Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
5 Space and Astrophysical Plasmas PRINCIPAL CONCLUSIONS 1. The increasing precision of measurements, numerical modeling, and theory applied to space plasma problems amounts to a revolution in technique relative to 10 years ago. As a result, the study of space plasmas has become one of the primary motivations and experimental arenas for basic plasma research. The solar system is the primary laboratory in which astrophysical plasma processes of great generality can be studied in situ. 2. Many practical systems, both civilian and defense, must operate in the highly variable and potentially hostile plasma environment of the Earth and solar system. Plasma processes in this environment also influence and even disrupt important ground-based systems over local and regional scales. 3. Because of the wealth of pertinent information flowing from solar-system plasma physics, and continuing advances in large-scale numerical modeling, magnetohydrodynamics and plasma physics are becoming central to the interpretation of many astronomical observa- tions. Studies of plasma behavior in extreme astrophysical environments, such as pulsars, enrich basic theory and may suggest future laboratory investigations and technology development. 4. Cosmic-ray observations provide important information about 243
244 PLASMAS AND FLUIDS space and astrophysical plasmas. The plasma physics of cosmic-ray acceleration and transport has made especially rapid progress in the past decade. The improved precision of cosmic-ray-composition measurements now makes possible quantitative tests of theories of nucleosynthesis and galactic chemical evolution. PRINCIPAL RECOMMENDATIONS To the federal agencies and advisory panels concerned with space and astrophysical plasma physics we make the following recommen- dations: 1. Observations, measurements, and experiments, in space and on the ground, are the key to productive research in space and astrophys- ical plasma physics. We recommend implementation of the comprehensive research strategy outlined in Solar-System Space Physics in the 1980's (Space Science Board, 1980~. These programs, and especially the International Solar-Terrestrial Physics Program and the Solar Optical Telescope, are the primary ones that will explicitly contribute to our knowledge of the physical processes in large-scale plasmas. We endorse the programs proposed in Astronomy and Astrophysics for the 1980's (Astronomy Survey Committee, 1982) because they could make significant contributions to many problems in plasma astrophysics. 2. We recommend a national computational program dedicated to basic plasma physics, space physics, and astrophysics that will main- tain the state of the art in the technology appropriate to large-scale theoretical models and simulations and provide access to users on the basis of peer review. To the academic community 1. We recommend that plasma physics become a regular part of the university science curriculum in view of the increasing precision of its experimental and theoretical techniques, and in view of its many applications to space physics, astrophysics, and technology.
SPACE AND ASTROPHYSICAL PLASMAS 245 INTRODUCTION The advances in understanding the plasmas in the laboratory, in space, and in astrophysics have reinforced one another throughout the twentieth century. In the 1920s, plasma oscillations were discovered in the laboratory and radio waves were reflected from the plasma in the Earth's ionosphere-the very edge of space. Between 1930 and 1950, the foundations of plasma physics were created as a by-product of ionospheric, solar-terrestrial, and astrophysical research, motivated by such diverse concerns as understanding how radio waves propagate in the ionosphere, how solar activity leads to magnetic storms and auroral displays at Earth, and the role of magnetic fields in the behavior of stars and galaxies. By the 1940s, it had become clear that, unlike ordinary gases, fully ionized plasmas at high temperatures are collision free an essential property that highlights the collective processes that are fundamental to plasmas. Modern plasma physics began in the l950s. Two events symbolizing the deeper intellectual currents of those years were the first launch of an artificial Earth satellite by the Soviet Union and the revelation, through declassification, that both the United States and the Soviet Union had been attempting to harness the energy source of the Sun- thermonuclear fusion for peaceful purposes. Then as now, the obsta- cles to achieving controlled fusion lay not in ignorance of nuclear physics but of plasma physics. By 1960, the Van Allen radiation belts and the solar wind had both been discovered by spacecraft. These discoveries demonstrated that future understanding of the space envi- ronment of the Earth and Sun would also be expressed in terms of plasma physics. Two powerful motivations stimulated the growth of plasma physics after 1960. Fusion research seeks a source of energy accessible to human use that will last for a time comparable with the present age of the Earth. Space research seeks useful comprehension of nature's processes on a global and, indeed, solar-system scale, in recognition of man's dependence on his environment. It is significant that the same discipline of physics plasma physics- is central to both fusion and space research. Moreover, the plasma phenomena in the solar system have proven to be examples of general astrophysical processes. Not only does plasma physics describe both solar-system and astrophysical phenomena, but the solar system has become a laboratory in which astrophysical processes of great gener- ality can be studied in situ.
246 PLASMAS AND FLUIDS RELATIONSHIP BETWEEN LABORATORY, SPACE, AND ASTROPHYSICAL PLASMA RESEARCH Definition of Space and Astrophysical Plasma Physics Space and astrophysical plasma physics comprise many subjects with distinct historical origins. Space plasma physics includes solar and solar-wind physics, planetary ionospheric and magnetospheric phys- ics, cometary physics, and the study of cosmic-ray acceleration and transport in the solar system. Solar research stands at the interface between space physics and astrophysics. The Sun's proximity makes it possible to make measurements, pertinent to the Sun's interior struc- ture and to the plasma phenomena in its surface layers, that are obtainable for no other star. The subject of plasma astrophysics includes the generation of magnetic fields in planets, stars, and galaxies; the plasma phenomena occurring in stellar atmospheres, in the interstellar and intergalactic media, in neutron-star magneto- spheres, in active radio galaxies, and in quasars; and the acceleration and transport of cosmic rays. Astrophysical questions motivate the study of relativistic plasmas. Each of these subjects depends on, and contributes to, laboratory plasma physics. Each has traditionally been pursued independently. Only recently has there been a tendency to view them as one unified discipline. Relationship Between Laboratory and Space Plasma Physics The Study Committee on Space Plasma Physics (Space Science Board, 1978) expressed this relationship as follows: Space and laboratory experiments are complementary. They explore different ranges of dimensionless physical parameters. Space plasma configurations usually contain a much larger number of gyroradii and Coulomb mean-free paths than is achieved in the laboratory plasma configurations. In the labora- tory, special plasma configurations are set up intentionally, whereas space plasmas assume spontaneous forms that are recognized only as a result of many single-point measurements. Space plasmas are free of boundary effects; laboratory plasmas are not, and often suffer severely from surface contamina- tion. Because of the differences in scale, probing a laboratory plasma disturbs it; diagnosing a space plasma usually does not. The pursuit of static equilibria is central to high-temperature laboratory plasma physics, whereas space physics is concerned with large-scale time-dependent flows....
SPACE AND ASTROPHYSICAL PLASMAS 247 Certain problems are best studied in space.... Certain problems could be more conveniently addressed in the laboratory.... Theory should make the results of either laboratory or space experiments available for the benefit of the whole field of plasma physics. The recent strengthening of theoretical space plasma physics, to- gether with the increasing capability of space plasma instrumentation and the superiority of the space environment for certain types of measurements, means that the experimental diagnosis and theoretical interpretation of some space-plasma processes now matches in preci- sion the best of current laboratory practice. This is especially true in the field of wave-particle interactions, where non-Maxwellian particle distributions, and the plasma waves they create, have been measured with such high resolution that theoretical instability models had to be increased significantly in precision. Relationship Between Space and Astrophysical Plasma Research The study of plasmas beyond the solar system has developed more slowly than space plasma physics for a fundamental reason: the microscopic plasma processes that regulate the behavior of astrophys- ical systems cannot be observed directly, as they can in space and in the laboratory. Now, however, the modern theoretical and computa- tional techniques developed to understand laboratory and space mea- surements have opened the door to modeling of the plasmas in the still larger and more exotic environments of astrophysics, where observa- tion suggests primarily the starting point in model development. The interplay between small- and large-scale processes is character- istic of space and astrophysical plasmas. Magnetohydrodynamics (MHD) describes large-scale fluid systems and identifies, locates, and characterizes the small-scale plasma processes that regulate their global dynamics. In general, the MHD flow and the associated plasma processes must be modeled simultaneously to achieve complete and self-consistent understanding. Many of the MHD systems studied in the solar system have important analogs in astrophysics. Space and astrophysical systems naturally involve similar plasma processes. We illustrate these remarks by discussing two types of MHD systems, winds and magnetospheres, and two plasma processes, particle acceleration and magnetic-field reconnection, that occur in them.
248 PLASMAS AND FLUIDS MAGNETOHYDRODYNAMIC ATMOSPHERES AND WINDS The outer layers of the Sun are a convective heat engine, whose motions produce both large- and small-scale magnetic fields. These magnetic fields do not spread uniformly over the solar surface, but instead concentrate into intense, small-scale flux tubes. The evolution of these surface magnetic fields occurs on time scales millions of times shorter than predicted by classical kinetic theory. The underlying processes compressible convection, the interaction of turbulent con- ducting fluids with magnetic fields in a convecting atmosphere, mag- netic buoyancy are ubiquitous and yet poorly understood. Most of the planets have magnetic fields whose origins are due to dynamo action in their interiors; the vast majority of stars in our Galaxy are now believed to have magnetic fields much like the Sun's; and our Galaxy contains interstellar magnetic fields that are also believed to arise because of dynamo action. The presence of magnetic fields in the Sun's outer layers has yet another consequence: the interaction between the surface magnetic fields and turbulent motions heats the solar corona. Although we do not yet know the precise mechanism responsible for the heating, we do know that most of the heated plasma is trapped by the solar magnetic fields and is observed to emit vigorously in the UV and in x rays; some of it, however, escapes into interplanetary space from open magnetic structures in the solar corona. This escaping hot gas is subsonic near the Sun but becomes supersonic as it flows outward to become the solar wind. l his wind carries outward not only plasma and energy but also the embedded magnetic fields and angular momentum. Thus, the solar wind carries energy away from the solar corona and decreases the Sun's mass, magnetic flux, and angular momentum. The transport of angular momentum is sufficiently vigorous to account entirely for the Sun's loss of angular momentum since it reached the main sequence. The solar wind is finally decelerated to subsonic speeds when it en- counters the interstellar medium. The solar-wind injects both nuclear- processed matter and magnetic fields into the interstellar medium. Magnetized atmospheres and winds of the kind just described are exceedingly common. Plasma streams out into space from the planets' polar ionospheres in miniature versions of the solar wind polar winds. The Einstein observatory has shown that stars with convecting outer layers have x-ray-emitting coronas, indicating surface magnetic activ- ity and winds much like the Sun's; these stars constitute the vast ma- jority of stars in our Galaxy. UV and x-ray observations have also shown that highly evolved stars with convecting outer layers do not
SPACE AND ASTROPHYSICA ~ PLASMAS 249 trap the heated plasma to form x-ray-emitting coronas, but rather eject the gas in the form of extremely massive winds. Indeed, much of the interstellar medium is filled with the blended wind material from these evolved stars and from supernova remnants. Because the densities, velocities, temperatures, and magnetic-field strengths in the interstellar medium are similar to those in the solar wind, many in situ observa- tions of the interplanetary medium are automatically relevant to astrophysics. The interstellar plasma may also expand out of our Galaxy as a wind. Winds that are confined by surrounding gas pressure take the form of collimated bipolar jets, which are observed to flow away from such diverse systems as stars in the early phases of formation, the exotic compact stellar system SS-433, and radio galaxies and quasars. Super- high-energy, relativistic plasma winds appear to flow away from pulsars and active galactic nuclei. The solar and atmospheric winds are the only astrophysical fluids accessible to detailed diagnostics and in situ measurement. Since the solar wind in particular has been as completely diagnosed as any laboratory plasma, a detailed theoretical understanding of it is being developed. PLANETARY AND ASTROPHYSICAL MAGNETOSPHERES The Earth has an atmosphere above the one we breathe that is made of plasma the magnetosphere. Beyond the magnetosphere, the plasma behavior is controlled by the solar wind. Within it, the Earth's magnetic field organizes the behavior of the plasma; it traps energetic particles to form radiation belts; and it transmits MHD stresses between the magnetosphere and atmosphere, a process that leads to auroras. The solar wind interacts with the magnetosphere to set the plasma inside in motion and to stretch the Earth's field into a long magnetic tail. Figure 5.1, a drawing of various regions of the magnetosphere alluded to above, does not convey how variable and dynamic this MHD system really is. Each planetary body in the solar system has a distinctive magneto- sphere, and we learn much by comparing their properties. A planet's size, rotation rate, magnetic field, satellites, and distance from the Sun influence the type of magnetosphere that it will have. Since the moon has no dynamo magnetic field, the solar wind interacts directly with its surface. The solar wind interacts with Venus' ionosphere. Mars may have a mixed magnetic and ionospheric interaction. Mercury has a magnetic field but no ionosphere. The Earth has a strong
250 PLASMAS AND FLUIDS i: C~ _ _ _ 3~ ~ ,. IONO~HE~ Bee CUSP in.. GEOM AGNETOSPHERE FIGURE 5.1 The Earth's magnetosphere. This imaginative drawing of the Earth's magnetosphere is a collective creation of the space research community, based on 25 years of measurements and theoretical modeling. It shows various features of the magnetosphere that will be discussed in this report. Standing ahead of the magnetosphere in the solar wind is a bow shock. The solar wind stretches the Earth's magnetic field into a long tail downstream. The northern and southern lobes of the tail are divided by a sheet of hot plasma. Impulsive plasma acceleration, probably due to reconnection, occurs in the tail and is probably related to violent disturbances in the pattern and strength of the auroras in the Earth's upper atmosphere and ionosphere. These disturbances are called substorms. The properties of the geomagnetically trapped energetic particles, the Earth's radiation belts, are determined by the balance of substorm acceleration and particle diffusion and transport. Such drawings cannot communicate the dynamism of this magnetohydrodynamic system.
. SPACE AND ASTROPHYSICAL PLASMAS 251 \ MERCURY \ / \ \ \ \ \ ~_ ~4 km PULSAR EARTH SATURN J UPITER - - ' ~ lOe kit ,,~ .~\ ~ .. ~__ ~ \ \ \ \ N G C 1265 1 1 -, V i, FIGURE 5.2 Planetary and astrophysical magnetospheres. This figure illustrates the enormous range of spatial scales to which the concepts of magnetospheric physics apply. Mercury's magnetosphere, the smallest in the solar system, has a size of a few thousand kilometers. It is about a factor of 10 smaller than the Earth's magnetosphere and pulsar magnetospheres. These, in turn, are about one hundredth the size of Jupiter's and Saturn's giant magnetospheres. All magnetospheres in the solar system are dwarfed by those of tailed radio galaxies, a trillion times larger than Jupiter' s. magnetosphere but it rotates slowly. Jupiter's rapidly rotating magnetic field couples with heavy-ion plasma from the satellite to to form a binary magnetosphere. Saturn is an aligned rotator, whose spin and magnetic dipole axes are parallel. Recent ultraviolet observations indicate that Uranus has auroras, a strong indicator of magnetospheric processes; its magnetosphere may be pole-on to the the solar wind, unlike all others known. Neptune's, if it exists, may be affected by interstellar neutral atoms. Finally, the solar wind interacts with neutral gases expanding from the nuclei of comets to form cometary magnetospheres. Figure 5.2 sketches the magnetospheres of Mercury, Earth,-Jupiter, and Saturn; they range in size from Mercury's (103 km) to Jupiter's (106 km), the largest MHD object in the solar system other than the solar wind itself. Figure 5.2 also shows two kinds of astrophysical /
252 PLASMAS AND FLUIDS magnetospheres. Pulsars are rapidly rotating highly magnetized neutron stars that generate and expel highly relativistic plasmas from their magnetospheres. There are striking similarities between pulsar magnetospheres and Jupiter's magnetosphere, both in their rotationally driven flows and in their pulsed, periodic radio emissions. Pulsar magnetospheres are comparable with the Earth's in size. At the opposite extreme in size are the so-called "tailed radio galaxies." It has been proposed that their magnetic fields might have been stretched into a long tail by the moving galaxy's interaction with intergalactic plasma, in roughly the same way that the Earth's magnetic tail is created. Our understanding of pulsars and tailed radio galaxies has certainly benefited from our awareness of analogous magnetospheric processes in the solar system. However, the parameters of space and astrophys- ical magnetospheres can differ so much that the day-to-day problems faced by researchers in these fields are quite different, and we observe them in very different ways. Nonetheless, the fact that both types of magnetosphere present similar questions about plasma dynamics gives us confidence that their physics is basically unified. MAGNETIC-FIELD RECONNECTION Suddenly the dark polar sky is pierced by a brilliant flash of light. Within minutes, a dazzling array of auroral forms stretches from horizon to horizon, million-ampere currents surge through the Earth's atmosphere and out into space, and 100 billion (10~) watts of power are dissipated in the Earth's atmosphere a magnetospheric substorm has begun (Figure 5.31. On the Sun, a burst of x rays near a dark sunspot signals the beginning of a catastrophic disruption of the solar corona a solar flare. Relativistic flare electrons heat the chromosphere to x-ray temperatures. A strong shock wave moves through the corona and begins a journey into interplanetary space that will carry it beyond all the planets of the solar system. The optical and x-ray luminosities start to build up in a distant quasar. Within a day, the quasar's luminosity will exceed the total power of a thousand galaxies. A sudden plasma loss occurs in a tokamak fusion device. These diverse phenomena seem unrelated. Nonetheless, they may share a common origin the release of stored magnetic energy by the mixed MHD and plasma process of reconnection. Violent reconnection can lead to spectacular events such as those above, but even in its more quiescent forms, reconnection can deter- mine the behavior of MOD systems. Consider the interaction between
SPACE AND ASTROPHYSICAL PLASMAS 253 FIGURE 5.3 The aurora from the ground. An observer in the Earth's polar regions can often look upward at the fiery auroral displays 100 km above him in the upper atmosphere. Their colors, complex patterns, and violent motions have fascinated observers for centuries. It has been given to this generation of space plasma physicists to understand how the aurora is made. Turbulent plasma processes some 5000 km above the Earth accelerate a beam of electrons downward. When they hit the upper atmo- sphere, these electrons cause the molecules and atoms there to radiate. The auroral acceleration processes may be activated by violent events in the Earth's magnetic tall, caused by reconnection. the magnetized solar wind and the Earth's magnetosphere. Reconnec- tion between solar wind and originally closed magnetospheric field lines opens some Earth field lines to interplanetary space. Energetic particles that ordinarily would not hit the Earth can be guided along open field lines into the Earth's polar atmosphere. Thus, reconnection changes the topology of the Earth's magnetic field. More importantly, reconnection enables the solar wind to do work on the magnetosphere, to set the plasma inside in motion. The basic energetics of the magnetosphere are in large part determined by the rate of reconnec- tion. Or consider the magnetic fields in the solar corona. It is thought that a balance is set by the creation of magnetic fields by turbulent convection below the solar surface and its destruction by reconnection
254 PLASMAS AND FL UlDS in microflares in the corona. Only when reconnection has been temporarily inhibited can the magnetic field increase enough to pro- duce a spectacular Hare by the reconnection that ultimately occurs. In sum, the effects of reconnection must always be considered in MHD models of space and astrophysical systems. Not only can reconnection influence the quiescent magnetic configurations of such systems, but it can cause sudden, dynamic reconfigurations of their magnetic fields. PARTICLE ACCELERATION AND COSMIC RAYS An astonishingly large fraction of the energy in space and astrophys- ical plasmas is in the form of energetic particles. For example, cosmic rays comprise about one third of the total energy density of the interstellar medium. The energetic particles themselves lead to important diagnostics of astrophysical systems. The observed cosmic-ray elemental and isoto- pic abundances are beginning to constrain models of nucleosynthesis and galactic evolution. We can infer the magnetic fields in many astrophysical systems by measuring the synchrotron radiation gener- ated by electrons accelerated to relativistic energies, a first step in constructing global MHD models of such systems. Analogs of cosmic acceleration processes are observed in solar- system plasmas. Energetic particles have been observed from probable reconnection events in the Earth's magnetic tail and from solar Hares. The ~ 10-keV electrons responsible for the terrestrial aurora are accelerated at 1000-5000 km altitudes above the Earth in regions that carry strong magnetic-field-aligned currents and, contrary to MHD assumptions, have parallel electric fields. Intense radio emissions are generated in the auroral acceleration region. Auroras and radio emis- sions have also been observed at Jupiter and Saturn, and auroras at Uranus, indicating that such acceleration occurs at these planets. Collisionless shock waves, either propagating in the solar wind or standing ahead of the planets, are observed to accelerate particles by processes similar to those now thought to generate cosmic rays. Shock waves of similar strength to those found in the interplanetary medium, propagating under similar conditions, and driven by supernova explo- sions are believed to be the primary source of galactic cosmic rays. Although it is not possible to detect the plasma processes responsible for particle acceleration in supernova shocks, theories of cosmic-ray acceleration can be tested by measurements of solar-system shocks together with measurements of galactic cosmic-ray energy spectra and
SPACE AND ASTROPHYSICAL PLASMAS 255 composition. On an even larger scale, collisionless shocks may be responsible for the acceleration of relativistic electrons in extragalactic radio sources. The Unifying Physical Problems Laboratory, space, and astrophysical plasma research is unified by a shared set of physical problems of true intellectual significance. The Committee on Space Plasma Physics (Space Science Board, 1978) identified six of these: 1. Magnetic-field reconnection, 2. The interaction of turbulence with magnetic fields, 3. The behavior of large-scale plasma flows and their interactions with magnetic and gravitational fields, The acceleration of energetic particles, 5. Particle confinement and transport, and. 6. Collisionless shocks. To the above six, we add two more: 7. Beam-plasma interactions and the generation of electromagnetic radiation and 8. Collective interactions between neutral gases and plasmas. Problem 3 and aspects of problem 2 are concerned with large-scale MUD systems, and the others relate to microscopic plasma processes. The fact that such problems emerge from a variety of contexts demonstrates their general significance and suggests that their solution will find further applicability in contexts that we cannot imagine today. The existence of such general problems provides a basis upon which a network of common interests, personal interactions, and ultimately a common discipline is being built. SPACE AND ASTROPHYSICAL PLASMA PHYSICS IN THE PAST 10 YEARS To communicate succinctly the flavor of space and astrophysical plasma research, we highlight here some of the progress made over the last decade on the eight problems defined above. Because Problem 3 is concerned with large-scale plasma systems, discussing it first places all the other problems in context.
256 PLASMAS AND FLUIDS Problem 3: The Behavior of Large-Scale Plasma Flows PLANETARY MAGNETOSPHERES Mariner 10 discovered a small highly active magnetosphere at Mercury that is energized by the intense solar wind near the Sun. The Pioneer-Venus Mission provided an impressive volume of information about the interaction of the solar wind with the Venusian ionosphere. Pioneer 10 made the first in situ measurements in Jupiter's magnetosphere. Pioneers 10 and 11 established the enormous scale and variability of Jupiter's magnetosphere, and Pioneer 11 made the first traversal of Saturn's magnetosphere. The Voyager 1 and 2 missions established that Jupiter's rotation powers a radial outflow of heavy-ion plasma injected by volcanic activity at the satellite lo. The Voyagers also found that Saturn's magnetic dipole and spin axes are aligned, a fact that challenges current theories of planetary-magnetic-field generation. Saturn's magneto- sphere proved to have an important interaction with the dense atmo- sphere of the satellite Titan. DYNAMICS OF THE EARTH S MAGNETOSPHERE Observations of intense plasma flows in the Earth's magnetic tail indicated that substorms are initiated by reconnection relatively near the Earth, which activates the aurora and expels hot plasma tailward. Large-scale convection cells, perhaps stimulated by a viscous inter- action with the solar wind, were discovered in the plasma sheet separating the two lobes of the Earth's magnetic tail. The field-aligned current systems in the magnetic tail proved to be consistent with those in the auroral acceleration region near the Earth. Strong MHD and plasma turbulence are associated with the currents in the tail. MAGNETOHYDRODYNAMIC STRUCTURES IN THE SUN S ATMOSPHERE AND IN THE SOLAR WIND Skylab measurements revealed the important difference between open and closed solar-magnetic-field structures. Open magnetic re- gions-solar coronal holes generate fast streams in the solar wind. Solar flares occur in closed magnetic structures solar coronal loops.
SPACE AND ASTROPHYSICAL PLASMAS 257 The Solar Maximum Mission ACRIM experiment showed that magnetic-field structures on the surface of the Sun are related to variations in the Sun's optical luminosity. The German/United States Hellos Mission extended measurements of the solar wind within the orbit of Mercury, and Pioneer 11 extended them past the orbits of Neptune and Pluto. The solar-wind magnetic field proved to reverse direction across a time-variable, warped neutral sheet, and strong, corotating shocks were found in the distant solar wind, thereby altering our picture of how galactic cosmic rays are modulated by the solar wind. MAGNETOSPHERES OF NEUTRON STARS Phase-resolved spectroscopy identified an x-ray line plausibly at the electron cyclotron frequency in two x-ray sources. If this identification continues to hold up, it will be the first proof that neutron stars can have super-strong magnetic fields, of the order of 10~2 gauss-a fundamental hypothesis of pulsar and galactic x-ray source theories. Detection of pulsed gamma-ray lines from the Crab and Vela pulsars proved that they generate super-high-energy particles and qualitatively supported the theoretical suggestion that electron-positron pair plas- mas are created in pulsar magnetospheres. MHD theories of radial transport in accretion disks suggested that processes similar to those in the solar corona may occur in accretion disk coronas. MAGNETOHYDRODYNAMIC JETS Active galactic nuclei frequently produce pairs of high-speed jets that propagate in opposite directions through the surrounding galaxy and out into the intergalactic medium, where they create the two- component radio emission characteristic of radio galaxies. Very-long- baseline interferometric measurements proved that the jets, which were theoretically anticipated, are accurately aligned on the light-year scale of the nuclei and on the 105-106 light-year scale of the double radio components (Figure 5.41. Some jets appear to expand faster than the speed of light, a kinematic effect that indicates that the flow speed can be relativistic. Jets were also associated with galactic objects, such as the energetic system SS-433, or stars in the early stages of formation.
258 PLASMAS AND FLUIDS in. FIGURE 5.4 A galactic jet. This figure shows three computer-constructed radio images of the radio galaxy NGC 6251, taken by different observational techniques and superposed to show structure on three different scales. Color denotes radio intensity, with white the most intense. The twin lobe pattern of radio emissions from extragalactic radio sources has been familiar since the 1950s. These lobes have a scale of millions of parsecs, much larger than the parent galaxy. (1 parsec = 3.26 light-years.) A bridge of radio emission appears to connect the galaxy to the upper right-hand lobe in the top panel. Higher-resolution measurements (middle panel) reveal that this bridge is a highly collimated jet that connects directly to a bright source at the center of the galaxy. (The stellar light of the galaxy does not appear in this radio image.) The very-long-baseline
:' SPACE AND ASTROPHYSICAL PLASMAS 259 GENERAL RELATIVISTIC ELECTRODYNAMICS It was suggested that the rotational energy of massive black holes, hypothesized to reside in the nuclei of active galaxies, may be extracted by electromagnetic torques. The plasma environments of these black holes would then be similar to pulsar magnetospheres. Problem 1: Reconnection Bursts of million-electron-volt particles associated with rapid plasma Hows in the Earth's magnetic tail indicated that inductive electromotive forces are important to magnetospheric dynamics and suggested that tail reconnection is unsteady and perhaps explosive. Rigorous analytical theory and numerical simulations established that a slow shock model proposed in 1964 is the correct description of reconnection in the MHD limit. The ISEE-3 spacecraft recently identified the slow shocks predicted by theory to exist in the geomagnetic tail. Theoretical understanding of the collisionless tearing limit of recon- nection was consolidated, and an explosive tearing instability was proposed analytically and simulated numerically. A laboratory experiment diagnosed with high precision the turbulent transport processes occurring in a strong guide field plasma regime appropriate, perhaps, to solar-flare conditions. Problem 2: Interaction of Turbulence with Magnetic Fields The Einstein observatory discovery that most stars exhibit solarlike surface activity proved that plasma processes are central to the physics of stellar chromospheres and coronas. The correlation of stellar interferometry (V~BI) technique enables the central source to be resolved on a scale of fractions of parsecs, less then the mean distance between stars. (Bottom panel.) This central source is still not resolved on this scale, but an extremely intense jet that is almost perfectly aligned with the features in the middle and top panels emerges from it. It is thought that mass accretion onto a massive black hole at the center of the galaxy powers an MHD jet, which eventually deposits its energy into the radio lobes in the top panel. The radio emissions are generated by electrons accelerated to highly relativistic energies, which emit synchrotron radiation in the magnetic field of the jet. (We thank the California Institute of Technology and the University of Cambridge for permission to reproduce this photograph.)
260 PLASMAS AND FF UlDS activity with stellar rotation observed for solar-type stars will constrain theories of dynamo generation of stellar magnetic fields and the dissipation of these fields as they emerge through the stellar surface. The solar-surface magnetic field proved to be concentrated in thin layers of strength of ~103 gauss separated by larger regions of strength of <10 gauss. There is essentially no evidence for a smooth uniform field at the solar surface. These remarkable observations challenge theories of turbulent magnetohydrodynamic convection. A coherent program of active and passive radar experiments, chemical releases, rocket measurements, analytic theory, and numer- ical simulations devoted to the equatorial ionosphere led to the most complete analysis of the nonlinear development of the Rayleigh-Taylor instability in plasma physics. Problem 4: Acceleration of Energetic Particles Measurements of the i0Be/9Be ratio in galactic cosmic rays showed that their age the time since they had been accelerated is about 10 million years, thereby defining the efficiency required of any cosmic- ray acceleration mechanism. Elemental and isotopic abundance measurements showed that cos- mic rays are accelerated not out of material freshly synthesized in supernovae but out of interstellar material whose composition, while broadly similar to solar, is significantly different, very possibly owing to chemical evolution of the interstellar medium since the birth of the solar system. The evidence that much of the volume of the interstellar medium is in a low-density plasma phase indicated that supernova shocks could propagate much further than originally thought. Together with the above cosmic-ray composition measurements, this revived the notion that supernova shocks Fermi-accelerate galactic cosmic rays out of the interstellar medium. The Fermi-acceleration particle energy spectrum was shown to be consistent with cosmic-ray observations. The discovery of an anomalous component of low-energy cosmic rays that has an unusual elemental composition, rich in oxygen, nitrogen, and helium and is modulated by the solar-cycle-like galactic cosmic rays, suggested the existence of another source of cosmic rays, very likely in the outer reaches of the heliosphere. Air-shower observations suggested that 10~4-10~6 eV cosmic rays are richer in heavier elements than those at lower energies and that cosmic rays above 10~7 eV energies may be anisotropic, unlike cosmic rays of galactic origin.
SPACE AND ASTROPHYSICAL PLASMAS 261 Measurements of field-aligned currents, electrostatic and electro- magnetic waves, and ion and electron distribution functions on auroral field lines were systematically assembled and combined with measure- ments of auroral light, ionization, motions, and structures, thereby set- ting the stage for comprehensive understanding of auroral acceleration. Impulsive-particle-acceleration events, probably associated with reconnection, were found to accompany rapid reconfigurations of the Earth's magnetic tail. Coordinated observations of energetic particles, x rays, and gamma rays from solar flares indicated that impulsive particle acceleration also occurs on the Sun. It was found that processes within Jupiter's magnetosphere acceler- ate nearly all of the ~10-MeV electrons found in the heliosphere. Problem 5: Particle Confinement and Transport Detailed models of the magnetic-mirror confinement, radial diffu- sion, and turbulent pitch-angle scattering of energetic ions and elec- trons were created and successfully tested by observations in the magnetospheres of Earth and Jupiter. Quantitative studies of the conduction of heat by electrons between the solar corona and chromosphere, which promise to make interpre- tation of chomospheric line emissions more secure, were initiated. Problem 6: Collisionless Shocks The strong dependence of the Earth's bow-shock structure on the parameters of the upstream solar wind was demonstrated by synopsis of individual detailed case studies, by the ISEE and other spacecraft. A clear understanding of the discontinuous change in shock structure at the so-called first critical Mach number was achieved by a combi- nation of spacecraft measurements, analytic theory, and numerical simulation. Quasi-parallel shocks, which propagate nearly parallel to the up- stream magnetic-field direction, proved to have extensive regions upstream that contain shock-accelerated particles and large-amplitude MHD turbulence, a property required by Fermi-acceleration theories. Detailed measurements of the energetic particle distributions and plasma turbulence associated with interplanetary shocks and planetary bow shocks began to be used to test self-consistent shock-acceleration theories. These results are providing a solid basis for theories of the acceleration of cosmic rays by supernova shocks.
262 PLASMAS AND FLUIDS Problem 7: Beam-Plasma Interactions, and the Generation of Radio Emissions Jovian auroral radio emissions were measured directly from space for the first time. They extend a factor 10 lower in frequency than is possible to measure from the ground, and their characteristic frequen- cy-time structure illuminates how they are modulated by the interac- tion between Jupiter's magnetosphere and its satellite Jo. Saturnian radio emissions were discovered and shown by the Voy- ager spacecraft to have a modulation entirely different from that of the Jupiter-Io system. Detailed studies of the intensity, spectrum, and polarization of Earth's auroral kilometric radiation motivated substantial development of theory. The auroral electron beam may linearly excite the kilometric radiation directly, in which case the aurora resembles a giant gyrotron in space. The magnetospheres of Earth, Jupiter, and Saturn confine a diffuse continuum of radio noise, whose lower cutoff frequency is the most sensitive indicator of the plasma number density. Detection of Jovian continuum radiation proved that Jupiter's magnetic tail extends to the orbit of Saturn and is the best vacuum so far encountered by man, with - a density of 10-: particles per cm3. Detailed studies of the fine structure of continuum radiation illumi nated how it is generated near the terrestrial plasmapause and magnetopause and by the interaction of Saturn's magnetosphere with its satellites. A nonlinear theory of Type III radio emissions was created and supported by observation in the solar wind. Detailed studies of pulsar radio emissions provided the most com- plete diagnosis to date of the microstructure of a highly relativistic plasma. Conclusive identification of orthogonal polarization states indicated that pulsar radio waves propagate in a magnetically birefring- ent medium similar to the ionosphere. The radio emissions of some pulsars were shown to consist entirely of discrete bursts, suggesting the action of nonlinear plasma processes. Problem S: Interactions Between Plasmas and Neutral Gases It was suggested in 1954 that a neutral gas would be rapidly ionized when its velocity through a plasma exceeds the so-called critical ionization velocity, given by equating an atom's kinetic energy to its ionization energy. Recent laboratory experiments, together with the
SPACE AND ASTROPHYSICAL PLASMAS 263 ory, showed that the rapid ionization results from collective electron heating by lower-hybrid plasma waves plus classical ionization. This effect is now thought important to rotating plasma devices such as centrifuges, the formation of minor bodies in the early solar system, comets, satellite-magnetosphere interactions, neutral-gas releases in space, and, possibly to the interaction of the Space Shuttle with the plasma through which it moves. The critical ionization phenomenon was incorporated in models of millimeter-wave radio emission from molecules behind the shock waves observed in magnetized molecular clouds, the sites of star formation. Chemical releases proved their worth in diagnosing the plasma flows and electric fields in the magnetosphere and in the auroral acceleration region. SPACE AND ASTROPHYSICAL PLASMA PHYSICS IN THE NEXT 10 YEARS We restrict ourselves to a general assessment of the state of our subject after the next 10 years. We assume that currently planned programs will be carried to completion because they lay the necessary foundations for the next generation of space missions. The initial exploration of solar-system plasmas will have been nearly completed. In 1986, an international consortium of spacecraft will fly by comets Halley and Giacobini-Zinner and provide the first in situ measurements of any comet. The International Solar Polar Mission will study the solar wind, and its effects on cosmic rays, in three dimen- sions for the first time. A Voyager spacecraft will make the first in situ measurements of the magnetospheres of Uranus and Neptune. The Galileo mission will diagnose Jovian magnetospheric plasma as com- pletely as any space plasma to date. A Pioneer or Voyager spacecraft might leave the heliosphere by the end of the decade and thus detect interstellar matter and galactic cosmic rays directly. In situ measure- ments of solar plasmas will be the primary unexplored problem in solar-system plasma research. The plasma environment of the Earth will be subjected to controlled study and, perhaps, to a measure of control, through the systematic use of active experiments and by synoptic measurements provided by the International Solar-Terrestrial Program. High-resolution optical measurements, by the Solar Optical Tele- scope, and radio observations, by the Very Large Array and by very-long-baseline interferometry, if funded, will have provided essen- tial information defining quantitative models of solar-surface magnetic
264 PLASMAS AND FL UlDS fields and dynamics, solar flares, and coronal heating, thereby creating the basis for general understanding of stellar activity. The growing ability to make series of detailed high-resolution observations in many wavelength bands will render many astrophysical objects increasingly subject to theoretical models that explicitly take plasma processes into account. Understanding of many space plasma processes will be sufficiently quantitative to make them reliable components of models of large-scale space and astrophysical systems. In addition to radiation-belt dynam- ics, the list of generally understood space plasma processes may include auroral acceleration, reconnection, collisionless shocks, and neutral-gas plasma interactions. The first generation of large-scale numerical models of space and astrophysical systems will have been completed. Foreseeable ad- vances in computing technology will lead to the creation of such models virtually simultaneously through space physics and astrophys- ics. The models will probably have made plasma physics central to the interpretation of many astronomical observations and have motivated new and different kinds of observations. One member of this class of models will have been tested by direct in situ measurements of the terrestrial magnetosphere's global dynamics, provided by the Interna- tional Solar-Terrestrial Physics Program. IMPACT OF RESEARCH ON SPACE AND ASTROPHYSICAL PLASMAS Research in this century has revealed a chain of interactions, almost all of which involve plasma physics, that connects activity at the surface of the Sun to the solar wind, and then on to the magnetosphere and atmosphere. The most spectacular manifestation of this solar- terrestrial interaction chain is the magnetic storm. The first evidence that a large solar flare might occur is the appearance of a complex sunspot group in the Sun's photosphere. Prompt electromagnetic radiation arrives at Earth a few minutes after the energy in the coronal magnetic fields associated with the sunspot group is suddenly released. Energetic solar-flare protons are guided by the solar wind and magnetospheric magnetic field into the polar atmosphere soon after- ward. The enhanced ionospheric plasma produced by the energetic protons attenuates the radio noise received from cosmic radio sources. A day or so later, a shock wave passes over Earth, enveloping it in dense, hot solar-flare plasma that compresses the magnetosphere. Substorms increase in frequency and strength and inject hot plasma
SPACE AND ASTROPHYSICAL PLASMAS 265 into the Earth's inner magnetosphere to form a ring current, which creates the geomagnetic field depression and activity that first moti- vated the name magnetic storm. The auroras intensify and move to unusually low latitudes, creating a dense highly disturbed ionospheric plasma that interferes with radio communication and on occasion blacks it out altogether. Intense wind systems, sometimes of a world- wide scale, are generated in the upper atmosphere. In its report Solar-Terrestrial Research for the 1980's, the Commit- tee on Solar-Terrestrial Research (1981) has identified four areas where research on the solar-terrestrial interaction chain can clarify important impacts on society and technology: 1. Predictions about the space environment; 2. Effects on stratospheric ozone, which shields the life at the Earth's surface from the harmful effects of solar ultraviolet radiation; 3. Effects on ionospheric physics and radio propagation; and 4. Elucidation of potential connection between solar variability, weather, and climate. Although items 2 and 4 have significant long-term implications, item 1 is at present the most important. Many practical systems, both civilian and defense, and all of our manned space endeavors, must operate in the highly variable and potentially hostile plasma environ- ment of the Earth and solar system. Plasma processes in this environ- ment also influence and even disrupt important ground-based systems over local and regional scales. Ground-based high-frequency commu- nication systems in the Earth's polar regions can be blacked out by magnetic storms. Entire spacecraft have been electronically disabled by violent electrical discharges that occur when hot ring-current plasma envelops the spacecraft. The risk of such disasters can be reduced only by continuing attention to the effects of the plasma environment on spacecraft systems. It is clear that to work in the space environment, we must understand it. Research on the solar-terrestrial interaction chain is the only way to learn how to predict the space environment. Because of the need to understand and predict its impact on military systems, the Air Force and Navy maintain significant research and monitoring programs in solar and solar-terrestrial physics. One of the most important by-products of today's research on solar-system plasmas will be a generation of students and practitioners familiar with the space environment whose work will enrich tomorrow's space science and technology. The Astronomy Survey Committee (1982) has characterized the motivations for astronomical research as follows:
266 PLASMAS AND FLUIDS Astronomy . . . is sustained by two of the most fundamental traits of human nature: the need to explore and the need to understand. Through the interplay of discovery, the aim of exploration, and analysis, the key to understanding, answers to questions about the Universe have been sought since the earliest times, for astronomy is the oldest of the sciences. Yet it has never been, since its beginnings, more vigorous or exciting than it is today. Our own branch of astronomy, astrophysical plasma physics, is driven by the need to understand the unusual plasmas surrounding some of the most exotic objects brought to light by recent astronomical research. Our imaginations are challenged, and we are forced to extend plasma physics to comprehend them. The enrichment of plasma physics, the most important by-product of astrophysical plasma re- search, strengthens the technological potential of the entire field. THE ROLE OF SPACE AND GROUND-BASED MEASUREMENTS AND OBSERVATIONS The basis of space and astrophysical plasma physics is large: federally funded programs that support measurements and observa- tions made from space and the ground. These programs have been studied extensively in previous reports sponsored by the National Research Council. Solar-System Plasma Physics Solar-System Space Physics in the 1980's, by the Committee on Solar and Space Physics (CSSP) of the Space Science Board (1980), gives a comprehensive strategy for experimental research in solar- system physics, nearly all of whose recommendations bear directly on solar-system plasma physics. Among other things, it proposes a four-spacecraft investigation of the transport of energy and plasma in the Earth's magnetosphere (the OPEN mission), high-resolution ob- servations of solar surface phenomena (the Solar Optical Telescope), a solar probe, and active plasma experiments performed from the Space Shuttle. The internationalization of the OPEN mission, now called the International Solar-Terrestrial Physics Program, adds an important possibility that more spacecraft will join the effort to understand the magnetosphere's global dynamics. Smaller-scale Explorer-class space missions were addressed in a subsequent CSSP report, A Strategy for the Explorer Program in Solar and Space Physics (Space Science Board, 19841. There it is recom- mended that "The size, complexity, and management of future Ex
SPA CE AND ASTROPHYSICAL PLASMAS 267 plorer missions should return to the original perceived philosophy of the Explorer program, i.e., relatively small, simple satellites." Fur- thermore, the CSSP found that the scientific opportunities in solar and space physics research merit an average of approximately one Ex- plorer satellite opportunity a year. We recommend implementation of these two CSSP strategies be- cause the programs that they recommend are the primary ones that will explicitly contribute to our knowledge of the physical processes in large-scale plasmas. The Geophysics Research Board in Solar-Terrestrial Research for the 1980's (Committee on Solar-Terrestrial Research, 1981) and in National Solar-Terrestrial Research Program (Committee on Solar- Terrestrial Research, 1984) recommends a program of ground-based measurements, designed to supplement the CSSP strategies, that will provide important information about the chain of physical interactions that links solar activity to changes in the Earth's magnetosphere, ionosphere, and atmosphere. We support the objectives of these reports. Finally, planetary research makes an important contribution to solar-system plasma physics. Because the planets and comets provide different tests of magnetospheric physics, it is essential that experimental and theoret- ical plasma investigations continue to be an integral part of each planetary and cometary mission. Astrophysical Plasma Physics Astronomy and Astrophysics in the 1980's, by the Astronomy Survey Committee (1982), proposed four major and seven moderate new programs designed to benefit all of astronomy and astrophysics. We have examined each of the proposed programs and find that the majority will contribute to the observational basis of astrophysical plasma physics. We now discuss four examples from the diversity of measurements that can contribute to research on astrophysical plasmas. The Ad- vanced Solar Observatory, whose central element will be the Solar Optical Telescope, will make high-resolution measurements of plasma structures on the surface of the Sun. The Advanced X-Ray Astrophys- ics Facility (AXAF) will diagnose the high-temperature plasmas sur- rounding stars, black holes, neutron stars, and active galactic nuclei, in supernova remnants, and in numerous other objects. The Very-Long- Baseline Array of radio telescopes will make exceedingly high- resolution measurements of the spatial structure of objects, such as
268 PLASMAS AND FF UlDS galactic nuclei, in which electrons are accelerated to high energies. The moderate-scale programs to measure cosmic rays proposed by the Astronomy Survey Committee will provide direct evidence about how and where cosmic rays are accelerated, about the propagation of cosmic rays in the Galaxy, and about galactic evolution. We endorse the programs proposed in Astronomy and Astrophysics for the 1980's because, in advancing the entirefield of astronomy, they will make significant contributions to many problems in plasma astrophysics. Given implementation of these recommendations, the- single most important step toward strengthening the role of plasma physics in the interpretation of astronomical observations will be to encourage theoretical and numerical models of large-scale plasma systems. In Situ Measurements near the Sun Several years ago, the National Aeronautics and Space Administra- tion (NASA) formed a project team, which, together with an external Science Working Group, studied a mission to the Sun called Starprobe. Starprobe would be an instrumented spacecraft designed to operate in the solar corona as close as 4 solar radii from the Sun's center. Starprobe would provide, for the first time, local sampling of the plasma, magnetic field, energetic particles, and turbulence in a stellar atmosphere. It would delineate the properties of the transition region where the subsonic coronal flow is converted into the supersonic and super-Alfvenic solar wind. The imaging available from close range would provide high-resolution information about solar surface fea- tures. It is possible to instrument the spacecraft to yield crucial information of the mass distribution in the solar interior, of importance to theories of stellar structure and gravitation. The final report of the Starprobe Science Working Group was never presented for further evaluation to the Space Science Board or to NASA's Space and Earth Science Advisory Committee. From the joint perspective of solar system and astrophysical plasma physics, in situ measurements of a stellar corona could be as significant as the discoveries of the solar wind or the Earth's magnetosphere. For this reason: We recommend that NASA ask its advisory committees to evaluate a mission to the Sun.
SPACE AND ASTROPHYSICAL PLASMAS 269 Concluding Remarks The brevity of our discussions of space and ground-based measure- ments and observations reflects only the extent to which these issues have been carefully discussed elsewhere. The large and complete studies that we referred to above are but one indication of the conviction shared by all researchers that measurements and observa- tions are a sine qua non of science. To maintain U.S. leadership in space and astrophysical research, a vigorous program of space and ground-based measurements is absolutely essential. THE ROLES OF LABORATORY AND ACTIVE SPACE EXPERIMENTS Laboratory Experiments Many experiments help to develop basic theory, and so contribute indirectly to space and astrophysical research. A few have been designed specifically to elucidate processes occurring in space and astrophysical plasmas. Such experiments usually cannot reproduce the large ratios of system scale to basic microscopic scale lengths charac- teristic of natural plasmas, but if the relative ordering of the various dimensionless plasma parameters is similar to that in the system of interest, the laboratory results can be extrapolated satisfactorily. Laboratory experiments and numerical simulations have similar advantages and limitations. The continuing refinement of detector technology and the great increase in precision and volume of data analysis made possible by the use of dedicated minicomputers now enable laboratory experiments to be diagnosed as thoroughly as a numerical simulation. Although an advanced computer can simulate many problems, and an experiment only one, experiments can be the method of choice for certain plasma problems and complement simu- lations for others. In these cases, experiments can give information as detailed as a simulation about real plasmas, rather than about mathe- matically idealized ones. Laboratory experiments have been particularly important in con- firming the existence of processes thought to occur in space plasmas. A terrella experiment, in which the flow of plasma past a magnetic dipole simulated the interaction of the solar wind with the Earth's magnetosphere, demonstrated the existence of the long postulated reconnection regions at the nose and tail of the magnetosphere several years before three-dimensional numerical simulations could do so.
270 PLASMAS AND FLUIDS Reconnection experiments in simple space-physics geometries, as well as in tokamaks, revealed the dynamic-magnetic-field reconfiguration and plasma acceleration generally expected by theory. Laboratory experiments first suggested, and later confirmed, the existence of electrical double layers and anomalous resistivity, which are involved in auroral particle acceleration. Laboratory experiments demonstrated how plasmas moving through a neutral background at the so-called critical ionization velocity suddenly bring the neutrals to full ionization and produce substantial momentum coupling. Past experience indicates that laboratory experiments will continue to contribute to space and astrophysical plasma physics. Examples of areas where experiments are needed include the following. Efforts to clarify the microstructure and the collective wave-particle interactions occurring in the reconnection process are under way for a limited set of laboratory parameters. Studies of reconnection in a range of different plasma conditions are needed. It has been 10 years since the last laboratory collisionless shock experiment. Studies of high-Mach- number shocks would provide information, important to astrophysics, about shocks stronger than those typically found in the solar system. The processes limiting strong heat conduction merit continued labora- tory study. Strong heat conduction takes place between the solar corona and chromosphere, the Earth's magnetosphere and ionosphere, the hot interstellar medium and cool interstellar clouds, and the surfaces and interiors of laser-fusion plasmas. Experiments on the interactions between relativistic electron beams and plasmas illuminate the physics of relativistic astrophysical objects such as pulsars. And, of course, the most important experiment of the next 10 years might well be one that finds an unforeseen new effect. Active Space Experiments Here we discuss some experiments performed on space plasmas. Very-high-frequency radar backscatter has been used for the past 20 years to diagnose both the large-scale properties and the small-scale plasma turbulence in the equatorial and auroral ionosphere. The nonlinear response of the ionosphere to intense high-frequency irradi- ation from ground-based radars has been studied at mid-latitudes for the past 10 years and more recently at auroral latitudes. Entirely analogous experiments have been performed on laboratory plasmas. Chemical releases in the ionosphere have clarified the interaction between flowing plasmas, magnetic and electric fields, and neutral gases. They also confirmed the existence of parallel electric fields in the
SPACE AND ASTROPHYSICAL PLASMAS 271 auroral acceleration region. The United States/German AMPTE space- craft mission is using chemical releases to provide similar information about conditions deep in the Earth's magnetosphere. Electron beams injected from rockets into the ionosphere have been used to study the radiation from space plasmas as well as to diagnose the properties of the geomagnetic field. The Space Shuttle is beginning to make possible large experiments, including chemical, beam, and wave injections, that will greatly increase the scope of active space experimentation. In sum, the past decade has seen the rapid growth of a new space-science discipline plasma physics in space in which experi- ence gained in the laboratory and the talents of laboratory experimen- talists have been used to advantage. The implications of this development are obvious. We have begun the long effort that may eventually result in a measure of useful control over our space environment. For this reason, we expect the already rapid growth to accelerate. THE ROLE OF THEORY The creative interplay between theory and experiment is the scien- tific method, and the quantitative prediction and understanding of experiment and observation is the central function of theory. Theory and theoreticians play another important role in our subjects. They express the unity in the diversity of plasma phenomena in the laboratory, in the solar system, and in the universe at large. They communicate conclusions, ideas, analogies, and techniques across disciplinary boundaries. Theoreticians have led the effort to unify space and astrophysical plasma physics. Space Plasma Theory The Study Committee on Space Plasma Physics (Space Science Board, 1978) spoke firmly about the role of theory in solar-system plasma research: The theoretical component of the space-plasma-physics effort needs to be strengthened by increased support and, most particularly, by encouraging theory to play a central role in the planned development of the field. The Committee on Solar and Space Physics of the Space Science Board (1980) echoed the above recommendation in forceful terms:
272 PLASMAS AND FLUIDS Theory has to play an increasingly central role in the planned development of solar-system space physics. Moreover, theory and quantitative modeling should guide its entire information chain data acquisition, reduction, dissem- ination, correlation, storage, and retrieval to a higher level of sophistication, to provide prompt availability of coordinated data of diverse origins. NASA's Solar-Terrestrial Theory Program, initiated in view of the above recommendations, has been one reason why solar-system plasma research has reached a new level of precision, whereby it now makes strong contributions to both general plasma physics and to the interpretation of space data. We heartily recommend continuance of the excellent support that space plasma theory has received in the past 5 years and, especially, of the Solar-Terrestrial Theory Program. Theoretical Astrophysics The Astronomy Survey Committee (1982) recommended as a pre- requisite for new research initiatives: [augmentation of theory and data analysis, to facilitate the rapid analysis and understanding of observational data; .... The Astronomy Survey Committee recommended a program like the Solar-Terrestrial Theory Program for theoretical astrophysics. The Theory Study Panel of the Space Science Board (1983) made a similar recommendation: We recommend that NASA establish independent theoretical re- search programs in planetary sciences and astrophysics, with objec- tives similar to those of the solar-system plasma-physics theory program. Experience suggests that such a theory program could be highly successful and, in particular, that it might transform plasma astrophys- ~cs. Theory and numerical modeling must both be strengthened in order that plasma physics play the central role in the interpretation of astronomical observations warranted by the fact that most of the universe is in the plasma state.
SPACE AND ASTROPHYSICAL PLASMAS 273 THE ROLE OF NUMERICAL MODELS AND SIMULATIONS Why Quantitative Models Are Essential We must picture the entire magnetosphere of the Earth before we can deduce where and how its plasma processes operate, yet we must understand what the processes do before we can determine the structure and dynamics of the magnetosphere. This essential difficulty is repeated through space and astrophysical plasma physics: plasma processes both determine, and are determined by, their parent system's global MHD configuration. Twenty years ago, imaginative drawings cartoons guided space plasma research. Despite their naivete, they were important. Even then, fairly detailed information about the local behavior of plasmas in the magnetosphere was being acquired. The significance of this infor- mation was evaluated with the help of drawings, which provided a conceptual link between local measurements taken at different points in space and time. As our picture of the magnetosphere grew more complete, so also did our grasp of the plasma processes regulating its behavior. Until this finally occurred, many scientific controversies would have been settled had it been possible to photograph the magnetosphere. We can photograph astrophysical systems. However, our photo- graphs detect photons that are usually generated by mechanisms indirectly related to the MHD and plasma processes that regulate the structure and energetics of the systems under study. We have no in situ measurements, as we do in the solar system, to tell us even the most basic plasma parameters. These must be inferred using our knowledge about how the light we observe was generated. Our photographs provide only a two-dimensional picture at one instant of time of three-dimensional, evolving systems. Thus, we study classes of related objects of different ages to deduce how they evolve in time, and we use drawings to elucidate the relationships between their structure, dynam- ics, evolution, and the radiation that we measure. To achieve quantitative agreement between theory and observation, it is essential to progress beyond the cartoon approximation to quan- titative models. It is less obvious, but no less true, that the process of model building is also a process of discovery. By constructing a series of models we are led to appreciate the relationships between the parts and the whole of the time-variable, three-dimensional systems that we observe and to perceive how microscopic processes regulate their
274 PLASMAS AND FLUIDS structure and behavior. Models also suggest new measurements that then clarify the physics underlying the models. In the past decade, our studies of solar-system plasmas have achieved a measure of quantitative understanding though the system- atic use of analytic and, more recently, numerical models. The first generation of numerical models of astrophysical plasma systems is being created at this time. Because we cannot detect the underlying astrophysical plasma processes directly, we believe that the best strategy will be to create numerical models at the system level that postulate plasma and radiation processes and iterate between the system and process levels until quantitative agreement with observa- tion is achieved. It is our perception that the present level of develop- ment of numerical technology, theory, and observations gives such a strategy a significant chance of success for the first time. The increasing urgency of the need for advanced numerical simula- tions and models may be perceived from the phrasing of successive recommendations of National Research Council and NASA panels. The Report on Space Science 1975 by the Space Science Board (1976) simply noted for all the space sciences that Results from . . . theoretical modeling have been of critical importance in planning and supporting space missions. without commenting on the needed computational facilities. The Advocacy Panels in their unified recommendations to the Study on Space Plasma Physics (Space Science Board, 1978) recommended the following: Strengthening theoretical solar-system plasma physics and, to aid in achieving this goal, support for computer modeling.... The International Magnetospheric Study Working Conference on Magnetospheric Theory made the following explicit recommendation for this field (Committee on Solar-Terrestrial Research, 1979~: Future theoretical progress must involve the use of plasma simulation and large-scale numerical modeling of magnetospheric dynamics in parallel with the development of pure theory. The Committee on Solar and Space Physics of the Space Science Board, in Solar System Space Physics in the 1980's (NAS, 1980), made a much more general recommendation: . . . theory and quantitative modeling should guide [the] entire information chain tof solar-system plasma physics] to a higher level of sophistication....
SPACE AND ASTROPHYSICAL PLASMAS 275 Recent recommendations for advanced computations for the broad field of astrophysics have been directed toward the computing facilities that will be needed. In A Strategy for Space Astronomy and Astro- physics for the 1980's, the Committee on Space Astronomy and Astrophysics of the Space Science Board (1979), among other recom- mendations for theory, advocated that: NASA should make available time on its largest computers for theoretical problems of great complexity, which are often beyond the capacity of university-scale computers. By 1982, the Astronomy Survey Committee felt it necessary to recommend, as a prerequisite for new research initiatives in astron- omy: Computational facilities, to promote data reduction, image processing, and theoretical calculations. The Initial Report of the NASA/University Relations Study Group (NASA, 1983) recommended that NASA should provide to researchers in fields sponsored by NASA, including space plasma physics and astrophysics: Major facilities [such as] . . . large, fast computer facilities of the Cray class, which would be used by several investigators and jointly by investigators at several institutions. The above recommendations reflect the increasingly widespread perception that theory, just as experiment, depends crucially on technology. System Models and Process Simulations in the Next Decade Realistic MHD system models will include the effects of collective plasma processes as regulating subelements; these processes can be individually simulated in idealized form. System models are conceptu- alized and executed at the fluid level, process simulations at the microscopic, kinetic level. Here we present examples of global models and local simulations that will be needed in the next 10 years. SYSTEM MODELS An entire space project, the International Solar-Terrestrial Physics Program (ISTP), has been designed and recommended with the idea
276 PLASMAS AND FL UlDS that an MHD magnetospheric model with realistic microscopic ele- ments will also be created and tested by the data. The multiple spacecraft and ground facilities associated with the ISTP will measure key parameters pertinent to the MHD model, together with the local plasma processes that regulate the dynamics of the magnetosphere. The ISTP project will provide the first systematic experimental test of a comprehensive magnetohydrodynamic model of a large-scale flowing system. Testing the model will force the development of innovative methods of data analysis and dissemination. The ISTP magnetospheric model will be the first MHD system model in space physics and astrophysics to include all known, pertinent plasma process elements. The ISTP magnetospheric model will be a prototype of what must be done if hydrodynamic and MHD models are to play their potentially powerful role in the interpretation of remote astronomical observa- tions. Solar-terrestrial models that successfully meet the test of de- tailed measurements at both large- and small-scale processes would substantially increase our confidence in models of more distant astro- physical systems. The first large-scale astrophysical plasma models are being created for solar physics, as hydrodynamic models of the turbulent convection zones of the Sun and similar stars have been extended to include MHD. These models will allow us to test our understanding of fluid-magnetic field interactions as the data from the next generation of high-resolution solar instruments become available. (The first and most important of these is the Solar Optical Telescope.) System models will be crucial to the interpretation of these anticipated high-resolution data because of the intimate connection between the processes that determine mor- phology and those that produce the photons that we observe. System models analogous to those just described are being devel- oped for accretion disks near neutron stars and black holes and are being used to interpret data from galactic x-ray sources and active galactic nuclei. Similarly, the first generation of models of bipolar jets is currently being constructed. PROCESS SIMUEATIONS Twenty years of experience in fusion plasma physics, and 10 years in space plasma physics, indicate that numerical simulations are one of the best ways to gain insight about nonlinear plasma processes. For example, simulations have illuminated how H + and O + ions are accelerated by auroral plasma turbulence and then ejected into space.
SPACE AND ASTROPHYSICAL PLASMAS 277 The output of such microscopic plasma calculations must be fed back into system models. In the above example, the H+ and O+ ions add mass to, and decelerate, the MHD flow in the magnetosphere. Advanced numerical simulations will be very important to reconnec- tion, where three-dimensional kinetic simulations in a time-dependent magnetically complex configuration are required to resolve our out- standing theoretical questions. Such problems as wave-particle inter- actions, generation of radiation in plasmas, collisionless shock struc- ture, strong heat conduction, and many more will continue to benefit from advanced numerical simulations. Another application of detailed modeling is to plasmas with super- high-energy densities. For example, the ultra-high plasma tempera- tures believed to prevail at the centers of quasars and active galaxies are far more extreme than those encountered either in the laboratory or the solar system and probably lead to some fascinating phenomena associated with the creation of electrons and positrons from heat energy. The electron-positron recombination line has been detected from the nucleus of our Galaxy, suggesting the existence of relativistic plasma processes there. The superstrong magnetic fields in neutron stars can lead to an unusual pair-production process that is thought to populate pulsar magnetospheres with positronic plasma. Detailed models of these processes can be checked by x- and gamma-ray observations of pulsars. From the meager theoretical work to date, it is already clear that an understanding of such plasmas, and the complex role played by electron-positron pairs, will provide novel theoretical constraints on the sizes, luminosities, and temperatures of some of the most energetic astrophysical objects. OVERALL CONCLUSIONS Our review of the models and simulations needed in the next 10 years has led us to the following conclusions: 1. Many problems in space and astrophysical plasma physics have evolved to the point where numerical system modeling is a next logical step. These problems include planetary magnetospheric structure, solar convection and coronal structure, three-dimensional structure of the solar wind, supernova remnants, astrophysical jets, pulsar magnetospheres, and accretion onto neutron stars and black holes. 2. Many microscopic plasma problems that arise in the study of space and astrophysical systems would benefit by coordinated simula- tion efforts.
278 PLASMAS AND Ft UlDS Proposal for a Dedicated, Advanced Computational Program Plasma physics was a pioneer in the successful utilization of large- scale computations, for fluid, MHD, hybrid, and kinetic models. Most of the progress since the late 1960s has been in fusion research and nuclear weapons phenomenology. The computational facility dedi- cated to magnetic fusion energy (MFE) has critically advanced under- standing of magnetic-confinement systems and of fusion and basic plasma processes. The establishment, in 1979, of NASA's theory program in solar-terrestrial plasma physics made numerical models and simulations regular tools in solar-system plasma research and has prepared that research community for the next, more advanced stage. The continued development of numerical technology will advance many branches of science. In our own fields, we foresee that many problems of a scale that requires today's national computing facilities will soon be addressable by local university and laboratory facilities. This will only increase the importance of numerical modeling to our subjects. Nonetheless, we believe that the leading research on many of the models discussed above will continue to be done on the most advanced computing facilities existing at any given time, because these models involve a complex interplay between large- and small-scale processes. Thus far, the responsibility for the maintenance and advancement of state-of-the-art computing facilities has been a national one, because it is beyond the capability of single institutions and because a national scope provides an adequate pool of users. America's existing advanced computational facilities, devoted to defense, fusion research, and meteorology, have been used on a piecemeal basis for space and astrophysics problems. These busy facilities do not have space physics and astrophysics as an institutional objective, and researchers in these fields must make individual agreements to secure access to advanced computing. In some cases, American researchers have had to journey to Europe or Japan in order to perform large-scale computations. In view of the arguments above, and in view of the many space and astrophysical systems ready for systematic modeling: We recommend a national computational program dedicated to basic plasma physics, space physics, and astrophysics, which will provide and maintain state-of-the-art technology appropriate to large- scale theoretical models and simulations. Such a program should ensure ready access to advanced computing on the basis of peer review.
SPACE AND ASTROPHYSICAL PLASMAS 279 We estimate that present U.S. expenditures on space plasma com- puting are about $2 million to $3 million per year. About $1 million per year is provided by the National Science Foundation (NSF), and a large fraction of NASA's approximately $2.2 million per year solar- terrestrial theory budget is devoted to numerical modeling. A some- what smaller sum is spent on astrophysical modeling. The effort represented by these expenditures provides a reasonable basis on which the more ambitious program that we are proposing could be constructed. Our role has been to point out that a large number of problems central to space and astrophysical plasma research are ready for advanced numerical modeling. If these problems are combined with others in hydrodynamics and general astrophysics that should be included in the program, in a few years the scientific demand, and more importantly, the scientific payoff, will justify the dedicated effort that we propose. We further recommend that a study be initiated forthwith that would address such issues as the following: 1. The scope and evolution of a national computational program for basic plasma physics, space physics, and astrophysics; 2. The institutional arrangements needed to provide strong scientific guidance to such a program and to ensure ready access to advanced computing on the basis of peer review; 3. The appropriate balance between large-scale and mid-scale com- putations and between national and local facilities; 4. The ability of existing national facilities to meet the needs of basic plasma physics, space physics, and astrophysics in the near future. Because problems in magnetic fusion are similar to those in space physics and astrophysics, the experience with the MFE-dedicated facility may prove valuable in considering the questions above. Be- cause the National Center for Atmospheric Research deals with large-scale hydrodynamic calculations, its experience may be equally useful. 4 THE ROLE OF PLASMA PHYSICS IN THE UNIVERSITY CURRICULUM Space Plasma Physics Studies of the space sciences are at present concentrated in a few major institutions and a somewhat larger number of smaller schools. For example, approximately 75 percent of graduate students in solar
280 PLASMAS AND FF UlDS terrestrial physics and the related field of aeronomy are enrolled in 24 United States institutions, which graduate 50 to 60 Ph.D.s each year. Within solar-terrestrial research, most students are supported in the subdisciplines of magnetospheric and solar physics. About 25 percent of the students involved in solar-terrestrial research appear to be doing theoretical plasma physics, and a larger fraction uses plasma concepts in the interpretation of experimental data. In aeronomy, which deals with the upper atmospheres and ionospheres of the Earth and other planets, about 15 percent of the graduate students are pursuing plasma-related topics. We believe that these sample figures illustrate the recent emergence of plasma physics as an important conceptual tool in the older fields of solar-terrestrial research and agronomy.* The place of solar-system plasma physics in the teaching curriculum differs from institution to institution. Courses are taught and graduate degrees are granted in Departments of Physics, Astronomy, Physics and Astronomy, Electrical Engineering, Space Sciences, Earth Sci- ences, and Atmospheric Sciences, among others. The course content and sequence differ from department to department, resecting the diverse historical origins and motivations for space plasma research. In those few institutions with major programs in both fusion and space plasma research, the two specialties are not always well integrated into a single course curriculum. We view the current fragmentation of the space plasma curriculum with concern but not with alarm. It appears to be a natural stage in the evolution of our new discipline. However, a more unified plasma- physics curriculum that takes into account the achievements of, and applications to, solar-system and astrophysical plasma physics is an important objective for the immediate future. Astrophysical Plasma Physics Because astrophysical plasma physics has not yet become a well- organized subdiscipline of astrophysics, it is difficult to pinpoint the number of graduate students working in astrophysics who are pursuing plasma research. However, an increasing number of research topics in astrophysics involves the use of plasma concepts wholly or in part. At many universities with graduate programs in astrophysics, including *The figures in this paragraph were compiled by D. S. Peacock, Program Director for Solar-Terrestrial Research at NSF, for a joint European-U.S. Workshop on Space Plasma Physics, held at Hilton Head, South Carolina, September 20-23, 1983.
SPACE AND ASTROPHYSICAL PLASMAS 281 several with distinguished programs, the teaching of plasma physics is inadequate to prepare students for research in plasma astrophysics. Because the role of plasma physics in astrophysics is destined to grow, this relative lack of university involvement limits both man- power and progress. To the extent that space and astrophysical plasma physics, at the graduate level, are not viewed as integral parts of plasma physics and astrophysics, the intellectual vitality of space science and astrophysics is bound to surer. We recommend that graduate teaching programs in space science and astrophysics include plasma physics as part of their basic course requirements. Plasma Physics in General The following remarks are meant to apply to all of plasma physics, and the recommendation is directed to colleges and universities whether or not they currently teach plasma physics. Well-developed scientific disciplines are characterized by deep philosophical motivations, a unified body of powerful theoretical and experimental techniques, and a wide range of applications. It is our conviction that because of the growing integration of space and astrophysics plasma physics with one another, and with laboratory and fusion research, plasma physics is maturing. When a scientific disci- pline matures, technological innovation soon follows. Plasma physics is only beginning to have its impact. It is only one generation since plasma physics became a highly developed discipline. During this time, a handful of universities, primarily those with federally funded projects in fusion or space physics, developed graduate programs in plasma physics. Because graduate training in plasma physics is excellent preparation for a variety of careers in science and technology, and in universities, government laboratories, and industry, it is now important to introduce undergraduate students to plasma physics, so that they may make an informed choice of graduate specialty. At present, it is primarily those universities with graduate programs in plasma physics that teach the subject at the undergraduate level. In view of the increasing precision of its experimental and theoret- ical techniques, and in view of its many applications to space physics, astrophysics, and technology, we recommend that plasma physics now become a regular part of the university science curriculum. A one-year junior- or senior-level elective course in plasma physics would be an excellent response to our recommendation.
282 PLASMAS AND FLUIDS REFERENCES Astronomy Survey Committee, National Research Council, G. B. Field, chairman, Astronomy and Astrophysics for the 1980's. Vol. 1, Report of the Astronomy Survey Committee, National Academy Press, Washington D.C.,1982. Committee on Solar-Terrestrial Research, Geophysics Research Board, The Interna- tional Magnetospheric Study: Report of a Working Conference on Magnetospheric Theory, F. V. Coroniti, conference chairman, National Academy of Sciences, Wash- ington, D.C., 1979. Committee on Solar-Terrestrial Research, Geophysics Research Board, National Re- search Council, Solar-Terrestrial Research for the 1980's, H. Friedman and D. S. Intriligator, study co-chairmen, National Academy Press, Washington, D.C., 1981. Committee on Solar-Terrestrial Research, Board on Atmospheric Sciences and Climate, National Research Council, D. S. Intriligator, chairman, National Solar-Terrestrial Research Program, National Academy Press, Washington, D.C., 1984. National Aeronautics and Space Administration, The Universities and NASA Space Sciences, Initial Report of the NASA/University Relations Study Group, T. Donahue and F. B. McDonald, co-chairmen, NASA, Washington, D.C., 1983. Space Science Board, National Research Council, Space Plasma Physics: The Study of Solar-System Plasmas, Vol. 1, Reports of the Study Committee and the Advocacy Panels, S. A. Colgate, chairman, National Academy of Sciences, Washington, D.C., 1978. Space Science Board, National Research Council, Committee on Space Astronomy and Astrophysics, P. Meyer and H. J. Smith, chairmen, A Strategy for Space Astronomy and Astrophysics for the 1980's, National Academy of Sciences, Washington, D.C., 1979. Space Science Board, National Research Council, Committee on Solar and Space Physics, C. F. Kennel, chairman, Solar-System Space Physics in the 1980's: A Research Strategy, National Academy of Sciences, Washington, D.C., 1980. Space Science Board, National Research Council, Theory Study Panel, A. G. W. Cameron, chairman, The Role of Theory in Space Science, National Academy Press, Washington, D.C., 1983. Space Science Board, National Research Council, Committee on Solar and Space Physics, L. J. Lanzerotti, chairman, A Strategy for the Explorer Program for Solar and Space Physics, National Academy Press, Washington, D.C., 1984.