Generation of Seismic-Related DC Electric Fields and Lithosphere-Atmosphere-Ionosphere Coupling

This paper reviews modeling of the influence of earthquake (EQ) preparation processes on the ionosphere through the electric field and electric current occurring in the global atmosphere–ionosphere electric circuit. Our consideration is based on the satellite-and ground-based experimental data of electric fields, plasma and electromagnetic perturbations obtained for several days before an EQ. We have ruled out the models which are not consistent with the experimental data on the electric fields in the ionosphere and also on the ground surface. There has then been proposed a new model of the generation of electric field on the basis of injection of charged aerosols into the atmosphere, and we discuss the mechanism of lithosphere-atmosphere-ionosphere coupling. It is then shown that such changes in the electric field within the ionosphere induce a variety of plasma and electromagnetic phenomena associated with an impending EQ.


Introduction
Numerous plasma and electromagnetic anomalies observed within the ionosphere above the regions of seismic activity are found as evidence that processes of earthquake (EQ) preparation effects take place in the ionosphere for several days before an EQ.Observations of anomalous plasma and electromagnetic phenomena in the ionosphere over the zones of seismic activity were extensively discussed in many reviews and books (Gokhberg et al., 1988;Liperovsky et al., 1992;Molchanov, 1993;Buchachenko et al., 1996;Varotsos, 2001;Hayakawa & Molchanov, 2002;Pulinets & Boyarchuk, 2004;Tronin, 2006;Sorokin, 2007;Molchanov & Hayakawa, 2008;Hayakawa, 2009Hayakawa, , 2012;;Uyeda et al., 2009;Sorokin & Chmyrev, 2010;Hayakawa & Hobara, 2010), and these phenomena are considered as manifestation for the existence of lithosphere-atmosphere-ionosphere (LAI) coupling or interaction.There are ionospheric effects as a result of the simultaneous actions of various factors such as acoustic waves, electric fields, electromagnetic radiation, chemically active substances, etc.An important role in the formation of these factors is played by aerosols of the lower atmosphere, which influence its conductivity and forms an external electric charge and a current by atmosphere dynamics.Seismic activity is accompanied by the injection of soil aerosols and radioactive substances into the atmosphere, so that the enhancement of such activity in seismic regions changes the state of ionospheric plasma and electromagnetic field at the temporal scale for a few days before an EQ.An analysis of satellite data showed the presence of electromagnetic perturbations over a wide frequency range.These perturbations are localized within the magnetic field tube conjugate with the seismic focus of an impending EQ.There are quite many papers on those satellite recordings of wave and plasma disturbances possibly associated with an individual EQ or several strong EQs (Parrot & Lefeuvre, 1985;Larkina et al., 1989;Chmyrev et al., 1989;Galperin et al., 1993;Molchanov et al., 1993;Pulinets et al., 1994;Parrot, 1994Parrot, , 2009Parrot, , 2011;;Chmyrev et al., 1997).The presence of electron density fluctuations in the ionosphere above seismic regions was substantiated by ample satellite data (Afonin et al., 1999), and there were recorded changes in the ionic composition and temperature of the plasma in the upper ionosphere and perturbations of the height profile of the ionospheric F region (Pulinets et al., 1994;Boskova et al., 1994).An analysis of satellite images of the Earth's surface in the infrared (IR) frequency range showed the presence of stable and unstable components of the anomalous IR radiation flux above active crust faults; this flux corresponded to an increase in the temperature of the near-Earth layer by several degrees (Qiang et al., 1999;Tronin, 1999;Tronin et al., 2002;Ouzounov et al., 2012).Simultaneously with electromagnetic and plasma phenomena in the ionosphere, there were observed an increase in the concentration of certain gases (e.g., H 2 , CO 2 , and CH 4 ) by several orders of magnitude, an increase in atmospheric radioactivity (related to such radioactive elements as radon, radium, uranium, thorium, and actinium and their decay products), and an increase in the injection of soil aerosols (Alekseev & Alekseeva, 1992;Virk & Singh, 1994;Heincke et al., 1995;Igarashi et al., 1995;Biagi, 2009;Yasuoka et al., 2012).
The ground-based observations which are aimed at searching electromagnetic phenomena related with processes of EQ preparation and evolution, have started in the last tens of years of XX century.The following phenomena were observed with a lot of hopes: ULF magnetic and electric emissions (Fraser-Smith et al., 1990;Molchanov et al., 1992;Kopytenko et al., 1993;Hayakawa et al., 1996a), acoustic emissions (Gorbatikov et al., 2002), amplitude and phase anomalies of subionospheric VLF/LF signals from powerful transmitters (Hayakawa et al., 1996b;Molchanov & Hayakawa, 1998;Rozhnoi et al., 2004), ionosphere perturbations measured by the ionospheric sounding (Pulinets et al., 1994;Liu, 2009), airglow anomalies (Gladychev & Fishkova, 1994) and some others.Uniform and global-size observations of possible ionospheric effects from many EQs can be carried out together with the estimation of the size of seismo-active region.
A joint analysis of observational results led us to conclude that seismic activity stimulated the development of intense processes in the lower atmosphere.Earth's surface seismic waves, chemically active and radioactive substances, and charged aerosols are likely to act simultaneously on the lower atmosphere.There then occur heating of the lower atmosphere, sharp changes in its electrophysical parameters, the generation of acoustic waves, and the formation of external electric currents.The acoustic action also appears on the ionosphere because of the upward propagation of infrasonic waves (Liperovsky et al., 1997).Processes in the lower atmosphere (seismic waves, atmosphere heating, and the injection of gases) result in the generation and upward propagation of internal gravity waves (IGWs), which might perturb the ionosphere (Gokhberg et al., 1996).The formation of ultralow-frequency radiation on the Earth's surface by lithospheric sources is considered in Molchanov and Hayakawa (1995), Molchanov (1999), Surkov and Pilipenko (1999), and Sorokin and Pokhotelov (2010), and the possibility of its penetration into the ionosphere is discussed in Molchanov et al. (1995).Numerous studies of the nature of atmosphere-ionosphere interactions aimed at determining their mechanism were performed.For instance, the physical processes of formation of currents in the lithosphere and propagation of their radiation into the ionosphere were considered in Fitterman (1979) and Pilipenko et al. (1999).Alperovich et al. (1979) discussed acoustic actions resulting in ionospheric perturbations and the generation of geomagnetic pulsations was discussed.Similar works were performed for numerous chains of processes between sources and measured parameters.Another approach to study EQ precursors consists in a joint analysis of a set of possible parameters observed.Such an analysis can be physically based on a model that makes it possible to interpret satisfactorily most of satellite-and ground-based observations as a manifestation of one cause.In this case measured parameters proved to be interrelated by certain regularities.One of the important problems of atmosphere-ionosphere interactions is the search for a chain of processes related to acting factors and identification of a set of observed effects of a common nature.It is considered that principal causes of lithosphere-ionosphere coupling are the generation of both acoustic waves and electric field in the seismic region.Below we discuss only one of these influence factors; namely, the purpose of this paper is to discuss the cause and consequences of electric field occurring at an eve of EQs.

Basic Properties of DC Electric Fields
The seismic-related DC electric fields in the ionosphere had been, for the first time, revealed by Chmyrev et al. (1989).They analyzed the vertical component of quasi-static (DC) electric field E z , and we show one example.They observed such an enhanced E z onboard the "Intercosmos-Bulgaria 1300" satellite within a 15-min interval before an EQ occurred on January 12, 1982 at 17.50.26UT.The quasi-static electric field with amplitude of 7-8 mV/m was observed in two zones: above the EQ focus and in its magnetically conjugate region, and the size of those zones was 1°~1.5° in latitude.
Subsequent investigations of DC electric field in the ionosphere based on direct satellite measurements over seismic regions were carried out by Gousheva et al. (2006Gousheva et al. ( , 2008Gousheva et al. ( , 2009)), who analyzed hundreds of seismic events in order to detect DC electric field enhancement in the ionosphere connected with EQs.Seismic events with different magnitudes in different tectonic structures at different latitudes were observed.They selected the orbits with distance less than 250 km with respect to the EQ epicenter, not crossing terminator and during low magnetic activity.Let us present one case study of their registration results.The DC electric field 5-10 mV/m was detected in the magnetic conjugate regions 11-13 hours before two EQs (magntitude around 5) occurring on 25.08.1981 at 16:54:39 UT and 17:29:07 UT correspondingly (Gousheva et al., 2008).Statistical analyses of the satellite data by Gousheva et al. (2008Gousheva et al. ( , 2009) ) led them to make a conclusion on the existence of seismic-related quasi-static electric field in the ionosphere.The duration of electric field disturbances with amplitude of the order of 10 mV/m can be up to 15 days, and the electric field disturbances in the daytime and nighttime ionospheres were on the same order.
Direct observations of quasi-static electric field in the ionosphere are confirmed by computational modeling of the ionospheric perturbation occurring at an eve of EQs.Spatial distributions of the total electron content (TEC) obtained by GPS receivers in the seismic region were analyzed (Liu, 2009;Pulinets, 2009a), and those TEC anomalies are tried to be interpreted with the use of global model of the upper atmosphere which describes the thermosphere, ionosphere and plasmasphere as an integrated system.The model is based on integration of the non-stationary three-dimensional equations of continuity, impulse and energy balance of multi-component gas simultaneously with the equation for electric field potential.In the frame of computer simulations there was sought an additional electric field which leads to the TEC perturbation coincident with the one observed in the EQ preparation region.For example, Zolotov et al. (2008) considered an EQ in Peru on 26.09.2005.The characteristics of TEC disturbances were given in Zakharenkova et al. (2008), and the TEC perturbation was observed during six days before the EQ from 21.09.2005till 26.09.2005.Based on the computer simulations Zolotov et al. (2008) have shown that the observable TEC perturbation is due to an additional electric field with an amplitude of 6 mV/m.It is further suggested by Klimenko et al. (2011Klimenko et al. ( , 2012) ) and Namgaladze et al. (2009) that a possible general cause of TEC perturbation is a vertical plasma drift by the zonal electric field.Computer simulations by Klimenko et al. (2012) have shown that the amplitude of electric field disturbance is required to be 3-9 mV/m.At the same time, observations of the quasi-static electric field on the Earth's surface in seismic regions were carried out by different workers (Kondo, 1968;Jianguo, 1989;Nikiforova & Michnovski, 1995;Vershinin et al., 1999;Hao et al., 2000;Rulenko, 2000).Analyses of those publications show that the local electric field surges with large amplitude reaching several kV/m are observed during the EQ preparation, but their duration is of the order of ten minutes.However, there are absent visible electric field disturbances with duration of several days observed simultaneously over the horizontal distance of hundreds of kilometers.
The indirect confirmation of electric fields occurring in the atmosphere is the observational results of VHF emissions propagating from the source located in the troposphere over a region of EQ preparation (Vallianatos & Nomicos, 1998;Ruzhin et al., 2000;Hayakawa et al., 2006;Ruzhin & Nomicos, 2007;Yonaiguchi et al., 2007a, b;Yasuda et al., 2009).VHF radiations are found to have occurred for several days before an EQ, and their duration reaches several days.If the VHF electromagnetic radiation propagated over a distance more than a wavelength, then the condition of optical propagation is fulfilled, so that it is possible to receive the signal at distance of the order of 300 km just in the case that its source is located in the atmosphere above Earth's surface.The region of generation of VHF electromagnetic radiation is found to be at the altitudes of the order of several kilometers above EQ epicenters located behind the horizon (Fukumoto et al., 2001;Yasuda et al., 2009).
Consequently, both the direct and indirect data of DC electric field observations in the atmosphere and ionosphere over a seismic region allow us to formulate its basic properties.The basic experimental results are summarized as follows: • The enhancement of seismic activity produces DC electric field disturbances in the ionosphere of the order of 10 mV/m.
• These disturbances occupy the region with horizontal spatial scale from hundreds to thousands km over the seismic region.
• DC electric field enhancements occur in the ionosphere from hours to 10 days before an EQ.
• DC electric field disturbances in the daytime and nighttime ionospheres have the same order of magnitude.
• DC electric field disturbances can reach the breakdown value during from hours to 10 days in the atmosphere at altitudes 1 to 10 km over the EQ zone.
• The quasi-stationary electric field on the Earth's surface does not exceed its background value simultaneously in the seismic area during several days.

Penetration of DC Electric Field Into the Ionosphere
Lithospheric activity stimulates the processes which are followed by the electric field generation.The enhancement in number density of charged aerosols by one-two orders and the increase in atmosphere radioactivity level by the injection of radon and other radioactive substances are observed during days and weeks before an EQ (Alekseev & Alekseeva, 1992;Virk & Singh, 1994;Voitov & Dobrovolsky, 1994;Heinke et al., 1995;Pulinets et al., 1997;Yasuoka et al., 2006Yasuoka et al., , 2012;;Omori et al., 2007;Biagi, 2009).Data on injection of the soil gases such as radon, helium, hydrogen, carbon dioxide in the surface atmosphere with horizontal spatial scale of 500 km during from several hours to several weeks before an EQ have been reported by King (1986).Igarashi et al. (1995) described the surge in five times of the radon concentration in the soil water, and the data on significant emissions of metallic aerosols Cu, Fe, Ni, Zn, Pb, Co, Cr and radon were given by Boyarchuk (1997).Quasi-static electric field disturbances in the ionosphere are observed at the same time as the injection of active substances in the lower atmosphere.
There are observed the local short-time releases of active substances along with large scale growth of the level of active substances in the lower atmosphere.They can generate the impulses of electric field near the Earth's surface, whose amplitude can reach tens kV/m but its duration does not exceed tens of minutes.A model of the generation of pulses of local electric fields with characteristic time scales of 1-10 min for the atmospheric conditions above fracture regions of EQs was considered by Liperovsky et al. (2005Liperovsky et al. ( , 2008)).They have proposed that aerosols, increased ionization velocity and upstreaming air flows occur at night-time conditions, and that water condensates at the aerosols at night when the temperature in the near-earth air is low and the relative humidity increases above earth-fracture regions.Then, the relatively large aerosol particles are mainly negatively charged, while the charge of smaller particles is overwhelmingly positive.It is assumed that aerosol clouds of small dimensions are suddenly injected into the locally heated surface atmosphere and move with the air up to higher altitudes.The vertical velocity of small particles is much smaller than that of large ones, which is equal to a few cm/s.As a consequence of the shift between the small and large particles, there occur the local pulses of the electric field in the atmosphere.The amplitude of such a field is estimated as 10 3 ~3×10 3 V/m, but the relaxation time of a cloud of aerosols is estimated 10 minutes.Anomalous emanation of radons preceding a large EQ was observed by Omori et al. (2007), who have analyzed atmospheric radon concentrations and estimated changes of electrical conditions in the atmosphere due to the preseismic radon anomaly.These authors used the model by Liperovsky et al. (2005), and they have shown that the radon emanation reduces the atmospheric electric field by 40%.Their estimation of field amplitude gives 10 4 ~10 5 V/m at the observable value of radon emanation, but unfortunately there are no calculations of electric field in the ionosphere in the above-mentioned works.Nevertheless it is assumed that this impulse electric field can be a source of lithosphere-ionosphere coupling.We should note that this impulse field is observed only in local regions.The duration of such impulses is 10 minutes, while ionospheric precursors and DC electric field in the ionosphere exist during a much longer interval of several days.The field occurs inside a dipole layer of charged aerosols cloud and the field slumps outside the dipole layer.Consequently, the local impulse electric field observed on the Earth's surface cannnot be a cause of the ionospheric effects and appearance of DC electric field in the ionosphere; that is, the radon injection in the frame of model does not affect the lithosphere-ionosphere coupling.
A generation mechanism of electric field in the lithosphere based on the result of laboratory experiments has been proposed by Freund et al. (2006Freund et al. ( , 2009) ) and Freund (2010).Their experiments show that when stresses are applied to one end of a block of igneous rocks, two currents flow out of the stressed rock volume.One current is carried by electrons and the other current is carried by p-holes.Positive electric potential, ionization of air molecules and corona discharge occur on the rock surface.It is assumed that air ionization is a cause of ionospheric disturbances, glows and IR emissions, but there are no calculations on the possible effect of this source to the ionosphere.This mechanism seems to be used to interpret the impulse phenomena because the source duration is over 10 minutes, but it seems to be an unlikely explanation of the existence of DC electric field over a long period of time.
Below we consider the generation mechanisms for quasi-static electric fields in the ionosphere.Spatial distribution of this field has a horizontal scale (100-1000) km and its duration is from tens hours to tens of days.
The field is quasi-static if its temporal variation exceeds considerably the relaxation time (τ) of charges in the surface atmosphere τ~ε 0 /σ~10~30 min (ε 0 is the permittivity of free space, and σ is the surface atmosphere conductivity).An equivalent circuit is often used to explain the generation of atmospheric electric field (Goldberg, 1984;Sapkota & Varshneya, 1990;Rycroft et al., 2000).The current flowing in the circuit is excited by a generator which is the resultant action of thunderstorms all over the world.The fair weather current density is of the order of 10 -12 A/m 2 in the closed circuit (e.g., MacGorman & Rust, 1998;Rakov & Uman, 2002).It is assumed that the conductivity of near-Earth atmosphere of 10 -14 S/m yield an electric field value on the Earth's surface of 10 2 V/m (Rakov & Uman, 2002).The fair weather current gives an electric field of 10 -3 mV/m in the ionosphere with conductivity 10 -6 S/m.Since the background electric field of magnetospheric and ionospheric origin has a value (0.1~1) mV/m in the middle-latitude ionosphere, then the field of atmospheric origin with intensity of 10 -3 mV/m is negligible in the ionosphere.The variation of DC electric field in the ionosphere over a seismic region can be realized in two different ways.First of all one can change the load resistance and in a different way one can include an additional EMF (electro-motive force) in the global circuit.
Let us consider the first way.The processes of EQ preparing impact take place in the lower atmosphere which contributes to over 80% of load resistance of the global circuit.The injection of radioactive and chemical substances and aerosols into the atmosphere, and the variation of aerosols size and atmospheric state result in a change of load resistance.In the final analysis, all of these processes change conductivity of the surface atmosphere.In Figure 1 there is depicted the circuit with selected part of current over a seismic region.The resistance of atmosphere over thunderstorms is denoted by R 1 , R 2 is the resistance of the region of thunderstorms activity, R 3 is the resistance of near-Earth atmosphere, R i is the resistance of ionosphere, and I is the fair weather current (Rycroft et al., 2000).The load resistance R is much smaller than all of these resistances.The disturbed part of circuit (designated by red color in Figure 1) consists of the following resistances: r 1 is the resistance of upper troposphere, r i is the resistance of ionosphere over a seismic region, and r 2 is the resistance of surface atmosphere.The disturbance of surface atmospheric conductivity results in the variation of r 2 and the electric current in this part of the circuit.For the first time, Sorokin and Yaschenko (2000) and Sorokin et al. (2001) have carried out the calculation of altitude dependence of DC electric field variation in the Earth-ionosphere layer produced by a source of ionization and a growth of conductivity in the lower atmosphere.They performed theoretical investigations of the atmospheric ionization by alpha particles and gamma quantum of the nuclear decay, and they calculated the altitude dependence of ionization rate and conductivity for different levels of radioactivity.It is shown that the electric field can be changed by 1.5-2 times in the ionosphere by the growth of conductivity in the surface atmosphere.Such a variation field does not impact onto the ionosphere because the amplitude of variations is considerably smaller than the background value.That is, the field variation is invisible in the ionosphere.This result is confirmed by Omori et al. (2008), who have shown that the quasi-static electric field is reduced by 1.5 times due to the growth of radioactivity and conductivity during the radon injection.In spite of evident results of continued unsuccessful attempts to explain the appearance of seismic-related quasi-static electric fields in the ionosphere due to the variation of conductivity in the lower atmosphere, for example, Pulinets (2009a) has assumed that an anomalous electric field in the ionosphere over an active fault occurs by the variation of conductivity in the near-earth atmosphere.The conductivity is varied due to the growth of additional radon ionization and the reduction in ions mobility by the generation of large clusters.There are missing both the proof of speculation and the calculation of field value in the ionosphere.According to Omori et al. (2007Omori et al. ( , 2008)), the radon surge with magnitude 10 Bq/m 3 leads to an increase in ionization rate up to (10 6 ~10 7 ) 1/m 3 s.As a result, conductivity of the near-earth atmosphere is increased in 1.5 times, and the field is varied approximately by 1.5 times in the ionosphere as well.Since the field of fair weather in the ionosphere is 10 -3 mV/m, then its variation by 1.5 times will be much smaller than the background value (0.1~1) mV/m.Harrison et al. (2010) have shown that an increase of ionization rate by radon in two times leads to a variation of the current flowing from the ionosphere to the Earth in 10%, and then the field is varied on the same quantity in the ionosphere.Thereby, any models based on the assumption that the ionization of lower atmosphere leads to a conclusion that the seismic-related electric field formation in the ionosphere is in apparent contradiction with experimental data that the electric field is up to 10 mV/m in the ionosphere.We consider a different way of DC electric field formation in the ionosphere.The electric current and electric field are varied due to the inclusion of a seismic-related EMF in the global circuit.The EMF can be located in the lithosphere, in the atmosphere and in the vicinity of boundary between the lithosphere and atmosphere.The scheme of altitude dependence of total electric current j = σE+j e (j e : EMF external current) in these three cases is depicted in Figure 2. The origin of coordinate system is located on the Earth's surface.We consider the case corresponding to the left panel of Figure 2. In the frame of this model it is assumed that the EMF is located in the lithosphere and the field is transferred through the atmospheric layer with specified altitude dependent electric conductivity.The vertical component of electric field disturbance is given on the Earth's surface, and the section of closed global electric circuit is depicted in Figure 3.The uniform Ohm's law for a subcircuit without the EMF is performed in the Earth-ionosphere layer.The nature of electric field source on the Earth surface and its characteristics are not discussed in the papers based on this model.The source of field is expected to create a quasi-static electric current in the circuit for several days.The field in the ionosphere is calculated at given spatial distributions of its vertical component on the Earth's surface (Kim & Hegai, 1999;Pulinets et al., 2000Pulinets et al., , 2003;;Grimalsky et al., 2003;Rapoport et al., 2004;Denisenko et al., 2008;Ampferer et al., 2010).Electric fields in the ionosphere are computed for different boundary conditions, shape and size of field horizontal distribution on the Earth's surface.Kim and Hegai (1999) showed that the field reaches (0.3~0.7) mV/m in the nighttime ionosphere if the field near the Earth has a value of 1000 V/m.Since the field in the seismic region does not exceed approximately 100 V/m (Kondo, 1968;Vershinin et al., 1999), then their calculated value of field in the ionosphere should be reduced to (0.03~0.07) mV/m.Taking into account that the conductivity of daytime ionosphere is larger than that of nighttime ionosphere by one-two order, the field value in the daytime ionosphere is approximately 10 -3 mV/m.Calculations fulfilled in Pulinets et al. (2000Pulinets et al. ( , 2003) ) show that electric field in the nighttime ionosphere can reach (0.1~1) mV/m if it reaches a value (10 3 ~10 4 ) V/m on the Earth's surface in a seismic area with horizontal scale 100 km.This value of field on the ground surface is required to remain during several days, but such a field is unlikely to exist.Calculations performed by Denisenko et al. (2008) confirm this conclusion.It is shown that the field reaches a value 10 -3 ~10 -4 mV/m in the ionosphere at the maximal field value E 0 = 100 V/m on the Earth surface.So, we can say that there exists, in the ionosphere, no static electric field of lithospheric origin.This can be obtained from a simple consideration of the continuity equation ∇•j = ∇•σE = 0 for the vertical conductivity current j = σ(z)E(z) in the conductive atmosphere.The estimate of maximum magnitude can be made simply in 1D (one dimensional) approximation dσE/dz = 0. Let σ 0 , σ 1 be the conductivity in the near-ground of the atmosphere and that in the ionosphere and E 0 , E 1 are the electric fields near the ground and in the ionosphere, then we obtain E 1 = E 0 (σ 0 /σ 1 ).Taking into account that σ 0 ≈ 10 -14 S/m; σ 1 ≈ 10 -6 S/m; E 0 = 100 V/m we find E 1 ≈ 10 -3 mV/m, which is four orders of magnitude lower than the background ionospheric field.Thus the considered model is found to contradict with the well-known experimental data The same situ into the ionosp he seismic-re rence between reparation.Nam erstorm cloud f EQ preparing ism for DC ele yakawa (1996) .shows that the 10 -8 A/m 2 .In ubstances, so th to a reduction with aerosols evel of atmosp o have obtaine gamma radiat mospheric rad on rate q = q( rth layer.

LAI Co
In order to study the p coupling m hypothese Hayakawa interpretation of observational results of EQ precursors based on the model of IGW propagation.One of them is as follows.These waves are propagated angularly to the Earth's surface, and the angle increases depending on wave period.So, IGW reaches the ionosphere at a distance of the order of 1000 km from the EQ epicenter, which seems to be in conflict with the localization of the plasma and electromagnetic disturbances in the vicinity of EQ epicenter.Though there have been recently published a few papers suggesting the important role of AGW channel in the LAI coupling (Korepanov et al., 2009;Hayakawa et al., 2011a), in which you can find a summary of findings in favor of this hypothesis.
Below we consider the mechanism of the influence of seismic-related electric fields on the ionosphere.The model of LAI coupling was described in Pulinets et al. (2000), which consists of two stages which are not related with each other.They considered the formation of electric field in the near-ground atmosphere due to the appearance of metallic aerosols and ionization source.The source of ionization produces positive and negative ions, and then heavy ions are formed by adhesion of water molecules to the light ions.They calculated the altitude dependence of electric field caused by diffusion, transfer of ions and aerosols by the electric field, gravitational sedimentation of the heavy particles and upward moving of the light particles by the atmosphere convection.The interaction between ions with different signs and their adhesion to the aerosols are taken into account.Their calculations show that during 50 seconds after the turning on the ionization source there is formed an electrode layer up to 30 cm altitude.The number density of the positive and negative ions is different and the electric field is reduced in 1.5 times in this layer.The value of electric field grows in three times above this layer.
As they show that this mechanism could be used to explain the electric field variation at fog occurring in the near-ground level.Further, those authors gave a solution of the problem on the electric field penetration into the ionosphere through the conducting atmosphere with exponentially upward increasing conductivity.By imposing the boundary condition on the horizontal distribution of vertical component of electric field on the Earth's surface, they calculated the horizontal component of electric field at the altitude 90 km based on the spatial scale of horizontal distribution of electric field on the Earth's surface.Obviously, the ionosphere conductivity in night time is less than that in day time, so that the electric field at night will be more enhanced than at day.Their calculation shows that even though the radius of disturbed region is 200 km and the field on the ground is 100 V/m, the magnitude of electric field will be 0.07 mV/m.This field is likely to be much smaller than the background field in the ionosphere and consequently it cannot have any effect on the ionosphere.Further those authors conclude that if the field on the ground will be 1000 V/m, then the effect of this field on the ionosphere will be possible.However, such a field in the seismic region with radius 200 km is considered to be implausible.So the above-mentioned work cannot be a basis of LAI coupling model.Pulinets (2009b) has then made an attempt to explain the possible ionosphere modification due to the atmospheric ionization during the radon injection in the vicinity of active faults.The process of local modification of global electric circuit and the ionosphere variability for tectonic activity is discussed.He supposes that the occurrence of any additional source of ionization has a double effect on the atmosphere conductivity.The appearance of additional ions increases the atmosphere conductivity, while the generation of heavy cluster ions leads to its reduction.However, there is no estimation on the resultant value of conductivity.
Further the author supposes that the anomaly of atmosphere conductivity leads to the variation of electric current in the local part of the global circuit, but no calculation of this field has been performed.One should keep in mind that there are theoretical investigations of the atmosphere conductivity modification during the course of ionization.In application to the seismic effect Sorokin et al. (2007) studied in details the processes of conductivity formation during the course of gamma and alpha decay based on the solution of a system of self-consistent nonlinear equations for electric field, atmosphere conductivity, density of ions and aerosols with taken into account their interaction.The well-known value of fair weather current is ~10 -12 A/m 2 and atmosphere conductivity is ~10 -14 mho/m, then the field on the ground has a value of ~100 V/m.The value of ionosphere conductivity is ~10 -6 mho/m, so that the field in the ionosphere for the current with the same density has a value of ~10 -3 mV/m.The variation of conductivity in the near-ground atmosphere due to the ionization in two times results in the variation of current density of the same order in the local part of circuit.So that, this additional electric field is on three-four orders less than the ionospheric field and its effect on the ionosphere and equatorial anomaly is negligible.Therefore, the hypothesis suggested is physically not well grounded and it cannot be a candidate for the creation of LAI coupling mode.On this reason the suggestion by Pulinets (2009b) is in contradiction with results obtained by Klimenko et al. (2012), who show that observed disturbances of TEC occur on the assumption that DC electric field reaches (3~9) mV/m in the ionosphere.After all, they use the work by Pulinets (2009b) to interpret the data in spite of the contradiction with obtained results.
A principally alternative physical idea based on the electrodynamic model of plasma and electromagnetic disturbances accompanying the processes of EQ preparing was developed in Sorokin et al. (2001Sorokin et al. ( , 2007) ) and Sorokin and Chmyrev (2010).First, this model allowed them to explain the results of observation of quasi-static electric field both in the ionosphere and on the Earth's surface in the seismic region, because other models could not explain the nature of such a field.In the frame of this model they found the mechanism for the enhancement of conducting electric current with altitude and the mechanism for limitation of electric field vertical component on the ground surface.The enhancement mechanism is realized by a decrease with altitude of EMF external current at the condition of conservation of the total current.This current is equal to the sum of conductivity and external currents.The external current of EMF is formed in the near ground atmosphere, as seen in Figure 5.In this case even at the growth in conductivity with altitude the field can reach an amplitude of 10 mV/m in the ionosphere.While the conductivity current is appeared by including an additional EMF in the global circuit.The EMF is formed during the injection of charged aerosols by soil gases in the atmosphere and their transfer in the convective atmosphere.The field is limited by a feedback between the external current of EMF and the electric field generated near Earth's surface.Calculations show that the amplitude of disturbed electric field does not exceed their background value on the Earth's surface.In one sense the above-mentioned model is similar to the model of AGW influence to the ionosphere, because the amplitude of AGW grows with altitude by a decrease in atmosphere density.By analogy, the value of conductivity current grows with altitude by a decrease of external current.This implies that the effects are becoming stronger in the ionosphere, but it is difficult to identify AGWs above the background in the near-ground atmosphere.Similarly it is difficult to select the disturbances of conductivity current because their amplitude on the ground does not exceed the background value which is equal to the fair weather current.Both of these effects have a unified source which are lithospheric gases injected into the atmosphere.One can suppose that both AGW and electric current can affect the ionosphere simultaneously, though the consequences of these effects can be different.
According to the electrodynamic model, the growth of electric field in the ionosphere is caused by the EMF formation and the corresponding variation of electro-physical characteristics of lower atmosphere as a result of injection of soil gases, aerosols and radioactive substances during EQ preparation.In the frame of our model, the theory of generation of quasi-static electric field in the atmosphere-ionosphere system was developed, and the methods for calculation of electric field spatial distribution were elaborated.Sorokin et al. (2001Sorokin et al. ( , 2005aSorokin et al. ( , 2006aSorokin et al. ( , 2007) ) and Sorokin and Chmyrev (2010) carried out the theoretical investigation of mechanisms of EMF formation in the lower atmosphere, who have shown that the quasi-static electric field reaches 10 mV/m in the ionosphere while their value is of the order of 100 V/m on the Earth's surface.Moreover, the field can reach a breakdown value in the layer with thin 1-2 km on the altitudes 5-10 km in the troposphere (Sorokin et al., 2011(Sorokin et al., , 2012a, b), b).Value of the external current of EMF can be approximately 10 -8 ~10 -6 A/m 2 near the ground.They have further investigated theoretically plasma and electromagnetic effects accompanying the generation of conducting current in the global circuit, and have shown that the appearance of EMF in the global circuit leads to the stimulation of a set of observed plasma and electromagnetic phenomena.An enhancement of the electric field might result in the instability of AGWs in the ionosphere (Sorokin et al., 1998), but the exponential growth of AGW amplitude by the electric field in the ionosphere is limited by vortex formation (Chmyrev & Sorokin, 2010).As a result, the horizontal irregularities of conductivity with scale of approximately 10 km are expected to take place in the E layer of ionosphere.This process is accompanied by field-aligned currents and plasma irregularities stretched along magnetic field lines in the upper ionosphere (Sorokin et al., 1998;Sorokin et al., 2000).Their appearance leads to ULF oscillations (Sorokin et al., 1998) and spectral broadening of VLF transmitter signals (Chmyrev et al., 2008) registered on satellites.The scattering of background electromagnetic emissions by horizontal irregularities of conductivity in the lower ionosphere results in the enhancement of electromagnetic ELF emissions registered on satellites (Borisov et al., 2001) and generation of gyrotropic waves propagated along E layer of the ionosphere.Their propagation forms line spectra of ULF oscillations (Sorokin et al., 2003;Sorokin & Hayakawa, 2008;Sorokin et al., 2009) and the change of resonance frequency of Schuman resonances (Hayakawa et al., 2005(Hayakawa et al., , 2011b)).Moreover, the appearance of irregularities in the nighttime ionosphere leads to depressions of ULF pulsations of magnetosphere origin (Sorokin et al., 2004;Schekotov et al., 2006;Hayakawa et al., 2013).The growth of electric field up to the breakdown value is caused by random electric discharges, which might generate VHF radio emission in the troposphere over the EQ epicenter (Sorokin et al., 2011(Sorokin et al., , 2012a, b), b).The generation of conductivity current in the global circuit is accompanied by the modification of ionosphere.Perturbations in the D region of the ionosphere may be generated by both the transfer of charged particles and electron heating (Laptukhov et al., 2009) The rison field observation in the ionosphere and in the atmosphere over a seismic region and on the ground.A principally different model is based on the assumption that the current source in the circuit connected with pre-EQ processes are situated not in the lithosphere or in the atmosphere, but in the near-ground atmospheric layer.The EQ preparation processes modify the atmosphere in this layer and form an EMF in the seismic zone.The additional source of electric current is generated in the global circuit at the stage of EQ preparing.The range of EMF is formed in the near-earth atmosphere and includes the boundary between the lithosphere and atmosphere.In this case the observable electric field on the surface is located inside the EMF range.Upward transfer of the charged aerosols by atmospheric convection and their gravitational sedimentation result in the EMF formation.Aerosols are injected into the atmosphere by soil gases during an increase in seismic activity.The external current of EMF is reduced with altitude while the current of conductivity increases with altitude, so the total current in the circuit is constant.The value of conductivity current near the surface can be of the order of the value of fair weather current, while the external current exceed the conductivity current by four-five orders.Therefore, conductivity current in the ionosphere is on the order of the external current of EMF near the surface.
The above-mentioned model is similar to the model for AGW influence to the ionosphere, because the amplitude of AGW grows with altitude by decreases of atmosphere density.By analogy, the value of conductivity current increases with altitude by decreases of external current.This implies that the effects become stronger in the ionosphere, but it is difficult to identify any AGW above the background in the near ground atmosphere.So that, it is difficult to identify the disturbances of conductivity current on the ground because their amplitude does not exceed the background value which equals the fair weather current.Both of these effects have a unified source which is injected lithospheric gases in the atmosphere, since one can suppose that both AGW and electric current can affect the ionosphere simultaneously.Lithosphere, atmosphere and ionosphere are an integrated environment in which physical phenomena are related with each other.On the basis of the above-mentioned model the intensive processes in the lithosphere and atmosphere such as EQs, volcanoes, typhoons, thunderstorms are the cause of electrodynamic effect onto the ionospheric plasma.All of these processes are accompanied by numerous electromagnetic and plasma phenomena, which will be discussed elsewhere.

Figure 1 .
Figure 1.Equivalent electric circuit of DC electric field formation in the ionosphere over a region of conductivity disturbance in the lower atmosphere.Black color denotes the conventional global circuit, and red color indicates the part of circuit over the region of disturbed conductivity Figure 2. D Figure Gravitati Figure 8 field near Figure 10.