Beyond the Atmosphere: Early Years of Space Science

[173] For want of a more appealing name the phrase particles and fields came into early use in the space program to denote the study of magnetic and electric fields in space and a variety of particle radiations. Among the last named were the extremely energetic cosmic rays, plasma radiations from the sun, and the electrons, protons, or whatever they were that were thought to cause the auroras. (Gravitational fields were not included, falling rather under geodesy, relativity, and cosmology, with which gravity studies were naturally associated.) The term magnetosphere denotes the region of space surrounding the earth where the earth's magnetic field plays a prominent, often controlling, role relative to various particle radiations found there. As will be seen, magnetospheric physics constituted an important aspect of the discipline of particles and fields.
The discovery of the magnetosphere began with Van Allen's discovery of the earth's radiation belt. At White Sands, New Mexico, Van Allen had traced the curve of cosmic ray intensity through the Pfotzer maximum to a more or less steady value at heights greater than 55 km that looked very [174] much as though it might be the free space value of the cosmic ray intensity.4 Cosmic rays, being charged particles, were affected by the earth's magnetic field, and fewer of them were able to get in over the geomagnetic equator than in the polar regions. The less energetic rays were the most affected by the magnetic field, making it difficult to determine what the lower end of the cosmic ray spectrum might be in interplanetary or inter stellar space. Since the total energy spectrum of the cosmic radiation in space would be an important factor in trying to figure out how and where cosmic rays were generated, Van Allen took a special interest in investigating the variation of the high-altitude cosmic ray intensity with geomagnetic latitude. For this purpose he took Aerobee rockets to sea aboard the U.S. Navy's seaplane tender Norton Sound , which had to be specially out fitted with an Aerobee launching tower. Van Allen's sounding ranged from the geomagnetic equator off the coast of Peru to Alaskan waters.5 The measured variations were sufficiently intriguing that Van Allen pursued the subject further with Rockoons-the small sounding rockets that he launched from Skyhook balloons in the stratosphere. These Rockoon experiments turned up a most interesting and puzzling phenomenon. In the auroral regions above 60 km was a rather soft-i.e., moderately penetrating-radiation that could be a mixture of charged particles and x-ray photons.6 This radiation was assumed to be in some way connected with the auroras, and efforts were begun to explore the connection.
At about this time the appearance of the International Geophysical Year satellite program gave Van Allen the chance to extend these investigations to even higher altitudes. When the first Explorer was launched (31 January 1958), Van Allen's counters appeared to show a zero counting rate at, certain locations, which didn't seem to make sense. Further study showed, however, that actually the counters were saturating because o ambient radiations far exceeding intensities with which the counters had been expected to cope. Explorer 3 (26 March 1958) pursued the question.
Soon Van Allen decided that he was observing a region of intense radiation surrounding the earth at high altitude, and on 1 May 1958 he announced his discovery.7 The region at once became known as the Van Allen Radiation Belt. Soviet measurements in Sputnik 3 (15 May 1958) confirm this discovery.
An explanation was quickly forthcoming. The radiations were attributed to charged particles caught in the earth's magnetic field, unable to escape because their energies were too low to allow them to cross the surrounding field lines. One thus visualized trapping regions within the earth's field and spoke of trapped radiations. Suddenly it was crystal clear that the earth's magnetic field, which could prevent some charged particles in interplanetary space from ever reaching the earth, could also prevent other particles already near the earth from leaving.
[175] In retrospect it seemed remarkable that the existence of the radiation belt had not been anticipated long before its discovery. Workers concerned with the problem of how gases escaped from the atmosphere understood that the magnetic field would hinder the escape of ions.8 More significantly, the experiments of K. Birkeland and E. Brilche with cathode rays aimed at small magnetized spheres and the half century of theoretical work by Carl Stormer and others on the influence of the earth's magnetic field on auroral particles and cosmic rays provided a substantial basis for predicting the existence of trapped radiations near the earth.9 Seeking an explanation for the auroras, Stormer had developed a theory of the motion of an electron approaching the earth's dipole magnetic field from the sun. He showed that such an electron would be deflected by the earth's field away from the equator to the polar regions, an action that appeared to him to explain the existence of auroral regions or zones at high latitude.
Stormer's calculations showed that there were regions inside the earth's magnetic field which such solar electrons could not reach, to which he gave the name "forbidden regions." Birkeland, with whom the theory had originated, had already demonstrated in the laboratory that electrons would be deflected to the polar regions, a fact Stormer's calculations nicely brought out.
Later, in the 1930s and after, theorists interested in explaining the geomagnetic-latitude effect observed in cosmic ray intensities, extended Stormer's work to much higher energy relativistic particles-i.e., particles approaching the speed of light-such as were to be found in the cosmic rays.10 Their calculations also revealed forbidden regions toward the geomagnetic equator and served to explain why cosmic rays increased in intensity with increasing geomagnetic latitude.
These investigations furnish an excellent example of how initial orientation can markedly bias an investigator's conclusions. To those seeking explanations of the auroras or the cosmic-ray-latitude effect, the orientation was from outside in. Their particles were approaching the earth from great distances. It was natural, then, that the regions which the earth's magnetic field prevented those particles from entering should be named forbidden regions. While the point was not missed, still the investigators did not focus on the fact that for a particle already within one of those regions, it could be the outside that was forbidden-in other words, a particle of too low an energy already within one of those regions couldn't get out. What were forbidden regions for particles approaching from the outside were trapping regions for some particles already there.
It was only a tiny step from this realization to the idea that these trapping regions might well be filled with trapped radiations forming a radiation belt around the earth. But no one paid any attention to this possibility until, on the eve of Van Allen's discovery, S. Fred Singer in discussing [176] magnetic storms touched upon the possibility that regions of trapped radiations might be found at high altitudes around the earth.11 Following Van Allen's announcement, this field of investigation blossomed forth as researchers vied with each other to learn about the fascinating trapped radiations.12
In the next half-dozen years a new paradigm emerged to characterize the magnetosphere and magnetospheric physics. Whereas before the spring of 1958 the space environment immediately surrounding the earth was thought to be relatively uncomplicated, it soon became clear that the recently discovered magnetosphere was extremely complex. Before the recognition of the radiation belts, there was no generally accepted picture of the space environment near the earth. Students of the earth's upper atmosphere and ionosphere tended to think of these as attenuating more or less exponentially with altitude, eventually merging at some considerable, but unknown, height with the medium of interplanetary space. Around the planet the earth's magnetic field was visualized as essentially that of a dipole, much as depicted in figure 3 in chapter 6. It was known that particles from the sun swept across the earth's atmosphere, some of them causing the auroras. Sidney Chapman, V. C. A. Ferraro, and others supposed that some of the solar particles impinging upon the earth's magnetic field would compress it, thereby causing the sudden increase in the surface field that had long been observed to follow flares on the sun. Such a theory implied, of course, that the earth's magnetic field would be distorted somewhat by the solar particles. Moreover, to explain the main phase of magnetic storms in which the field dropped well below normal for a day or more, Chapman and Ferraro thought of the cloud of solar particles as somehow setting up a ring current around the earth; the current generated a magnetic field that caused the considerable drop in field intensity an hour or so after the sudden increase of the initial phase of the storm. The cloud of solar particles was presumably a plasma; that is, a gas composed of equal numbers of positively and negatively charged particles. Thus, the plasma, though neutral in the large, would be highly conducting. Also, since the positive particles would be deflected in one direction by the earth's magnetic field, the negative particles in the opposite, one could sense intuitively how a current might be set up around the earth although there were formidable difficulties to overcome in developing such a theory. The period of one to several days required for the field to return to normal would then be the time required for the ring current to dissipate.
Chapman and Ferraro visualized the ring current as flowing on the surface of a huge cavity which the earth's magnetic field carved out of the plasma cloud as it swept by the earth. There were, of course, two sides to this coin. From one point of view the earth's magnetic field generated a cavity in the flowing plasma. From the other point of view, however, one could think of the plasma cloud as confining the earth's field to the cavity [177] region. The discovery of the radiation belt focused attention on the second point of view, and the region within the Chapman-Ferraro cavity became known as the magnetosphere (fig. 31).
Because of the intense interest in the new topic, many of NASA's early spacecraft-and those of the USSR, also-were instrumented to make measurements of the particles and fields in the vicinity of the earth and in interplanetary space. By the end of 1964 a highly detailed picture of the magnetosphere had been worked out, a picture that was still evolving.13
Explorer 1 measurements put the radiation belt at about 1000 km above the equator, and Explorer 3 and Sputnik 3 confirmed this observation. From Explorer 4 and the space probe Pioneer 3, Van Allen could show that, at least for particles that could penetrate one gram per square centimeter of material, there were two radiation belts, an inner zone and an outer zone as shown in figure 32. Pioneer 4, which eventually went into orbit around the sun, gave additional information about the extent of the radiation belts. It appeared that the belts extended to about 10 earth radii from the center of the earth, but the exact location of the outer edge appeared to be variable.
The variability was quickly tied to conditions in interplanetary space, which in turn were controlled by solar activity. A major factor influencing the earth's space environment was shown to be the solar wind. In 1958 Eugene Parker had shown theoretically that the sun's corona had to be expanding continuously, and that a continuous wind from the sun should be blowing through interplanetary space.14 Highly conducting and virtually free of collisions among the constituent particles, this solar wind should entrap and draw out magnetic field lines of the sun. Such interplanetary plasma fluxes of about 108 particles per square centimeter per second were measured by Gringauz on Lunik 2 and 3. 15 With a probe on Explorer 10, H. Bridge and coworkers at the Massachusetts Institute of Technology confirmed the fluxes detected by the Luniks and found that the wind came from the general direction of the sun at about 300 km per second.16 More definitive measurements from Mariner 2 and Explorer 18 showed a very gusty wind, nearly radial from the sun, to be blowing at all times with velocities of roughly 300 to 500 km per second. Protons and helium nuclei appeared to be present in the wind.17

Chapman-Ferraro cavity
Figure 31. Chapman-Ferraro cavity. Ring currents set up around the earth were assumed to be the cause of magnetic field effects observed during magnetic storms.

Radiation Belts
Figure 32. Radiation belts. Van Allen's picture of the inner and outer zones of the radiation belt made after Pioneer 3 data returns. J. A. Van Allen and L. A. Frank, from Nature 183 (1959). 430; copyright Macmillan Journals Ltd., 1959.

Meantime more information had been collected on the structure of the radiation belts. The inner zone was shown to be largely high-energy protons, many of which could be accounted for by the decay of neutrons splashed back from the atmosphere.18 The neutrons were generated by cosmic rays colliding with nitrogen or oxygen nuclei of the air; being neutral, the neutrons could move upward unhindered by the magnetic field. But the neutrons decayed quickly and produced protons and electrons which, being charged, were trapped to form a part of the radiation belt. Detailed measurements revealed that both protons and electrons existed throughout the altitude range from the bottom of the so-called inner zone to the far edge of the outer zone. The apparent existence of two belts had been due to the insensitivity of some early instruments to lower-energy particles.
The boundary of the magnetosphere was first definitely located with instruments on Explorer 10, which was launched on 25 March 1961. The spacecraft was projected at an angle of roughly 130 degrees from the direction to the sun, that is, quartering away from the sun. Between the distances of 22 earth radii and the apogee of 47 earth radii, the satellite appeared to cross the boundary at least six times, suggesting that the boundary wavered in the wind. Inside the boundary the magnetic field was 20 to 30 gammas and steady, and there was no detectable plasma. Outside the boundary, however, the field weakened to between 10 and 15 gammas, and plasma [179] was always observed. Data from Explorer 12 in the direction of the sun showed a very sharp outer limit to the geomagnetic field, a limit that came to be called the magnetopause. Beyond the magnetopause was a region in which the magnetic fields were variable in direction and intensity, and the ambient radiation isotropic.19
Thus, by about the beginning of 1962, scientists began to envision a magnetosphere much as shown in figure 33. A continuous solar wind blowing against the earth's magnetic field was pictured as sweeping around the earth, confining the field to an immense cavity which extended to about 10 earth radii in the direction toward the sun, and to considerably more than this in the opposite direction. Inside the cavity lay the Van Allen Radiation Belt which showed considerable structure, with high intensities of energetic protons in the inner portions and large quantities of electrons in the outer reaches. Outside the magnetopause that is, outside the boundary of the cavity-lay a region of turbulent magnetic fields and plasma. It was suggested that surrounding the turbulent region would be found a huge shock wave produced in the solar wind by the earth's magnetic field, which would act upon the high-speed plasma much as a blunt body would act upon a supersonic flow of gas in ordinary aerodynamics. By analogy with aerodynamics, estimates were made of where the bow shock might be found.
The bow shock was first detected by instruments in the Interplanetary Monitoring Platform, Imp 1, otherwise known as Explorer 18, which was launched in November 1963 into an orbit with an apogee at 30 earth radii.20 In the course of its lifetime the spacecraft's instruments provided clearcut evidence that Imp 1 had crossed the magnetopause and the bow shock many times. The data from a magnetometer installed by Norman Ness of the Goddard Space Flight Center were most convincing.21 Figure 34 shows magnetic field data from orbit 11 of Imp 1. Inside 13.6 earth radii, a well-ordered field was noted, but from 13.6 to 20 earth radii the field was quite turbulent. Beyond 20 earth radii the field became quite steady at about 4 gammas, with some fluctuation in direction. The turbulent region from 13.6 radii to 20 earth radii was interpreted as a transition region between the shock wave in the solar wind and the magnetopause bounding the geomagnetic field. Plasma data from MIT and Ames Research Center instruments were consistent with this interpretation.22 Beyond 20 earth radii the MIT instruments showed large fluxes in only one of six energy channels, presumably that due to the solar wind, whereas in, the transition region the plasma probe indicated considerable turbulence, showing appreciable fluxes on all six channels of the instrument.
In December 1963 Imp 1 found the interplanetary magnetic field, which was usually quite steady, to be disturbed, rising to about 10 gammas for a day or more. On the first day of this disturbance, 14 December, the moon was close to lying between the satellite and the sun. Ness originally....

Visualized magnetosphere
Figure 33. The magnetosphere as visualized early in 1962. Here and in figure 35, the lines emanating, from earth represent magnetic field lines. Although the general structure was emerging many features were still to be delineated.

....attributed this unusual disturbance to a wake produced by the moon in the solar wind.23 That the moon with almost no magnetic field should produce a wake detectable so close to the earth at once suggested that the much larger earth with a strong magnetic field would produce a similar wake reaching certainly to the orbit of the moon, and most likely well beyond. It began to appear that the earth's magnetospheric tail should extend to very large distances in the antisolar direction.
As investigation of the magnetosphere proceeded, it was clear that this region was intimately involved in many familiar phenomena, such as magnetic storms and auroral displays, serving in some way as a connecting link between the original solar radiations and the ultimate terrestrial effects. But the precise mechanisms involved eluded explanation. It was shown that both electrons and protons produced the auroras, with electrons of energies below 25 kiloelectron volts contributing most to the auroral emissions.24 Stormer's theory that these particles came directly from the sun into the auroral regions of the earth had to be abandoned when both Soviet and U.S. deep-space probes showed that the fluxes of such particles in interplanetary space were insufficient. An alternate theory....

Figure 34. Magnetospheric bow shock as revealed by space-probe measurements. Magnetic field data from orbit 11 of Imp 1. The magnetopause is at 13.6 earth radii. The second transition ' at 20 earth radii to an ordered field outside is the location of the bow shock wave. C. S. Scearce and J. B. Seek, Journal of Geophysical Research 69 (1964). 3531-69; copyright American Geophysical Union, 1964.
...that the particles were accumulated in the trapping regions of the magnetosphere and then dumped or dribbled into the auroral zones to produce the auroras also ran into difficulties. Although both Soviet and U.S. measurements showed that the fluxes at the altitudes from which the particles could spiral along the field lines into the auroral regions were adequate to produce an aurora, the quantity of radiation was too low. The particles would be drained away in a few seconds, whereas auroras often lasted for hours.25 Brian O'Brien observed, however, from instruments in Injun satellites of the State University of Iowa that trapped electrons in the radiation belt, electrons precipitated into the atmosphere of the auroral zone, and auroral light emissions all increased simultaneously.26 One could conclude that the disturbances ultimately causing the auroras somehow also replenished the radiation belt, perhaps in this way making it possible to sustain a long-duration auroral display. Whether these additional electrons were inserted into the radiation belt from outside or came from lower energy electrons already existing within the belt and accelerated by some mechanism to the necessary higher energies was not known. Indeed, while many clearcut relationships between auroras and radiation belt activity had been [182] established, at this stage the actual mechanism producing the auroras remained a mystery.
Also unexplained was the immediate cause of the main phase of magnetic storms. A ring current around the earth continued to be the most likely candidate, but how such a current was generated remained a puzzle. It could be shown that charged particles in the magnetosphere, in addition to spiraling around magnetic field lines bouncing back and forth between northern and southern reflection points, would also tend to drift longitudinally, the electrons drifting eastward and the protons westward.27 Thus, these drift motions produced in effect a ring current, which S. Fred Singer suggested as the cause of the main phase of magnetic storms.28 By the end of 1964, however, no spacecraft measurements had been able to locate the postulated ring current.
By the mid-1960s a very detailed, though by no means complete, picture of what the magnetosphere was like had evolved, as illustrated in figure 35. In the magnetospheric paradigm of 1964 the existence of the solar....

Figure 35. The magnetosphere as visualized in the mid-1960s. Space-probe measurements have provided a wealth of detail. The principal research problems r, shifting from describing the phenomenon to explaining the relationships and processes.
[183] ....wind had been established. The wind consisted of protons mostly, with some alpha particles (helium nuclei), both of which had been observed. To be neutral the wind had to include equal numbers of electrons, but these had not been detected as yet. Embedded in the solar wind was an interplanetary magnetic field pulled out of the sun by the solar wind plasma. Near the earth the interplanetary field intensity was between five and six gammas. Blowing against the earth's magnetic field, the solar wind produced a huge shock wave sweeping around the earth much as an aerodynamic shock wave accompanies a supersonic airplane. But, whereas an aerodynamic shock wave is produced by compression of a gas consisting of air molecules all colliding with each other, the magnetospheric shock wave was set up by deflection of the individual plasma particles by the earth's magnetic field and was referred to as a collisionless shock wave.
Behind the shock was a region of turbulence. Here the magnetic fields became highly disordered; particle velocities, which in the solar wind were usually confined to a rather narrow range, suddenly varied widely. Closer to earth this transition region was bounded by the magnetopause enclosing the geomagnetic field now grossly distorted from the simple dipole configuration that would have existed in the absence of a solar wind. Some of the field lines that would otherwise have lain on the sun-ward side of the earth were swept backward in the antisolar direction and along with field lines on the night side were extended into a magnetospheric tail. The magnetic field lines that still enveloped closed regions near the earth contained the Van Allen Radiation Belt, which paradoxically appeared to be more limited in extent on the night side of the earth than on the daytime side, where the field was compressed by the solar wind. On the dayward side, toward the poles, where some of the field lines were swept out into the tail, appeared a cusp or dimple in the magnetopause. It was thought that where magnetic field lines of opposite direction came together near the equatorial plane of the tail, they might cancel each other producing a neutral sheet. Along this neutral sheet one could envision charged particles leaking from interplanetary space into the zones closer to earth, where they could then be steered by the field toward the poles.
In the steady state this magnetospheric configuration drifted slowly around the earth, always keeping the tail away from the sun as the earth revolved around the sun. The nose of the shock wave was about 14 earth radii from the center of the earth, and the nose of the magnetopause typically at about 10 earth radii. The extent of the magnetospheric tail was a matter of speculation, but it appeared certain to reach at least to lunar distances.
At times when the sun was disturbed, the magnetosphere and the radiation belts were affected. The spatial extent of the magnetosphere varied appreciably and trapped radiations were enhanced. following solar storms. [184] There was a question as to whether during these disturbed conditions new particles were injected into the radiation belt or energy was transferred by hydromagnetic waves from the interplanetary plasma to particles already in the magnetosphere.
Many problems, of course, auroras appeared tantalizingly remained unsolved. An explanation of the close, yet elusive. The immediate cause of the main phase of magnetic storms was still to be found. How energy and particles were inserted from the, interplanetary medium into the magnetospheric regions had yet to be explained. The existence of the neutral sheet had not been established, nor had its precise role in magnetospheric physics been described. How the field lines in the magnetospheric tail closed again also had yet to be described. Did they perhaps connect with magnetic field lines in interplanetary space, as some surmised? Related questions concerned the sun. How did the sun manage to eject the streams and clouds of highly energetic particles and magnetic fields that from time to time upset the normal conditions in the solar wind? There was reason to suppose that solar magnetic fields were the ultimate source of the energy conveyed to these clouds, but there was as yet no generally accepted explanation.
Most of the early research on the magnetosphere was directed toward describing it. As the subject became more familiar, more and more attention was devoted to achieving a coherent explanation of the magnetosphere and its relationship to the sun and interplanetary medium on the one hand, and to terrestrial phenomena on the other. By 1964 the major interest of the scientists lay in trying to understand the various processes in magnetospheric physics. There was, of course, still much to learn about what the magnetosphere and its most important phenomena were. But enough of the what had been learned that now investigators could profitably spend much of their time on the how , the immediate and ultimate causes of the auroras, magnetic storms, radiation belts, and the magnetospheric tail, and on the processes that related causes with effects. To understand these processes would be the principal objective of magnetospheric research in the years ahead.