The Ionosphere and Upper Atmospheric Electrodynamics

Adapted with modifications from the article "Space weather research prompts study of ionosphere and upper atmospheric electrodynamics", by A.D. Richmond, published in EOS, Trans. Am. Geophys. Union}, Vol. 77, No. 07, March 12, 1996, pp. 101-104, copyright 1996 by the American Geophysical Union. Further electronic distribution is strictly prohibited.

Introduction

The upper atmosphere contains free electrons and ions produced by ionizing radiation from the Sun and from the Earth's space environment. It comprises a weakly ionized plasma, called the ionosphere, that is a conductor of electricity. Above 60 km altitude electrons are sufficiently dense to influence the propagation of radio waves, giving the ionosphere much of its practical importance. The ionosphere lies at the base of the magnetosphere, which encompasses those regions of space where the Earth's magnetic field has a dominant influence on charged particles. The electrodynamical behavior of the ionosphere is strongly influenced both by the neutral atmosphere within which it is embedded and by the magnetosphere. Global electric currents flow throughout the ionosphere and magnetosphere, connecting into currents in interplanetary space that are carried by the plasma of the solar wind. The highly variable currents and their associated electric fields have a major impact on the energetics, dynamics, and structure of the upper atmosphere and the space environment.

This article briefly describes the electrical characteristics of the upper atmosphere and highlights some important areas of ionospheric research.

Uses and impacts of the ionosphere

The ionosphere's influence on radio-wave propagation is sometimes useful, but sometimes bothersome. Over-the-horizon transmissions for telecommunications or surveillance usually rely on ionospheric reflection at radio frequencies below about 30-MHz (10-meter wavelength). The maximum usable frequency depends, among other things, on the maximum electron density of the ionosphere, which is highly variable. Lower-frequency waves are subject to absorption in the lower ionosphere, where the electrons oscillating in the wave's electromagnetic field lose energy to air molecules through collisions. The radio-wave absorption depends on the electron density, and is strongest during the day. The absorption can increase to the point of radio blackout during sporadic ionization enhancements associated with solar-flare X-rays and with energetic protons that precipitate into the high-latitude upper atmosphere.

Because the ionosphere is a magnetized plasma, its study contributes to the field of plasma physics. A variety of natural plasma instabilities occur that are observed with radars and other radio-wave techniques, as well as with rockets and spacecraft. Active experiments are carried out by modifying ionospheric properties with high-power radio waves, with chemical releases, or with space-based energetic electron beams. Unlike laboratory plasmas, the ionosphere has no chamber walls to interfere with the experiments or to complicate interpretation of the data.

Radio transmissions between the Earth and spacecraft operate at frequencies that are not reflected by the ionosphere and that do not suffer much absorption. However, these transmissions are subject to degrading scintillation when they refract through ionospheric irregularities. They also undergo phase-path changes and propagation delays in traversing the ionosphere, necessitating adjustments to precise measurements like satellite-based radar altimetry of ocean and land surfaces, positioning with the Global Positioning System, and radioastronomy. To some extent, the electron density and the irregularities have predictable variations with the time of day, season, phase of the 11-year sunspot cycle, and geographic location. However, they are also subject to irregular variations due to influences coming from the magnetosphere and from the lower atmosphere.

Ionospheric electric currents, especially those strong currents that occur during magnetic storms, can have a number of impacts. The magnetic field produced by the currents induces additional electrical currents in the Earth, which can flow through grounded electrical power grids and harm their transformers or trip circuit breakers. On occasion, large-scale disruptions of power grids have resulted, as happened in Quebec during the magnetic storm of 1989 March 13. Even during less-disturbed periods, the magnetic perturbations associated with ionospheric currents complicate geomagnetic surveys that attempt to derive accurate models of the Earth's internal field or to determine the nature and geologic significance of subtle spatial structure in the field. Electrical heating of the upper atmosphere above 120 km during storms raises the temperature, thereby reducing the rate of exponential density fall-off with increasing altitude so that the density at high altitude is greatly increased.

Satellites orbiting the Earth below 1000 km then experience perceptible alterations of their trajectories owing to the increased atmospheric drag. They can become temporarily lost to satellite-tracking services. The heating also changes the wind patterns and the composition of the upper atmosphere, which influence the plasma density distribution.

The Conducting Upper Atmosphere

The earliest suggestions that the rarefied upper atmosphere might conduct electricity came from eighteenth-century experimenters who were struck by the similarity of polar lights (auroras) and electrical glows produced in evacuated containers. Benjamin Franklin formulated a theory of the aurora that invoked accumulations of electrical charge at polar latitudes. Experimenters of that time also noticed fluctuations of magnetic compass needles, which led nineteenth-century scientists like Carl Friedrich Gauss to suggest that electric currents might flow in the upper atmosphere. Around 1880, Balfour Stewart proposed that the currents could be driven by upper atmospheric winds that would generate electromotive forces in the conducting medium as they move it through the geomagnetic field, in effect acting as an electric dynamo. In 1908 Kristian Birkeland suggested that strong currents in the auroral ionosphere during magnetic disturbances are caused by charged particles coming from distant space that are forced to flow primarily along geomagnetic field lines until they reach the high-latitude ionosphere. Marconi's demonstration in 1901 that radio waves could propagate across the Atlantic led to suggestions by Kennelly and Heaviside that the waves might be reflected by a conducting layer in the upper atmosphere. In 1924 experiments by Appleton and Barnett and by Breit and Tuve demonstrated the direct reflection of radio waves from about 100 km altitude, thus clearly establishing the existence of the ionosphere.

Conventionally, the main ionosphere is divided into D, E, and F regions, at altitudes of roughly 60-90 km, 90-140 km, and 140-1000 km, respectively, based on features of the electron-density profile with altitude. The primary ionization sources are solar ultraviolet and X-ray radiation at wavelengths shorter than 103 nm striking the day side of the Earth, and energetic electrons precipitating from the magnetosphere into the auroral regions.

Solar ultraviolet and X-ray radiation vary over the 11-year solar activity cycle, with sporadic enhancements during solar flares. Precipitating auroral electrons vary dynamically in association with magnetospheric disturbances. The ionospheric electron density is highly variable, depending not only on the ionization sources, but also on ion-neutral chemical transformations, ion-electron recombination, and plasma transport by neutral winds, electric fields, and diffusion. The maximum density in a vertical profile usually occurs in the F region between about 200 km and 500 km, with values between 4x1010 and 4x1012 m-3 which correspond to natural resonant plasma frequencies of 1.8-18MHz. Radio waves are totally reflected at frequencies below the maximum resonant frequency, called the critical frequency. Reflection can also occur at higher frequencies for waves obliquely incident on the ionosphere. Obviously, long-distance terrestrial radio transmissions that reflect from the ionosphere must rely on frequencies below or near the critical frequency, while radioastronomy and communications with spacecraft can only use those frequencies that penetrate the ionosphere.

The ionospheric electrical conductivity is highly anisotropic, owing to the strong influence of the geomagnetic field on charged-particle motion. At high altitudes, where collisions between ions and neutral air molecules are infrequent, the ions and electrons are constrained to gyrate around magnetic-field lines, though they are free to move parallel to the field. The direct-current conductivity along the magnetic field can be as large as 100 S/m, while the conductivity perpendicular to the field is usually less than 10-4 S/m above 150 km. (For comparison, seawater has a conductivity of about 4 S/m.) The large parallel conductivity almost completely shorts out any electric field that might otherwise tend to be established along the magnetic field, so that for most situations the geomagnetic field lines can be considered to be electric equipotentials. At lower altitudes collisions between the ions and neutrals become more frequent, decoupling the electron and ion motions in the plane perpendicular to the magnetic field, so that more significant amounts of current can flow in that plane. The direct-current conductivity perpendicular to the magnetic field is largest at heights of 90-150 km during the day and in the nighttime auroral zone, with maximum values of the order 10-3 S/m. In addition to the parallel/perpendicular anisotropy of the conductivity, further anisotropy occurs perpendicular to the geomagnetic field: the Hall effect causes the direction of the current to deviate by as much as 88% from that of the electric field, an effect maximizing around 100 km altitude. For alternating currents at radio frequencies, the conductivity is frequency-dependent. It becomes nearly isotropic at frequencies well above the gyrofrequency of electrons in the geomagnetic field (on the order of 1MHz), and it decreases with increasing frequency.

Sources of ionospheric electric fields and currents

The two main sources of global-scale electric-field generation in the ionosphere are the ionospheric wind dynamo and the solar-wind/magnetospheric dynamo. A third source, thunderstorm activity, is believed to contribute only in a minor way to the global-scale fields, though at night it may be locally important.

The ionospheric dynamo is essentially that mechanism proposed by Stewart: winds in the thermosphere (90-500km) move the conducting medium through the geomagnetic field, producing an electromotive force (emf) that drives currents and that sets up polarization electric fields. Electric-potential differences of 5kV to 10kV between different parts of the globe are produced by this mechanism. The emf interacts only with the conductivity component transverse to the geomagnetic field, so that dynamo action is weighted toward the 90-150 km height range during the day. At night, however, the E-region transverse conductivity is greatly diminished, so that F-region dynamo action above 200 km becomes relatively more important. The ionospheric currents are strongest on the day side of the Earth, where they typically form two large horizontal current vortices, clockwise in the southern hemisphere and counterclockwise in the northern hemisphere. The currents in the two hemispheres are connected by magnetic-field-aligned current when the dynamo effects in the two hemispheres are unbalanced.

The solar-wind/magnetospheric dynamo draws its energy from the kinetic and thermal energy of the solar-wind and magnetospheric plasmas, generating electric fields and currents that connect to the high-latitude auroral and polar ionosphere along geomagnetic-field lines as suggested by Birkeland. In a recent EOS article, Cowley [1995] described the nature of the solar-wind/magnetosphere interaction. It depends strongly on the direction of the interplanetary magnetic field that is embedded in the solar wind, since the direction of that field determines the topology of its connection with the Earth's magnetic field. The ionospheric electric fields and currents produced by the solar-wind/magnetospheric dynamo are usually much stronger than those of the ionospheric wind dynamo, and are highly variable in time. On the average, a high electric potential is established around 70-75 º magnetic latitude on the morning side of the Earth, and a low potential at around the same latitude on the evening side. The potential drop varies from 20kV to 200kV.

Electric-power generation by the dynamos involves extraction of energy from the thermospheric wind and from the solar-wind and magnetospheric plasmas, modifying these in the process. For example, thermospheric winds experience a significant drag force as the electric currents they generate flow through the geomagnetic field; this force is known as ``ion drag'' because it results microscopically from collisions between ions and neutral molecules moving at different mean velocities. The solar wind is also retarded by the dynamo currents it generates. These energy losses, as well as the rates of electric energy transfer between the ionosphere and the hot magnetospheric plasma, are dependent on the ionospheric conductivity. The ionospheric conductivity is itself dependent on the dynamo electric fields, since those fields cause transport and redistribution of the F-region plasma. Furthermore, the electrical circuits of the ionospheric-wind dynamo and of the solar-wind/magnetospheric dynamo are intercoupled, so that the two dynamos react to each other. For example, the strong high-latitude currents driven by the solar-wind/magnetospheric dynamo, flowing through the geomagnetic field, force high-speed thermospheric winds by a motor effect, winds that in turn influence the electric fields and currents. Realistic modeling of dynamo processes quickly becomes very complicated when the various feedback effects are considered. This is an active area of current research in ionospheric electrodynamics.

The electric power is used up in a number of ways. Much of it is dissipated as resistive heating in the thermosphere, especially at high latitudes. Some of it is transferred through the ionospheric circuit from the solar-wind source to the magnetospheric plasma, which is heated as it is transported toward the Earth into regions of stronger magnetic field. Some of it goes into acceleration of energetic electrons that precipitate into the high-latitude thermosphere to produce the polar lights (aurora). A fraction of the electric power goes into forcing the strong high-latitude thermospheric winds.

Inferences from observations of dynamo effects

Observations of ionospheric electric fields and of the magnetic perturbations produced by ionospheric currents give us important information about thermospheric winds and about the interaction of the solar wind with the magnetosphere. Direct observations of thermospheric winds are relatively limited, but observations of magnetic perturbations exist for long periods of time at many locations around the Earth. When interpreted with the aid of simulation models of the ionospheric wind dynamo, magnetic data from sites at middle and low latitudes can provide a wealth of information about the distribution and variability of thermospheric winds on the sunlit side of the Earth. At high magnetic latitudes, observations of ionospheric electric fields and of magnetic perturbations on the ground and on satellites reveal characteristics of solar-wind/magnetospheric dynamo processes.

At middle and low latitudes, winds in the ionospheric dynamo region tend to be dominated by global oscillations. Above 140 km, daily wind oscillations with magnitudes over 100 m/s are driven primarily by the absorption of far-ultraviolet solar radiation. Between 90 km and 140 km the oscillations are strongly influenced by upward propagating global waves, called atmospheric tides, that are generated by solar heating at lower altitudes: in the upper ozone layer (30-60 km) and in the troposphere (below 10 km). Gravitational tidal forcing by the Moon and Sun also contribute, but only in a minor way. As the tides propagate into regions of exponentially decreasing air density, their amplitudes can grow, reaching values of 100 m/s or so in the lower thermosphere before the waves are eventually dissipated. The generation and propagation conditions for these waves tends to favor the arrival of semidiurnal (12-hour) tides over diurnal (24-hour) tides in the dynamo region. Upward-propagating planetary waves with periods of 2-20 days are also believed to influence winds in the lower thermosphere, but their relative importance there has not yet been established. At high latitudes, electric currents drive thermospheric winds that at times can reach 1000 m/s or more in the upper thermosphere, both by the motor effect mentioned earlier, and by resistive heating of the gas that affects the pressure-gradient forces on the air. Variations in the sources of the winds, as well as variations in the propagation conditions of tides and planetary waves through the middle atmosphere (10-100 km), are responsible for variability of the thermospheric winds on a day-to-day, seasonal, and solar-cycle basis. Analyses of geomagnetic variations have revealed many properties of the winds and their variations. Since many of the geomagnetic measurements extend back in time for many decades, studies related to possible long-term global atmospheric change are feasible.

The high-latitude ionosphere provides a window to the outer magnetosphere, since electric fields, electric currents, and energetic charged particle populations in the magnetosphere readily project along magnetic-field lines down to the ionosphere, where they are more easily measured. Throughout much of space, the electric field that would be measured in the frame of reference moving with the plasma nearly vanishes, because the charged particles readily adjust to cancel an electric field. If E and v are the electric field and the plasma velocity that would be measured in a reference frame fixed to the Earth, and B is the magnetic field, then the electric field in the frame moving with the plasma is E + v x B (for a non-relativistic Lorentz transformation). Then

E + v x B ~ 0

meaning that the electric field and the plasma velocity are closely linked. As electric fields project along the magnetic field, so do plasma motions. Here is an example of electric potential patterns in the northern and southern polar regions deduced from syntheses of data from many different ground- and satellite-based instruments. Both hemispheres are viewed from above the north magnetic pole (the southern hemisphere is thus viewed through the Earth), so that for both hemispheres the geomagnetic field is directed into the page. the above equation implies that plasma convects along electric-potential contours, counterclockwise around the potential highs (+) and clockwise around the lows (-). The interplanetary magnetic field at this time has a dusk-to-dawn component. In the northern hemisphere, magnetic field lines emanating from the Earth (coming out of the page) above about 78 º on the day side eventually bend to the left at great distance to join the interplanetary field, while in the southern hemisphere magnetic-field lines emanating from the Earth (going into the page) above about 78 º on the day side eventually bend to the right. Magnetic-field tension tends to pull the plasma towards dusk in this region in the northern hemisphere, and toward dawn in the southern hemisphere, while the solar wind also drags the plasma on these field lines across the pole towards midnight. At lower latitudes the geomagnetic field lines are no longer connected to the interplanetary field, and the plasma flow can return towards the day side. When the interplanetary field changes direction, as it frequently does, the plasma convection in the polar ionospheres also changes. Currently there is intensive research into understanding magnetospheric processes at the boundaries between geomagnetic field lines that interconnect with the interplanetary field and those that do not, corresponding to roughly 80 º magnetic latitude in the dayside ionosphere and 70 º on the night side. At these boundaries the approximation of the above equation breaks down, and plasma processes become much more complex.

Research on the behavior of the ionosphere, its electrodynamics, and its coupling with the neutral atmosphere and the magnetosphere, is revealing the underlying causes of that behavior. Because our society is becoming increasingly dependent on technological systems that can be impacted by ionospheric phenomena, those phenomena are being studied as part of a coordinated program of ``space weather'' research. That research seeks to characterize the variability of ionospheric density and of electric currents during magnetic storms more accurately, and to determine to what extent valid predictions of those phenomena and their effects can be made.


Acknowledgements

I thank Steve Suess, Bruce Tsurutani, and Robert Stening for helpful comments. This work was supported by the NASA Space Physics Theory Program and by NASA Order No. W-17,385.

References and Further Reading

Cowley, S.W.H., The Earth's magnetosphere: A brief beginner's guide, EOS Trans., AGU, 76, 525, 1995

Hargreaves, J. K., The Solar-Terrestrial Environment: An Introduction to Geospace - the Science of the Terrestrial Upper Atmosphere, Ionosphere, and Magnetosphere, Cambridge Univ. Press, New York, 1992

Kelley, M. C., The Earth's Ionosphere: Plasma Physics and Electrodynamics, Academic Press, San Diego, 1989

Kivelson, M. G., and C. T. Russell, (Eds.), Introduction to Space Physics, Cambridge University Press, 1995

Lu, G., A. D. Richmond, B. A. Emery, P. H. Reiff, O. de la Beaujardière, F. J. Rich, W. F. Denig, H. W. Kroehl, L. R. Lyons, J. M. Ruohoniemi, E. Friis-Christensen, H. Opgenoorth, M. A. L. Persson, R. P. Lepping, A. S. Rodger, T. Hughes, A. McEwin, S. Dennis, R. Morris, G. Burns, and L. Tomlinson, Interhemispheric asymmetry of the high-latitude ionospheric convection pattern, J. Geophys. Res. 99, pp. 6491-6510, 1994

Rees, M. H., Physics and Chemistry of the Upper Atmosphere, Cambridge Univ. Press, New York, 1989

Richmond, A. D., The ionospheric wind dynamo: effects of its coupling with different atmospheric regions, in The Upper Mesosphere and Lower Thermosphere, edited by R. M. Johnson and T. L. Killeen, pp. 49-65, Am. Geophys. Union, Washington DC, 1995.

Volland, H., Atmospheric Tidal and Planetary Waves, Kluwer Academic Publishers, Dordrecht, Netherlands, 1988.

Volland, H., ed., Handbook of Atmospheric Electrodynamics, Vol. II, CRC Press, Boca Raton, 1995


Revised, Thursday, January 3, 1997 by bill@ucar.edu