Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 195
Upper-Atmosphere Electric-Field
Sources
14
ARTHUR D. RICHMOND
National Center for Atmospheric Research
The Earth's space environment is filled with electrons
and positive ions, comprising a plasma of very low den-
sity. These charged particles collide only infrequently
and are strongly influenced by magnetic and electric
fields. In turn, the charged particles affect the distribu-
tions of the magnetic and electric fields in space. The
space plasma environment, therefore, is dominated by
electrodynamic processes. The regions of space involved
in creating upper atmosphere electric fields are illus-
trated in Figure 14.1 and are described below.
The solar wind is a plasma with electron and ion
number densities of order 5 X 106 m~3 flowing contin-
ually outward from the Sun at a speed of 300-1000 km/
sec. Imbedded within it is the interplanetary magnetic
field (IMF), which is maintained by electric currents
flowing throughout the solar-wind plasma. The IMF
strength at the orbit of the Earth is roughly a factor of
10-4 smaller than the strength of the surface geomag-
netic field. Most of the time, an interplanetary field line
near the Earth can be traced back to the surface of the
Sun, where magnetic fields are ubiquitous. As the Sun
rotates (once every 27 days) different magnetic regions
influence the IMF near the Earth. The combination of
solar-rotation and outward solar-wind flow produce a
roughly spiral IMF pattern. In addition, the solar-wind
velocity can change dramatically and produce both
large-scale and small-scale distortions of the IMF so that
195
the field direction and strength vary greatly. These
changes have been found to influence the electrical state
of the magnetosphere and ionosphere.
The magnetosphere is the region of space where the
geomagnetic field has a dominant influence on plasma
properties. As the charged particles of the solar wind are
deflected by the geomagnetic field, an electric current
layer is formed at the boundary between the solar wind
and the magnetosphere, called the magnetopause. This
current layer distorts the geomagnetic field from the di-
pole-like configuration that it would otherwise have
and helps to create a long magnetized tail trailing the
Earth. Although the full extent of this tail has not yet
been determined, it is known to be more than 500 Earth
radii. The magnetosphere contains the radiation belt,
composed of energetic charged particles trapped in the
magnetic field. The number density of electron-ion
pairs in the magnetosphere is highly variable, ranging in
order of magnitude from a low of 106 m ~ 3 in parts of the
tail up to 10~2 m - 3 in the densest portions of the dayside
ionosphere.
The ionosphere is the ionized component of the
Earth's upper atmosphere. It is not distinct from the
magnetosphere, but rather forms the base of the magne-
tosphere in terms of electrodynamic processes. The
lower boundary of the ionosphere is not well defined but
can be taken as about 90 km altitude for the present pur
OCR for page 196
196
FIGURE 14.1 Configuration of the magnetosphere. The magnetic
field is shown by continuous lines, and locations of important magne-
tospheric features are pointed out.
poses, representing the level where the density of elec-
tron-ion pairs falls to roughly 10~° m ~ 3 and below which
electric currents become relatively small. The ioniza-
tion is formed largely by the effect on the upper atmo-
sphere of solar extreme-ultraviolet and x-ray radiation
at wavelengths shorter than 102.6 nm, but energetic
particles impacting the upper atmosphere from the
magnetosphere also create important enhancements.
The ionospheric plasma has a temperature on the order
of 1000 K, which is much cooler than the energetic
plasma farther out in the magnetosphere. Collisions be-
tween charged particles and neutral atmospheric mole-
cules become important below 200 km altitude and
strongly affect the electrodynamic characteristics of the
ionosphere.
At high latitudes, where magnetic-field lines connect
the ionosphere with the outer magnetosphere, the iono-
spheric features are quite complex. Ionospheric phe-
nomena become better organized in a coordinate system
based on the geomagnetic field than in geographic co-
ordinates, with the difference arising mainly from the
11° tilt of the dipolar field from the Earth's axis. Differ-
ent magnetic coordinate systems exist, but for descrip-
tive purposes the differences are not crucial, and the
simple terms "magnetic latitude" and "magnetic local
time" will suffice here.
At high magnetic latitudes, aurora are produced as
energetic charged particles, mainly electrons, precipi-
tate into the upper atmosphere from the outer magneto-
sphere, creating both visible emissions and ionization
enhancements. The aurora form a belt around the mag-
netic pole, called the auroral oval. The oval is in fact
ARTHUR D. RICHMOND
roughly circular, of variable size, and has a wider latitu-
dinal extent on the nightside than on the dayside of the
Earth. The entire oval is shifted toward the nightside, so
that aurora appear at lower magnetic latitudes at night
(roughly 67°) than during the day (roughly 78°~. The
nighttime particle precipitation also tends to be more
intense and widespread than on the dayside. Contained
within the auroral oval is the polar cap, where auroras
are less frequent but where on occasion very energetic
protons from solar flares enter and penetrate relatively
deep into the upper atmosphere.
A more detailed description of the Earth's space envi-
ronment can be found in several of the references listed
at the end of this paper, especially in the book of Akasofu
and Chapman (1972~.
ELECTRODYNAMIC PROCESSES IN SPACE
The charged particles of a plasma react strongly to
electric and magnetic fields. There is a strong tendency
for the particles to short out any electric fields, so that it
is often a good approximation to treat the electric field
in the frame of reference of the plasma as vanishing:
Ep~asma= 0. (14.1)
This is often called the magnetohydrodynamic (or
MHD) approximation (e. g., Roederer, 1979~. The
frame-of-reference choice is important if the plasma is
moving and if a magnetic field is present, because the
electric field observed in a different reference frame is
not the same. If we let E be the electric field in an Earth-
fixed reference frame, V be the velocity of the plasma
with respect to the Earth, and B be the magnetic-field
vector, then a (nonrelativistic) Lorentz transformation
yields
Ep~asma = E + V X B. (14.2)
where the vector product V X B results in a vector di-
rected perpendicular to both V and B. The electric field
is then simply related to the plasma velocity and mag-
netic field by the approximate relation
E = - V X B. (14.3)
Alternatively, the velocity component perpendicular
to B can be related to E and B as
V = (E X B)/B2. (14.4)
Equations (14.3) and (14.4) express the same fact from
two different points of view: the electric field and
plasma velocity are closely interrelated and help to de-
termine each other. In some cases, as in the solar wind
where plasma momentum is high, the electric field
quickly adjusts toward the value given by Eq. (14.3~. In
other cases, as in the upper ionosphere where electric
OCR for page 197
UPPER-ATMOSPHERE ELECTRIC-FIELD SOURCES
fields tend to be imposed on the plasma as a result of the
dynamo processes to be discussed, the plasma quickly is
set into motion at the velocity given by Eq. (14.4~.
A further important consequence of the MHD ap-
proximation, when combined with the Faraday law of
magnetic induction, is the following: all plasma parti-
cles lying along a common magnetic-field line at one
instant of time will forever remain on a common field
line. The magnetic field may vary both temporally and
spatially, and the plasma may move from one region of
space to another, but plasma ions and electrons will con-
tinue to share a field line with the same partners. This
result affects both the magnetic-field configuration and
the plasma velocity. In the solar wind, the magnetic
field is distorted to follow the motions of the plasma.
Nearer the Earth, where the magnetic field is so strong
that it is not easily distorted, the constraint means that
all particles on a dipolar field line must move simultane-
ously together to another field line. Convection of
plasma thus can be mapped between the outer magneto-
sphere and the ionosphere along magnetic-field lines.
The electric field similarly maps along the magnetic
field.
The MHD approximation is useful in interrelating
plasma motions with electric and magnetic fields, but it
breaks down under a number of important circum-
stances, especially where electric current densities are
large. Furthermore, the approximation does not explain
the distribution of currents, which are a central element
in the processes giving rise to upper-atmospheric electric
fields. Current flow across magnetic-field lines exerts a
force on the medium, a force that must either be bal-
anced by other forces, like pressure gradients, or else
result in acceleration of the medium. Consideration of
force and momentum balance thus is an important part
of understanding currents in plasmas.
One place where the MHD approximation breaks
down is in the lower ionosphere, below 150 km, where
collisions between ions and the much more numerous air
molecules are sufficiently frequent to prevent the ions
from maintaining the velocity given by Eq. (14.4~. In
this region electric current readily flows across mag-
netic-field lines. As we shall see, neutral-air winds in the
lower ionosphere lead to generation of electric currents
and fields, and for this reason the height range of
roughly 90-150 km is called the "dynamo region" of the
ionosphere.
Unlike the outer magnetosphere, the ionosphere be-
haves as an Ohmic medium, with the current density
linearly related to the electric field under most circum-
stances. The conductivity, however, is highly aniso-
tropic owing to the presence of the geomagnetic field.
The conductivity in the direction of the magnetic field is
very large, so that the electric-field component in this
197
direction is almost entirely shorted out, and magnetic-
field lines are nearly electric equipotential lines at all
altitudes above 90 km. The conductivity characteristics
perpendicular to the magnetic field depend on the rate
of ion-neutral collisions, and they change with altitude
as the neutral density varies. Figure 14.2 shows typical
mid-latitude conductivity profiles for day and night
conditions. In the nighttime auroral oval the conductiv-
ity is more akin to the "Day" profile in Figure 14.2 than
to the "Night" profile because of ionization production
associated with the aurora. Although the ionospheric
plasma density typically maximizes at around 300 km
altitude, the conductivity perpendicular to the mag-
netic field maximizes at around 110 km, with a large
day-night difference owing to the day-night difference
in ionospheric density. The current component perpen-
dicular to the magnetic field flows in a direction as
much as 88° different from the electric-field direction,
as also shown in Figure 14.2. This effect results from the
fact that electrons are relatively little influenced by col-
lisions above 80 km and move perpendicular to the elec-
tric field as given by Eq. (14.4), while positive ions are
strongly affected by collisions below 130 km and are un-
able to cancel current carried by the drifting electrons.
SOLAR WIND/MAGNETOSPHERE DYNAMO
In an analogy with a dynamo-electric machine that
generates electricity by rotating a conducting armature
through a magnetic field, the motion of a plasma
through a magnetic field produces an electromotive
300
-
y
-
~ 200
Cl
00
o
ANGLE BETWEEN
CONDUCTIVITY MAGNITUDE CURRENT AND
PERPENDICULAR TO MAGNETIC FIELD ELECTRIC FIELD
400 1 1 1 1
/ Night \Day
10-4 10-3 Go 45O 90O
10-7 10-6 10-5
S/m
FIGURE 14.2 Typical profiles of the ionospheric conductivity com-
ponent perpendicular to the geomagnetic field for day and night con-
ditions (left) and angle between the current and electric-field compo-
nents perpendicular to the geomagnetic field (right).
OCR for page 198
198
force and current flow and can also be considered a dy-
namo. The solar wind/magnetosphere dynamo results
from the flow of the solar wind around and partly into
the magnetosphere, setting up plasma motion in the
magnetosphere as well an electric field and currents
(e.g., Stern, 1977; Hill, 1979; Roederer, 1979; Cowley,
1982~. All details of this process are not yet understood,
but a number of features have become clear.
Strong evidence exists that the magnetosphere is
partly open, that is, that some magnetic-field lines
traced from the Earth extend indefinitely into inter-
planetary space (Hill and Wolf, 1977; Stern, 1977; Ly-
ons and Williams, 1984~. The amount of magnetic flux
that connects to the interplanetary magnetic field may
be as great as the entire flux passing through the polar
caps. Figure 14.3 shows a schematic figure of the mag-
netic-field configuration for the simple case where the
IMF is directed southward. If the IMF had an east-west
component, as it usually does, we would require a three-
dimensional representation of the magnetic-field inter-
connection. Details of the magnetic-field configuration
are not yet resolved, so Figure 14.3 should be treated
more as a conceptual tool than as a true representation
of the magnetosphere. The essential features are the
four classes of magnetic-field lines denoted in Figure
14.3: (1) closed field lines connected to the Earth in both
northern and southern hemispheres; (2) interplanetary
field lines unconnected to the Earth; (3) open field lines
connecting the northern polar cap to interplanetary
space; and (4) open field lines connecting the southern
polar cap to interplanetary space.
The interplanetary electric field, obtained from Eq.
(14.3), is directed out of the page in Figure 14.3. To the
extent that the MHD approximation is valid and electric
fields map along the magnetic field, the polar iono-
sphere is also subject to an electric field out of the page,
t ~
\ ~ ~
.~
::=
14
l ~ ~ ~ ~
FIGURE 14.3 Schematic diagram of magnetic field and plasma
flow in the solar wind/magnetosphere environment (from Lyons and
Williams, 1984). Continuous lines show the magnetic field for the case
where the IMF is purely southward. Open arrows show the plasma
velocity direction. Numbers 1-4 denote magnetic regions of different
topology, as discussed in the text.
ARTHUR D. RICHMOND
causing ionospheric plasma to convect antisunward.
The magnitude of the ionospheric electric field is greater
than that of the interplanetary electric field because the
bundling of magnetic-field lines at the ionosphere
causes electric potential gradients to intensify. On the
other hand, the plasma drift velocity in the upper iono-
sphere, given by Eq. (14.4), is much less than the solar-
wind velocity because of the inverse dependence on
magnetic-field strength. The polar-cap electric field is
typically 20 mV/m, giving an ionospheric convection
velocity of roughly 300 miser. Other directions of the
IMF than shown in Figure 14.3 result in a somewhat
altered pattern of polar-cap ionospheric convection, but
a usual feature is the general antisunward flow. More
about IMF influence on the high-latitude electric field is
discussed in a later section.
The physical processes that determine the amount of
magnetic flux that interconnects the geomagnetic field
and the IMF are not well understood and are the subject
of much study (e.g., Cowley, 1982~. They clearly in-
volve a violation of the MHD approximation since the
ionospheric and solar-wind plasmas coexisting on an
open magnetic-field line at one instant of time could not
have lain on a common magnetic-field line throughout
their entire histories. The violations of this approxima-
tion occur to some extent throughout the magnetosphere
but are particularly important in at least two regions: at
the sunward magnetopause and somewhere in the mag-
netospheric tail. At the sunward magnetopause plasma
flows together through unconnected interplanetary and
magnetospheric magnetic fields and flows out north-
ward and southward on interconnected magnetic-fields
lines (Figure 14.3~. In the tail plasma flows together on
interconnected field lines and flows out through uncon-
nected magnetospheric and interplanetary magnetic
fields. In the closed portion of the magnetosphere
plasma flows generally toward the Sun, passing around
the Earth on the morning and evening sides (out of the
plane of Figure 14.3~. However, some of the outermost
portions of the closed-field region convect away from
the Sun because of momentum transfer from the nearby
solar wind (e. g., Hones, 1983) .
Magnetospheric plasma convection has a number of
important consequences, one of which is the energiza-
tion of plasma and particle precipitation into the iono-
sphere. As plasma flows from the tail toward the Earth
it is compressionally heated because the volume occu-
pied by plasma on neighboring magnetic-field lines is
reduced as the magnetic-field strength increases and as
the length of field lines decreases. Other particle accel-
eration processes also help to energize the plasma. Some
of the energized particles precipitate into the ionosphere
and create ionization enhancements, especially in the
auroral oval. The energized plasma also has an impor
OCR for page 199
UPPER-ATMOSPHERE ELECTRIC-FIELD SOURCES
tent influence on the flow of electric currents and on the
distribution of electric fields (e.g., Spiro and Wolf,
1984~. Energetic particles drift in the Earth's magnetic
field-electrons toward the east and positive ions to
ward the west so that a westward current flows within
the hot plasma. This westward current, flowing in the
geomagnetic field, essentially exerts an electromagnetic
force on the plasma directed away from the Earth and
thus tending to oppose the earthward] convection.
Charge separation associated with the current tends to DUSK
create an eastward electric-field component, opposite to
the nightside westward convection electric field, largely
canceling the convection electric field in the inner mag
netosphere.
The overall pattern of magnetospheric convection
tends to map along the magnetic field into the iono
sphere even though this mapping is imperfect because of
net electric fields that tend to develop within the non
uniform energetic plasma. In the upper ionosphere,
where Eq. (14.4) is valid, the general convection pat
tern looks something like that shown in Figure 14.4.
There is antisunward flow over the polar cap and sun
ward flow in most of the auroral oval. The dayside
magnetopause maps perhaps somewhere near the polar
cap auroral oval boundary on the dayside, while the
most distant closed field lines in the tail map perhaps
somewhere near the polar cap boundary on the night
side (the mapping of these outer magnetospheric regions
into the ionosphere is not yet well determined). The
flow lines in Figure 14.4 correspond to lines of constant
electrostatic potential in a steady state. There is a poten
tial high on the dawn side of the polar cap and a low on
the dusk side, with a potential difference of the order of
50,000 V. The electric-field strength in the auroral oval
tends to be somewhat larger than the polar-cap electric
field.
Currents are an integral part of the electrical circuit
associated with the solar wind/magnetosphere dynamo
(e.g., Banks, 1979; Roederer, 1979; Stern, 1983; Aka
sofu, 1984~. Figure 14.5 shows schematically the cur
rent flow near the Earth. Currents flowing along the
direction of the magnetic field (field-aligned currents)
couple the auroral oval with outer portions of the mag
netosphere. The upward and downward currents are
connected by cross-field currents in the ionospheric dy
namo region. The anisotropy of the dynamo-region con
ductivity gives rise to strong current components per
pendicular to the electric field in the auroral oval, in the
form of eastward and westward auroral electrojets.
IONOSPHERIC WIND DYNAMO
Winds in the dynamo region have the effect of mov
ing an electric conductor (the weakly ionized plasma)
199
SUN
~ :_ ~ ~ AURORAL
8 kV - ~ - - OVAL
00 HRS.
FIGURE 14.4 Schematic diagram of the magnetic north polar re-
gion showing the auroral oval and ionospheric convection (from
Burch, 1977~. The convection contours also represent electric potential
contours, with a potential difference of order 8 kV between them.
through a magnetic field (the geomagnetic field), which
results in the production of an electromotive force and
the generation of electric currents and fields (e. g., Aka-
sofu and Chapman, 1972; Wagner et al., 1980~. The
effective electric field driving the current, E ', is related
to the electric field in the earth frame, E, the wind ve-
locity u, and the geomagnetic field, B. by
E' = E + uXB. (14.5)
Thus we may consider that two components of current
exist, one driven by the "real" (measurable) electric field
E and the other driven by the "dynamo electric field"
FIELD ALIGNED ELECTROJET
CURRENTS COW:\ ~///
CURRENTS DRIVEN-it\ ~ ~ \
BY SOLAR HEATING ~ ~ \/
EQUATORIAL i=:= / AURORAL OVAL
ELECTROJET '5
SOLAR RADIATION
~ RELD AU - ED
/ ~1CURRFNTR
FIGURE 14.5 Schematic diagram of electric currents in the iono-
sphere and inner magnetosphere.
OCR for page 200
200
u x B. These two components are not independent,
however, because it turns out that the electric field E
itself depends on the dynamo electric field u X B.
To see how ionospheric winds cause electric fields to
be set up, let us for the moment ignore the effects of the
solar wind/magnetosphere dynamo. The dynamo elec-
tric field associated with the wind will drive a current.
In general, this current would tend 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 directed from the positive
toward the negative regions, 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. Actually, the
time scale for this equilibrium to be achieved is ex-
tremely rapid, so that the electric field is effectively al-
ways in balance with the wind. The magnitude of the
electric field is of order 1 mV/m. A net current flows in
the entire ionosphere owing to combined action of the
wind and electric field, especially on the sunlit side of
the Earth (e.g., Takeda and Maeda, 1980, 1981~. Figure
14.5 shows a large-scale current vortex at middle- and
low-latitudes flowing counterclockwise in the northern
hemisphere. A corresponding clockwise vortex flows in
the southern hemisphere. These vortices are known tra-
ditionally as the Sq current system because of the nature
of the ground-level magnetic variations that they pro-
duce: S for solar daily variations and q for quiet levels of
magnetic activity. At high latitudes the electric fields
and currents produced by the ionospheric wind dynamo
are relatively weak in comparison with those of the solar
wind/magnetospheric dynamo.
Winds in the upper atmosphere change strongly
through the course of the day. They are driven in one
ARTHUR D. RICHMOND
form or another by the daily variation in absorption of
solar radiation. Atmospheric heating causes expansion
and the creation of horizontal pressure gradients, which
drive the global-scale upper-atmospheric winds. Solar-
ultraviolet radiation absorption at the height of the dy-
namo region drives a major portion of the winds. Ab-
sorption by ozone lower down (30-60 km altitude) also
affects the clynamo-region winds by generating atmo-
spheric tides that can propagate upward as global-scale
atmospheric waves (e. g., Forbes, 1982a, 1982b). Figure
14.6 gives an example of three days of wind measure-
ments at 100-130 km above Puerto Rico, showing the
strong effects of propagating tides. Important contribu-
tions to the dynamo also come from altitudes above 130
km, where the conductivity is smaller (Figure 14.2) but
where winds tend to be stronger and to vary less in
height.
Figure 14.7 shows the average global electrostatic po-
tential generated by the ionospheric wind dynamo, ex-
pressed in magnetic coordinates. The zero potential is
arbitrarily defined here such that the average iono-
spheric potential over the Earth is zero. The total poten-
tial difference of the average pattern (4.7 kV) is smaller
than what can usually be expected on any given day and
is much smaller than that associated with the solar
wind/magnetospheric dynamo. This pattern was de-
rived from observations of plasma drifts on magneti-
cally quiet days at an altitude of about 300 km. The elec-
tric-field maps along magnetic-field lines between
hemispheres, providing the symmetry about the mag-
netic equator even when the wind dynamo action in op-
posite hemispheres is asymmetric. Magnetic-field lines
peaking below 300 km in the equatorial region are not
represented in Figure 14.7. Electric fields in the equato-
rial lower ionosphere have a localized strong enhance-
ment of the vertical component associated with the
strong anisotropy of the conductivity in the dynamo re
AUG 10, 1 974
E 130
20
-
110
100
. .
' ' ' ' ' ' '1
AUG 12, 1974
~ arm// ~ ~ _ -
.~
~ =
. . . . . . . . .
oG 10 1 2 14 16 08 10
T I ~ E (AST)
AUG 13, 1974
~-
2 14 16 08 10 12 14 16
Contour level: 20 m/s
FIGURE 14.6 Observed eastward (unshaded) and westward (shaded) winds above Puerto Rico curing three daytime periods in August 1974 (from
Harper, 1977). The contour level is 20 miser.
OCR for page 201
UPPER-ATMOSPHERE ELECTRIC-FIELD SOURCES
-
~n
60
Cal
30
-
77~) ~
O
- -60~ ~1~==
Z 0 3 6 9 12 tO 18 21 24
~ MAGNET IC LOCAL T Il\/IE ( hours )
FIGURE 14.7 Average quiet-day ionospheric electrostatic potential
at 300 km altitude as a function of magnetic local time (from Rich-
mond et al., 1980~. The contour level is 500 V, and extreme relative to
the global average are labeled in kilovolts.
"ion. This enhanced electric field drives an eastward
daytime current along the magnetic equator called the
equatorial electrojet, as seen in Figure 14.5 (e.g.,
Forbes, 1981~.
VARIABILITY
The preceding sections discussed the average patterns
of ionospheric electric fields. Substantial deviations
from these patterns occur, on a global scale as well as on
a localized scale, and with a wide range of time scales.
Because the solar wind and particularly the IMF un-
dergo large variations (e.g., Hundhausen, 1979), it is
not surprising that the electric fields associated with so-
lar wind/magnetosphere dynamo action similarly show
large variations (e. g., Cowley, 1983; Rostoker, 1983~.
The interconnection of the interplanetary and magneto-
spheric magnetic fields maximizes when the IMF is
southward, as in Figure 14.3. As the IMF direction ro-
tates out of the plane of Figure 14.3 the amount of inter-
connected magnetic flux appears to lessen. For a north-
ward-directed IMF the pattern of magnetic-field
interconnection must be quite different from that
shown in Figure 14.3, and it is possible that interconnec-
tion becomes insignificant, so that the magnetosphere is
closed. The east-west component of the IMF affects the
interconnection morphology and consequently also the
pattern of high-latitude magnetosphere convection.
The third IMF component, in the plane of Figure 14.3
but directed toward or away from the Sun, seems to
have a less important role in the solar wind/magneto-
sphere dynamo than the other two components. How-
ever, the toward/away component correlates strongly
with the west/east IMF component because of the ten
201
dency for the IMF to assume a spiral pattern around the
Sun. Some discussion of IMF effects has therefore taken
place in reference to the toward/away component
rather than the west/east component.
Figure 14.8 shows the average patterns of electric po-
tential above 60° magnetic latitude deduced from
ground magnetic variations for four different directions
of the IMF. Bz is the northward component of the IMF;
By is the eastward component (toward the east for an
observer on the sunward side of the Earth). The electric
fields are stronger for a southward IMF (Bz < 0) than
for northward IMF (Bz ~ 0), in accord with the concept
that magnetic field interconnection is greater and mag-
netospheric convection is stronger for a southward IMF.
There are also clear differences in the patterns between
the westward (By 0) IMF
cases, especially on the sunward side of the Earth.
Magnetic storms are dramatic disturbances of the en-
tire magnetosphere lasting a time on the order of 1 day,
usually produced by a strong enhancement of the solar-
wind velocity, density, and/or southward IMF compo-
nent (e. g., Akasofu and Chapman, 1972~. The enhance-
ments often come from explosive eruptions of plasma
ELECTRIC POTENTIAL
BOO 12
FIGURE 14.8 Average electric potential patterns above 60°N mag-
netic latitude as a function of magnetic local time deduced from mag-
netic variations at the ground, for four directions of the IMF (see text).
The contour level is 4 kV (from Friis-Christensen et al., 1985).
OCR for page 202
202
\
near the Sun's surface but sometimes are a long-lasting,
relatively localized feature of the solar wind that sweeps
past the Earth as the Sun rotates. During a magnetic
storm magnetospheric convection varies strongly but is
generally enhanced; plasma energization and precipita-
tion into the ionosphere are greatly increased; and elec-
tric current flow is much stronger, also varying rapidly
in time. Figure 14.9 gives an example of ground-level
magnetic fluctuations caused by magnetospheric and
ionospheric currents during a large storm. Quiet-day
variations in the declination (D), vertical component
(Z), and horizontal intensity (H) are seen up until 8:30
UT, when a shock wave in the solar wind hit the magne-
tosphere to produce a storm sudden commencement.
Perturbations up to a few percent of the total geomag-
netic-field strength occurred during the subsequent
hours. Storms are composed of a succession of impulsive
disturbances lasting 1-3 hours, called substorms (e.g.,
Akasofu, 1977; Nishida, 1978; McPherron, 1979~. Im-
pulsive disturbances with similar characteristics also oc-
cur on the average a few times a day even when no storm
is in progress, and these are also called substorms. The
characteristics of substorms can vary quite considerably
from one to another but are often associated with what
appears to be a large-scale plasma instability in the mag-
netospheric tail. The disturbed electric fields extend be-
yond the auroral oval and can even be seen at the mag-
netic equator (e. g., Fejer, 1985~.
Magnetic storms are predominantly a phenomenon of
the solar wind/magnetosphere dynamo, but they affect
the ionospheric wind dynamo as well. In addition to
strong auroral conductivity enhancements, conductivi-
ties can also be altered at lower latitudes at night by over
D
>
=,
in
AS
o Z
to
j
H
_ . _
ARTHUR D. RICHMOND
an order of magnitude (Rowe and Mathews, 1973),
though they still remain well below daytime values. The
nighttime ionospheric layer above 200 km altitude can
be raised or lowered in response to stormtime electric
fields and winds, which changes its conductive proper-
ties. During major storms the entire wind system in the
dynamo region can be altered by the energy input to the
upper atmosphere, so that the pattern of electric-field
generation is modified (Blanc and Richmond, 1980~.
This effect can raise the potential at the equator by sev-
eral thousand volts with respect to high latitudes.
Regular changes in the electric fields and currents oc-
cur over the course of the 11-year solar cycle and with
the changing seasons. Ionospheric conductivities change
by up to a factor of 2 as the ionizing solar radiation
waxes and wanes along with the trend of sunspots.
Ionospheric winds also change as the intensity of solar
extreme-ultraviolet radiation increases and decreases
(e.g., Forbes and Garrett, 1979~. We know that iono-
spheric dynamo currents change by a factor of 2 to 3
with the solar cycle (e.g., Matsushita, 1967), but the
variation of electric fields is not yet so extensively docu-
mented. At Jicamarca, Peru, however, ionospheric elec-
tric fields have been measured on an occasional basis for
well over a solar cycle, and the average behavior of the
east-west equatorial electric-field component at 300 km
altitude is shown in Figure 14.10, as represented by the
vertical plasma drift that it produces. The presence of a
strong upward drift after sunset at solar sunspot maxi-
mum (1968-1971) is not usually present at solar mini-
mum (1975-1976~. Clear seasonal variations appear at
Jicamarca as well as at higher latitudes. Day-to-day
changes in middle- and low-latitude electric fields and
FRE
)ERICKS
BERG V,
~ - 22!
4_ ~ - F _
Izl'~
Fang
IZ
.
~4 T 6 U.T. 2 O ~2 4
FIGURE 14.9 Magnetogram from Fredericksburg, Virginia, on March 22, 1979, showing variations of the magnetic declination (D)' vertical
magnetic field (Z), and horizontal magnetic intensity (H) during a magnetic storm. Scale values are not shown.
OCR for page 203
UPPER-ATMOSPHERE ELECTRIC-FIELD SOURCES
mars
40
20
~ I ~ · ~ ~
JlCAMARCa VERTICAL DRIFTS
WINTER SOLSTICE
8 May-7Aug
. . ·~
EQUINOX
-20 - 4 hb-7May
8 Aug-9 Nov
MA
R I\ _ ,_ ~` _ j,~
BE _` _ my'
20
o
SUMMER SOLSTICE
10 Nov-3 Feb
.~ ~ ~
At .
, ·, · ~ ~ ., · 1, ·, · ~ . ~ I · ~ ~ ~ ~
16 20 00 04 08
LOCAL TIME (75 W)
08 12
FIGURE 14.10 Average quiet-day vertical component of the plasma
drift velocity caused by east-west electric fields over Jicamarca, Peru,
as a function of local time (from Fejer et al., 1979~. 1968-1971 (a) are
sunspot cycle maximum years, while 1975-1976 (b) are sunspot cycle
minimum years.
currents also occur, produced by corresponding changes
in the ionospheric winds.
Rapid variations with time scales ranging from sec-
onds to hours are a common feature of ionospheric elec-
tric fields and currents. The magnitude of the fluctua-
tions is often as large as that of the regular daily
variations, generally increasing with magnetic latitude
up to the auroral oval. Besides storm and substorm phe-
nomena, global-scale disturbances in the electric fields
and currents also occur with the arrival of solar-wind
shocks, with fluctuations of the IMP, and with rapid
ionosphere conductivity changes during solar flares. Lo-
calized ionospheric electric-field fluctuations may be as-
sociated with wind or conductivity irregularities or with
small-scale magnetospheric processes. For example, lo-
calized quasi-periodic oscillations in the electric fields
and currents with periods ranging from a fraction of a
second to minutes, called pulsations, are often observed
at high latitudes (e. g., Nishida, 1978~.
EFFECTS OF UPPER-ATMOSPHERIC
ELECTRIC FIELDS AND CURRENTS
Electric fields and currents interact strongly with the
upper atmosphere and help determine its behavior
203
(e.g., Banks, 1979~. The drifting ions, as they interact
collisionally with neutral molecules, exert a force on the
air and tend to bring it toward the ion motion. Above
200 km altitude this effect can be important: at high
latitudes winds are common that approach the rapid ve-
locity of the convecting plasma (e.g., Meriwether,
1983), while at low latitudes, where plasma drifts are
much smaller, the collisional interaction tends to retard
the winds driven by pressure gradients. Of even greater
importance is the heating of the upper atmosphere
caused by currents in the auroral region. The heating
can make a significant contribution to the upper-atmo-
spheric energy budget and can even be the dominant
heat source above 120 km during magnetic storms. As
the temperature increases the upper atmosphere ex-
pands, and the drag on near-Earth satellites is in-
creased, changing their orbits (e. g., Joselyn, 1982) .
The ionosphere is affected in many ways by electric
fields. Above 200 km, where the chemical lifetimes of
ions range from several minutes to hours, rapid convec-
tion of ionization at high latitudes can bring dense day-
sicle plasma to the nightside of the Earth in some places,
cause stagnation and prolonged nighttime decay of ion-
ization at other places, and generally produce highly
complex patterns of ionization density (e. g., Sojka et al.,
1983~. Plasma temperatures and chemical reaction rates
are also affected by the rapid ion convection through the
air. Even at middle and low latitudes plasma drifts have
an important influence on the upper ionosphere, pri-
marily by raising or lowering the layer into regions of
lower or higher neutral density, so that chemical decay
is retarded or accelerated. During magnetic storms the
plasma-drift effects on the ionosphere are not only in-
tensified but are also supplemented by indirect effects
through modification of the neutral atmosphere (e.g.,
Prolss, 1980~. Auroral heating induces atmospheric con-
vection that alters the molecular composition of the up-
per atmosphere and leads to more rapid chemical loss of
the ionization, even at middle latitudes. Winds gener-
ated by the magnetic storm impart motion to the ioniza-
tion along the direction of the magnetic field, causing
redistribution of the plasma as well as further modifica-
tion of the loss rate. All these ionospheric phenomena
affect radio-wave transmissions that reflect off the iono-
sphere.
Radio waves at frequencies greater than about 30
MHz do not normally reflect off the ionosphere and are
much less influenced by large-scale plasma density vari-
ations than are lower frequencies. However, these
waves are affected by small-scale plasma irregularities
that cause radio signals to scintillate, undergoing sub-
stantial amplitude and phase modulations (e. g.,
Aarons, 1982~. The scintillations are bothersome for
transmissions between satellites and the ground. Elec
. . .
OCR for page 204
204
trio fields are involved in producing the ionospheric ir-
regularities, often by plasma instabilities resulting from
drifts or from steep density gradients created by nonuni-
form convection (e.g., Fejer and Kelley, 1980~. The ir-
regularities tend to be strongest at high latitudes, where
electric fields and their consequences are greatest, but
the equatorial region is also subject to strong irregulari-
ties, both in the equatorial electrojet and in the night-
time upper ionosphere.
Rapid magnetic fluctuations caused by changing
ionospheric currents disturb geophysical surveys of
magnetic anomalies in the Earth's crust. In addition,
the fluctuations induce electric currents in the Earth.
Analysis of the Earth currents can be useful for geophys-
ical studies, but when they enter large man-made struc-
tures like electric transmission lines and pipelines they
can cause disruptive electrical signals and corrosion.
Lanzerotti and Gregori discuss these problems in more
detail in Chapter 16, this volume.
Finally, it should be noted that the ionospheric elec-
tric field extends down into lower atmospheric regions.
In fact, the presence of the field at lower altitudes has
been used for many years to measure ionospheric elec-
tric fields from stratospheric balloons (e.g., Mozer and
Lucht, 1974) . Roble and Tzur (Chapter 15, this volume)
discuss the relation of the ionospheric electric field to the
global atmospheric electric circuit.
SOME OUTSTANDING PROBLEMS
The entire range of subjects discussed in this chapter is
undergoing scientific investigation, but there are cer-
tain topics that currently present questions of funda-
mental importance to the understanding of upper-at-
mospheric electric-field sources.
Since the beginning of the artificial satellite era mea-
surements of space plasmas, energetic particles, and
electromagnetic fields have provided a broad picture of
the interaction of the solar wind with the magneto-
sphere. Yet theory still is unable to explain quantita-
tively, without the use of ad hoc parameterizations,
how much magnetic flux interconnects the Earth with
the solar wind or what the details of the magnetic- and
electric-field configurations are in the vicinity of the
magnetopause and in the distant magnetospheric tail.
Measurements alone cannot answer these questions be-
cause of the impossibility of measuring simultaneously
the entire spatial structure of the continuously changing
magnetosphere. However, some future observational
programs may aid in the solution of the problem by
combining continuous monitoring of electrodynamic
features of the solar wind and of the entire auroral oval
ARTHUR D. RICHMOND
ionosphere with spot measurements in the poorly ex-
plored high-latitude magnetopause and distant tail re-
gions.
Because solar wind/magnetosphere dynamo processes
intimately involve magnetospheric plasma energiza-
tion, ionospheric conductivity alterations, and coupling
of magnetospheric and ionospheric electric fields and
currents, the dynamo cannot be fully understood with-
out taking account of the magnetosphere-ionosphere in-
teract~ons. Theoretical models incorporating the mu-
tual interactions have made great strides of progress in
recent years (e. g., Spiro and Wolf, 1984), but much re-
mains to be learned. Observational programs that can
measure a large number of the important phenomena
involved in the interactions are a necessary adjunct to
theoretical studies.
The distribution of ionospheric conductivity is impor-
tant in determining the distributions of both electric
fields and currents in the ionosphere and magneto-
sphere, yet our understanding of nightside conductivi-
ties is still rudimentary. We know that the conductivi-
ties vary greatly but they are difficult to measure when
the ionization density is low, and we do not yet fully
understand the nature of nighttime ionization sources.
In the auroral oval and polar cap the nightside conduc-
tivities can be highly structured, having features correl-
ated with structured features of the electric fields and
currents. Because the conductivity features are so vari-
able, useful models are difficult to construct, although a
number of encouraging developments in modeling au-
roral-oval conductivities have been made in the past few
years (Reiff, 1984~.
Concerning the ionospheric wind dynamo, theoreti-
cal models and observations are in general agreement,
except that the morphology of nighttime electric fields is
not yet well explained, a problem related to the conduc-
tivity uncertainties. The nature of day-do-day variabil-
ity in the electric fields and currents, as well as smaller-
scale variations, are only poorly understood at present.
Winds in the dynamo region, especially those of a tidal
nature, exhibit day-to-day variability that has not been
fully explained and is currently unpredictable. It ap-
pears that day-to-day variability in the distribution of
atmospheric heating as well as changes in atmospheric
tidal propagation characteristics and interactions with
other wave motions are largely to blame, but these con-
ditions are not now monitored on a global scale. Ad-
vances in this area may be forthcoming in the wake of
intensified study of the middle atmosphere (10-120 km
altitude). In addition, further clarification of upper-at-
mospheric circulation changes during magnetic storms
is needed in order to understand how storms affect the
ionospheric wind dynamo.
OCR for page 205
UPPER-ATMOSPHERE ELECTRIC-FIELD SOURCES
ACKNOWLEDGMENT
Most of this chapter was prepared while the author
was employed at the NOAA Space Environment Labo
ratory.
REFERENCES
Aarons, J. (1982). Global morphology of ionospheric scintillations,
Proc. ISEE 70, 360-378.
Akasofu, S.-I. (1977~. Physics of Magnetospheric Substorms, D.
Reidel, Dordrecht, Holland.
Akasofu, S.-I. (1984~. The Magnetospheric currents: An introduction,
in Magnetospheric Currents, T. A. Potemra, ea., American Geo-
physical Union, Washington, D.C., pp. 29-48.
Akasofu, S.-I., and S. Chapman (1972~. Solar-Terrestrial Physics,
Clarendon Press, Oxford.
Banks, P. M. (1979~. Magnetosphere, ionosphere and atmosphere in-
teractions, in Solar System Plasma Physics, Vol. II, C. F. Kennel, L.
J. Lanzerotti, and E. N. Parker, eds., North-Holland Publ. Co.,
Amsterdam, pp. 57-103.
Blanc, M., and A. D. Richmond (1980~. The ionospheric disturbance
dynamo, J. Geophys. Res. 85, 1669-1686.
Burch, J. L. (1977~. The magnetosphere, in The Upper Atmosphere
and Magnetosphere, NRC Geophysics Study Committee, National
Academy of Sciences, Washington, D.C., pp. 42-56.
Cowley, S. W. H. (1982~. The causes of convection in the Earth s mag-
netosphere: A review of developments during the IMS, Rev.
Geophys. Space Phys. 20, 531-565.
Cowley, S. W. H. (1983~. Interpretation of observed relations be-
tween solar wind characteristics and effects at ionospheric altitudes,
in High-Latitude Space Plasma Physics, B. Hultqvist and T.
Hagfors, eds., Plenum Press, New York, pp. 225-249.
Fejer, B. G. (1985~. Equatorial ionospheric electric fields associated
with Magnetospheric disturbances, in solar Wind-Magnetosphere
Coupling, Y. Kamide, ea., D. Reidel, Dordrecht, Holland.
Fejer, B. G., and M. C. Kelley (1980~. Ionosphere irregularities, Rev.
Geophys. Space Phys. 18, 401-454.
Fejer, B. G., D. T. Farley, R. F. Woodman, and C. Calderon (1979~.
Dependence of equatorial F region vertical drifts on season and solar
cycle, J. Geophys. Res. 84, 5792-5796.
Forbes, J. M. (1981~. The equatorial electrojet, Rev. Geophys. Space
Phys. 19, 469-504.
Forbes, J. M. (1982a). Atmospheric tides, 1. Model description and
results for the solar diurnal component, J. Geophys. Res. 87, 5222-
5240.
Forbes, J. M. (1982b). Atmospheric tides, 2. The solar and lunar semi-
diurnal components, J. Geophys. Res. 87, 5241-5252.
Forbes, J. M., and H.13. Garrett (1979). Theoretical studies of atmo-
spheric tides, Rev. Geophys. Space Phys. 17, 1951-1981.
Friis-Christensen, E., Y. Kamide, A. D. Richmond, and S. Matsushita
(1985). Interplanetary magnetic field control of high-latitude elec-
tric fields and currents determined from Greenland magnetometer
data, J. Geophys. Res. 90, 1325-1338.
Harper, R. M. (1977). Tidal winds in the 100- to 200-km region at
Arecibo, J. Geophys. Res. 82, 3243-3250.
Hill, T. W. (1979). Generation of the Magnetospheric electric field, in
Quantitative Modeling of Magnetospheric Processes, W. P. Olson,
ea., American Geophysical Union, Washington, D.C., pp. 297-
315.
Hill, T. W., and R. A. Wolf (1977). Solar-wind interaction, in The
Upper Atmosphere and Magnetosphere, NRC Geophysics Study
205
Committee, National Academy of Sciences, Washington, D. C., pp.
25-41.
Hones, E. W., Jr. (1983~. Magnetic structure of the boundary layer,
Space Sci. Rev. 34, 201-211.
Hundhausen, A. J. (1979). Solar activity and the solar wind, Rev.
Geo phys. S pace Phys. 17, 2034-2048.
Joselyn, J. C., ed. (1982~. Workshop on Satellite Drag, NOAA Space
Environment Laboratory, Boulder, Colo., 251 pp.
Lyons, L. R., and D. J. Williams (1984~. Quantitative Aspects of Mag-
netospheric Physics, D. Reidel Publishing Co., Dordrecht, Holland.
Matsushita, S. (1967~. Solar quiet and lunar daily variation fields, in
Physics of Geomagnetic Phenomena, Vol. 1, S. Matsushita and W.
H. Campbell, eds., Academic Press, New York, pp. 301-424.
McPherron, R. L. (1979). Magnetospheric substorms, Rev. Geophys.
Space Phys. 17, 657-681.
Meriwether, J. W., Jr. (1983~. Observations of thermospheric dy-
namics at high latitudes from ground and space, Radio Sci. 18,
1035-1052.
Mozer, F. S., and P. Lucht (1974~. The average auroral zone electric
field, J. Geophys. Res. 79, 1001-1006.
Nishida, A. (1978~. Geomagnetic Diagnosis of the Magnetosphere,
Springer-Verlag, New York, 256 pp.
Prolss, G. . (1980~. Magnetic storm associated perturbations of the
upper atmosphere: Recent results obtained by satelliteborne gas an-
alyzers, Rev. Geophys. Space Phys. 18, 183-202.
Reiff, P. H. (1984~. Models of auroral-zone conductances, in Magneto-
.spheric Currents, T. A. Potemra, ea., American Geophysical
Union, Washington, D.C., pp. 180-191.
Richmond, A. D., M. Blanc, B. A. Emery, R. H. Wand, B. G. Fejer,
R. F. Woodman, S. Ganguly, P. Amaynec, R. A. Behnke, C.
Calderon, and J. V. Evans (1980~. An empirical model of quiet-day
ionospheric electric fields at middle and low latitudes, J. Geophys.
Res. 85, 4658-4664.
Roederer, J. G. (1979~. Earth s magnetosphere: Global problems in
Magnetospheric plasma physics, in solar System Plasma Physics,
Vol. II, C. F. Kennel, L. J. Lanzerotti, and E. N. Parker, eds.,
North-Holland Publ. Co., Amsterdam, pp. 1-56.
Rostoker, G. (1983~. Dependence of the high-latitude ionospheric
fields and plasma characteristics on the properties of the interplane-
tary medium, in High-Latitude Space Plasma Physics, B. Hultqvist
and T. Hagfors, eds., Plenum Press, New York, pp. 189-204.
Rowe, J. F., Jr., and J. D. Mathews (1973~. Low-latitude nighttime E
region conductivities, J. Geophys. Res. 78, 7461-7470.
Sojka, J. J., R. W. Schunk, J. V. Evans, and J. M. Holt (1983). Com-
parison of model high-latitude electron densities with Millstone Hill
observations, J. Geophys. Res. 88, 7783-7793.
Spiro, R. W., and R. A. Wolf (1984~. Electrodynamics of convection
in the inner magnetosphere, in Magnetospheric Currents, T. A. Po-
temra, ea., American Geophysical Union, Washington, D.C., pp.
247-259.
Stern, D. P. (1977). Large-scale electric fields in the Earth s magneto-
sphere, Rev. Geophys. Space Phys. 15, 156-194.
Stern, D. P. (1983). The origins of Birkeland currents, Rev. Geophys.
Space Phys. 21, 125-138.
Takeda, M., and H. Maeda (1980). Three-dimensional structure of
ionospheric currents-1. Currents caused by diurnal tidal winds, J.
Geophys. Res. 85, 6895-6899.
Takeda, M., and H. Maeda (1981). Three dimensional structure of
ionospheric currents-2. Currents caused by semidiurnal tidal
winds, J. Geophys. Res. 86, 5861-5867.
Wagner, C. U., D. Mohlmann, K. Schafer, V. M. Mishin, and M. I.
Matveev (1980). Large-scale electric fields and currents and related
geomagnetic variations in the quiet plasmasphere, Space Sci. Rev.
26, 391-446.
Representative terms from entire chapter:
electric field
