Review of electromagnetic coupling between the Earth's atmosphere and the space environment

ARTICLE IN PRESS

1364-6826/$ - se

doi:10.1016/j.ja

�CorrespondGeomagnetism,

Tel.: +9198920

E-mail addr

[email protected]

Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658

www.elsevier.com/locate/jastp

Review of electromagnetic coupling between the Earth’satmosphere and the space environment

Devendraa Siingha, R.P. Singhb, A.K. Kamraa, P.N. Guptab, Rajesh Singhb,�,V. Gopalakrishnana, A.K. Singhc

aIndian Institute of Tropical Meteorology, Pune 411 008, IndiabDepartment of Physics, Banaras Hindu University, Varanasi 211 005, India

cDepartment of Physics, Bundelkhand University, Jhansi 284 128, India

Received 3 May 2004; received in revised form 9 August 2004; accepted 1 September 2004

Abstract

We review our understanding of the electrical properties of the lower and upper atmosphere along with various

possible sources of the electromagnetic energy near and far above the Earth’s surface. The transport of electromagnetic

energy from the atmosphere to the ionosphere and then to the magnetosphere and back to the Earth’s surface via

ionosphere and lower atmosphere is discussed. The electromagnetic coupling of various regions is also discussed.

r 2005 Elsevier Ltd. All rights reserved.

Keywords: Fair-weather atmospheric electrical conductivity; ELF/VLF wave; Ionosphere; Lightning discharge; Magnetosphere;

Optical emissions (Sprites, blue jet, elves); Schumann resonance; Thunderstorm; ULF waves; Whistler waves

1. Introduction

The conducting Earth is surrounded by comparatively

non-conducting atmosphere of thickness 60–80 km. The

layer above it is the ionosphere extending up to

�1000 km. The region outside the ionosphere known

as the magnetosphere is filled with tenuous neutral gas

(mainly hydrogen), and an ionized gas with proton as

the dominant positive ion. The collision rate in this layer

is so low that for many purposes the region can be

considered as collisionless in which charged particle’s

dynamics is governed by the Earth’s magnetic field. The

e front matter r 2005 Elsevier Ltd. All rights reserve

stp.2004.09.006

ing author. Present address: Indian Institute of

New Panvel, Navi Mumbai 410 218, India.

96886; fax: +9122 27480762.

esses: [email protected],

.res.in (R. Singh).

large-scale morphology of the magnetosphere is con-

trolled by the interaction between the solar wind and the

Earth’s magnetic field (Dungey, 1978). The outer

boundary of the magnetosphere called the magneto-

pause is located at a distance of about 10–12 Earth radii

(Re) in the direction of the sun. In the anti-sun direction,

magneto-tail is extended up to undefined length, due to

the dragging of the geomagnetic field lines by the solar

wind (Dungey, 1978). Magnetopause boundary layer is

the site where energy and momentum are exchanged

between the solar wind plasma and the magnetospheric

plasma. This energy is dissipated by several complex

current systems arising due to the solar wind–magneto-

sphere interaction. A sketch of the Earth’s magneto-

sphere is shown in Fig. 1.

In electromagnetic coupling one must consider the

source of electromagnetic energy in one region and its

transmission to the other region. The major source of

d.

ARTICLE IN PRESS

Fig. 1. A sketch of the noon-midnight meridian plane view of

the Earth’s magnetosphere.

D. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658638

electromagnetic energy in the lower atmosphere is the

lightning discharges radiating electromagnetic waves in

the wide frequency range. Other sources are current

system due to dynamo action, geomagnetic storm/

substorm, wave–particle interaction in the magneto-

sphere, earthquake, nuclear explosion, etc. The trans-

mission of electromagnetic energy from one region to

the other depends upon the electrical behavior of the

medium which is governed by the presence of charged

particles and its dynamics through electrical conductiv-

ity. At most places, the Earth behaves as a perfectly

conducting medium for electromagnetic waves of less

than a few megahertz (Kamra and Ravichandran, 1993).

The conductivity increases exponentially with height in

the lower and middle atmosphere and causes attenua-

tion and dispersion of the low-frequency waves. At radio

frequencies, the lower and middle atmosphere behaves

like a vacuum. Thus, DC fields and currents along with

low-frequency waves could control the electrodynamics

of the lower and middle atmosphere.

The electric field changes communicate with almost

velocity of light and coupling take place much faster

than the coupling of electromagnetic energy in different

frequency bands. For example, in VLF band the group

velocity lies between 105 and 106m/s and coupling

process will be comparatively slower. ULF waves

propagate at Alfven velocity making the coupling

process still slow. Similarly, coupling due to energetic

particle movement will depend upon the kinetic energy

of the particle involved in the process. Acoustic gravity

waves and tidal waves act on much longer time-scale.

Thus, different processes of transport of electromagnetic

energy and hence electromagnetic coupling act at

different time-scales.

The upper atmosphere including the ionosphere and

the magnetosphere is anisotropic, inhomogeneous and

contains transient fields along with wide variety of wave

modes. The amplitude, phase and frequency of waves

propagating through such a media are modified. There-

fore, for the proper interpretation of the observed wave-

features precise knowledge of the intervening medium

which behaves as a transmission path to the electro-

magnetic signal is required.

The sources of electromagnetic energy and electrical

behavior of the ambient medium are also controlled by

the space weather changes such as solar-flares and sun-

spots affect the occurrences and characteristics of

thunderstorms (Tinsley, 2000), the cosmic-ray-produced

ions affect the nucleation and growth characteristics of

cloud particles (Carslaw et al., 2002; Harrison and

Carslaw, 2003). High-energy particles penetrating to the

lower altitudes, increases the electrical conductivity of

the lower atmosphere (Markson, 1978; Tinsley, 2000).

Markson and Muir (1980) suggested how solar varia-

bility moderates the Earth’s electric field and electrical

potential of the ionosphere which is maintained by the

world thunderstorm activity. Such a link supports the

mechanism in which solar control of ionizing radiation

modulates atmospheric electrification, cloud physical

processes and atmospheric energetics. On the other

hand, some tropospheric disturbances are known to

influence the ionospheric phenomena. For example,

several theoretical and experimental studies show that

the lightning activity in thunderstorms influence the

temperature, ion densities, composition and electrical

potential of the ionosphere (Inan et al., 1991; Taranenko

et al., 1993; Pasko et al., 1997). Recent observations of

optical phenomenon such as sprites, elves, blue jets and

blue starters propagating from the top of active

thunderstorms generate radiations in the ULF and

VLF range and contribute to the maintenance of

potential of the ionosphere in the global electric circuit

(GEC) (Rycroft et al., 2000; Su et al., 2003). The

variable solar activity also affect the weather and climate

(Markson, 1978), thus leading a connection between

electrical behavior of the medium and weather and

climate.

Is there any effect of optical emissions (sprites, elves

and jets) over the thunderstorms in our environment in

important ways or just beautiful natural phenomena like

rainbows? This question is a challenge to our scientific

community to find the possible influences of space

process on weather and climate. It has motivated a

reexamination of our understanding of the electrical

processes and properties of the Earth environment. A

possible connection between electrical environment and

climate of the Earth atmosphere, including the modula-

tion of electrical conductivity and cloud nucleation rates

by cosmic radiation have been discussed by Carslaw et

al. (2002) and Singh et al. (2004a). Recently, Hiraki et al.

(2002) have suggested that sprites would change

chemically the concentration of NOx and HOx in the

mesosphere and lower atmosphere. These chemical

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 639

changes may impact on the global cooling or heating in

the middle atmosphere. However, more work is required

in future on this topic.

While summarizing the electrical behavior of different

regions of the Earth’s environment, Singh et al. (2004a)

briefly discussed sources of the electric field and the

electrodynamics involved in each region. They argued

that the GEC model could provide short-and long-term

variations in the electrical processes of various regions

and their intra-coupling (Lakhina, 1993). Some impor-

tant features of the GEC are summarized in Table 1. In

Table 1

Some physical properties of the global electric circuit

S.

No.

Physical parameters

1 Number of thunderstorm acting at one time over the globe

2 Currents produced above thunderstorm

(A)

(a) Range

(b) Average

(c) Global current

3 Fair-weather electric field (V/m)

(a) Ground

(b) At 20 km altitude

(c) At 50 km altitude

4 Electrical conductivity (mho/m)

(a) Sea level

(b) Tropopause

(c) Stratosphause

(d) Ionosphere

(i) Pedersen conductivity

(ii) Parallel conductivity

5 Current density (A/m2)

(a) Inhabited and industrialized area

(b) Vegetated ground and deserts

(c) South pole station

(d) Fair-weather

6 Total energy associated with global electric circuit (J)

7 Total resistance (O) (including decreases by mountain)

8 Ionospheric potential (kV)

(a) Range

(b) Mean

9 Columnar resistance at sea level (O/m2)

(a) Low latitude

(b) High latitude

(c) Antarctic and Tibet plateau

10 Electrical relaxation times

(a) 70 km

(b) 18 km

(c) 0.01 km

(d) Earth surface

11 Average charge on the Earth surface (C)

12 Average charge transfer over the entire world (Ckm�2Yr�1)

this model, thunderstorms charge the ionosphere to a

potential of several hundred thousand volts (Roble and

Tzur, 1986) which drives a vertical current downwards

from the ionosphere to the ground. The fair-weather

current depends upon the potential difference and

conductivity of the medium between the ionosphere

and the ground. Horizontal currents flow freely along

the highly conducting Earth’s surface and in the

ionosphere. The circuit is closed by the current flowing

from the ground in to the thunderstorm generator and

from thunderstorm cloud top towards the ionosphere.

Typical

value

References

�1500–2000 Roble and Tzur (1986), Rycroft et al. (2000)

and Singh et al. (2004a)

Gish and Wait (1950), and Singh et al. (2004a)

0.1–6.0

0.5–1.0

700–2000

Rycroft et al. (2000) and Singh et al. (2004a)

102

1

10�2

Volland (1987) and Singh et al. (2004a)

10�14

10�13

10�10

10�4–10�5

10

Rycroft et al. (2000) and Muhleisen (1977)

10�12

2.4� 10�12

2.5� 10�12

2� 10�12

2� 1010 Rycroft et al. (2000)

230 Muhleisen (1977)

200

Muhleisen (1977) and Roble and Tzur (1986)

150–600

280

Gish and Wait (1950)

1.3� 1017

3� 1017

2� 1016

Roble and Tzur (1986)

10�1 s

4 S

5–40min

10�5 s

5� 105 Roble and Tzur (1986)

+90 Roble and Tzur (1986)

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658640

There are temporal variations on time-scales varying

from microseconds (lightning discharges) to milliseconds

(sprites), minutes to an hour (thunderstorm

regenerator), hour to a day (diurnal variations), and

months (seasonal variations) and to a decade (solar cycle

effect).

In this paper, we briefly review the electrodynamic

and electromagnetic properties of different regions of

the Earth’s atmosphere and it is near space in order to

understand the electromagnetic linkages of various

regions. Both types of sources of electromagnetic energy,

natural and man made are discussed. Natural sources

include thunderstorm, ionospheric dynamo, earth-

quakes, plasma waves, magnetospheric convection. In

man-made sources, we consider power line harmonic

radiation, VHF and HF transmitters and nuclear

explosions. Several other artificial sources such as

world’s rail lines, green house gases, supersonic jets,

rockets and satellites, gas released in the atmosphere etc

are not discussed for want of space. A brief discussion

on the upward and downward transmission of dc electric

fields and the transient and wave energy follows.

Fig. 2. Electrical conductivity, neutral temperature and elect

2. Electrical conductivity of the Earth’s atmosphere

The fair-weather atmospheric electrical conductivity

alongwith the electron density and temperature distribu-

tions with height are shown in Fig. 2. The electrical

conductivity near the Earth’s surface is of the order of

10�14mho/m and increases nearly exponentially with

altitude up to 60 km with a scale length of �7 km. Above

80 km, the conductivity becomes anisotropic with the

Pedersen conductivity (sp) parallel to an E-field and

orthogonal to B0; the Hall conductivity (sh), orthogonalto E and B0; and the field-aligned conductivity (sF)parallel to B0; because of the influence of the geomag-

netic field and shows diurnal variation due to solar

photo-ionization process. Pedersen and Hall conductiv-

ity peaks in the height range between 100 and 150 km,

the so-called dynamo region.

The electrical conductivity in clean atmosphere is

inversely proportional to aerosol particle content of the

air. Electrical conductivity over open ocean is therefore

considered as an index of atmospheric aerosol loading

and has been used to estimate global changes in

ron density profiles of the Earth’s atmospheric regions.

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 641

background air pollution levels (Cobb and Wells, 1970;

Kamra and Deshpande, 1995).

The electrical conductivity is determined by mobile

ions and electrons. Since mobility of the large ions is two

orders of magnitude smaller than the mobility of the

small ions, the electrical conductivity is mainly due to

the small ions. The major source of ionization in the first

few meters above the ground surface is the radon gas

which is the decay product of Uranium-238 present in

the Earth’s crust. Upto 60 km the main source of

ionization is Galactic Cosmic Rays (GCR) flux. The

MeV electrons and associated X-rays produce ionization

in the stratosphere. Above 60 km solar photon are the

main sources of ionization. Below 60 km, the main

charge carriers are small positive and negative ions

whereas above 60 km free electrons become more

important. The high mobility of electrons abruptly

increases the conductivity throughout the mesosphere.

Cho and Rycroft (1998) present a simple model profile

for the atmospheric conductivity ranging from

10�13mho/m near the surface to 10�7mho/m at 80 km

altitude in lower ionosphere. Hale (1994) presents a

more complex profile depicting variation in both space

and time. The conductivity is 3 orders of magnitude

higher at the height of 35 km as compared to that at the

Earth’s surface whereas the air density at 35 km is 1% of

that at the Earth’s surface.

Solar activity influences the conductivity on the day-

to-day basis and decadal time-scale with relative

amplitude of 3–20%. With increase in solar activity,

the GCR flux reduces in mid-latitude causing reduction

in conductivity in this region, while during the same

period solar proton may be ‘funneled’ by the Earth’s

magnetic field to polar regions resulting in an increased

atmospheric conductivity there. The interaction of solar

wind with the Earth’s magnetic field also causes a dawn-

to-dusk potential difference across the polar region

(Tinsley and Heelis, 1993). During the geomagnetically

active periods, the energetic charged particles precipitat-

ing from the inner and the outer Earth’s magnetospheric

radiation belts interact with the middle and lower

atmosphere by depositing their energy in the atmosphere

and producing ionization directly or via Bremmstrah-

lung radiation, thereby influencing the dynamics of

storm and atmosphere (Tinsley and Heelis, 1993;

Tinsley, 2000). Radioactivity of the ground and its

emanations cause significant variations in electrical

conductivity near the ground both in space and time in

an unpredictable way (Hoppel et al., 1986; Volland,

1987).

As a result of large field aligned conductivity, the

geomagnetic field lines behave like electric equipotential

lines and hence electric field parallel to B0 breaks down

within a fraction of a second. Significant current flows if

electric fields orthogonal to B0 exist and Pedersen and

Hall conductivities are large. The finite conductivity and

its variation in space and time modify the transmission

characteristics of the electromagnetic energy which is

necessary for the interpretation of observed wave forms

with respect to their original wave structure at the source

(Volland, 1987).

3. Sources of electromagnetic energy in the Earth’s

atmosphere

The sources of the electromagnetic energy could be

categorized under natural sources (e.g. thunderstorm

and lightning discharge, earthquake, ionospheric dyna-

mo, magnetospheric plasma convection and waves from

the magnetosphere, etc) and artificial sources (e.g. ULF/

VLF/VHF transmitters, world power grid system,

nuclear explosions, etc). These sources inject energy

into the Earth’s atmosphere in the form of direct current

(dc), quasi-dc and wave form. The same source could

generate energy in all the frequency ranges starting from

zero frequency (dc) to high frequency. For example,

thunderstorm/lightning discharges are the major source

of dc and transient energy as well as the waves from

ULF to microwave frequency and higher frequencies. In

the following we shall discuss some of the processes in

brief.

3.1. Thunderstorms/lightning discharge

The major source of dc energy is the thunderstorm/

lightning discharge. Convective clouds usually accumu-

late a net negative charge in the lower regions and a

positive charge at the top. When electric field strength

locally exceeds �400 kV/m, electric break down occurs

which we observe as lightning.

The widely accepted model of thunderstorm electrifi-

cation is shown in Fig. 3. The current density consisting

of convection, conduction, precipitation and displace-

ment currents varies with altitude between the negative

charge layer and ground. Between the bottom and top of

thunderstorm, the cloud charge separation current

density varies in space and time. Above thunderstorm,

the current consists mainly of conduction and displace-

ment. All lightning currents are considered as discontin-

uous charge transfers (Roble, 1991). In the fair-weather

regions far away from the thunderstorm, only conduc-

tion current flows downward. The current flowing

upward from the top of clouds charges the ionosphere

which behaves as an equipotential surface having a

potential of �300 kV with respect to the Earth (Gish and

Wait, 1950; Kasemir, 1979; Roble and Tzur, 1986).

The total current flowing through the thunderstorm/

lightning discharge is called Maxwell current. The

average Maxwell current density is usually not affected

by lightning discharges and varies slowly throughout the

ARTICLE IN PRESS

Fig. 3. Various currents that flow in the vicinity of an active

thundercloud, there are five contributions of the total current

(Maxwell current), JM ¼ JE þ JC þ JL þ JP þ @D=@t; below

the thunderclouds. Above the thundercloud, JM ¼ JE+qD/qt.

Here, JE ¼ s E is the field-dependent current. JC is the

convection current, JL is the lightning current, JP is the

precipitation current, and qD/qt is the displacement current

(Roble, 1991).

D. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658642

evolution of storm (Krider and Musser, 1982). Since the

Maxwell current remains steady at time when the electric

field both at the ground and aloft undergoes large

changes in amplitude, and some times even polarity,

Krider and Musser (1982) inferred that the cloud

electrification processes may be substantially indepen-

dent of the electric field.

Lightning activity is mainly concentrated in three

distinct zones—East Asia, Central Africa and America.

Lightning is more prevalent in the northern than the

southern hemisphere and mostly occur over the land

surface. Lightning is relatively rare over the oceans and

over the poles. The variation of lightning activity with

latitude as observed from space shows that two of every

three lightning flashes occur in tropical region (Williams,

1992). In addition to the tropical lightning, extra-

tropical lightning activity plays a major role in the

summer season in the northern hemisphere, resulting in

the global lightning activity having a maximum from

June to August. The similarity of the diurnal variations

of the electric field over the oceans and of the worldwide

thunderstorm activity supports the hypothesis that

thunderstorms are electrical generators in the GEC.

Latitudinal and longitudinal distributions of lightning

activity and VLF wave activity recorded at the Earth’s

surface correlate, leading to the suggestions that VLF

wave (whistler) activity find their origin in lightning

discharges (Singh, 2003).

3.1.1. Optical emissions during lighting discharges

Some observations suggest that the return current in

the lightning discharge from the ground does not end in

the cloud, but continues to move upward and terminate

in the lower ionosphere (Sentman et al., 1995; Lyons,

1996 with references therein). This transient current/field

is associated with optical emissions (sprites, elves, blue

jets, blue starters) in the space between the top of the

cloud and the lower ionosphere. Sprites appear as cluster

of short-lived (�50ms) pinkish red luminous columns,

stretching from �30 to 90 km altitude having width less

than 1 km (Sentman et al., 1995; Lyons, 1996; Neubert,

2003) and the maximum brightness at 66 km altitude

(Wescott et al., 2001). The upper portion of the sprites is

red, with wispy, faint blue tendrils extending to 40 km or

lower. Boccippio et al. (1998) showed that about 80% of

sprites are associated with ELF transient events and

positive CG lightning return strokes have large peak

current (435 kA) (Barr et al., 2000; Singh et al., 2002)

and large DMQ (total charge moment change of the

thunderstorm) values. Some sprites associated with

negative CG lightning have also been observed (Bar-

rington-Leigh et al., 1999). Sprites produce detectable

ELF/VLF transients (Price et al., 2002) and a vertical

electric field perturbation of 0.73V/m in stratosphere

(Bering et al. 2002).

Sprites have been observed in Africa, South, Central

and North America, Australia, and recently in Europe

and Japan also. The evidence to date suggests that

sprites may occur over any area, as long as energetic

thunderstorms are present (Rodger, 1999; Barr et al.,

2000; Rycroft et al., 2000; Singh et al., 2002). We find

that upward escape of the lightning signal a phenomena

popularly known in early literature as ‘blue’ or ‘green’

pillars and rocket discharges like columns of optical

emissions (Boys, 1926; Malan, 1937; Wood, 1951).

Cho and Rycroft (1998), using electrostatic and

electromagnetic codes simulated the electric field struc-

ture from the cloud top to the ionosphere and explained

the observation of a single red sprite. To explain the

clusters of sprites, they suggested that the positive

charges may have been distributed in spots so that a

single discharge may lead to clusters of red-sprites. The

redistribution of charge and the electromagnetic pulse

during lightning discharge may produce acceleration of

electrons, heating and ionization of atmosphere. This

may lead to strongly non-linear situation and runway

electrons/electrical breakdown of the atmosphere may

occur (Rycroft and Cho, 1998; Rowland, 1998). Nagano

et al. (2003) evaluated the modification in electron

density and collision frequency of the ionosphere by the

electromagnetic pulse of the lightning discharge and

explained the generation of elves.

The above discussion suggests that the electrical

conductivity of the atmosphere above thunderstorms

could be different from the surrounding atmosphere.

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 643

Enhancement in conductivity by upto a factor of 2 from

ambient values have been observed in the case of strong

thunderstorm events (Holzworth and Hu, 1995). These

changes in conductivity could be due to thunderstorm-

produced gravity wave or X-rays from lightning induced

electron precipitation (Hu et al., 1989).

3.1.2. ELF/VLF wave radiation

Lightning discharges generate transient electromag-

netic pulses. The pulse duration of return strokes

(responsible for the generation of ELF/VLF waves) is

of the order of 100–200 ms, which implies that the

maximum spectral energy is in the frequency range

5–10 kHz. The peak pulse amplitude is �10–100 kA and

typical lightning channel length is 5–10 km. Thus, the

return stroke is a very powerful generator of ELF/VLF

waves in the atmosphere. The wave amplitude is

�1–10mV.

The VLF waves propagating in whistler mode along

the dipolar geomagnetic field line interact with counter

streaming energetic electrons in the equatorial region

(Tsurutani and Lakhina, 1997; Singh, 1999; Singh and

Singh, 2002; Singh et al., 2003). During interaction

process, energy and pitch angle of interacting electrons

decreases leading to their precipitation to the lower

atmosphere usually known as whistler-induced electron

precipitation/trimpi precipitation/lightning-induced

electron precipitation. These precipitated energetic

electrons produce additional ionization (Rycroft, 1973)

leading to change in electrical conductivity (Hu et al.,

1989) and hence modify the flow of electric currents and

distribution of electric fields.

3.1.3. Excitation of Schumann resonance (SR)

SRs are the eigen frequencies of the Earth-ionosphere

cavity oscillation excited by global lightning activity and

lie in the lower ELF band between 5 and 60Hz. The

resonant frequencies are 8, 14, 20, 26, etc. Hz, where the

8—Hz mode represents a wave with wavelength equal to

the Earth’s circumference (Fullekrug and Fraser-Smith,

1996; Barr et al., 2000; Rycroft et al., 2000; Singh et al.,

2002). Singh et al. (2002) have discussed the principal

features of SR, which are being used to monitor global

lightning activity (Heckman et al., 1998; Rycroft et al.,

2000), global variability of lightning activity (Satori,

1996; Nickolaenko et al., 1998) and sprite activity

(Boccippio et al., 1995; Cummer et al., 1998; Rycroft

et al., 2000; Singh et al., 2002). The amplitude of the

Schumann modes is determined by the temporal and

spatial distribution of global lightning which is intense

over tropics. The variations in solar activity or nuclear

explosions produce disturbances in the ionosphere and

affect SR (Schlegel and Fullekrug, 1999). Solar proton

events cause increase in frequency, Q-factor (i.e. band

width of the resonance mode) and amplitude of the SR

mode (Schlegel and Fullekrug, 1999).

Since the main source of the SR phenomenon is

thunderstorm, both the SR and the GEC could be linked

to weather and climate (Williams, 1992; Price, 1993;

Price and Rind, 1994). A positive correlation between

the monthly means of the tropical surface-air-tempera-

ture anomaly and the magnetic field amplitude of the

fundamental mode of the SR has been demonstrated

(Williams, 1992). SR has also been closely linked to the

upper atmosphere water vapor, which plays important

role in tropical cirrus clouds, stratospheric water vapor

and tropospheric chemistry (Price, 2000). Such links in

the electromagnetic, thermodynamic, climate and cli-

mate-change characteristics of the atmosphere have

greatly enhanced the interest in monitoring of electro-

magnetic waves and their mapping and propagation

properties in different regions of the atmosphere.

3.2. Earthquakes

Earthquakes are the explosions inside the Earth due

to movement and interaction of tectonic plates, which

can be characterized by the location of epicenter as well

as the main parameters of the rupture (magnitude,

seismic moment, source mechanism, orientation of the

fault plane and direction of motion). These parameters

are measured by using global network of seismometers.

These networks also allow the study of mechanical

properties of the seismic rupture and the detection of

heterogeneities in the crust. Apart from mechanical

properties, efforts are also being made to study the

changes induced in the surrounding electric and

magnetic fields associated with seismic activity. Analyz-

ing magnetometer data, Kalashnikov (1954) for the frist

time suggested the association of magnetic field and

electrical field with seismic or volcanic activity. There is

ample evidence to show the ionospheric perturbations

are caused by Earthquakes (Hayakawa, 1999). Even the

electromagnetic emissions (ULF, ELF, VLF and HF

ranges) emitted during earthquakes modify the iono-

sphere while propagating through it (Parrot et al., 1993;

Parrot, 1995).

3.2.1. Transient magnetic and electric fields

Johnston (1989) has reviewed the characteristics of

transient magnetic and electric fields near active faults

and volcanoes. These are generally interpreted as the

results of piezomagnetic and piezoelectric effects related

to stress variations during microfracturing of rocks

under the ground (Davies et al., 1980; Enomoto and

Hashimoto, 1990). In some cases the observed magnetic

signal could be due to electrokinetic effects related to the

water circulation system in to this massive system

(Zlotnicki and Le Mouel, 1990). Based on Japanese

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658644

observations, Rikitake (1987) suggested that a telluric

precursor could be observed even at distances several

times larger than the effective radius of the epicentral

area.

3.2.2. Ionospheric perturbations

The ionospheric effects due to earthquakes are known

since the large Alaskan earthquakes of 1964 (Davies and

Baker, 1965). The critical frequencies, f0E and f0F2 at a

distance between 30 and 50 km have been reported to

increase before earthquakes (Sobolev and Husamiddi-

nov, 1985; Parrot et al., 1985). Alimov et al. (1989) have

reported perturbations of the Es layer during night of

the earthquakes and also 2 days before 5 earthquakes of

magnitudes between 4.5 and 6. Such anomalies also

perturb the propagation of VLF waves in the Earth-

ionosphere wave guide. Gokhberg et al. (1989) presented

cases of phase variation when an earthquake occurs on

one path, while other paths were used as tests for

detecting anomalies. Night air-glow observations have

also been reported during the hours preceding an

earthquake. Fishkova et al. (1985) have reported

significant increase in the emission of the oxygen green

line at 5577 A, at an altitude of 80–110 km, whereas the

intensity of the oxygen red line at 6300 A decreases by

15% above 250 km of altitude. All these observations

support the perturbation of the ionosphere due to

earthquakes.

The ionospheric perturbations may be caused by the

emission of radioactive gases (radon) during the

preparatory stage of earthquake. Such emissions can

modify the air conductivity and cause enhanced fair-

weather current leading to the perturbation of iono-

sphere. The other possibility could be enhancement of

Earth’s eigen oscillations during the preparatory stage

which may generate internal gravity waves. These waves

while propagating through the atmosphere via acoustic

mode could modify the ionospheric parameters (Negoda

et al., 1999).

3.2.3. Electromagnetic emissions

The first observations of electromagnetic emissions

associated with earthquake were made by Gokhberg et

al. (1982) based on OGO-6 satellite data and by Larkina

et al. (1983) based on the analysis of INTERKOSMOS-

19 satellite data. They have reported an increase in VLF

wave intensity few hours before and after the earth-

quake. Anomalies in the electromagnetic noises in the

ULF, ELF, VHF and HF bands before and after seismic

shocks have been extensively reported and studied from

time-to-time using various satellite data (Parrot et al.,

1993; Parrot, 1995; Molchanov and Hayakawa, 1995;

Borisov et al., 2001; Gotoh et al., 2002; Tronin et al.,

2002). Chmyrev et al. (1997) presented simultaneous

measurements of ELF emissions and plasma density

inhomogeneities (with dN/N�(3–8)% and horizontal

characteristic scales dL�(4–10) km above the Spitak

earthquake zone in Caucasus. Koons and Roeder (1999)

measured ULF/ELF magnetic field within a few kilo-

meters of the fault involved in the earthquakes and

found that the earthquake-associated spectra decrease

inversely with the square of the frequency or faster. Dea

and Boerner (1999) have presented ULF data with

signal strength approximately twice the background

level some 18 days preceding the Northride earthquake

of January 17, 1994. The signal strength continued to be

the same and it returned to normal level after the quake.

However, Rodger et al. (1996) could not establish any

correlation between the enhancement of electromagnetic

radiation in the ionosphere and seismic activity. This

shows that the effect is rather complicated and probably

depends on many geophysical parameters.

Details of the generation mechanism are out of the

scope of the present paper. It will suffice to say that there

are two main hypotheses regarding the generation

mechanism of these waves: (a) electromagnetic waves

are directly emitted from the earthquake focal region;

(b) the emission is a result of electric charge redistribu-

tion in the Earth’s atmosphere (Singh et al., 2002). The

propagation of the wave from the Earthquake’s focus to

the atmosphere is not well understood.

3.3. Ionospheric dynamo

Ionospheric potential is governed not only by

thunderstorm generator in the troposphere, but also by

the ionospheric and magnetospheric dynamos (Fig. 4)

(Roble and Tzur, 1986). Solar heating causes different

pressure and temperature during the day and night

region of the atmosphere. To compensate it, winds are

set in motion. These perturbations are termed as tides,

which may be either solar or lunar depending upon

whether it is related to the solar or lunar day. The solar

diurnal tide has a period of 24 h and covers the Earth’s

circumference at the observer’s latitude in 24 h, where as

the semi-diurnal tide has a period of only 12 h and also

covers the circumference in 24 h. The tides generated in

the lower atmosphere propagates upwards and when

they reach upper mesosphere/lower thermosphere alti-

tudes, they undergo dissipation by turbulence (which is

believed to be generated by gravity waves and diurnal

tides) and deposit momentum and heat energy there.

The gravitational tidal forces exerted by the moon on

the atmosphere excite lunar tides of much smaller

amplitude. The regular tidal wind system drives iono-

spheric plasma at dynamo layer heights and pushes it

against the geomagnetic field. Ions and electrons,

however, are affected differently by these winds. While

the ions still move essentially with the neutral, the

geomagnetic field already controls the motion of the

ARTICLE IN PRESS

Fig. 5. Schematic diagram of the field—aligned electric currents

flowing in the ionosphere/inner magnetosphere (Richmond,

1986). These field-aligned currents couple the auroral oval with

the outer magnetosphere and are also responsible for sustaining

the auroral electrojets. The solar quite current system and the

equatorial electrojet current system are also shown.

Fig. 4. Schematic diagram of various electrical processes in the

global electric circuit (Roble and Tzur, 1986). The model is

based on an atmosphere divided into four coupled regions (i.e.,

troposphere, middle atmosphere, ionosphere, and magneto-

sphere) and also takes into account the orography of the Earth.

Ionosphere and magnetosphere are treated as the passive

elements of the circuit. The vector B shows the direction of

the Earth’s geomagnetic field, and arrows show the direction of

the current flow in the regions of the tropospheric, ionospheric

and magnatospheric generators.

D. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 645

electrons at dynamo layer heights. The differential

motion of ions and electrons within the dynamo layer

is responsible for horizontally flowing electric currents.

Moreover, charge separation causes an electric polariza-

tion field, which is constrained by the condition of

source free currents (Volland, 1987) and has been

observed indirectly from backscatter measurement

(Richmond, 1976). Lunar variations are usually less

than 10% of magnitude of solar variation (Matsushita,

1967). They depend not only on latitude, solar time,

season and solar cycle, but also on lunar phase. Global

analyses of geomagnetic lunar effects have also found

significant longitudinal variations. The seasonal varia-

tions of the lunar magnetic perturbation tend to be

greater than those for the solar perturbation.

The dynamo electric field associated with the wind

drives a current, which tends to converge in some

regions of space and cause an accumulation of positive

charge, while in other regions of space it would diverge

and cause negative charge to accumulate. These charges

would create an electric field, which would cause current

to flow tending to drain the charges. An equilibrium

state would be attained when the electric-field-driven

current drained charge at precisely the rate it was being

accumulated by the wind-driven current. A net current

flows in the ionosphere owing to the combined action of

the wind and electric field (Takeda and Maeda, 1980).

The large-scale vortex currents at the middle and the

low-latitudes flow counter clockwise in the northern

hemisphere, and clockwise in the southern hemisphere

(Fig. 5). Traditionally these vortices are known as the

solar quiet Sq current system because of the nature of

the ground-level magnetic field variations that they

produce. Currents and electric fields produced by the

ionospheric wind dynamo are relatively weak in

comparison with those of the solar wind/magnetospheric

dynamo at high latitudes. Electric field in the equatorial

lower ionosphere has a localized strong enhancement of

the vertical component associated with the strong

anisotropy of the conductivity in the dynamo region.

This enhanced electric field drives an eastward daytime

current along the magnetic equator called equatorial

electrojet (Forbes, 1981; Richmond, 1986). Efforts are

being made to understand the changes in equatorial

electrojet in response to the electrodynamic processes

involved in the coupling between the solar wind,

magnetosphere and ionosphere. This is due to dynamo

region electric fields being communicated to higher

latitudes along the geomagnetic field lines. Coherent and

incoherent backscatter radar observations of the upper

atmosphere have confirmed that the distortions in the

dynamo region electric fields at equatorial latitudes

originate in the corresponding electrodynamic distur-

bances at high latitudes (Somayajulu et al., 1985).

Studies based on the surface magnetic data have shown

consistent and near instantaneous response of the

equatorial electrojet variations to geomagnetic distur-

bances at high latitudes (Rastogi and Patel, 1975).

Ionospheric dynamo is also affected by the absorption

of ozone at the lower altitudes (30–60 km) and presence

of stronger winds at higher altitudes (4130 km).

3.4. Magnetospheric plasma convection

Fig. 6 shows the topology of the magnetic field and

plasma flow during the interaction of solar wind with the

ARTICLE IN PRESS

Fig. 6. Schematic diagram of Earth’s magnetic field and plasma

flow in the solar wind/magnetosphere. Solid lines show the

magnetic field, open arrows show the plasma velocity direction.

(a)–(d): four classes of the magnetic field line; (a) closed

magnetic field lines connected to the Earth in both northern and

southern hemisphere, (b) interplanetary field lines unconnected

to the Earth, (c) open field lines connecting the northern polar

cap to interplanetary space, and (d) open field lines connecting

the southern polar cap to interplanetary space (Lyons, L.R.,

Williams, D.J., 1984. Quantitative Aspects of Magnetospheric

Physics, D Reidal Publishing Co., Dordrecht, Holland).

Fig. 7. Schematic diagram of the magnetic north pole region

showing the auroral oval and ionospheric convection. The

convection contours also represent electric potential contours,

with a potential difference of the order of 8 kV between them

(Burch, J.A., 1977. The magnetosphere. In: Upper Atmosphere

and Magnetosphere, NRC Geophysics Study committee,

National Academy of Science, Washington DC, pp. 42–56).

D. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658646

Earth’s magnetosphere. In the tail region of the

magnetosphere, where magnetic field reconnection takes

place (Dungey, 1961; Gonzalez et al., 1994) plasma

directly flows along the field lines. This is more likely to

occur when interplanetary magnetic field (IMF) is

directed southward. In this case 5–10% of solar wind

energy is transferred in the Earth’s magnetosphere

(Lester and Cowley, 2000). During northward IMF

intervals, the energy injection due to magnetic reconnec-

tion is considerably reduced and cross field transport

due to scattering across closed field line becomes

important (Lee et al., 1994) and about 0.1–0.3% of the

solar wind energy gets transferred to the magnetosphere

(Tsurutani and Gonzalez, 1995). Several other processes,

like impulsive penetration of the magneto-sheath plasma

elements with an excess momentum density (Owen and

Cowley, 1991), plasma entry due to solar wind

irregularities (Schindler, 1979), the Kelvin–Helmholtz

instability (Miura, 1987) and plasma percolation due to

overlapping of a large number of tearing islands at the

magnetopause (Galeev et al., 1986) have been suggested

for the plasma transport across the magnetopause.

In the magneto-sheath region plasma is accelerated to

high energy and as a consequence of drift motion charge

separation takes place, establishing a polarization

electric field from dawn to dusk. Discharging currents

flow along the geomagnetic field lines down in to the

ionosphere on the dawnside and upward from the

ionosphere on the dusk side, both foot points being

electrically connected via the dynamo region. This

process is equivalent to a huge hydromagnetic generator

situated in the magnetosphere with a load in the

ionosphere; linked to each other via a field-aligned

current (Strangeways and Raeder, 2001).

The overall pattern of the magnetospheric convection

tends to map along the magnetic field line to the

ionosphere. This mapping is imperfect because of net

electric field that tends to develop within non-uniform

energetic plasma. However, in the upper ionosphere, we

may consider a steady-state case where Eplasma ¼ 0 ¼

Eþ v� B:E is the electric field in the Earth’s fixed frame

of reference, v is the velocity of plasma and B is

geomagnetic field vector. The general convection pattern

derived for this condition is shown in Fig. 7, where the

flow lines correspond to lines of constant potential. The

potential is high on the dawnside and low on the dusk

side, with a potential difference �50 kV. The electric

field strength in the auroral oval tends to be somewhat

larger than the polar cap electric field. During magnetic

storm period, the entire magnetosphere is disturbed and

affects the ionospheric dynamo system. This results into

the modification of electric field generation and dis-

tribution pattern (Richmond, 1986; Roble and Tzur,

1986).

3.5. Mangetospheric plasma waves

A good source of electromagnetic energy is the plasma

waves generated in the magnetosphere by various types

of instabilities. The plasma waves could be electromag-

netic or electrostatic in nature. Some of them may

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 647

propagate to the ground and the others can be received

locally from spacecraft. Typical electromagnetic modes

are hydromagnetic wave, ion cyclotron wave, whistler

mode wave, free space wave, whereas electrostatic

modes are ion acoustic wave, electron acoustic wave,

electrostatic ion cyclotron wave, electrostatic electron

cyclotron wave, lower hybrid wave and upper hybrid

wave. These wave modes are excited by free energies of

plasma through wave–particle interaction processes. The

free energies in the plasma exist when charged particles

in plasma have anisotropic temperature/non-Maxwel-

lian velocity distribution/density of the plasma is

spatially inhomogeneous. A typical example with

anisotropic temperature is electrons/protons trapped in

the radiation belt of the magnetosphere, where as

auroral electrons have non-Maxwellian velocity distri-

bution.

3.5.1. Plasma waves generated by trapped particles

The high energy particles injected in the radiation belt

from the nightside plasma sheet bounce back and forth

along the dipolar geomagnetic field lines, which can

excite electromagnetic mode plasma waves by the

cyclotron resonance interaction with parallel propagat-

ing electromagnetic waves in the presence of cold

plasma. In this process Doppler-shifted wave frequency

must be equal to the cyclotron frequency of the particles.

In addition, for the growth of the wave, temperature

anisotropy of particles should exist. The parallel

propagating waves could be the lightning-generated

whistler-mode waves or artificially transmitted waves or

waves from local sources. Both an increase in the

number of resonant particles and temperature aniso-

tropy increase the wave growth rate. The cold plasma is

supplied by diffusion from the ionosphere, where the

plasma density is much higher in the daytime than in the

night time due to sunlit effect. The cold plasma density

in the magnetosphere is also high in the dayside region.

The high-energy particles injected in the nightside region

drift towards the dayside and excite waves in the

enhanced plasma density region. The electrons drifting

eastward excite ELF/VLF emissions (whistler mode

waves) in the morning-to-noon sector, whereas protons

drifting westward excite Pc 1-2 emissions (ion cyclotron

waves) in the evening-to-noon sector.

The emitted whistler mode waves including hiss,

risers, fallers, hooks and emissions of complex dynamic

spectra (Singh, 1999; Singh et al., 2003) in the ELF/VLF

range could also be received on the Earth’s surface. The

observed wave-form on the Earth’s surface depends

upon the electrical condition of the intervening medium.

These waves have been used as an effective diagnostic

tool of the magnetosphere (Singh et al., 1998, 1999a).

Pc1-2 emissions are emitted in the frequency range

0.1–5Hz during cyclotron resonance interaction be-

tween protons (10–100 keV) and left-hand polarized

electromagnetic waves. The proton gyro-frequency at

the radial distance of 4 Earth radii is about 7Hz and

2Hz at about 6 Earth radii.

3.5.2. Plasma waves generated by auroral electrons

Auroral electrons excite lower and upper hybrid

frequencies by the Landau resonance process. The lower

and upper hybrid frequencies are about 1–2 kHz and

100–600 kHz, respectively, in the altitude range

3000–10000 km. Auroral hiss is whistler mode wave

excited by auroral electrons in the frequency range

between the lower hybrid frequency and the electron

plasma frequency. The free-space wave excited in the

frequency range higher than the electron gyrofrequency

or the electron plasma frequency (180–500 kHz) is called

auroral kilometric radiation (AKR).

Auroral hiss occurs in association with auroral

curtains or arcs and is characterized by a broad band

spectrum with a clear low-frequency cutoff. The wave

intensity is maximum near the lower cutoff frequency.

The average intensity of an auroral hiss is about

10�12–10�11W/m2Hz (Singh et al., 2001; Singh and

Singh, 2002). The characteristics of auroral hiss ob-

served in the dayside region is different from those

observed in the nightside region. The dayside hiss

usually consists of short duration bursts whereas the

nightside hiss is characterized by continuous and long

duration bursts. This difference may be due to different

behavior of auroral electrons precipitated in the dayside

and nightside regions.

AKR is the plasma wave observed from the satellite,

when passing through the auroral arcs. Since its

radiation occurs in kilometer wave length range, it is

known as Auroral Kilomeric radiation. AKR is

generated by the auroral electrons in the high altitude

region either in right-hand extraordinary (R–X) mode or

left-hand ordinary (L–O) mode. The wave is originally

generated by the auroral electrons in the downward

direction like auroral hiss and initially propagates

downward. In the generation region, wave frequency is

greater than the local electron gyrofrequency. As the

wave propagates downward, it encounters a medium

with increasing electron gyrofrequency and plasma

frequency. The wave is reflected back from the region

where electron gyrofrequency becomes equal to wave

frequency (R–X mode) or wave frequency becomes

equal to plasma frequency (L–O) mode. The reflection

region is estimated to be at 3000–20000 km altitude.

Thus, finally the wave is upward propagating electro-

magnetic mode. The average intensity observed by the

NASA Hawkeye satellite was 10�18–10�13W/m2Hz at

the radial distance of 7 Earth radii. The total power is

estimated to be 102–106 kW. This shows that AKR

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658648

carries away sufficiently large amount of energy from

the magnetosphere.

Another plasma wave phenomena associated with

auroral electrons is VLF saucer, V-shaped and funnel-

shaped hiss. The spectral structure of V-shaped hiss and

funnel-shaped hiss could be explained by the propaga-

tion effect—of the whistler mode waves if the wave is

excited near the resonance angle. Such waves could be

excited by low-energy electron (�several eV). A possible

source of the low-energy electron beam is ionospheric

upward electrons that carry the downward field-aligned

current located at the equator side of the upward field-

aligned current. This model is consistent with the fact

that a saucer, V-shaped and funnel-shaped auroral hiss

are usually emitted upward (LaBelle and Treumann,

2002).

3.5.3. ULF wave generation at the magnetospause

Pc1-2 may also be generated by the coupling of wave

energy propagating through the magnetosphere with the

field line resonances (Kivelson and Southwood, 1985,

1986; Zhu and Kivelson, 1988; Walker, 2000) either in

the solar wind/ magneto-sheath or at magnetopause

boundary layer (Southwood, 1974, 1975; Tsurutani et

al., 1998a, b) by one or more of the processes such as

magnetopause boundary motions, Kelvin–Helmholtz

instabilities (Lakhina and Schindler, 1996; Lakhina,

2003).

Inspite of theoretical developments such as discrete

mode excitation in the cavity between the magnetopause

and the upper region of the atmosphere where plasma

from the ionospheric and magnetospheric origin merge

with each other (i.e. turning point) (Kivelson and

Southwood, 1985, 1986) and excitation of field line

resonance by the cavity/wave guide modes (Zhu and

Kivelson, 1988; Walker, 2000), the observational fea-

tures such as the absence of VLF waves on the nightside

of the magnetosphere could not be explained. Further,

properties of observed pulsations that result from warm

plasma effect, finite Larmor radius effects, trapped

particle effects in the realistic magnetospheric geometry

are yet to be explained.

3.6. Man-made influences on the ionosphere and

magnetosphere

Humans are perturbing the Earth’s environment in

various aspects such as by sending rockets and satellites,

transmitting VHF/HF waves in the atmosphere for

communication and exploring of the upper atmospheric

phenomena. Urban railway system could induce ELF

waves in the range 0.01–5Hz (Ho et al., 1979). The

power line of the world’s power grid is the source of

ELF/VLF waves in the ionosphere/magnetosphere.

Acoustic gravity wave produced by large ground

explosions contributes to the heating of the upper

atmosphere. Nuclear explosions in the upper atmo-

sphere produce acoustic and electromagnetic waves,

perturbations of the ionosphere and the geomagnetic

field, artificial radiation belt, air-glows (Blanc, 1985;

Longmire, 1995). The green house gases can modify the

ionospheric layers either directly (Bremer, 1992) or with

indirect mechanisms because the global warming will

increase the lightning activity (Price and Rind, 1994).

We shall only be concentrating to the power line

harmonic radiation, VLF/HF transmitter and nuclear

explosion.

3.6.1. Power line harmonic radiation

Bullough (1995) obtained magnetic induction from

unbalanced harmonic currents flowing in the power lines

with ground return. The radiation field obtained in free

space at a height of 100 km–few gammas. Using this

idea, Molchanov et al. (1991) tried to explain the weekly

variation of ELF data recorded by a low-altitude

satellite. They considered that the attenuation of PLHR

penetrating into the ionosphere leads to the modification

of ionospheric conductivity and the ionospheric current

which results into a change of magnetospheric currents

through the ionsphere–magnetosphere coupling. The

magnetospheric current generates ELF turbulence which

could be transformed into propagating electromagnetic

emissions due to the inhomogenity of the plasma. The

variation of computed ionospheric current is similar to

the observed intensity of ELF emissions (Molchanov et

al., 1991; Parrot and Zaslavski, 1996).

3.6.2. VLF transmitters

The VLF waves transmitted from the ground-based

transmitters in the frequency range of 10–30 kHz are

used for radio navigation, communications and iono-

spheric/magnetosphereic investigations. VLF waves

propagating through the ionosphere produce triggering

of new waves (Omura et al., 1991), ionospheric heating

(Inan, 1990), wave–particle interactions, particle pre-

cipitation and wave amplifications.

The VLF wave propagating along the geomagnetic

field interacts with the counter streaming energetic

electron flux, which is effective in the equatorial region

when the Doppler-shfited wave frequency seen by the

particles is close to the electron gyrofrequency. During

interaction, particles undergo pitch angle diffusion

leading to precipitation of the particle into the atmo-

sphere (Singh et al., 1996). Inan et al. (1984) have

presented maps of global precipitation zones due to the

main VLF transmitters. The precipitated particles have

energy �keV (Singh and Singh, 2002).

Triggered emissions are related to non-linear wave

growth caused by resonant particle trapping in a non-

uniform magnetic field (Omura et al., 1991; Singh et al.,

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 649

2003). It is important to determine the motion of the

resonant particles and the associated resonant current,

which creates triggered emissions. The inhomogeneous

magnetic field forces the particles to have second-order

resonance condition which can be expressed as the

change in parallel velocity of the particle with time and

should match with the time change in resonance velocity

of the resonant particle. Further, the interaction length

should be larger than the trapping length (Singh et al.,

2003).

VLF transmitters are also known to decrease the wave

emissions in a frequency band just below the pulse

frequency under certain conditions (Raghuram et al.,

1977). This has also been explained by the non-linear

interaction process (Mattews et al., 1984). Other

phenomena such as parametric interaction and cross-

modulation can also be achieved using VLF transmit-

ters. Cross-modulation causes changes in collision

frequency and electron density of the ionosphere. For

cross-modulation, usually HF transmitters are used.

3.6.3. HF transmitters

Intense radiowaves have been launched from the

Earth’s-based transmitters into the ionosphere to

simulate electromagnetic emissions, artificial spread F ;field aligned density irregularities, air-glow emissions,

ionospheric cavity excitation, electrons and ions heating,

etc. (Gurevich, 1978; Fejer, 1979; Kuo, 1993). The

collisional heating of electrons and ions can also be

caused by the absorption of energy from the high-power

transmitted radio wave. This heating causes the electron

temperature and hence the electron collision frequency

to fluctuate at the modulation frequency of the high-

power radio wave. The effect is mostly observed in the

D-region of the ionosphere (Parrot and Zaslavski, 1996).

3.6.4. Nuclear explosion

Nuclear explosion is characterized by the release of

extremely high internal energy, which cause the gas to be

ionized resulting in to free electrons and highly charged

nuclei. Free electrons collide with the nuclei and emit

X-rays (photon), which may carry out 75% of the

energy. Most of the remaining energy is in kinetic energy

of the exploding device debris, which is transferred to air

mass. The X-ray energy is also transferred to the air

mass, loading to the expansion of hot air (fireball). The

conducting fireball radiates heat in the form of photons

with wavelengths in the ultraviolet, visible and infrared.

Gamma-rays released during nuclear explosion either

in the fission process or in inelastic scatter or capture of

neutrons in the device materials, while traveling through

the air collide with electrons and produce Compton

recoil electrons. These recoil electrons produce trans-

verse current density which become the source of

electromagnetic pulse (EMP) (Longmire, 1978). The

transverse Compton current radiates both outgoing and

incoming waves. These EMP propagate to large

distances from the explosion site (Blanc, 1985; Long-

mire, 1995) and thus help in electromagnetic coupling of

regions.

Other interesting phenomena associated with nuclear

explosion is lightning (Salanave, 1980). The discharges

are believed to have formed a sharp metallic structure

and were seen to grow upward at an apparent speed of

�10 km/s. The circular shape of the discharges indicates

that they are driven by the EMP in the quasi-static phase

(Gardner et al., 1984).

The motion of electrically conducting fireball and

strongly shocked air (during nuclear explosion in the

atmosphere) in the presence of geomagnetic field

generates electric fields, currents and magnetic perturba-

tions. The ionospheric conductivity is also modified

because X-ray bursts are absorbed in this region and

enhance the ionization and conductivity.

The acoustic and gravity waves produced by nuclear

explosions can trigger other waves (Pokhotelov et al.,

1994) and their attenuation contributes to the heating of

the upper atmosphere. Thus, the electromagnetic cou-

pling of the upper atmosphere and near space during

nuclear explosion in the atmosphere involves acoustic

and electromagnetic waves, perturbations of the iono-

spheric and geomagnetic field, artificial radiation belts,

air-glow process etc. (Tolstoy and Herron, 1970; Price,

1974; Blanc, 1985; Longmire, 1995).

4. DC electric field mapping in the Earth’s atmosphere

The dc electric fields generated by different sources

form a closed circuit which can be described by the GEC

model (Singh et al., 2004a) in which the ionosphere and

the ground behave as two conducting paths closed by

the upward and downward currents/fields (Fig. 4). In the

following, we shall briefly discuss upward and down-

ward mapping of the field separately.

4.1. Upward mapping of the field

The primary source of upward current is a total of

about 2000 thunderstorms which are active at any time

around the globe each contributing to an average of

about 0.5A. The efficiency of a thundercloud to supply a

fraction of its discharge current to GEC depends on

whether a cloud is a current generator or a voltage

generator (Willett, 1979). Further, the ionospheric

magnetic field line configuration modifies the vertical

current output from a thunderstorm. Tzur and Roble

(1985) have simulated the distribution of regional

vertical current from a dipole current source under both

the conditions when magnetic field is horizontal or when

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658650

magnetic field is vertical. In the lower atmosphere at low

latitudes, geomagnetic field is inclined whereas at the

high latitudes it may be considered as vertical. In Fig. 8

we have plotted the results of Tzur and Roble (1985).

From the figure it is noted that for a vertical magnetic

field, the flow of current is concentrated in the

immediate vicinity of the source, and most of the

upward current flows to the magnetically conjugate

point along the field lines before it spreads in the

dynamo region. This situation may correspond to high-

latitude regions. The lower portion of the figure

(computed for horizontal magnetic field) may approxi-

mately represent low latitude case, where the field

aligned currents are interrupted at about 100 km

altitude, and spreading starts above the source (thun-

dercloud).

Blakeslle et al. (1989) measured air conductivity and

electric field with a high altitude NASA U-2 airplane

flying over thunderstorms and had reported Wilson

current varying between 0.09 and 3.7A with an average

of 1.7A, and area-averaged Maxwell current varying

between 0.09 and 5.9A with an average of 2.2A. They

have also reported that the relative efficiency of a

thunderstorm to supply current to the GEC is inversely

related to the storm flash rate. The current generated

within the cloud is divided between production of

lightning and maintenance of the Wilson current.

Intra-cloud discharges do not support Wilson current.

Since the ratio of intracloud to cloud-to-ground

discharges increases from about 0.1 in the equatorial

Fig. 8. Contour of log10 J in (A/m2), the regional vertical

current flow toward the ionosphere from a thunderstorm model

considering an ionosphere with Horizontal magnetic field lines

(lower panel) and vertical magnetic field lines (upper panel)

(Tzur and Roble, 1985).

region to about 0.4 near 501 latitude and thunderstorm

activity is the maximum near the equator and decreases

with latitudes, the supply of Wilson current varies with

latitude having a peak at low latitudes.

4.2. Downward mapping of the filed

The downward mapping of the electric fields depends

upon the height variation of electrical conductivity and

its anisotropic nature at higher altitudes alongwith the

ground conductivity. The ground conductivity depends

upon the Earth’s orography and hence downward

mapping will be controlled by the Earth’s orographic

surface and associated changes in columnar resistance.

To illuminate this point, Roble and Hays (1979)

simulated the distribution of electric field and current

density along the Earth’s orographic surface. In the

computation, the effect of ionospheric dynamo and

magnetospheric plasma convection were added linearly

and the effects of the tilted geographic and geomagnetic

poles were considered. Fig. 9 shows the contours of the

downward mapping of the ionospheric potential pattern.

The imposed ionospheric potential pattern as the

boundary condition at 1900UT along the constant

conductivity (4.54� 10�6mho/m) surface at approxi-

mately 105 km altitude is shown in Fig. 9a (Roble and

Hays, 1979). The computed electric fields and air-earth

current density along the Earth’s orographic surface is

shown in Fig. 9b and c from which it is noted that under

the maximum positive ionospheric potential, the calcu-

lated surface electric field is +20V/m (positive iono-

spheric potential region is considered when the air-earth

current flows into the ground) and under the minimum

negative potential the calculated surface electric field is

�20V/m (Roble and Hays, 1979). Further, the max-

imum ground current density occurs over the mountai-

nous regions of Antarctica and Rocky mountains.

The calculated air-earth current shows considerable

variations as a result of the Earth’s orography that

is associated with changes in the columnar resistance.

In the vicinity of the high Antarctic mountain

plateau, the air-earth current has positive perturba-

tions of �0.2� 10�12 A/m2 on the dawnside and

��2� 10�12A/m2 on the duskside. This potential

pattern should move over the Earth’s surface during

the day, rotating about the geomagnetic pole in a

complex but systematic manner. Due to orographic

variations, the globally integrated ground current varies

between net an upward and downward value, which in

turn causes difference in the fair weather ionospheric

potential to vary between positive and negative values in

a diurnal cycle.

The conductivity variations control the efficiency of

mapping of the electric field. Horizontal electric fields of

small scale sizes (o500 km) are rapidly attenuated as

ARTICLE IN PRESS

Fig. 9. Contour illustrating the downward mapping of the

ionospheric potential pattern, (a) imposed ionospheric potential

(kV) at ionospheric height (4100 km), (b) calculated electric

field (V/m) along the Earth’s orographic surface, and (c)

calculated ground current (A/m2� 10�13) along the Earth’s

orographic surface. All figures are plotted at 1900UT in

geomagnetic coordinates (Roble and Hays, 1979).

D. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 651

they map downward to the ground from the ionospheric

heights whereas electric fields of large horizontal scale

sizes (500–1000 km) map effectively right down to the

Earth’s surface.

5. Wave energy transfer in the Earth’s atmosphere

The wave energy is transferred during the propagation

of waves from one region to the other. The propaga-

tional features and hence wave energy transfer depends

upon both the ambient medium parameters (such as

electron/ion densities, temperature, magnetic field and

its orientation with respect to wave vector, inter-particle

collision frequency, etc.) and wave parameters (ampli-

tude, polarization, frequency and phase). This shows

that as the wave propagates from one region to the other

region because of the change in electrodynamic proper-

ties of the medium, propagational features (such as

transmission, reflection, scattering, attenuation and

amplification) of the wave also changes. In the

following we shall discuss how different regions are

linked through the transfer of wave-energy and wave

propagation.

5.1. Wave energy transfer in the Earth-ionosphere

waveguide

The conductivity of the Earth’s surface depends upon

its orographic nature and wave frequency. At the ELF

and VLF ranges both the Earth’s surface and ionosphere

behave as a conductor, acting as the lower and upper

boundaries of a waveguide. The presence of finite

conductivity, which increases with height, makes the

lower and middle atmosphere to act as dissipative

waveguide. In this height region, energetic electrons

are precipitated and complicate the transmitting proper-

ties of the medium (Imhof et al., 1978). Electromagnetic

waves in the ELF/VLF range generated from a source

on or near the Earth’s surface such as lightning/man

made transmitters, propagate over long distances

through the wave-guide by the process of internal

reflection (Budden, 1985; Barr et al., 2000). Attenuation

of the wave also takes place as it propagates. For short

distance propagation the ray theory can be used whereas

for large distance wave mode theory should be appro-

priate, which should account for a dipolar geomagnetic

field, features of boundaries such as the day–night

terminator at the ionosphere, land–sea distribution and

orographic structure (Tolstory et al., 1982). Thus, the

waves, while propagating in the Earth-ionosphere

waveguide redistribute electromagnetic energy in the

lower/middle atmosphere from the source to the whole

waveguide volume.

Another class of waves which propagate in the

Earth-ionosphere wave guide is SRs which are the

oscillations of the Earth-ionosphere cavity—can

propagate globally with extremely low (o 1 dB/

103 km) attenuation rates (Jones, 1999). The continuous

SR background noises of amplitudes of �1mV/m or

3 pT are a superposition of individual pulses arriving

from random lightning strokes whose occurrence

rate is about 100 strokes/s all over the world (Belyaev

et al., 1999).

ARTICLE IN PRESS

Fig. 10. Dynamic spectra (upper plate) and Fine structure

component of dynamic spectra (lower plate) of a whistler wave

recorded at Varanasi on 19 February, 1997.

D. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658652

5.2. Whistler mode wave energy distribution along dipolar

field line

Part of the wave energy generated by the natural and

man-made sources on and near the Earth’s surface

penetrate the ionosphere and when wave normal angle

with respect to geomagnetic field lines is small, these

waves are trapped in the ducts formed by the electron

density irregularities. Transient electric field of lightning

helps in the formations of ducts. These waves propagate

to the conjugate points without appreciable attenuation

and spreading in wave energy. Depending upon the

location of source, the waves propagate through

different regions of the magnetosphere. Although, the

source may remain the same, the wave may follow

different path and propagate through the different

regions of the magnetosphere (Singh et al., 2004b).

Whistler mode waves, while propagating through the

magnetosphere interact with the beam of energetic

electrons and lead to wave generation and precipitation

of electrons in keV range in the atmosphere. The

precipitated electrons may produce enhanced ionization,

heat and optical emissions such as aurora/borealis in the

lower ionosphere, as well as X-rays detectable down to

about 13 km altitude and modify the electrical con-

ductivity (Rycroft, 1973; Burke, 1992; Rycroft et al.,

2000). The interacting waves may also accelerate the

electrons under suitable geophysical conditions.

The whistler mode waves finally disappear in the

medium after many bounces. Thus, the electromagnetic

energy from a localized source is distributed into various

parts of the magnetosphere. Lightning and other sources

are continuously injecting electromagnetic energy which

is being distributed/dispersed by the wave resulting in

the alteration of electrodynamic properties of the

medium.

When the wave is propagating through the medium,

its amplitude/phase gets changed. Dispersion analysis of

whistler waves recorded at the ground station has been

widely used to probe the magnetosphere (Sazhin et al.,

1992; Singh et al., 1998, 2004b). Usually a lightning may

illuminate more than one duct in the magnetosphere and

the resulting whistler trace consists of multiple compo-

nents. Sometimes a single duct may contain a number of

fine structure ducts. In such cases, high-resolution

analysis of whistler trace may consist of several

components separated in time due to differences in

travel time through different ducts (Figs. 10a and b).

Thus, an analysis of wave can yield information about

fine structures present in electron density irregularities.

5.3. Downward wave energy transport

Electromagnetic waves excited within the ionosphere

and the magnetosphere/magnetopause, propagate

downward and are received either on the ground surface

or onboard rockets/satellites in the ionosphere/magneto-

sphere. These waves have wide variety of tonal

characteristics and propagate in whistler mode but are

distinctly different than the whistlers. Some of the waves

are absorbed in the medium and others can propagate

along the geomagnetic field line and are finally observed

on the Earth’s surface.

The waves in the space plasma are generated through

the process of wave–particle interaction. Location of

interaction region and propagation mechanism control

the shape of dynamic spectrum such as hiss, chorus

(riser, faller, hook), etc. The waves are generated at

different altitudes in the equatorial region and in the

auroral region. If they propagate along field line, we can

receive them at different locations on the Earth’s

surface. The intensity of the wave varies from event-

to-event. The average hiss intensity in the equatorial

region and auroral region onboard satellite is about

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 653

10�12–10�11W/m2Hz (Singh et al., 1999b, 2001). If we

assume that the band width of an auroral hiss is 30 kHz

and the radiation region is a doughnut-like region of

width of 2000 km, the total power of the auroral hiss

(including V-shaped hiss) is estimated to be 103–104 kW.

In the case of equatorial or low latitude hiss, the

bandwidth is less than 10 kHz and radiating volume will

also be less. Hence the generated power in low-latitude

region would be less than 103 kW. This shows that

substantial amount of wave energy is generated in the

magnetosphere and transported downward.

Lion roar signals are the packets of whistler mode

waves in the frequency range between 90 and 160Hz

associated with magnetospheric substorms and propa-

gate along the magnetic field lines down the cusp

(Volland, 1987). Another type of wave observed only in

cusp region is the electromagnetic wave excited by

plasma instabilities in hot magnetospheric turbulent

plasma having frequencies near the half-harmonics of

the electron gyrofrequency ([f ¼ (2Z+1)fe/2), where

Z ¼ 1; 2; . . .]. Electrostatic ion–cyclotron waves in the

cusp region heat the protons and change proton

temperature (Shawhan, 1979).

Another example of downward propagating wave is

the ULF waves produced during the coupling of eigen

modes of magnetospheric cavity with geomagnetic field

line resonances involving complex coupled boundary

region (Kivelson and Southwood, 1985, 1986; Walker,

2000). These boundary layer waves play an important

role in the cross-field transport processes. In fact, these

waves can diffuse the magneto-sheath plasma across the

closed magnetospheric field lines at a rate rapid enough

to create the low latitude boundary layer itself (Tsur-

utani and Lakhina, 1997). This provides a specific

mechanism for viscous interaction (Tsurutani and

Gonzalez, 1995) in which the solar wind energy is

transferred to the magnetosphere. For the up-stream

generated waves, one important task is to establish

linkage between waves external and internal to the

magnetosphere and to find the site of transfer of energy.

One model would have compressional oscillations in the

fore shock to propagate directly through the shock,

sheath, and sub-solar magnetopause into the lower

magnetosphere, while the other would have wave

entering along cusp/cleft/ boundary layer field lines

transferring energy to the interior dayside magneto-

sphere via an ionospheric process.

Song and Vasyliunas (2002) have discussed the

propagation of the perturbations of different frequencies

from the magnetopause to the ionosphere after recon-

nection at the magnetopause has been switched on.

Reconnection produces perturbations over a broad

frequency range. It is found that the highest frequencies

reach the ionosphere first followed by the whistler mode

waves. MHD waves arrive last at the Alfven speed. This

result can explain recent observations indicating the

presence of some processes in which the magnetosphere

and the ionosphere communicate faster than the Alfven

wave travel time (Lockwood and Cowley, 1999;

Watanabe et al., 2000). The highest frequencies carried

electric field perturbations where as the current pertur-

bations are mostly carried by the whistler mode. The

strongest magnetic and plasma velocity perturbations

are carried by the MHD modes (Song and Vasyliunas,

2002).

Thus, ULF waves derive their energy from the solar

wind and while propagating from the magnetopause

towards the Earth’s surface transport energy to the

ionosphere/atmosphere/ground.

5.4. Upward wave energy transport

VLF saucer emissions (name is based on the spectral

structure on the frequency–time display-V-shape) are

generated by low-energy electrons (�eV) in the auroral

region about an altitude of 1400 km. They propagate

upward with vector near resonance angle. These waves

could not propagate for long distances and are mostly

observed by the satellites (LaBelle and Treumann, 2002).

A possible source of the low-energy electron beam is

ionsopheric upward electrons that carry the downward

field aligned current both on the nightside and dayside.

Saucers occur in a doughnut-like region surrounding the

geomagnetic pole similar to the auroral hiss emission

region, although, the size of the emission region is much

smaller than the V-shaped downward propagating hiss.

Hence, the energy transferred will be smaller than that

of auroral hiss.

AKR is another wave mode in the frequency range of

about 50–500 kHz which propagates upwards. The brief

description of the AKR is given in Section 3.5.2.

Another upward propagating mode is acoustic gravity

waves generated during ground explosions and the ascent

of large rockets (Pokhotelov et al., 1995). These waves

propagate away from the source and carry momentum

and energy which will be transferred to the atmosphere

through interactions. Klostermeyer (1972) had calculated

electron density perturbations and heating caused by

acoustic gravity waves in the F-2 layer. The presence of

these waves also affects the ionization rate due to the

change of the electron density (Nagorskiy, 1985). Thus,

acoustic gravity waves also affect the electrodynamic

properties of the lower atmosphere by transporting

energy and momentum from the man made or natural

sources lying near the Earth’s surface.

6. Summary

In this paper, we have briefly summarized electrical

nature of the Earth’s atmosphere, natural and man-

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658654

made sources of electromagnetic energy near the Earth’s

surface and in the ionosphere/magnetosphere. The

transport of energy from one region to the other region

is also discussed. It is clearly seen that the Earth’s near

environment (lower atmosphere to the magnetosphere)

is electrodynamically coupled and perturbation intro-

duced at one place is effectively communicated to the

remote places and alters the electrical properties of the

whole system. The generation and transport phenomena

related with the DC fields along with low- and high-

frequency fields are considerably complex and needs

further studies. The interpretation of observed wave-

form requires complete knowledge of electromagnetic

coupling of lower and upper atmosphere along with its

frequency-dependent transmission (reflection, refraction

and scattering) characteristics.

Thunderstorm is an efficient and effective generator of

ELF/VLF waves. The exact mechanism of generation

and its efficiency is not well understood specially under

the conditions when discharges are cloud-to-ground.

Cloud–to-cloud or intra-cloud. Even the coupling of

energy from the lightning to the ducts present in the

magnetosphere and then exit of the electromagnetic

energy from the duct and propagation to the Earth’s

surface is not well understood. Recently (Ferencz et al.,

2002; Singh et al., 2004b), various cases of anomalistic

VLF wave propagation has been presented which need

proper explanation.

Earthquake is one of the source to modify the

electrodynamics of the Earth’s near space, which is

unexplored. Recently, huge amount of literature related

with wave observations from ULF to VHF associated

with various Earthquakes have been reported. However,

physical mechanism of the wave generation during

Earthquake processes in the various frequency ranges

are not known. Even the processes involved in the

modification of the Earth’s ionosphere during the

Earthquake is not known. The propagation of waves

from the source to the ionosphere/magnetosphere

requires detail study. Unless these processes and many

other processes are understood properly, it is very

difficult to evaluate exact coupling of electromagnetic

energy from one region to the other region in the Earth’s

near environment.

References

Alimov, O.A., Gokhberg, M.B., Likerovskaia, E.V., Gufeld,

I.L., Liperovsky, V.A., Roubtov, L.N., 1989. Anomalous

characteristics of the middle latitude Es layer before

earthquakes. Physics of the Earth and Planetary Interiors

57, 76–81.

Barrington-Leigh, C.P., Inan, U.S., Stanley, M., Cummer, S.A.,

1999. Sprites triggered by negative lightning discharges.

Geophysical Research Letters 26, 3605–3608.

Barr, R., Llanwyn, J.D., Rodger, C.J., 2000. ELF and VLF

radio waves. Journal of Atmospheric and Solar-Terrestrial

Physics 62, 1689–1718.

Belyaev, G.G., Schekotov, Yu.A., Shvets, A.V., Nickokaeko,

A.P., 1999. Schumann resonances observed using pointing

vector spectra. Journal of Atmospheric and Solar-Terres-

trial Physics 61, 751–763.

Bering, E.A., Benbrook, J.R., Garret, J.A., Paredes, A.M.,

Wescott, E.M., Moudry, D.R., Sentman, D.D., Stenback-

nielsen, H.C., 2002. The electrodynamics of sprites.

Geophysical Research Letters 29.

Blakeslle, R.J., Chritian, H.J., Vonnegnt, B., 1989. Electrical

measurements over thunderstorms. Journal of Geophysical

Research 94, 13135–13140.

Blanc, E., 1985. Observations in the upper atmosphere of

infrasonic waves from natural or artificial sources: a

summary. Annales Geophysicae 3, 673–688.

Boccippio, D.J., Williams, E.R., Heckman, S.J., Lyons, W.A.,

Baker, I.T., Boldi, R., 1995. Sprites, ELF transients, and

positive ground strokes. Science 269, 1088–1091.

Boccippio, D.J., Wong, C., Williams, E.R., Boldi, R., Christian,

H.J., Goodman, S.J., 1998. Global validation of single-

station Schumann resonance lightning location. Journal of

Atmospheric and Solar-Terrestrial Physics 60, 701–712.

Borisov, N., Chmyrev, V., Rybachek, S., 2001. A new

ionospheric mechanism of electromagnetic ELF precursors

to earthquakes. Journal of Atmospheric and Solar-Terres-

trial Physics 63, 3–10.

Boys, C.V., 1926. Progressive lightning. Nature 118,

749–750.

Bremer, J., 1992. Ionospheric trends in mid latitudes as a possible

indicator of the atmospheric greenhouse effects. Journal of

Atmospheric and Terrestrial Physics 54, 1505–1511.

Budden, K.G., 1985. The Propagation of Radio Waves.

Cambridge University Press, Cambridge, UK.

Bullough, K., 1995. Power–line harmonic radiation. In: Vol-

land, H. (Ed.), Hand Book of Atmospheric Electrody-

namics. vol. II. second ed., CRC Press, Boca Raton, FL, pp.

292–332.

Burke, C.P., 1992. Early trimpi events from lightning-induced

electric field in the ionosphere: an alternative explana-

tion. Journal of Atmospheric and Terrestrial Physics 54,

243–250.

Carslaw, K.S., Harrison, R.G., Kirkby, J., 2002. Cosmic rays,

clouds and climate. Science 298, 1732–1737.

Chmyrev, V.M., Isaev, N.V., Serebryakova, O.N., Sorokin,

V.M., Sobolev, Ya.P., 1997. Small-scale plasma inhomo-

geneities and correlated ELF emissions in the ionosphere

over the earthquake region. Journal of Atmospheric and

Solar-Terrestrial Physics 59, 967–974.

Cho, M., Rycroft, M.J., 1998. Computer simulation of electric

field structure and optical emission from cloud top to

ionosphere. Journal of Atmospheric and Solar-Terrestrial

Physics 60, 871–888.

Cobb, W.E., Wells, H.J., 1970. The electrical conductivity of

oceanic air and its correlation to global atmospheric

pollution. Journal of Atmospheric Science 27,

814–819.

Cummer, S.A., Inan, U.S., Bell, T.F., Barrington-Leigh, C.P.,

1998. ELF radiation produced by electrical currents in

sprites. Geophysical Research Letters 25, 1281–1285.

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 655

Davies, K., Baker, D., 1965. Ionospheric effects observed

around the time of the Alaskan earthquake of March 28,

1964. Journal of Geophysical Research 70, 2251–2253.

Davies, P.M., Jackson, D.D., Johnston, M.J.S., 1980. Further

evidence of localized geomagnetic field changes before the

1974 thanksgiving days earthquake, Hollister, California.

Geophysical Research Letters 7, 513–516.

Dea, J.Y., Boerner, W., 1999. Observations of anomalous ULF

signal preceding the Northride earthquake of January 17,

1994. In: Hayakawa, M. (Ed.), Atmospheric and Inospheric

Electromagnetic phenomena Associated with Earthquakes.

Terra Scientific Publication, Tokyo, Japan, p. 137.

Dungey, J.W., 1961. Interplanetary magnetic field and the

auroral zones. Physical Review Letters 6, 47–48.

Dungey, J.W., 1978. The history of the magnetopause

region. Journal of Atmospheric and Terrestrial Physics 40,

231–234.

Enomoto, Y., Hashimoto, H., 1990. Emissions of charged

particle from indentation fracture. Nature 346, 641–643.

Fejer, J.A., 1979. Ionospheric modification and parametric

instabilities. Review of Geophysics and Space Physics 17,

135–153.

Ferencz, C.s., Hamar, D., Lichtenberger, J., Ferencz, O.E.,

Steinbach, P., Singh, R.P., 2002. Anomalistic phenomena in

whistler propagation. In: Proceeding of URSI Conference,

Maastrichet, 17–24 August 2002, XXVIIth General Assem-

bly, Netherlands, pp. 1–4.

Fishkova, L.M., Gokhberg, M., Pilipen, V.A., 1985. Relation-

ship between night air-blow and seismic activity. Annales

Geophysicae 3, 689.

Forbes, J.M., 1981. The equatorial electrojet. Review of

Geophysics and Space Physics 19, 469–504.

Fullekrug, M., Fraser-Smith, A.C., 1996. Further evidence for a

global correlation of the Earth-ionospheric cavity reso-

nance. Geophysical Research Letters 23, 2411–2414.

Galeev, A.A., Kuznetsova, M.M., Zelenyi, L.M., 1986.

Magnetopause stability threshold for patchy reconnection.

Space Science Reviews 44, 1–41.

Gardner, R.L., Frese, M.H., Gilbert, J.L., Longmire, C.L.,

1984. A physical model of nuclear lightning. Physics of

Fluids 27, 1694.

Gish, O.H., Wait, G.R., 1950. Thunderstorms and the

Earth’s electrification. Journal of Geophysical Research

55, 473–484.

Gokhberg, M.B., Morgunov, V.A., Yoshino, T., Tomizawa, I.,

1982. Experimental measurement of electromagnetic emis-

sions possibly related to earthquakes in Japan. Journal of

Geophysical Research 87, 7824–7828.

Gokhberg, M.B., Gufeld, I.L., Rozhnoy, A.A., Marenko, V.F.,

Yampolsky, V.S., Ponomarev, E.A., 1989. Study of seismic

influence on the ionosphere by super long-wave probing of

the earth ionosphere waveguide. Physics of the Earth

Planetary Interiors 57, 64–67.

Gonzalez, W.D., Joselyn, J.A., Kamide, Y., Krochl, H.W.,

Rostoker, G., Tsurutani, B.T., Vasyliunas, V.M., 1994.

What is a geomagnetic storm? Journal of Geophysical

Research 99, 5771–5792.

Gotoh, K., Akinaga, Y., Hayakawa, M., Hattori, K., 2002.

Principal component analysis of ULF geomagnetic data for

Izu islands earthquakes in July 2000. Journal of Atmo-

spheric Electricity 22, 1–12.

Gurevich, A.V., 1978. Non-Linear Phenomena in the Iono-

sphere. Springer, New York.

Hale, L.C., 1994. The coupling of ELF/VLF energy from

lightning and MeV particle to the middle atmosphere,

ionosphere and global circuit. Journal of Geophysical

Research 99, 21089–21096.

Harrison, R.G., Carslaw, K.S., 2003. Ion-aerosol-cloud pro-

cesses in the lower atmosphere. Reviews of Geophysics 41.

Hayakawa, M., 1999. Atmospheric and Ionospheric Electro-

magnetic Phenomena Associated with Earthquake. Terra

Scientific Publication, Tokyo, Japan.

Heckman, S.J., Williams, E.R., Boldi, R., 1998. Total global

lightning inferred from Schumann resonance measurements.

Journal of Geophysical Research 103, 31775–31779.

Hiraki, Y., Lizhu, T., Fukunishi, H., Nambu, K., Fujiwara, H.,

2002. Development of a new numerical model for investiga-

tion the energetic of sprites. Eos. Trans. AGU 83 (47), 0 Fall

Meet. Suppl. Abstract A11 C-0105.

Ho, A.M., Fraser-Smith, A.C., Villard, O.G., 1979. Large-

amplitude ULF magnetic fields produced by a rapid transit

system: close-range measurements. Radio Science 14,

1011–1015.

Holzworth, R.H., Hu, H., 1995. Global electrodynamics from

super pressure balloons. Advances in Space Research 16,

131–140.

Hoppel, W.A., Anderson, R.V., Willett, J.C., 1986. Atmo-

spheric electricity in the planetary boundary layer. In: Study

in Geophysics—The Earth Electrical Environment. Na-

tional Academy Press, Washington, DC, pp. 149–165.

Hu, H., Li, Q., Holzworth, R.H., 1989. Thunderstorm related

variations in stratosphere conductivity measurements.

Journal of Geophysical Research 94, 16429–16435.

Imhof, W.L., Reagan, J.B., Gainess, E.E., Larsen, T.R., Davis,

J.R., Moler, W., 1978. Coordinated measurements of ELF

transmission anomalies and the precipitation of energetic

particles in to the ionosphere. Radio Science 13, 717–727.

Inan, U.S., 1990. VLF heating of the lower ionosphere.

Geophysical Research Letters 17, 729–732.

Inan, U.S., Chang, H.C., Helliwell, R.A., 1984. Electron

precipitation zones around major ground-based VLF signal

sources. Journal of Geophysical Research 89, 2891.

Inan, U.S., Bell, T.F., Ridriguez, J.V., 1991. Heating and

ionization of the lower ionosphere by lightning. Geophysi-

cal Research Letters 18, 705–708.

Johnston, M.J.S., 1989. Review of magnetic and electric field

effect near active faults and volcanoes in USA. Physics of

the Earth and Planetary Interiors 57, 47.

Jones, D.L., 1999. ELF sferics and lightning effects on the

middle and upper atmosphere. Modern Radio Science,

171–189.

Kalashnikov, A.D., 1954. Potentialities of magnetospheric

methods for the problem of earthquakes forerunners.

Transactions of the Geofizical Institute of Akademiya Na

Naukite. SSSR 25, 180–182.

Kamra, A.K., Deshpande, C.G., 1995. Possible secular change

and land–to-ocean extension of air pollution from measure-

ments of atmospheric electrical conductivity over the

Bay of Bangal. Journal of Geophysical Research 100,

7105–7110.

Kamra, A.K., Ravichandran, M., 1993. On the assumption

of the Earth’s surface as perfect conductor in atmo-

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658656

spheric electricity. Journal of Geophysical Research 98,

22875–22885.

Kasemir, H.W., 1979. The atmospheric electric global circuit.

In: Proceedings of Workshop on the Need for Lightning

Observations from Space, NASA CP-2095. pp. 136–147.

Kivelson, M.G., Southwood, D.J., 1985. Resonant ULF waves:

a new interpretation. Geophysical Research Letters 12,

49–52.

Kivelson, M.G., Southwood, D.J., 1986. Coupling of global

magnetospheric MHD eigen modes to field line resonances.

Journal of Geophysical Research 91, 4345.

Klostermeyer, J., 1972. Numerical calculations of gravity wave

propagation in a realistic thermosphere. Journal of Atmo-

spheric and Terrestrial Physics 34, 765–774.

Koons, H.C., Roeder, J.L., 1999. A comparison of ULF/VLF

measurements associated with earthquakes. In: Hayakawa,

M. (Ed.), Atmospheric and Ionospheric Electromagnetic

Phenomena Associated with Earthquake. Terra Scientific

Publication, Tokyo, Japan.

Krider, E., Musser, J.A., 1982. The Maxwell current density

under thunderstorm. Journal of Geophysical Research 87,

11171–11176.

Kuo, S.D., 1993. Generation of ELF and VLF waves by a

thermal instability excited in the HF heater-modulated

polar electrojet. Radio Science 28, 1019–1027.

LaBelle, J., Treumann, R.A., 2002. Auroral radio emissions 1.

hiss, roara and bursts. Space Science Reviews 101, 295–440.

Lakhina, G.S., 1993. Electrodynamic coupling between

different regions of the atmosphere. Current Science 64,

660–666.

Lakhina, G.S., 2003. Solar wind magnetosphere coupling. In:

Singh, R.P., Singh, R., Singh, A.K. (Eds.), Solar Terrestrial

Environment: Space Weather. Allied Publisher, New Delhi,

India, pp. 69–104.

Lakhina, G.S., Schindler, K., 1996. Tearing modes at the

magnetopause. Journal of Geophysical Research 101,

2707–2712.

Larkina, V.I., Nalivayko, A.V., Dershenzon, N.I., Gokhberg,

M.B., Liperovskiy, V.A., Shalimov, S.L., 1983. Observa-

tions of VLF emissions related with casmic activity, on the

Interkosmos-19 satellite. Geomagnetism and Aeronomy 23,

684.

Lee, L.C., Johnson, J.R., Ma, Z.W., 1994. Kinetic Alfven

waves as a source of plasma transport at the dayside

magnetopause. Journal of Geophysical Research 99,

17405–17412.

Lester, M., Cowley, S.W.H., 2000. The solar terrestrial

interaction and its importance for space-weather. Advances

in Space Research 26, 79–88.

Lockwood, M., Cowley, S.W.H., 1999. A statistical study of the

ionospheric convection response to changing interplanetary

field condition using the assimilative mapping of iono-

spheric electrodynamics techniques. Journal of Geophysical

Research 104, 4387–4391.

Longmire, C.L., 1978. On the electromagnetic pulse produced

by nuclear explosions. IEEE Transmission Antennas

Propagation AP 26, 3.

Longmire, C.L., 1995. Electromagnetic effects of nuclear

explosions. In: Volland, H. (Ed.), Hand book of Atmo-

spheric Electrodynamics. vol. II. second ed., CRC Press,

Boca Raton, FL, pp. 135–153.

Lyons, W., 1996. Sprites observations above the US high plains

in relation to their parent thunderstorms systems. Journal of

Geophysical Research 101, 29641–29652.

Malan, D.J., 1937. Physics of Lightning. Academic of Science,

Paris (in English).

Markson, R., 1978. Solar modulation of atmospheric electrifi-

cation and possible implications for the Sun–weather

relationship. Nature 273, 103–109.

Markson, R., Muir, M., 1980. Solar wind control of the Earth’s

electric field. Science 206, 979–990.

Matsushita, S., 1967. Solar quiet and lunar daily variation

fields. In: Matsushita, S., Campbell, W.H. (Eds.), Physics of

Geomagnetic Phenomena. Academic Press, New York, p.

301.

Mattews, J.P., Omura, Y., Matsumoto, H., 1984. A study of

particle trapping by whistler mode waves in the geomagnetic

field: the early development of the VLF quiet band. Journal

of Geophysical Research 89, 2275–2287.

Miura, A., 1987. Simulation of Kelvin–Helmholtz instability at

the magnetospheric boundary. Journal of Geophysical

Research 92, 3195–3206.

Molchanov, O.A., Hayakawa, M., 1995. Generation of ULF

electromagnetic emissions by microfracturing. Geophysical

Research Letters 22, 3093–3094.

Molchanov, O.A., Parrot, M., Mogilevsky, M.M., Lefeuvre, F.,

1991. A theory of PLHR emissions to explain then weakly

variations of ELF data observed by a low-altitude satellite.

Annales Geophysicae 9, 669–680.

Muhleisen, R., 1977. The global circuit and its parameters. In:

Dolezalek, H., Reiter, R. (Eds.), Electrical Process in

Atmospheres. Steinkopff, Darmstadt, p. 467.

Nagano, I., Yagitani, S., Migamura, K., Makino, S., 2003.

Fullwave analysis of Elves created by lightning generated

electromagnetic pulses. Journal of Atmospheric and Solar-

Terrestrial Physics 65, 615–625.

Nagorskiy, P.M., 1985. Perturbations of the electron concen-

tration in the ionosphere induced by ground-level explo-

sions. Izvestiya Akademic of Science USSR—Phys. Solid

Earth 22, 860–863.

Negoda, A., Korepanov, V., Fedorov, M., 1999. Warning

mission-present state and development. In: Hayakawa, M.

(Ed.), Atmospheric and Ionospheric Electromagnetic Phe-

nomena Associated with Earthquakes. Terra Scientific

Publication, Tokyo, Japan, pp. 711–716.

Neubert, T., 2003. On sprites and their exotic kin. Nature 300,

747–748.

Nickolaenko, A.P., Satori, G., Zieger, B., Rabinowicz, B.,

Kudintseva, I.G., 1998. Parameters of global thunderstorm

activity deduced from the long-term Schumann resonance

records. Journal of Atmospheric and Solar-Terrestrial

Physics 60, 387–399.

Omura, Y., Nunn, D., Matsumoto, H., Rycroft, M.J., 1991. A

review of observational, theoretical and numerical studies of

VLF triggered emissions. Journal of Atmospheric and

Terrestrial Physics 53, 351–368.

Owen, C.J., Cowley, S.W.H., 1991. Heikkila’s Mechanism for

impulsive plasma transport through the magnetopause: a re-

examination. Journal of Geophysical Research 96, 5565.

Parrot, M., 1995. Use of the satellites to detect seismo-

electromgnetic effects. Advances in Space Research 15,

27–35.

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658 657

Parrot, M., Zaslavski, Y., 1996. Physical mechanism of man-

made influences on the Magnetosphere. Survey in Geophy-

sics 17, 67–100.

Parrot, M., Lefeuvre, F., Corcuff, Y., Goderefroy, P., 1985.

Observations of VLF emissions at the time of earth-

quakes in the Kerguelen Islands. Annales Geophysicae 3,

736–737.

Parrot, M., Achache, J., Berthelier, J.J., Blanc, E., Deschamps,

A., Lefeuvre, F., Menville, M., Plantet, J.L., Tarits, P.,

Villain, J.P., 1993. High-frequency seismo-electromagnetic

effects. Physics of the Earth and Planetary Interiors 77,

65–83.

Pasko, V.P., Inan, U.S., Bell, T.F., Taranenko, Y.N., 1997.

Sprites produced by quasi-electrostatic heating and ioniza-

tion in the lower ionosphere. Journal of Geophysical

Research 102, 4529–4561.

Pokhotelov, O.A., Pilipenko, V.A., Fedorov, E.N., Stenflo, L.,

Shukka, P.K., 1994. Induced electromagnetic turbulence in

the ionosphere and magnetosphere. Physica Scripta 50,

600–605.

Pokhotelov, O.A., Parrot, M., Fedorov, E.N., Pilipenko, V.A.,

Surkov, V.V., Gladychev, V.A., 1995. Response of the

ionosphere to natural and main made acoustic sources.

Annales Geophysicae 13, 1197–1210.

Price, G.H., 1974. The electromagnetic pulse from nuclear

detonations. Reviews of Geophysics and Space Physics 12,

389–400.

Price, C., 1993. Global surface temperature and the atmo-

spheric electric circuit. Geophysical Research Letters 20,

1363–1366.

Price, C., 2000. Evidence for a link between global lightning

activity and upper tropospheric water vapour. Nature 406,

290–293.

Price, C., Rind, D., 1994. Possible implication of global climate

change and global lightning distributions and frequencies.

Journal of Geophysical Research 99, 10823–10831.

Price, C., Asfur, M., Lyons, W., Nelson, T., 2002. An improved

ELF/VLF method for globally geolocating sprites produ-

cing lightning. Geophysical Research Letters 29 (3).

Raghuram, R., Bell, T.F., Helliwell, R.A., Katsufrakis, J.P., 1977.

A quiet band produced by VLF transmitters signal in the

magnetosphere. Geophysical Research Letters 4, 199–202.

Rastogi, R.G., Patel, V.L., 1975. Effect of interplanetary

magnetic field on ionosphere over the magnetic equator.

Proceeding of the Indian Academy of Sciences (Earth and

Planetary Sciences) 82, 121–141.

Richmond, A.D., 1976. Electric field in the ionospheric and

plasmaspheric on quiet days. Journal of Geophysical

Research 81, 1447–1450.

Richmond, A.D., 1986. Upper-atmospheric electric field

sources. In: Study in Geophysics—The Earth’s electrical

Environment. National Academy Press, Washington, DC,

pp. 195–205.

Rikitake, T., 1987. Magnetic and electric signals precursory to

earthquakes; an analysis of Japanese data. Journal of

Geomagnetism and Geoelectricity 39, 47–61.

Roble, R.G., 1991. On modeling component processes in the

Earth’s global electric circuit. Journal of Atmospheric and

Terrestrial Physics 53, 831–847.

Roble, R.G., Hays, P.B., 1979. A quasi-static model of global

atmospheric electricity, 2: electrical coupling between upper

and lower atmosphere. Journal of Geophysical Research 84,

7247–7254.

Roble, R.G., Tzur, I., 1986. The Global Atmospheric Electrical

Circuit. In Study in Geophysics—The Earth’s Electrical

Environment. National Academy Press, Washington, DC,

pp. 206–231.

Rodger, C.J., 1999. Red sprites, upward lightning and VLF

Perturbations. Reviews of Geophysics 37, 317–336.

Rodger, G.J., Thomson, N.R., Dowden, R.L., 1996. A search

of ELF/VLF activity associated with earthquakes using

ISIS satellite data. Journal of Geophysical Research 101,

13369–13378.

Rowland, H.L., 1998. Theories and simulations of Elves, sprites

and blue jets. Journal of Atmospheric and Solar-Terrestrial

Physics 60, 831–844.

Rycroft, M.J., 1973. Enhanced energetic electron intensities at

100 km altitude and a whistler propagating through the

plasmasphere. Planetary and Space Science 21, 239–251.

Rycroft, M.J., Cho, M., 1998. Modeling electric and magnetic

fields due to thunder cloud and lightning from cloud tops to

the ionosphere. Journal of Atmospheric and Solar-Terres-

trial Physics 60, 889–893.

Rycroft, M.J., Israelsson, S., Price, C., 2000. The global

atmospheric electric circuit, solar activity and climate

change. Journal of Atmospheric and Solar-Terrestrial

Physics 62, 1563–1576.

Salanave, L.E., 1980. Lightning and its Spectrum. University of

Arizona Press, Tucson.

Satori, G., 1996. Monitoring Schumann resonance—II: daily

and seasonal frequency variations. Journal of Atmospheric

and Terrestrial Physics 58, 1483–1488.

Sazhin, S.S., Hayakawa, M., Bullough, K., 1992. Whistler

diagnostics of magnetospheric parameters: a Review.

Annales Geophysicae 10, 293–308.

Schindler, K., 1979. On the role of irregularities in plasma entry

into the magnetosphere. Journal of Geophysical Research

84, 7257.

Schlegel, K., Fullekrug, M., 1999. Schumann resonance

parameter changes during high energy particle precipitation.

Journal of Geophysical Research 104, 10111–10118.

Sentman, D.D., Wescott, E.M., Osborne, D.L., Hampton,

D.L., Heavner, M.J., 1995. Preliminary results from the

sprites 94 aircraft campaign, 1 Red sprite. Geophysical

Research Letters 22, 1205–1208.

Shawhan, S.D., 1979. In: Lanzerotti, L.J., et al. (Eds.), Solar

Plasma Physics, Vol-III. North-Holland Publication, Am-

sterdam.

Singh, D.K., 1999. The study of whistler waves & VLF

emissions observed at Varanasi, India. Ph.D. Thesis,

Banaras Hindu University, India.

Singh, R.P., 2003. Lightning-whistlers, trimpis and optical

emissions. In: Singh, R.P., Singh, R., Singh, A.K. (Eds.),

Solar Terrestrial Environment: Space Weather. Allied

Publisher, New Delhi (India), pp. 366–388.

Singh, D.K., Singh, R.P., 2002. Hiss emissions during quiet and

disturbed period. PRAMANA—Journal of Physics 59,

563–573.

Singh, U.P., Singh, A.K., Singh, D.P., Singh, R.P., Singh, R.N.,

1996. Wave–particle interaction and charge particle pre-

cipitation. Indian Journal of Radio and Space Physics 25,

352–355.

ARTICLE IN PRESSD. Singh et al. / Journal of Atmospheric and Solar-Terrestrial Physics 67 (2005) 637–658658

Singh, R.P., Singh, A.K., Singh, D.K., 1998. Plasmaspheric

parameters as determined from whistler spectrograms: a

review. Journal of Atmospheric and Solar-Terrestrial

Physics 60, 495–508.

Singh, R.P., Singh, D.K., Singh, A.K., Hamar, D., Lichten-

berger, J., 1999a. Application of matched filtering and

parameter estimation technique to low latitude whistlers.

Journal of Atmospheric and Solar-Terrestrial Physics 61,

1081–1092.

Singh, D.K., Singh, Ashok, K., Patel, R.P., Singh, R.P., Singh,

Abhay, K., 1999b. Two types of ELF hiss observed at

Varanasi, India. Annales Geophysicae 17, 1260–1267.

Singh, D.K., Patel, R.P., Singh, R.P., 2001. Auroral and low

latitude VLF hiss. Indian Journal of Radio and Space

Physics 30, 24–30.

Singh, R.P., Patel, R.P., Singh, A.K., Das, I.M.L., 2002.

Lightning generated ELF, VLF, optical waves and their

diagnostic features. Indian Journal of Physics 76, 235–249.

Singh, R.P., Patel, R.P., Singh, D.K., 2003. Triggered emission

observed at Varanasi, India. Planetary and Space Science

51, 495–503.

Singh, D.K., Singh, R.P., Kamra, A.K., 2004a. The electrical

environment of the Earth’s atmosphere: a review. Space

Science Reviews 113 (3–4), 375–408.

Singh, R.P., Singh, K., Hamar, D., Lichtenberger, J., 2004b.

Matched filtering analysis of whistler and its propagation

characteristics at low latitudes. Journal of Atmospheric and

Solar-Terrestrial Physics, Revised.

Sobolev, G.A., Husamiddinov, S.S., 1985. Pulsed electromag-

netic earth and ionosphere field disturbances accompanying

strong earthquakes. Earthquake Prediction Research 3, 33.

Somayajulu, V.V., Reddy, C.A., Viswanathan, K.S., 1985.

Simultaneous electric field changes in the equatorial electrojet

in phase with polar caps latitude changes during a magnetic

storm. Geophysical Research Letters 12, 473–475.

Song, P., Vasyliunas, V.M., 2002. Solar wind–magnetospher-

e–ionosphere coupling: signal arrival and perturbation

relations. Journal of Geophysical Research 107, 1358.

Southwood, D.J., 1974. Some features of field line resonances in

the magnetosphere. Planetary and Space Science 22, 483–491.

Southwood, D.J., 1975. Comments on field line resonances and

micropulsations. Geophysical Journal of the Royal Astro-

nomical Society 41, 425.

Strangeways, R.J., Raeder, J., 2001. On the transition from

collision to collisional magnetohydrodynamics. Journal of

Geophysical Research 106, 1955–1960.

Su, H.T., Hsu, R.R., Chen, A.B., Wang, Y.C., Hsiao, W.S.,

Lai, W.C., Lee, L.C., Sato, M., Fukunishi, H., 2003.

Gigantic jets between a thundercloud and the ionosphere.

Nature 423, 974–976.

Takeda, M., Maeda, H., 1980. Three dimensional structure of

ionospheric currents—1: currents caused by diurnal tidal

winds. Journal of Geophysical Research 85, 6895–6899.

Taranenko, Y.N., Inan, U.S., Bell, T.F., 1993. Interaction with

the lower ionosphere of electromagnetic pulses from light-

ning: heating, attachment and ionization. Geophysical

Research Letters 20, 1539–1542.

Tinsley, B.A., 2000. Influence of the solar wind on the global

electric circuit, and inferred effects on cloud microphysics,

temperature, and dynamics of the troposphere. Space

Science Reviews 94, 231–258.

Tinsley, B.A., Heelis, R.A., 1993. Correlations of atmospheric

dynamics with solar activity. Evidence for a connection via

the solar wind, atmospheric electricity, and cloud micro-

physics. Journal of Geophysical Research 98, 10375–10384.

Tolstoy, I., Herron, T.J., 1970. Atmospheric gravity waves from

nuclear explosions. Journal of Atmospheric Science 27, 55–61.

Tolstory, V., Rosenberg, T.J., Carpenter, D.L., 1982. The

influence of localized precipitation-induced D-region ioni-

zation enhancements on subionospheric VLF propagation.

Geophysical Research Letters 9, 563–566.

Tronin, A.A., Hayakawa, M., Molchanov, O.A., 2002. Thermal

IR Satellite data application for earthquake research in

Japan and China. Journal of Geodynamics 33, 519–534.

Tsurutani, B.T., Gonzalez, W.D., 1995. The efficiency of

‘‘viscous interaction’’ between the solar wind and the

magnetosphere during intense northward IMF events.

Geophysical Research Letters 22, 663–666.

Tsurutani, B.T., Lakhina, G.S., 1997. Some basic concepts of

wave–particle interactions in collisionless plasmas. Reviews

of Geophysics 35, 491–502.

Tsurutani, B.T., Lakhina, G.S., Ho, C.M., Arballo, J.K.,

Galvan, C., Boonsiriseth, A., Pickett, J.A., Gurnett, D.A.,

Peterson, W.K., Thorne, R.M., 1998a. Broadband plasma

waves observed in the polar cap boundary layer: polar.

Journal of Geophysical Research 103, 17351–17366.

Tsurutani, B.T., Arballo, J.K., Lakhina, G.S., Ho, C.M., Buti,

B., Pickett, J.S., Gurnett, D.A., 1998b. Plasma waves in the

dayside polar cap boundary layer: bipolar and monopolar

electric pulses and whistler mode waves. Geophysical

Research Letters 25, 4117–4120.

Tzur, I., Roble, R.G., 1985. The interaction of a dipolar

thunderstorm with its global electrical environment. Journal

of Geophysical Research 90, 5989–5999.

Volland, H., 1987. Electromagnetic coupling between lower and

upper atmosphere. Physica Scripta T18, 289–297.

Walker, A.D.M., 2000. Coupling between waveguide models

and field line resonance. Journal of Atmospheric and Solar-

Terrestrial Physics 62, 779–813.

Watanabe, M., Sato, N., Greenwald, R.A., Pinnock, M.,

Hairston, M.R., Rairden, R.L., McEwen, D.J., 2000. The

ionospheric response to interplanetary magnetic field varia-

tions: evidence for rapid global change and the role of

preconditioning in the magnetosphere. Journal of Geophy-

sical Research 105, 22955–22978.

Wescott, E.N., Stenback-Mielsen, H.C., Sentman, D.D.,

Heavner, M.J., Moudry, D.R., Sabbas, F.T.S., 2001.

Triangulation of sprites, associated halos and their possible

relation to causative lightning and micrometeors. Journal of

Geophysical Research 106, 10467–10478.

Willett, J.C., 1979. Solar modulation of the supply current for

atmospheric electricity? Journal of Geophysical Research

84, 4999–5002.

Williams, E.R., 1992. The Schumann resonance: a global

thermometer. Science 256, 1184–1197.

Wood, C.A., 1951. Unusual lightning. Weather 6, 64.

Zhu, X., Kivelson, M.G., 1988. Analytic formulation and

quantitative solutions of the coupled ULF wave problem.

Journal of Geophysical Research 93, 8602–8612.

Zlotnicki, J., Le Mouel, J.L., 1990. Possible electrokinetic

origin of large magnetic variations at La Fournaise volcano.

Nature 343, 633.