A1. THE MAGNETOSPHERE, IONOSPHERE, AND ATMOSPHERE
Appendix A1 was largely developed by the following workshop group: James Slavin (Group Leader), R. Clauer, M. Engebretson, D. Herzog, R. McPherron, J. Olson, V. Papitashvili, V. Patel, F. Rich, A. Richmond, M. Teague, R. Walker, L. Zanetti.
A complete and accurate knowledge of the geomagnetic field and how it changes with time is critical to many aspects of solar-terrestrial physics and applications affected by conditions in near-Earth space. It is the interaction of the solar wind with the geomagnetic field that creates the Earth's magnetosphere. The magnetosphere, in turn, stores energy from the solar wind and dissipates it sporadically in geomagnetic storms and substorms that accelerate large fluxes of energetic charged particles and drive large electrical currents; these currents, in part, are diverted down into the ionosphere. Upper-atmosphere winds also generate global currents through dynamo action as the conducting ionosphere is moved through the geomagnetic field. In addition to receiving steady electrical currents, the Earth is also bathed in a variety of magnetohydrodynamic waves, or pulsations, produced at much higher altitudes in the magnetosphere. The physical processes leading to these phenomena are being actively investigated by the international space physics community.
Geomagnetic storms and substorms are also important from the standpoint of the frequently deleterious effects they have on a variety of critical civilian and government equipment and functions. Energetic particles trapped in the geomagnetic field produce malfunctions and rapid aging in a variety of commercial, scientific, and military satellite subsystems. The currents induced in power and communication cables by geomagnetic storms and substorms can cause considerable damage in ground systems such as transformers. The monitoring, and when scientific advances permit, the prediction of “space weather” is an
Page 76important goal of agencies such as the National Oceanic and Atmospheric Administration (NOAA) and the U.S. Department of Defense (DOD).
At altitudes from 80 to 1,000 kilometers (km), the motion of ionospheric ions caused by the electric fields imposed by the magnetosphere drives large-scale winds in the thermosphere. Joule heating associated with these currents is an important energy source, and it can result in dramatic temperature and density increases of the upper atmosphere during magnetic storms, affecting the orbits of near-Earth satellites and dynamically altering the global ionosphere. Relativistic charged-particle precipitation may influence some aspects of the neutral atmosphere, for example, noctilucent clouds, as far down as the middle atmosphere. Understanding the coupling between the solar wind, the magnetosphere, the ionosphere, and the thermosphere is a major objective of the solar-terrestrial physics community supported by NASA, the National Science Foundation (NSF), NOAA, DOD, and other federal agencies.
As is discussed below, measurements of the direction and intensity of the local magnetic field by low-altitude satellites and at ground stations located around the world are essential for scientific progress in many areas of solar-terrestrial physics. In addition to contributing to the geomagnetic field models that are needed to describe the flow of energy and momentum between the magnetosphere and upper atmosphere, these observations provide important measures of the strength and intensity of the horizontal currents flowing overhead in the ionosphere. Careful analysis of ground magnetic variations can tell us a great deal about the nature of solar wind-magnetosphere-ionosphere couplings at high latitudes, and about the nature of the upper-atmosphere winds that drive the ionospheric dynamo at all latitudes. Following a more thorough discussion of these scientific and technological issues, this section concludes with a summary of pertinent requirements regarding ground-based and satellite-based magnetic field measurements with respect to instrumentation, ground station location, data collection, data processing, and data archiving.
The magnetosphere is defined as the region of space above the Earth's ionosphere in which charged-particle motion is dominated by the geomagnetic field. Collisions are so infrequent in most of this region that its electrical conductivity is nearly infinite. Because of this condition, magnetic fields and very-low-energy particles are frozen together, and when the plasma moves, so too do the magnetic field lines that thread it. Finite energy particles also drift in gradients of the magnetic field. These drifts are energy- and charge-dependent and thus must produce electrical currents. These currents create magnetic fields that add to the geomagnetic field originating within the Earth. Because of these currents, the total field in the magnetosphere departs radically from what is calculated by extrapolating the Earth's internal field to distances greater than several Earth radii.
The outer boundary of the magnetosphere is sharp and well defined at a distance in the sunward direction of about 10 Earth radii. It is produced by a sheet of electrical current that cancels almost all of the Earth's field outside the boundary while doubling it inside. This current layer is termed the magnetopause, and it separates the geomagnetic field from the field and plasma of the solar wind. At the magnetopause the solar wind interacts with the Earth's field by two major processes: viscous interaction and magnetic reconnection. The viscous interaction occurs along the flanks of the magnetosphere transferring solar wind momentum to closed field lines inside the magnetopause. These field lines flow tailward and create boundary layers with flows internal to the magnetosphere. Magnetic reconnection, on the other hand, allows the interplanetary magnetic field (IMF) to merge with the geomagnetic field near the subsolar point. These field lines are then transported by the solar wind over the poles and laid back behind the Earth as a long, cometlike tail. In the tail these field lines eventually reconnect and return by internal flows to their origin. These flows produced by the viscous and the dominant reconnection process are termed magnetospheric convection.
Page 78The stretching of the field lines through convection produces the tail current that flows partly on the magnetopause and partly across the interior of the magnetic tail.
The rate at which the magnetic field in the solar wind reconnects with the geomagnetic field depends on the angle between the solar wind field and the Earth's field. Only when the IMF is southward, that is, antiparallel to the Earth's field, does a strong interaction occur. Because this angle is constantly changing, the level of magnetospheric convection also changes. These fluctuations make it possible to energize and trap particles in the inner portions of the magnetosphere, creating the radiation belts. The rapid drift of protons to the west and electrons to the east creates a torus of westward current around the Earth, known as the ring current.
The electric field produced in the magnetosphere by convection is projected onto the ionosphere along magnetic field lines. However, the ionosphere has finite electrical conductivity and, consequently, an electrical current flows parallel to the electric field. This is called the Pedersen current. Since magnetospheric convection is a closed cycle, some field lines flow tailward and some sunward. The boundary between these flow regions projected onto the ionosphere is called the convection reversal. The electric field is divergent along these flow lines; therefore, so is the ionospheric current. This is possible only if field-aligned currents flow in or out of the reversal regions to satisfy current continuity. These field-aligned currents are known as Region 1 Birkeland currents. They link the magnetopause and tail currents to the Earth's ionosphere and are the primary source of energy that drives ionospheric phenomena.
A second source of electric field in the magnetosphere is the rotation of the Earth. The plasma polarizes so that the resulting electric field is everywhere orthogonal to the magnetic field. In the absence of other effects, the combination of electric and magnetic fields would cause charged particles to corotate eastward with the Earth. However, when added to the dawn-dusk magnetospheric electric field, this second field source imposes a fundamental asymmetry on the magnetosphere. A result of this is that electrons drift closer to the Earth on the dusk side and ions closer on the dawn side.
The electric field projected onto the ionosphere causes the plasma at the feet of convecting field lines to drift just as the magnetospheric plasma does. However, the ionosphere is not fully ionized, and drifting charges collide with atmospheric neutrals. The collision rate relative to the gyrofrequency is much higher for positive ions than for electrons, and therefore they drift more slowly than electrons. This produces a current in a direction opposite the drift. This current is orthogonal to both the magnetic and electric fields; it is called a Hall current. The geometry of the Hall current system is the same as that of ionospheric convection, but with the opposite sense. It is a two-cell system flowing sunward across the polar cap and antisunward in the dusk and dawn sectors of the auroral oval. The concentrations of current in these sectors are known as the eastward (dusk) and westward (dawn) electrojets. Most of the magnetic disturbance seen on the ground at high latitudes is produced by this Hall current system. Because of their solenoidal geometry, the field-aligned currents do not create large ground disturbances.
Fluctuations in the interplanetary magnetic field carried by the solar wind affect the rate of dayside reconnection, and hence cause corresponding fluctuations in the convection electric field and all of the current systems mentioned above. These changes are highly filtered and delayed because of the large inductance of the field-aligned current systems. Nightside reconnection occurs in the center of the tail current sheet, but its onset is delayed relative to the dayside. When reconnection begins in the nightside, it explosively releases the energy stored in the tail field. This release is accompanied by a new current system that temporarily diverts a fraction of the cross-tail current along field lines that close westward across the midnight ionosphere. This current is known as the substorm current wedge. Multiple substorms driven by a long interval of fluctuating southward IMF cause many particles to be energized and trapped in the radiation belts. These longer intervals of enhanced convection and intense ring currents are called magnetic storms.
Page 80Ionosphere and Atmosphere
At heights of 80 to 1,000 km, the ionosphere and upper atmosphere respond strongly to the electric fields and currents of magnetospheric origin. The rapidly drifting ions set neutral winds in motion through collisional coupling, and the resistive dissipation of electrical energy, or Joule heating, can be the dominant energy input to the high-latitude upper atmosphere. During magnetic substorms, winds approaching sonic speeds can be generated, and the heating is sufficient to raise the temperature of the upper atmosphere by hundreds of kelvins. The expanded atmosphere results in order-of-magnitude increases in the drag on near-Earth satellites, altering their orbits and affecting their orientations. The global upper-atmosphere circulation during magnetic substorms and storms is altered, resulting in a changed chemical composition and large changes in ionospheric density. The high-speed winds can significantly feed back upon the magnetospheric electrodynamics through ionospheric dynamo action. Hence, the ionosphere cannot be viewed simply as a passive sink for magnetospheric energy. Accurately explaining these various effects requires complex simulation models of upper-atmosphere and ionospheric dynamics that take into account the characteristics of electrodynamic energy input from the magnetosphere.
Even when magnetospheric energy inputs to the upper atmosphere and ionosphere are weak, electric fields and currents are generated in the ionosphere by the dynamo action of the ever-present neutral atmosphere winds. Because the ionosphere is electrically conducting, its motion through the geomagnetic field produces an electromotive force (EMF). This EMF results in electrical current flow and the establishment of space charges and electric fields. Typically, two large cells of current flow in the sunlit ionosphere, a counterclockwise cell in the northern magnetic hemisphere and a clockwise cell in the southern hemisphere, each containing on the order of 105 amperes. Current also flows along geomagnetic field lines between the two hemispheres because of the extremely high conductivity along the magnetic field. Most of the horizontal current in the ionosphere flows below 200 km, although
Page 81measurable currents can also be found at higher altitudes when the plasma and neutral densities are large, as at solar maximum in the afternoon.
A portion of the upper-atmosphere wind that drives the dynamo results from the diurnal variation in absorption of solar ultraviolet radiation in the thermosphere. Another portion is driven by upward propagation of global atmospheric waves, particularly atmospheric solar tides with 24- and 12-hour periods, but also planetary waves of longer periods and lunar tides of 12.4-hour periods. Geomagnetic data from the global network of stations thus record effects of these winds, which are dependent on the state of the middle atmosphere and can provide information about spatial and temporal variations in middle-atmospheric conditions.
The highly anisotropic nature of ionospheric conductivity can result in strong Hall polarization electric fields, especially at the magnetic equator where the horizontal geometry of magnetic field lines inhibits the discharge of these fields. The electric field drives a strong current along the magnetic equator on the dayside of the Earth, the equatorial electrojet, that produces an enhancement in magnetic perturbations on the ground and at satellite altitudes. The latitudinal variation of the currents is closely tied to the mean eastward electric field and to the vertical variation of winds in the thermosphere, so that analysis of latitudinal profiles gives important information about the dynamical state of the ionosphere and upper atmosphere at low latitudes.
The external sources such as ionospheric current systems also induce electrical currents to flow in the conducting lithosphere. Studies by magnetotelluric and global geomagnetic methods have investigated the Earth's electrical conductivity to depths of about 1,000 km. The resulting fields add to the magnetic field measured at the Earth's surface and at low-Earth orbit. It has recently become possible to refine the calculations of magnetospheric and ionospheric current processes by accounting for these induced contributions once the external ionospheric and magnetospheric magnetic fields are known. Implementation of these corrections will allow for more accurate measurements of the currents flowing overheard in the ionosphere.
Page 82Geomagnetic Pulsations and Transients
Sinusoidal variations in the Earth's magnetic field with time scales on the order of seconds to several minutes have been observed for over a century. More recently, it has been recognized that these pulsations, or hydromagnetic waves, may be used as diagnostics for magnetospheric plasma processes at high altitudes and as means of dissipating solar wind and magnetospheric energy at ionospheric altitudes. Geomagnetic pulsations with periods of minutes include field resonances and transients in the global current system. Short-period pulsations, on the other hand, are thought to be electromagnetic ion cyclotron waves. These waves are recognized as having a significant role in transporting energy between various constituents of magnetospheric plasmas.
Beginning in the solar wind, energy is coupled to magnetospheric pulsations through several mechanisms. Wave power is generated by turbulence outside the magnetopause near noon and by Kelvin-Helmholtz waves caused by velocity shear along the flanks of the magnetosphere. This wave power propagates inward and couples to toroidal (azimuthally polarized) field-line resonances. Ions energized in the nightside magnetosphere are injected into the ring current during periods of increased magnetic activity and drift westward through the late evening and dusk toward noon, providing energy for poloidal (radially polarized) field-line resonances, storm-time compressional waves, and electromagnetic ion cyclotron waves. Other, less structured waves are focused along magnetic field lines directly into the polar cusp and cleft regions.
The study of pulsations has been pursued for several reasons in addition to seeking an understanding of their underlying mechanisms. First, it allows detailed comparisons between plasma physics theory and observation in ways not possible in laboratory experiments. Examples include drift and ballooning-mode instabilities, bounce resonances, shear-driven instabilities (such as Kelvin-Helmholtz instabilities), nonlinear wave-wave coupling, and ion cyclotron wave instabilities. Pulsation studies also allow for a better understanding of mechanisms of wave mode coupling (for example, from either traveling or globally resonant compressional waves to transverse waves and back again), of
Page 83feedback effects due to local, diurnal, and seasonal variations in ionospheric parameters, and of the importance of spatial boundaries as sources and/or transmitters of plasma waves.
Second, such studies can provide diagnostics of dynamic and static features within the Earth's plasma environment. Many dayside high-latitude magnetic pulsations appear to be related to plasma interactions occurring in the magnetospheric cusp/cleft regions, which link the high-latitude ionosphere to the outer boundary of the Earth's magnetosphere, or to interactions occurring directly at these remote boundaries. Because satellite orbits allow only brief crossings through this complex and critical region, ground-based magnetometers provide the only means of obtaining synoptic coverage of cusp/cleft processes.
Third, geomagnetic pulsations are used for electromagnetic sounding of the solid Earth. By sensing the induction effects of external pulsations, one can infer considerable detail about properties of the underlying crust and mantle.
In the following sections we discuss the present-day knowledge and important unresolved questions associated with magnetic pulsation activity, beginning with the lowest frequencies.Long-Period and Impulsive Events
Long-period (3- to 15-minute) waves observed at latitudes connected magnetically to the outer boundary of the magnetosphere are designated as being in the Pc 5 period range. These waves, with amplitudes of tens of nanoteslas, include both solitary and continuous-wave trains, with an occurrence distribution spread throughout dayside local times from dawn to near dusk. As a result of intensive study of ground-based observations, many of the magnetic pulsations in this category have been found to be associated with traveling ionospheric plasma vortices produced by moving filamentary field-aligned currents that appear to originate near the inner edge of the low-latitude boundary layer. The generation of solitary vortices appears to be related to transients generated by changes in the interplanetary magnetic field or solar wind pressure. The more continuous events, on the other hand, appear to be caused by velocity shear
Page 84instabilities at the magnetospheric boundary (most likely the Kelvin-Helmholtz instability). In this interpretation, these pulsations appear to be the ground signature of driven magnetospheric reorientations traveling tailward along the magnetopause at magnetosheath speeds. How frequently these occur and under what circumstances remain open questions. Irregular Variations
Variations of large-scale field-aligned and ionospheric currents can also cause large, less structured variations in magnetic fields near the Earth's surface. Those with longer periods, known as Pi 2 pulsations, often resemble an impulsive ringing and are associated with auroral electron precipitation. Near local midnight, Pi 2s are associated with auroral substorm onsets and can reach amplitudes of tens of nanoteslas. At times these signatures can be observed worldwide. How they propagate to equatorial latitudes and to all local times is not yet understood. Higher-frequency irregular pulsations (Pi 1), extending to and even past local noon, appear to be much more highly localized, and are apparently produced by transient currents in the ionosphere due to variations in ionization and conductivity. These latter pulsations are clearly associated with the aftermath of substorm activity.Pc 3-4 Pulsations
Pulsations in this period range (20 to 50 seconds) are a ubiquitous feature of the dayside magnetosphere, occur over a wide range of latitudes, and typically have amplitudes of from 1 to 10 nanoteslas. Their dependence on IMF orientation and magnitude suggests a driving mechanism related to “upstream waves” and/or processes at the subsolar bow shock, which exhibits oscillatory behavior when its shock geometry becomes quasi-parallel. The mechanism that relates upstream phenomena to dayside field-line resonances in this same frequency band has been studied for many years but still is not understood. However, there is
Page 85growing evidence that processes in the cusp/cleft may play a central role in the transmission process.
Multipoint studies have shown that Pc 3 pulsations are most intense in the dayside cleft regions, and recent observations of relations between magnetic, optical, and very low frequency (VLF) modulations at Pc 3 frequencies, with clear IMF control of all three types of signals, suggest that at least along field lines connecting to cusp/cleft latitudes there are some fundamental unexplored interactions between various kinds of waves and plasma populations. Simultaneous and carefully synchronized information from magnetometer arrays at a variety of longitudes and latitudes will be especially valuable in determining whether cleft and/or cusp sources are involved in the transmission of these pulsations to the ground at lower latitudes, and can help to provide further tests of the proposed wave-entry mechanisms from an upstream wave source.Pc 1-2 Emissions
Pulsations in this frequency range (0.1 to 10 hertz) are the result of instability in the hot, anisotropic ion distributions that are regularly set up within the magnetosphere and within the boundary layers, magnetosheath, cleft, and cusp. As these ions become unstable to the growth of electromagnetic ion cyclotron waves, their energy and anisotropy decrease as waves are emitted along magnetic field lines. These waves (with amplitudes usually less than 1 nanotesla) serve two functions in magnetospheric research. First, they can function as diagnostics of remote processes and particle populations if their source regions and particle populations can be independently documented. Second, they are indicative of processes responsible for precipitating medium-energy ions out of trapped orbits.
Of particular interest are high-latitude Pc 1-2 bursts, frequently observed in association with traveling vortex events, which appear to originate on cusp or cleft field lines. It is possible that the field compressions often associated with these burst events increase the magnetic field strength and/or alter the particle distributions sufficiently to make ion cyclotron waves grow. It has also been proposed that Pc 1-2 observations
Page 86can be used as indicators of local field lines that map to magnetospheric boundaries. Recent reports indicate that, for example, unstructured 0.15-to 0.4-hertz noise is seldom observed more than a few degrees from the cusp.
Scientific Issues To Be Addressed
Solar Wind-Magnetosphere-Ionosphere Coupling. Many aspects of dayside solar wind-magnetosphere-ionosphere coupling are still poorly understood. The physical processes to be elucidated include various forms of reconnection, particle diffusion, and wave penetration of the magnetopause. Two dimensional magnetometer arrays at high magnetic latitude (approximately 60° to 80°) could provide data to map the space-time development of ionospheric currents produced by reconnection and other energy/momentum transfer processes in the dayside and nightside magnetosphere.
Substorm Electrodynamics. The detailed coupling of the magnetotail to the nightside ionosphere is not adequately modeled. Even the relative importance of the two regions and the peculiar quasi-periodic development of the aurora and westward electrojet during substorm expansion phase remain to be explained. An all-digital, real-time auroral network magnetometer array is needed to define the auroral electrojet indices at all levels of activity and to track the details of electrojet development. Observations from a smaller number of stations should be monitored in real time to detect Pi 2 pulsations, which signal the onset of substorm activity.
Global Magnetosphere Models. Improved representations of the sources of external magnetic field are important for further progress of the geomagnetic community. Because the solar wind is changing with time, these models must be parameterized to include solar wind dynamic pressure and substorm phase as well as dipole tilt. Magnetic data from past, present, and future spacecraft collected in a unified data base and annotated with time, location, solar wind conditions, and internal state of the magnetosphere would provide the basis for such a model.
Global Ionospheric Dynamo. More realistic modeling of the global ionospheric dynamo, using realistic simulations of thermospheric dynamics, would allow full advantage to be taken of the wealth of information contained in geomagnetic data relating to quiet-day variations. Not only would these global data place strong constraints on the uncertain model inputs, but they would guide the evaluation of day-to-day, seasonal, and long-term variations in the wind systems that propagate upward from the middle atmosphere. In addition to improving our understanding of thermospheric and middle-atmospheric dynamics, such studies may also shed light on long-term changes in the global atmosphere.
Ionospheric Electric Fields. An improved model of ionospheric electric fields, taking into account their dependence on magnetospheric and thermospheric conditions, is a critical need for a variety of upper-atmospheric studies and applications. Simulation models of thermospheric and ionospheric dynamics have shown that winds, temperatures, and electron densities depend very sensitively on the electrodynamic conditions. It is also becoming apparent that there is significant feedback on the magnetospheric electrodynamics from winds generated at high latitudes. Further modeling of coupled magnetosphere-ionosphere dynamics is essential to help clarify the nature of these interactions.
Magnetic Pulsations as Magnetospheric Diagnostics. There is great promise that magnetic pulsations can be used as diagnostics to provide real-time monitoring of plasma densities throughout large regions of geospace, to identify magnetic field lines connecting to the magnetospheric boundary, and to warn of substorm onset. Measurements from dense magnetometer arrays deployed at auroral latitudes will be essential if further progress is to be made in pulsation research.
Applications of Geospace Studies
The interaction of the solar wind with the Earth's magnetic field produces a variety of phenomena collectively known as geomagnetic activity. Magnetospheric substorms, magnetic storms, and pulsations are the three main classes of geomagnetic activity. Each of these produces
Page 88measurable and often significant effects on human activities. Occasionally these effects are so severe that expensive systems are significantly degraded or even destroyed. Repair or replacement can cause long delays in service and enormous expense.
To properly assess the causes of malfunctions and anomalous operations and to prepare systems so they will not fail in extreme circumstances, NOAA and the U.S. Air Force (USAF) cooperate in the acquisition of environmental data. At the NOAA/Space Forecast Center (in Boulder, Colorado) these data are used by space weather forecasters to alert civilian and military customers of current and expected magnetic activity. At the Air Force Space Forecast Center (near Colorado Springs), these and other data are fed to a variety of models to specify the current state of space weather at locations lacking direct measurements and to forecast space weather at given locations.
Examples of systems affected by space weather are extremely varied. Shortwave radio communications are interrupted. Navigation systems for ships and aircraft experience degraded accuracy. Electrical utilities encounter large transients that may damage transformers or trip overload protection. Long-distance pipelines undergo accelerated corrosion. Airline personnel and passengers on transpolar flights are subjected to increased levels of penetrating radiation. Small computers experience memory failures due to the same particles. Low-altitude spacecraft are subjected to increased drag caused by ionospheric heating. Other spacecraft that use the magnetic field for orientation suffer degraded attitude control because of large changes in field orientation. Spacecraft at higher altitudes often experience “single event upsets” from solar cosmic rays and energetic particles trapped in the Van Allen radiation belts. Solar cell arrays experience serious degradation due to energetic particle bombardment. Satellites at geosynchronous orbit are bathed in 10- to 100-kiloelectronovolt electrons raising the potential of the spacecraft surface relative to its interior to the point that sparks jump between electrical components causing anomalous signals.
Such problems require practical applications of the knowledge acquired about the space environment. As technology continues to reduce the size of electronic components and as society becomes increasingly
Page 89dependent on electronic systems that operate in inclement space weather, it becomes critical to maintain adequate monitoring of this weather so that protective measures can be taken. For these reasons as well as for scientific reasons discussed elsewhere in this appendix, solar wind, and synchronous and polar orbiting environmental platforms measuring magnetic field variations have important applications.
Operational ConsiderationsNeed for Comprehensive Magnetic Field Measurements
Magnetospheric physics is primarily concerned with the study of processes through which the solar wind interacts with the Earth's magnetic field to create and change the various current systems described previously. The present understanding of these currents has been obtained through innumerable studies utilizing data taken at ground stations and by spacecraft orbiting the Earth. Today little progress can be made unless magnetic field data are simultaneously available from the solar wind, geosynchronous orbit, the magnetotail, low-Earth orbit, and ground stations distributed around the globe.
Magnetic indices derived from the ground data are used to characterize the level and type of activity so that certain time intervals can be chosen for detailed study and long-term changes in the magnetosphere and ionosphere can be detected. These indices are also used to select measurements for the construction of average models of the external field. A primary use of these average models is to extrapolate the observations taken at various ground observatories and by remote-sensing spacecraft (such as those that image the aurora) into the magnetosphere for comparison with higher-altitude satellite observations. Another major use of magnetic data is to accurately model the currents that produce the ground magnetic variations. The ability to quantitatively remove these external contributions to the field measured at the Earth's surface would greatly enhance the ease and accuracy with which a variety of geomagnetic mapping tasks could be carried out.
Page 90Observational Concerns
For the purpose of measuring and understanding the magnetic fields and currents in the magnetosphere and ionosphere, a suite of magnetometers on spacecraft and on the ground is required. This requirement can be satisfied only by maintaining the existing platforms, augmenting them with modern instrument packages, and adding some new platforms.Spacecraft Observations
The source of all energy to the magnetosphere and much of the energy to the ionosphere is the solar wind, with its imbedded interplanetary magnetic field. In order to support research and operational activities, it is essential that solar wind data (with at least 5-minute temporal resolution) be continuously available to users in near real time. Presently Interplanetary Monitoring Platform (IMP)-8 provides such data, but the data are collected by NASA only about one-third of the time and are not available to the users until months after they are taken. The Wind satellite will soon be launched by NASA to provide IMF and solar wind data, in support of the International Solar-Terrestrial Physics (ISTP) Program, but only a limited amount of data will be available in real time. A permanent platform for monitoring the IMF and solar wind that could be deployed following the ISTP Program is clearly needed.
Inside the magnetosphere the geomagnetic field is now monitored at geosynchronous orbit by magnetometers on the Geostationary Operational Environmental Satellites (GOES). Unfortunately the normal complement of two GOES satellites does not allow for a complete coverage in local time. Another magnetometer at geosynchronous orbit is needed 6 to 8 hours in local time from both GOES East and GOES West.
Between geosynchronous orbit and the solar wind, there is no requirement for continuous magnetic field measurements, but significant scientific and operational benefits would result if more satellites in this sector of space carried a magnetometer suitable for scientific measurements of the magnetospheric magnetic field.
Existing satellite data sets provide a foundation for the development of models of the magnetospheric field, but the effectiveness of addressing critical scientific problems depends on the acquisition of new measurements. Between geosynchronous and low-Earth orbit, available scientific magnetic field data are inadequate for developing accurate models of the magnetospheric field. The Combined Release and Radiation Effects Satellite (CRRES) mission was providing suitable data when its mission was terminated prematurely by a subsystem failure. When new scientific and operational missions in this region of space are developed, they should include a research-grade magnetometer to support modeling of the magnetospheric field as a function of IMF direction, substorm activity, and other effects. Low-Earth orbit measurements of the electrical currents flowing between the magnetosphere and the ionosphere are also essential to the modeling of both regions. Since the characteristic time scale of the individual substorm phases is approximately 20 minutes, a measurement of the disturbance field is needed at intervals of 20 minutes or less. Present plans call for two magnetometers on Defense Meteorological Satellite Program (DMSP) satellites. To obtain the required time resolution, a magnetometer on at least one more spacecraft is required. The NOAA/TIROS (Television Infrared Observation Satellite) are excellent platform candidates to fulfill this need.
In low-Earth orbit there is also a critical need for a dedicated geomagnetic field-mapping payload to update the models of the internal magnetic field of the Earth. The Earth's magnetic field is changing rapidly enough that a new model is required every 5 to 10 years. Ground-based magnetic observatory data are important for this updating, but for the best possible model over the whole Earth, satellite data are an absolute requirement. Magsat (1979-1980) was the last satellite that completely met this need. Magnetometers on board the Dynamics Explorer (DE)-2 (1981-1983) and the Polar Orbiting Geomagnetic Satellites (1990-1993), for example, provided data that were useful. However, their utility was limited by the lack of a high-accuracy attitude transfer system and an absolute scalar magnetometer. Magnetometers on board future satellites of the DMSP series will provide useful data sets, but again, with limited accuracy. There are possibilities for future NASA
Page 92missions with fully capable instrument packages, but none of these has progressed beyond feasibility and design studies. Ground Stations
In order to characterize the disturbance magnetic field induced in the solid Earth due to currents in both the ionosphere and magnetosphere, data from ground-based geomagnetic observatories are essential. These data are used both for modeling the complete current system and for the creation of magnetic indices that give a measure of the level of large-scale magnetic disturbances. In addition to the standard ground-based observatories, temporary observatories are also necessary for characterizing the disturbance with the finer detail required for research purposes. In order to meet future research and operational requirements, all existing stations need to be maintained. If a particular country is having difficulty in supporting its existing stations, the international community should take steps to preserve these vital resources.
Many of the existing observatories also need to be upgraded. All observatories must be able to record the magnetic field digitally at a sampling rate of at least 1 vector per minute, and the data must be put into a data base accessible in real time through a computer network such as the one being developed for the International Real-Time Geomagnetic Observatory Network (INTERMAGNET). At least half of the high-latitude stations should have the capability to record data at higher sample rates for studies of magnetic pulsations and other rapidly varying phenomena. Many types of pulsations, for example, Pc 3-4, Pc 5, and Pi 2, can be adequately resolved with 1-second resolution measurements. The study of Pc 1-2 class pulsations, however, requires sample rates of at least 2 vectors per second.Station Spacing
For adequate spatial resolution (by data from permanent magnetic observatories) of the main field and field variations due to magnetospheric and ionospheric sources, the latitudinal spacing of the stations should be
Page 93no greater than 10° at equatorial and middle latitudes. In the auroral region and the polar cap, a latitudinal spacing of 3° or less is desirable. Where possible, the spacing between stations in local time should be no more than 2 hours. At present, large gaps exist in coverage, such as in the Pacific Ocean and in Siberia. For specific research purposes, temporary stations are needed with a spacing of approximately 100 km in both latitude and longitude. Such stations would be in place for periods of 1 to 5 years. In the past, such temporary stations have been custom-built by various scientific groups and later dismantled. With the development of modern fluxgate magnetometers and the advent of low-cost data retrieval, storage, and communications hardware, appropriate agencies should procure reusable geomagnetic observatory stations. The availability of several dozen would support many important studies at reasonable cost. Data Retrieval
It is desirable to have as much data as possible returned in near real time and deposited in a central processing facility or cataloged into a distributed system. For example, the Alaska meridian chain data from 10 sites are telemetered, via the GOES satellite, to the NOAA/Space Environment Lab where they are available with a delay of less than 15 minutes. While disk and tape storage should be used at field sites as backups to the normal data flow, users should be able to access data sets in near real time. In cases where users need to monitor activity from a large number of stations but do not need every data point, near real time values of the power spectral density of the data could be calculated either at the station or at the central facility.Data Archival
Quality control, storage, and distribution of data from magnetic observatories are vital to many research projects. In the past, many magnetograms have been available only in analog form—a form that is cumbersome to use and expensive for a single researcher to digitize. It
Page 94would enhance many studies and make other studies possible, if all magnetograms were made available in digital form. In the past, digital data distribution meant use of reel-to-reel data tapes. It is now desirable to store and distribute massive amounts of data on compact disks. It would also be very helpful to users if the World Data Center would organize such distribution disks with magnetometer data from all stations for a given week or month instead of the present practice of distributing long data sets from a single station on individual tapes or disks.
General RecommendationsData Management Requirements
A quantitative understanding of the magnetosphere and the thermosphere requires the acquisition of a comprehensive suite of spacecraft and ground-based observations that must be analyzed in association with complex models. It is essential that the outstanding research questions identified earlier be addressed in the context of an efficient and flexible data management environment capable of presenting a wide variety of data to a physically diverse science community in readily usable fashion. The data sources are very diverse. Most modern spacecraft missions are international in nature, with data being processed and analyzed at multiple locations. Ground observations, particularly those relating to the magnetic field, are by their nature multipoint, crossing many agency and national boundaries. Not all collaborative programs involving multiagency and multinational participation give appropriate emphasis to the importance of data exchange and data management planning. The purpose of this section is to identify the data management requirements necessary to support the development of answers to the outstanding research questions posed earlier.
A key requirement for the scientific community is to have timely and effective access to the complete suite of readily usable data necessary for the study of a particular scientific problem. For the proposed initiative, “timely” has a variety of meanings. Certain problems may be approached
Page 95only if real-time geomagnetic data are available. The U.S. Geological Survey (USGS) INTERMAGNET program is endeavoring to establish a real-time data base containing 1-minute ground-based magnetometer observations from around the world. The NOAA Space Environment Laboratory Data Acquisition and Display System (SELDADS) program provides real-time spacecraft and ground data that support a variety of forecasting activities relevant to this initiative. This initiative should endeavor to maximize use of these systems and should advocate the expansion of the data-base contents through the inclusion of such data sets as the DMSP magnetometer.
Three essential aspects of “readily usable” are generally not well addressed by modern data systems. First: “Where are the data?” Particularly for geomagnetism, there is a requirement for a comprehensive catalog that is on-line accessible through a diverse set of computer hardware. Second: “Can I use the data easily?” The answer relates to documentation and format and the use of Standard Data Formats within the geomagnetic community. It is not necessary to constrain the community to the use of a single standard, since conversion between one standard and another is relatively simple. The National Office of Standards and Technology (NOST) at NASA/Office of Space Science and Applications (OSSA), for example, supports such activities. Third: “Can existing data be used?” A large body of data exists in the geomagnetic arena. Some components of this are in danger of extinction, or at least of underutilization as a result of a lack of proper maintenance and support. Ground-based magnetometers fall into this category because the number of sites is decreasing and certain data sets are available only in analog form. The U.S. geomagnetism community would benefit greatly from digitization of selected existing data sets.
In summary, the data management requirements to support the present initiative are as follows:
extensive collaboration with existing and planned data systems;
real-time and on-line interactive data bases;
data systems that catalog, display, and deliver data in on-line fashion;
extensive use of Standard Data Formats; and
a data rescue program.
Magnetic variations on the Earth's surface are produced by a variety of electrical currents in the magnetosphere and ionosphere. The patterns of disturbance are characterized by magnetic indices that crudely represent the important physical properties of these current systems. The Dst (disturbance storm time) index is proportional to the total energy in the drifting radiation belt particles that make up the torus called the ring current. The AE (auroral electrojet) indices are roughly proportional to the density of horizontal overhead currents that flow near midnight in the auroral oval. The Polar Cap Index measures the current density in the overhead currents flowing across the polar cap. These indices provide essential tools that allow researchers to identify quiet and disturbed intervals and to select different phases of magnetospheric substorms and magnetic storms for detailed study. Improvement of these indices through the addition of ground stations for improved spatial coverage, increased time resolution, better removal of uninteresting currents, real-time generation, and rapid dissemination to users are matters of great importance to the geomagnetic community. Accordingly, the relevant agencies should work together to meet the following goals:
upgrade the current network of geomagnetic observatories used for the AE index, replacing analog stations with digital and adding real-time satellite links;
expand the AE network with additional stations at crucial points in eastern Canada, eastern Siberia, and possibly the Southern Hemisphere auroral oval; stations located at higher and lower latitudes to record electrojet activity for expanded or contracted auroral ovals should also be established when resources permit;
calculate the Dst index with 1-minute resolution and make it available in real time via network access;
obtain real-time digital data from appropriately located polar cap stations and compute a real-time Polar Cap Index; and
continuous monitoring of Pi 2 pulsations from three ground stations spaced evenly in longitude around the globe should be undertaken and an index representing Pi 2 pulsations should be created and distributed.
A variety of temporary (1- to 5-year) arrays of magnetometers is needed to make progress in the research problems described above. At high latitudes, dense (approximately 100- to 200-km separation) two-dimensional arrays are required to investigate the detailed properties of magnetic pulsations and the electrodynamics of high-latitude current systems. Where possible, arrays should be deployed conjugately in the Northern and Southern hemispheres and should take advantage of complementary measurements from other ionospheric instruments such as radars, riometers, and optical imagers and spectrometers. Because of the synergism between such arrays and the even denser arrays needed for electromagnetic induction studies, these efforts should also be coordinated whenever possible. Sampling rates should range from approximately 5 seconds to 0.5 seconds to investigate Pc 5 to Pc 1-2 scale variations. At middle or low latitudes, magnetometers are required at about 30° longitudinal increments to investigate variations due to solar wind dynamic pressure changes, substorm current systems, the ring current, and the ionospheric equatorial electrojet. These scientific objectives can be reached if the following actions are taken:
deploy dense (100- to 200-km separation) two-dimensional arrays of temporary observatories at high latitudes and one-dimensional arrays at lower latitudes;
deploy conjugate magnetometer arrays at appropriate locations;
deploy a small number of new permanent stations in poorly covered regions of the globe; and
coordinate the siting of arrays for magnetospheric studies and electromagnetic induction studies whenever possible.
Satellites provide the only platform that allows the observation of the flow of energy from the solar wind to the magnetosphere and the ionosphere. Near-real-time data from satellites are needed to warn of major geomagnetic disturbances that affect commercial systems and to support scientific investigations such as rocket and balloon campaigns. In short, satellite measurements provide the only way to “image” the magnetic field across the vast volume of the magnetosphere. To accomplish these goals the following requirements are established:
The interplanetary magnetic field and the solar wind in the vicinity of the Earth must be measured continuously. These data must be made available to users on a near-real-time basis.
Measurements of the magnetospheric magnetic field at three or more equi-spaced geosynchronous spacecraft must be made available to users in near real time.
At least three of the available operational and research satellites in polar low-Earth orbit (for example, TIROS, DMSP) must be instrumented with magnetometers to “image” the large-scale three-dimensional ionospheric current system.
Measurements of the main field of the Earth with a high-accuracy vector/scalar magnetometer system (for example, Magsat) capable of updating the International Geomagnetic Reference Field (IGRF) model must be obtained at least once per decade.
A major need exists for improved quantitative models of the magnetic fields generated by currents external to the Earth. Near-Earth external field models must be time dependent and include the effects of both ionospheric currents and magnetospheric dynamics. In the case of the magnetospheric models, the effects of Birkeland currents must be included and representations of the magnetopause currents and of the near-Earth magnetosphere must be improved. The models must include the changes that occur in magnetospheric configuration in response to changes in the solar wind plasma and interplanetary magnetic field. A goal should be to model the perturbations in the geomagnetic field at low and middle latitudes to within a total error budget of 10 nanoteslas. Conversely, studies of the magnetospheric magnetic field require improved models of the internal magnetic field for analyzing the ground signatures of magnetospheric activity and pulsations. These scientific and operational objectives can be met only if the measurement requirements identified above are met and strong interagency support is provided for theoretical and empirical modeling of ionospheric and magnetospheric electrical currents and magnetic fields.