaks on the data curves mean an absence of data. Short arrows at the abscissa axis of the panel (b) mark the time when in [69] anomalies were reported in the ionosphere at Boulder ISVS: F-spread (arrow from above) and a powerful sporadic layer Es (arrow from below). The intervals when current values of foF2 exceed the boundaries of foF2med ± 0.75∙IQR are shaded [68].

Figure 9. Left panel—differential LT map of the ∆foF2 (%) distribution based on Alouette-1 topside sounder data [69] for day-time passes of the satellite on 27 March 1964. Black circle—EQ-2 epicenter location; right panel—differential GIM TEC map for local day-time hours on 27 July 2021. Asterisk corresponds to the EQ-1 epicenter location. White line—Dobrovolsky earthquake preparation zone boundary.

Figure 10. Same as Figure 2, but for the EQ-2 earthquake conditions.

Figure 11. Same as Figure 4, but for EQ-2 conditions.

References

1. Rishbeth, H. Day-to-day ionospheric variations in a period of high solar activity. J. Atmos. Terr. Phys.; 1993; 55, pp. 165-171. [DOI: https://dx.doi.org/10.1016/0021-9169(93)90121-E]

2. Rishbeth, H.; Mendillo, M. Patterns of F2-layer variability. J. Atmos. Sol.-Terr. Phys.; 2001; 63, pp. 1661-1680. [DOI: https://dx.doi.org/10.1016/S1364-6826(01)00036-0]

3. Kim, V.P.; Hegai, V.V. On the variability of the ionospheric F2-layer during the quietest days in December 2009. J. Astron. Space Sci.; 2016; 33, pp. 273-278. [DOI: https://dx.doi.org/10.5140/JASS.2016.33.4.273]

4. Tramutoli, V.; Marchese, F.; Falconieri, A.; Filizzola, C.; Genzano, N.; Hattori, K.; Lisi, M.; Liu, J.-Y.; Ouzounov, D.; Parrot, M. et al. Tropospheric and Ionospheric Anomalies Induced by Volcanic and Saharan Dust Events as Part of Geosphere Interaction Phenomena. Geosciences; 2019; 9, 177. [DOI: https://dx.doi.org/10.3390/geosciences9040177]

5. Pulinets, S.A.; Davidenko, D.V.; Pulinets, M.S. Atmosphere-ionosphere coupling induced by volcanoes eruption and dust storms and role of GEC as the agent of geospheres interaction. Adv. Space Res.; 2022; 69, pp. 4319-4334. [DOI: https://dx.doi.org/10.1016/j.asr.2022.03.031]

6. Pulinets, S.A.; Boyarchuk, K.A. Ionospheric Precursors of Earthquakes; Springer: Berlin/Heidelberg, Germany, 2004.

7. Leonard, R.S.; Barnes, R.A., Jr. Observation of ionospheric disturbances following the Alaska earthquake. J. Geophys. Res.; 1965; 70, pp. 1250-1253. [DOI: https://dx.doi.org/10.1029/JZ070i005p01250]

8. Row, R.V. Evidence of long-period acoustic-gravity waves launched into the F region by the Alaskan earthquake of March 28, 1964. J. Geophys. Res.; 1966; 71, pp. 343-345. [DOI: https://dx.doi.org/10.1029/JZ071i001p00343]

9. Calais, E.; Minster, J.B. GPS detection of ionospheric perturbations following the January 17, 1994, Northridge earth-quake. Geophys. Res. Lett.; 1995; 22, pp. 1045-1048. [DOI: https://dx.doi.org/10.1029/95GL00168]

10. Ducic, V.; Artru, J.; Lognonné, P. Ionospheric remote sensing of the Denali earthquake Rayleigh surface waves. Geophys. Res. Lett.; 2003; 30, SDE 8-1–8-4 [DOI: https://dx.doi.org/10.1029/2003GL017812]

11. Artru, J.; Farges, T.; Lognonné, P. Acoustic waves generated from seismic surface waves: Propagation properties determined from Doppler sounding observations and normal-mode modelling. Geophys. J. Inter.; 2004; 158, pp. 1067-1077. [DOI: https://dx.doi.org/10.1111/j.1365-246X.2004.02377.x]

12. Astafyeva, E.I.; Afraimovich, E.L. Long-distance traveling ionospheric disturbances caused by the great Sumatra- Andaman earthquake on 26 December 2004. Earth Planets Space; 2006; 58, pp. 1025-1031. [DOI: https://dx.doi.org/10.1186/BF03352607]

13. Rolland, L.M.; Lognonné, P.; Munekane, H. Detection and modeling of Rayleigh wave induced patterns in the ionosphere. J. Geophys. Res.; 2011; 116, A05320. [DOI: https://dx.doi.org/10.1029/2010JA016060]

14. Maruyama, T.; Yusupov, K.; Akchurin, A. Ionosonde tracking of infrasound wavefronts in the thermosphere launched by seismic waves after the 2010 M8.8 Chile earthquake. J. Geophys. Res.; 2016; 121, pp. 2683-2692. [DOI: https://dx.doi.org/10.1002/2015JA022260]

15. Hegai, V.V.; Kim, V.P.; Legen’ka, A.D. Ionospheric F2-layer perturbations observed after the M8.8 Chile earthquake on February 27, 2010, at long distance from the epicenter. J. Astron. Space Sci.; 2017; 34, pp. 1-5. [DOI: https://dx.doi.org/10.5140/JASS.2017.34.1.1]

16. Pulinets, S.; Ouzounov, D. The possibility of earthquake forecasting. Learning from Nature; IOP Publishing: Bristol, UK, 2018.

17. Liu, J.Y.; Chen, Y.I.; Chuo, Y.J.; Chen, C.S. A statistical investigation of pre-earthquake ionospheric anomaly. J. Geophys. Res.; 2006; 111, A05304.

18. Oyama, K.I.; Kakinami, Y.; Liu, J.Y.; Kamogawa, M.; Kodama, T. Reduction of electron temperature in low latitude ionosphere at 600 km before and after large earthquakes. J. Geophys. Res. Space Phys.; 2008; 113, A11317. [DOI: https://dx.doi.org/10.1029/2008JA013367]

19. Zhao, B.; Wang, M.; Yu, T.; Wan, W.; Lei, J.; Liu, L.; Ning, B. Is an unusual large enhancement of ionospheric electron density linked with the 2008 great Wenchuan earthquake?. J. Geophys. Res. Space Phys.; 2008; 113, A11304. [DOI: https://dx.doi.org/10.1029/2008JA013613]

20. Liu, J.Y.; Chen, Y.I.; Chen, C.H.; Hattori, K. Temporal and spatial precursors in the ionospheric Global Positioning System (GPS) total electron content observed before the 26 December 2004 M9.3 Sumatra–Andaman earthquake. J. Geophys. Res.; 2010; 115, A09312. [DOI: https://dx.doi.org/10.1029/2010JA015313]

21. Kuo, C.L.; Huba, J.D.; Joyce, G.; Lee, L.C. Ionosphere plasma bubbles and density variations induced by pre-earthquake rock currents and associated surface charges. J. Geophys. Res. Space Phys.; 2011; 116, A10317. [DOI: https://dx.doi.org/10.1029/2011JA016628]

22. Liu, J.Y.; Le, H.; Chen, Y.I.; Chen, C.H.; Liu, L.; Wan, W.; Su, Y.Z.; Sun, Y.Y.; Lin, C.H.; Chen, M.Q. Observations and simulations of seismoionospheric GPS total electron content anomalies before the 12 January 2010 M7 Haiti earthquake. J. Geophys. Res. Space Phys.; 2011; 116, A04302.

23. Oyama, K.-I.; Kakinami, Y.; Liu, J.Y.; Abdu, M.A.; Cheng, C.Z. Latitudinal distribution of anomalous ion density as a precursor of a large earthquake. J. Geophys. Res.; 2011; 116, A04319. [DOI: https://dx.doi.org/10.1029/2010JA015948]

24. Li, M.; Parrot, M. Statistical analysis of an ionospheric parameter as a base for earthquake prediction. J. Geophys. Res. Space Phys.; 2013; 118, pp. 3731-3739. [DOI: https://dx.doi.org/10.1002/jgra.50313]

25. Kuo, C.L.; Lee, L.C.; Huba, J.D. An improved coupling model for the lithosphere-atmosphere-ionosphere system. J. Geophys. Res. Space Phys.; 2014; 119, pp. 3189-3205. [DOI: https://dx.doi.org/10.1002/2013JA019392]

26. Daneshvar, M.R.; Freund, F.T. Remote sensing of atmospheric and ionospheric signals prior to the Mw 8.3 Illapel earthquake, Chile 2015. Pure Appl. Geophys.; 2017; 174, pp. 11-45. [DOI: https://dx.doi.org/10.1007/s00024-016-1366-0]

27. Liu, H.F.; Ding, B.; Zhao, J.; Li, L.; Hu, L.; Wan, W.; Ning, B. Ionospheric response following the Mw 7.8 Gorkha earthquake on 25 April 2015. J. Geophys. Res. Space Phys.; 2017; 122, pp. 6495-6507. [DOI: https://dx.doi.org/10.1002/2016JA023079]

28. Denisenko, V.V.; Nesterov, S.A.; Boudjada, M.Y.; Lammere, H. A mathematical model of quasistationary electric field penetration from ground to the ionosphere with inclined magnetic field. J. Atmos. Solar Terr. Phys.; 2018; 179, pp. 527-537. [DOI: https://dx.doi.org/10.1016/j.jastp.2018.09.002]

29. Khegai, V.V. Analytical model of a seismogenic electric field according to data of measurements in the surface layer of the midlatitude atmosphere and calculation of its magnitude at the ionospheric level. Geomagn. Aeron.; 2020; 60, pp. 507-520. [DOI: https://dx.doi.org/10.1134/S0016793220030081]

30. Pulinets, S.A.; Khegai, V.V.; Legen’ka, A.D.; Korsunova, L.P. New parameter for analysis of ionospheric disturbances and the search for ionospheric precursors of earthquakes based on Barbier’s formula. Geomagn. Aeron.; 2022; 62, pp. 255-262. [DOI: https://dx.doi.org/10.1134/S001679322203015X]

31. Balasis, G.; Mandea, M. Can electromagnetic disturbances related to the recent great earthquakes be detected by satellite magnetometers?. Tectonophysics; 2007; 431, pp. 173-195. [DOI: https://dx.doi.org/10.1016/j.tecto.2006.05.038]

32. De Santis, A.; De Franceschi, G.; Spogli, L.; Perrone, L.; Alfonsi, L.; Qamili, E.; Cianchini, G.; Di Giovambattista, R.; Salvi, S.; Filippi, E. et al. Geospace perturbations induced by the Earth: The state of the art and future trends. J. Phys. Chem. Earth Parts A/B/C; 2015; 85–86, pp. 17-33. [DOI: https://dx.doi.org/10.1016/j.pce.2015.05.004]

33. De Santis, A.; Balasis, G.; Pavón-Carrasco, F.J.; Cianchini, G.; Mandea, M. Potential earthquake precursory pattern from space: The 2015 Nepal event as seen by magnetic Swarm satellites. Earth Planet. Sci. Lett.; 2017; 461, pp. 119-126. [DOI: https://dx.doi.org/10.1016/j.epsl.2016.12.037]

34. De Santis, A.; Marchetti, D.; Pavón-Carrasco, F.J.; Cianchini, G.; Perrone, L.; Abbattista, C.; Alfonsi, L.; Amoruso, L.; Campuzano, S.A.; Carbone, M. et al. Precursory worldwide signatures of earthquake occurrences on Swarm satellite data. Sci. Rep.; 2019; 9, 20287. [DOI: https://dx.doi.org/10.1038/s41598-019-56599-1] [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31889060]

35. Sasmal, S.; Chowdhury, S.; Kundu, S.; Politis, D.Z.; Potirakis, S.M.; Balasis, G.; Hayakawa, M.; Chakrabarti, S.K. Pre-Seismic Irregularities during the 2020 Samos (Greece) Earthquake (M = 6.9) as Investigated from Multi-Parameter Approach by Ground and Space-Based Techniques. Atmosphere; 2021; 12, 1059. [DOI: https://dx.doi.org/10.3390/atmos12081059]

36. Hegai, V.V.; Kim, V.P.; Liu, J.Y. On a possible seismo-magnetic effect in the topside ionosphere. Adv. Space Res.; 2015; 56, pp. 1707-1713. [DOI: https://dx.doi.org/10.1016/j.asr.2015.07.034]

37. Kondo, G. The variation of the atmosphere electric field at the time of earthquake. Mem. Kakioka Magn. Obs.; 1968; 13, pp. 11-23.

38. Hao, J. The anomalous of atmospheric electric field at the ground level and earthquakes. Acta Seismol. Sin.; 1988; 10, pp. 207-212.

39. Hao, J.; Tang, T.M.; Li, D.R. Progress in the research of atmospheric electric field anomaly as an index for short-impending prediction of earthquakes. J. Earthquake Pred. Res.; 2000; 8, pp. 241-255.

40. Smirnov, S. Association of the negative anomalies of the quasistatic electric field in atmosphere with Kamchatka seismicity. Nat. Hazards Earth Syst. Sci.; 2008; 8, pp. 745-749. [DOI: https://dx.doi.org/10.5194/nhess-8-745-2008]

41. Vershinin, E.F.; Buzevich, A.V.; Yumoto, K.; Saita, K.; Tanaka, Y. Correlations of seismic activity with electromagnetic emissions and variations in Kamchatka region. Atmospheric and Ionospheric Electromagnetic Phenomena Associated with Earthquakes; Hayakawa, M. Terra Scientific Publishing Company: Tokyo, Japan, 1999; pp. 513-517.

42. Choudhury, A.; Guha, A.; De, B.K.; Roy, R. A statistical study on precursory effects of earthquakes observed through the atmospheric vertical electric field in northeast India. Ann. Geophys.; 2013; 56, R0331.

43. Law, J. The Ionization of the Atmosphere near the Ground in Fair Weather. Q. J. R. Met. Soc.; 1963; 89, pp. 107-121. [DOI: https://dx.doi.org/10.1002/qj.49708937908]

44. Pestova, O.; Pestov, D.; Shishenya, A.; Kupovykh, G.; Redin, A.; Klovo, A. Electrode layer structure generating under radon-222 transfer across land-atmosphere interface. Proceedings of the XV International Conference on Atmospheric Electricity; Norman, OK, USA, 15–20 June 2014; 215.

45. Kupovykh, G.V. Electric volumetric charge in the surface layer. Izvestiya Taganrog State Radio Technology University; Taganrog State Radio Technology University Publishing: Taganrog, Russia, 1998; pp. 202-205.

46. Pulinets, S.A.; Davidenko, D.V. The Nocturnal Positive Ionospheric Anomaly of Electron Density as a Short-Term Earthquake Precursor and the Possible Physical Mechanism of its Formation. Geomagn. Aeron.; 2018; 58, pp. 559-570. [DOI: https://dx.doi.org/10.1134/S0016793218040126]

47. Freund, F. Charge generation and propagation in igneous rocks. J. Geodyn.; 2002; 33, pp. 543-550. [DOI: https://dx.doi.org/10.1016/S0264-3707(02)00015-7]

48. Hegai, V.V.; Kim, V.P.; Nikiforova, L.I. A possible generation mechanism of acoustic-gravity waves in the ionosphere before strong earthquakes. J. Earthq. Pred. Res.; 1997; 6, pp. 584-589.

49. Hegai, V.V.; Kim, V.P.; Liu, J.Y. The ionospheric effect of atmospheric gravity waves excited prior to strong earthquake. Adv. Space Res.; 2006; 37, pp. 653-659. [DOI: https://dx.doi.org/10.1016/j.asr.2004.12.049]

50. Kelley, M.C.; Swartz, W.E.; Heki, K. Apparent ionospheric total electron content variations prior to major earthquakes due to electric fields created by tectonic stresses. J. Geophys. Res.; 2017; 122, pp. 6689-6695. [DOI: https://dx.doi.org/10.1002/2016JA023601]

51. Pulinets, S.A.; Legen’ka, A.D. First simultaneous observations of the topside density variations and VLF emissions before the Irpinia earthquake, 23 November 1980 in magnetically conjugated regions. Proceedings of International Workshop on Seismo Electromagnetics; University of Electro-Communications Publ: Chofu, Japan, 1997; pp. 56-59.

52. Chmyrev, V.M.; Isaev, N.V.; Serebryakova, O.N.; Sorokin, V.M.; Sobolev, Y.P. Small-scale plasma inhomogeneities and correlated ELF emissions in the ionosphere over an earthquake region. J. Atmos. Solar-Terr. Phys.; 1997; 99, pp. 967-974. [DOI: https://dx.doi.org/10.1016/S1364-6826(96)00110-1]

53. Zhang, X.; Shen, X.; Parrot, M.; Zeren, Z.; Ouyang, X.; Liu, J.; Qian, J.; Zhao, S.; Miao, Y. Phenomena of electrostatic perturbations before strong earthquakes (2005–2010) observed on DEMETER. Nat. Hazards Earth Syst. Sci.; 2012; 12, pp. 75-83. [DOI: https://dx.doi.org/10.5194/nhess-12-75-2012]

54. Zhang, X.; Shen, X.; Zhao, S.; Yao, L.; Quyang, X.; Qian, J. The characteristics of quasistatic electric field perturbations observed by DEMETER satellite before large earthquakes. J. Asian Earth Sci.; 2014; 79, pp. 42-52. [DOI: https://dx.doi.org/10.1016/j.jseaes.2013.08.026]

55. Pulinets, S.A.; Boyarchuk, K.A.; Hegai, V.V.; Karelin, A.V. Conception and model of seismo-ionosphere-magnetosphere coupling. Seismo-Electromagnetics: Lithosphere-Atmosphere-Ionosphere Coupling; Hayakawa, M.; Molchanov, O.A. Terrapub: Tokyo, Japan, 2002; pp. 353-361.

56. Zhima, Z.; Yan, R.; Lin, J.; Wang, Q.; Yang, Y.; Lv, F.; Huang, J.; Cui, J.; Liu, Q.; Zhao, S. et al. The Possible Seismo-Ionospheric Perturbations Recorded by the China-Seismo-Electromagnetic Satellite. Remote Sens.; 2022; 14, 905. [DOI: https://dx.doi.org/10.3390/rs14040905]

57. Dobrovolsky, I.P.; Zubkov, S.I.; Miachkin, V.I. Estimation of the size of earthquake preparation zones. Pure Appl. Geophys.; 1979; 117, pp. 1025-1044. [DOI: https://dx.doi.org/10.1007/BF00876083]

58. Bowman, D.D.; Ouillon, G.; Sammis, C.G.; Sornette, A.; Sornette, D. An observational test of the critical earthquake concept. J. Geophys. Res. Solid Earth; 1998; 103, pp. 24359-24372. [DOI: https://dx.doi.org/10.1029/98JB00792]

59. Toutain, J.-P.; Baubron, J.-C. Gas geochemistry and seismotectonics: A review. Tectonophysics; 1998; 304, pp. 1-27. [DOI: https://dx.doi.org/10.1016/S0040-1951(98)00295-9]

60. Baumgaertner, A.J.G.; Thayer, J.P.; Neely, R.R., III; Lucas, G. Toward a comprehensive global electric circuit model: Atmospheric conductivity and its variability in CESM1(WACCM) model simulations. J. Geophys. Res. Atmos.; 2013; 118, pp. 9221-9232. [DOI: https://dx.doi.org/10.1002/jgrd.50725]

61. Rycroft, M.J.; Odzimek, A.; Arnold, N.F.; Füllekrug, M.; Kułak, A.; Neubert, T. New model simulations of the global atmospheric electric circuit driven by thunderstorms and electrified shower clouds: The roles of lightning and sprites. J. Atmos. Solar-Terr. Phys.; 2007; 69, pp. 2485-2509. [DOI: https://dx.doi.org/10.1016/j.jastp.2007.09.004]

62. Park, C.G.; Dejnakarintra, M. Thundercloud electric fields in the ionosphere. Electrical Processes in Atmospheres; Dolezalek, H.; Reiter, R. Steinkopff: Darmstadt, Germany, 1977; pp. 544-551.

63. Larkina, V.I.; Migulin, V.V.; Molchanov, O.A.; Kharkov, I.P.; Inchin, A.S.; Schvetcova, V.B. Some statistical results on very low frequency radiowave emissions in the upper ionosphere over earthquake zones. Phys. Earth Planet. Inter.; 1989; 57, pp. 100-109. [DOI: https://dx.doi.org/10.1016/0031-9201(89)90219-7]

64. Parrot, M.; Mogilevsky, M.M. VLF emissions associated with earthquakes and observed in the ionosphere and the magnetosphere. Phys. Earth Planet. Inter.; 1989; 57, pp. 86-99. [DOI: https://dx.doi.org/10.1016/0031-9201(89)90218-5]

65. Kim, V.P.; Hegai, V.V. On possible changes in the midlatitude upper ionosphere before strong earthquakes. J. Earthq. Predict. Res.; 1997; 6, pp. 275-280.

66. Kim, V.P.; Pulinets, S.A.; Hegai, V.V. Theoretical model of possible disturbances in the nighttime mid-latitude ionospheric D-region over an area of strong-earthquake preparation. Radiophys. Quantum Radioelectron.; 2002; 45, pp. 262-268. [DOI: https://dx.doi.org/10.1023/A:1016353929416]

67. Davies, K.; Baker, D.M. Ionospheric effects observed around the time of the Alaskan earthquake of March 28, 1964. J. Geophys. Res.; 1965; 70, pp. 2251-2253. [DOI: https://dx.doi.org/10.1029/JZ070i009p02251]

68. Khegai, V.V.; Legen’ka, A.D.; Pulinets, S.A.; Kim, V.P. Variations in the Ionospheric F2 Region Prior to the Catastrophic Earthquake in Alaska on March 28, 1964, According to the Data of the Ground-Based Stations of the Ionospheric Vertical Sounding. Geomagn. Aeron.; 2002; 42, pp. 344-349.

69. Pulinets, S.A.; Legen’ka, A.D. Spatial-Temporal Characteristics of Large-Scale Distributions of Electron Density Observed in the Ionospheric F-region before Strong Earthquakes. Cosm. Res.; 2003; 41, pp. 221-229. [DOI: https://dx.doi.org/10.1023/A:1024046814173]

70. Pulinets, S.A.; Legen’ka, A.D.; Zelenova, T.I. Local-Time Dependence of Seismo-Ionospheric Variations at the F-Layer Maximum. Geomagn. Aeron.; 1998; 38, pp. 400-402.

71. Pulinets, S.A.; Davidenko, D.V.; Budnikov, P.A. Method for cognitive identification of ionospheric precursors of earthquakes. Geomagn. Aeron.; 2021; 61, pp. 14-24. [DOI: https://dx.doi.org/10.1134/S0016793221010126]

72. Pulinets, S.; Ouzounov, D.; Karelin, A.; Davidenko, D. Lithosphere–atmosphere–ionosphere–magnetosphere coupling a concept for pre-earthquake signals generation. Pre-Earthquake Processes: A Multidisciplinary Approach to Earthquake Prediction Studies; Dimitar, O.; Sergey, P.; Katsumi, H.; Patrick, T. AGU/Wiley: Hoboken, NJ, USA, 2018; pp. 77-98.

73. Liu, J.-Y.; Chao, C.-K. An observing system simulation experiment for 1 FORMOSAT-5/AIP detecting seismoionospheric precursors. Terr. Atmos. Ocean. Sci.; 2017; 28, pp. 117-127. [DOI: https://dx.doi.org/10.3319/TAO.2016.07.18.01(EOF5)]