National Academies Press: OpenBook

Space Plasma Physics: The Study of Solar-System Plasmas (1978)

Chapter: Shock Systems in Collisionless Space Plasmas

« Previous: Kenetic Process in the Solar Wind
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 54
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 55
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 56
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 57
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 58
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 59
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 60
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 61
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 62
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 63
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 64
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 65
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 66
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 67
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 68
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 69
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 70
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 71
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 72
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 73
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 74
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 75
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 76
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 77
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 78
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 79
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 80
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 81
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 82
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 83
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 84
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 85
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 86
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 87
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 88
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 89
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 90
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 91
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 92
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 93
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 94
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 95
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 96
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 97
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 98
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 99
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 100
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 101
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 102
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 103
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 104
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 105
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 106
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 107
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 108
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 109
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 110
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 111
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 112
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 113
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 114
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 115
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 116
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 117
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 118
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 119
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 120
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 121
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 122
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 123
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 124
Suggested Citation:"Shock Systems in Collisionless Space Plasmas." National Research Council. 1978. Space Plasma Physics: The Study of Solar-System Plasmas. Washington, DC: The National Academies Press. doi: 10.17226/18481.
×
Page 125

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.

SH0CK SYSTEMS IN COLLISIONLESS SPACE PLASMAS by E. W. Greenstadt and R. W. Fredricks Space Sciences Staff TRW Defense 6 Space Systems Group 807

808 INTRODUCTION A gas in motion must find a way of flowing around an impenetrable object in its path. Gas particles flowing directly into the object are reflected from it, i.e., they bounce off the object's surface, and in turn collide with other particles farther away, about as far as the mean free path; these collide with others, and so on, until a sufficient amount of the gas is deflected to form a flow pattern around the object. This suc- cessive colliding of particles is essentially the transmission of sound through the gas and the transmission travels at the natural sound speed in the gas, "warning" the gas that it is approaching an obstacle. But what if the gas is supersonic? That is, what if it flows toward the object at high speed, faster than sound can travel backward to warn of the oncoming barrier? In that case the gas particles or, equivalently, the sound waves, pile up ahead of the object, forming a new object, a "shock" wave which is not impenetrable, but through which the gas can travel. On passing through the shock, however, the gas is slowed down and heated, i.e., the particles are scattered by collisions, so that between the shock and the original object the gas is no longer supersonic and flow deflection can take place as it would have subsonically. But what if the gas is so tenuous as to be col 1isionless? And what if the gas is a plasma consisting almost entirely of ionized hydrogen, which is really two gases, one of protons, one of electrons? And what if the plasma is magnetized so that particle motion is related to the field and there is not one sound wave velocity but many types of waves with different velocities dependent on frequency and on direction in the plasma with respect to the magnetic field?

809 Then there is formed a collisionless, plasma shock: a new, "penetrable object" fashioned in a complex way by the interactions of charged particles, magnetic gradients, and wave electric fields. The problems posed in the preceding paragraphs are far from academic. The gas they describe is the solar wind that occupies the entire known solar system. Moreover, spacecraft instruments have disclosed a col 1isionless plasma shock in front of every one of the five planets visited so far, and still more of them are continually found traveling through the solar wind away from solar flares and ahead of certain plasma streams originated in the sun. The existence of others has been inferred in the solar atmosphere. The first importance of col 1isionless shocks derives therefore from their ubiquity in the solar system, where few interactions with or within the solar wind or solar atmosphere can be conceptualized or comprehended without including an appropriate shock "surface." Shocks are also of great value for study of their physical construction, however, which incorporates some of the fundamental processes common to plasmas. The paramount interest in plasmas in general, whether for under- standing or application, in the laboratory, in near-earth environment, or in astrophysical extrapolations, lies in their responses to nonequi1ibrium, particularly nonlinear, conditions. The collisionless shock provides a nonlinear perturbation that "exercises" the gas flowing through it with many of the wave-particle calisthenics plasmas everywhere are expected to perform. Moreover, the col 1isionless shock in space performs its exercises on such a large dimension that one or more spacecraft can make detailed mea- surements within its nonequi1ibrium scale length. This gives opportunity for really authoritative observations, almost like the idealized concept il-

810 lustrated in the motion picture "Fantastic Voyage" in which a crew pilots a microscopic craft among the cells of the human anatomy. 0rdinarily, the discovery of so interesting an entity as a natural col Us ion less shock would be expected to inspire intensive study under the controlled conditions of the laboratory. There have indeed been laboratory- produced shocks and much has been learned from them. In this case, however, nature offers such an excellent observation site that much of the follow-up laboratory work is better done by spacecraft. There are of course advantages C and disadvantages inherent in both laboratory and in spacecraft measurements. The advantage of doing shock experiments in a laboratory is the obvious one of control on conditions and consequent repeatability of the experimental measurements. Unhappily, finite dimensions and plasma lifetimes place severe constraints on the ability to measure the full evolutionary charac- teristics of the shock. Also, density of the plasma customarily is so high that shock scale lengths are smaller than the dimension of the diagnostic probes used, thus obviating acquisition of space-resolved fine structures. The greatest advantage of spacecraft measurements Fn either the Earth's bow shock or interplanetary shocks is potential resolution of the fine structure by probes of dimensions much smaller than shock scale lengths. The offsetting disadvantage has been the restriction to single-point measurements of passing structures whose motion relative to the probe is usually unknown, so that ambiguous Doppler-shifted frequencies and scale lengths result. Further com- plications arise from unknown or poorly-known temporal changes in upstream plasma parameters, with the result that characteristics of magnetohydrodynamic (MHD) fluid elements in the shock cannot be connected to the real characteri- istics of the same fluid elements upstream at an earlier time.

811 The single-point restriction has been partially overcome by taking ad- vantage of unplanned but advantageous conjunctions and separations of existing spacecraft to answer some of the large-scale (macroscopic) questions of shock structure. The main benefit has been characterization of the unshocked fluid element. Fortuitous conjunctions have not provided separation distances suitable for isolating microscopic phenomena. Even so, some inferences have been made about the modes of wave-particle interaction responsible for re- arranging the MHD properties of the plasma within the shock transition layer. The "exercises" through which a perturbed plasma restores itself to equilibrium are a marvel of complex instabilities, meaning electric and electromagnetic wavemodes that grow and decay in such a way that their fields can deflect the ions and electrons of the plasma and establish new average MHD parameters of the gas without particles colliding mechanically or electrically. The growth, maintenance, and decay of the wavemodes depend on the numerous quantities it takes to characterize an ionized MHD gas. A few of these are the effective "temperatures" T.. and TI parallel and perpendicular to the magnetic field B; their ratio Tp/Tj; the ratio of electron to ion temperature ' -L T /T.; the average density M = N. » M ; the ratio of thermal to magnetic C 1 1*6 energy density, 6 » 8irNtc(T.+T )/B2; and the magnetosonic Mach number of the flowing gas, M » V//C*+ C *, where C. and C_ are the Alfven and sonic wave velocities. While a few inferences have been justified regarding the insta- bilities occurring within some shock profiles under some upstream conditions, the complex of possible interactions has been but partially explored, and a full accounting of shock processes corresponding to all parameter sets has yet to be posted, let alone explained theoretically.

812 The following sections present the status of our knowledge of shock structure from a plasma physics point of view, based on direct observations in space. We do not cover the formation or propagation of extraterrestrial shocks via the fluid approximation and we do not intend or attempt a critical review of the details, arguments, or contributions that have yielded the picture we present. 0ur objective is a timely, reasonably documented picture of the col 1 isionless shock and our methods of observing it in space, in which image of the present and future of the subject can be discerned without spectacles. We begin by describing more technically the objects of our attention, and follow with a succinct history of the subject. We then des- cribe what is known of shock structures in space on both large and small scales by synopsizing the documented features of the Earth's bow shock sys- tem. The ensuing discussion touches on the properties of shocks elsewhere in the solar system and on the status of shock theory and then describes in some detail an example of how properties of the bow shock system are investi- gated with paired satellite measurements. We close with some recommendations for the future.

813 BRIEF HIST0RY 0F C0LLISI0NLESS SH0CK EXPERIENCE "Col 1 is ion less shock" is a title applied, by analogy with the familiar irreversible shock in neutral gases, to the case of a plasma sufficiently tenuous such that any classically calculated mean-free-path for particle- particle (coulomb) collisions is much greater than the characteristic length of field, density, or velocity jumps usually associated with "shock fronts." Such col 1isionless shocks can occur in a bewildering variety in both the laboratory and in space plasmas. This is because the structures of such plasma shocks can depend critically on the parameters of the plasma, e.g., magnitude of flow velocity, directions and magnitudes of intrinsic magnetic fields, details of upstream ion and electron velocity distributions, plasma betas, and of course on the size and configuration of either obstacles to the flow or of sources of driver plasmas. In this review we deal only with hydromagnetic shocks; that is, we talk only about col 1isionless plasmas in which intrinsic magnetic fields are present and have an influence on shock structures that cannot be neglected. There are two types of col 1isionless shocks of interest here. 0ne is a bow shock produced when an obstacle is placed in a "supersonic" plasma flow. The obstacle may be a solid body, an intrinsic magnetosphere, or even a planetary ionosphere. The term "supersonic" may refer to the ratio of flow velocity to some characteristic velocity (sound, ion-acoustic, Alfven, magnetosonic, etc.) with which information about the presence of the obstacle would be transmitted upstream.

814 A second type of col 1 isionless shock is produced by a secondary plasma (i.e., a driver plasma) injected into a primary plasma of different parame- ters). The secondary plasma acts as a piston to produce shock waves in advance of it. Typical examples are interplanetary shocks produced by solar flare plasmas injected into the evaporating solar corona (solar wind). Such shocks, if they are configured to illuminate the earth, may command our attention by producing sudden commencements and sometimes magnetic storms which affect radio communications, power distribution networks, and even telephone networks, not to mention often deleterious effects in communications satellites. Apparently the first theoretical work on MHD shocks was done by D_e_ 14 Hoffman and Teller who derived the MHD analog of the Rankine-Hugoniot re- lations for gas-dynamic shocks. Montgomery investigated the development of MHD shocks from large amplitude Alfven waves. Papers by Zhigulev ^ and Zhigulev and Romishevskii appear to be the first serious conjectures that the continuous solar wind plasma, upon encounter- ing the intrinsic magnetic field of t;he earth, should produce a standing bow shock front. No dissipation or entropy-changing mechanism was specified in any detai1. General theoretical work on MHD col 1isiorless shocks was published by 25 Fishman, Kantrowitz and Petschek , in which non-linear wave-wave interactions were postulated as a dissipation mechanism. In 1962, Axford and Kellogg published independent papers postulating a bow shock around the earth due to a solar wind plasma flowing, at a speed greater than the Alfven speed, by earth's magnetospheric obstacle.

815 In 1961, Auer, Hurwitz and Kilb published a computer study of large- amplitude compressional waves in col 1isionless, magnetized plasma, and pro- duced shock-like solutions. They attributed dissipation to particle orbit- crossings. The geophysica11y oriented theoretical work between about 1962 and 1968 proceeded largely along the lines of the "gas dynamic analog," in • which the col 1isionless MHD bow shock was studied grossly by reducing the MHD jump conditions for 8, v, p and pressure to a set of jump conditions analogous to those for shocks in gases. An excellent bibliography of such 98 efforts is contained in the review by Spreiter and Alksne . While such gas» dynamic analogs enjoyed great popularity and some success in predicting gross configurational characteristics of earth's bow shock, they were in- capable of predicting any detailed structure, and in fact gave misleading information on parameters of the plasma flow behind (earthward of) the shock. Dissipation mechanisms remained a mystery, as did even gross struc- tural properties of the real MHD shock. Meanwhile, other plasma physicists concerned themselves with theoret- ical details of MHD col 1isionless shocks, for various reasons. Collision- less shocks were proposed to heat CTR plasmas, so that joint interest was 56 exhibited by both laboratory and space physicists. Kellogg , noting that magnetic field gradients inherent in the soli ton solutions of Ad 1am and Allen corresponded to currents which should produce unstable plasma oscil- lations within the gradient (Langmuir waves, Buneman waves and ion-acoustic waves), postulated that such instabilities would produce turbulent wave- particle interactions and consequent heating, a dissipation mechanism.

816 Or qq ]00 Sagdeev pursued this same reasoning, Tidmarr ' proposed both ion- electron and ion-ion counterstreaming instabilities in magnetic field gradients as the dissipation mechanism and presented a partly quantita- tive but largely qualitative model for earth's bow shock. In 1968 and 1969, theoreticians were presented for the first time with two experiments which indicated that microinstabi1ities were indeed present in shock structures, and were somehow Important to dissipative mechanisms. The first of these almost simultaneous experimental results came from the plasma wave detector aboard NASA's 0rbiting Geophysical 0bservatory (0G0-5)- The second came from experiments in the TARANTULA device at Culham Labora- tories in England under the direction of J. Paul. The 0G0-5 results, published by Fredricks et al.^ '" showed clearly that strong electrostatic waves were present in the larger magnetic field gradi- ents within the structure of earth's bow shock. Because the probe size was small compared to local Debye lengths as well as gradient scale lengths, these experiments were quite conclusive. However, they suffered from some defects, such as inability to measure field polarization, and unknown Doppler shifting caused by relative motion between the spacecraft and the flowing solar wind. Also, it is clear that the shock structures themselves •th had intrinsic motion. Although Fredricks et al. were forced to make some unverifiable interpretations of shock thickness to model their shock, the presence of electrostatic wave turbulence was unequivocally demonstrated. The laboratory experiments by Pau1 et a I. and Doughney et al. also clearly showed electrostatic waves in magnetic field gradients in labora- tory shocks. Measurements were carried out by laser scattering diagnostics.

817 Unfortunately, the laser scattering came from a volume of dimensions compar- able to the scale length of the magnetic gradient representing the shock front. This did not allow adequate spatial resolution in the experiment to determine unambiguously the exact electrostatic mode responsible for the observed turbulence. In spite of such drawbacks, these two results stimulated a significant theoretical thrust to investigate just what electrostatic instabilities could be produced by the current systems which produce the detailed mag- netic field profiles in MHD col 1isionless shocks. This was further rein- forced by the publication of a more comprehensive study of selected bow shock structures made by instruments aboard 0G0-5, and has been confirmed by instruments aboard subsequent flights, such as NASA's IMP-6,7,8 series, ESA's HE0S-1,2 and the USSR's Prognoz series. However, with respect to resolving exactly the details of the plasma wave modes responsible for shock dissipation mechanisms, little substantial has been added to the 0G0-5 results. The reason for this is that in order to advance the state of know- ledge further, coordinated measurements between two, or among several, satellites at known separation are required. A number of studies of such measurements have been conducted with relatively little impact on dissipation modes, but with appreciable success in defining the overall properties of shock structural forms. The next section summarizes the results to date.

818 EARTH'S BOW SHOCK SYSTEM Shock Macrostructure Figure 1 is a conceptual composite of the overall characteristics of the earth's bow shock system as we now perceive it in an imagined steady- state condition. The shock is defined in the figure by the signature of the magnetic field magnitude carried through it by the solar wind. The nominal shock is not planar, but curved, so that its relationship to a uniform solar wind flow with velocity V-.,, carrying uniform field B... at an average 45° angle to ysw, is necessarily asymmetric. That is, the local normal of the nominal shock varies continuously from point to point in its orientation with respect to both the uniform upwind (unshocked) flow and the uniform field. The nomenclature attached to the shock in the figure is defined by the orientation of the local normal to the upstream field, with the guidance of theory and data from laboratory and space observations: perpendicular is taken to mean that the local orientation of n to B-w, represented by angle 1 /2 9 _, is within arctan (M /M.) ' * 1?8 of 90°. This criterion is adopted no e i directly from laboratory and theoretical nomenclature ' and is associated with certain specialized shock properties which have not yet had comprehen- sive verification in space. The orthogonal geometry at the left side of the shock in the figure, where 8 _ » 0°, is called parallel, in agreement with no 58 theoretical designation Every geometrical condition other than perpendicular was, before mea- surements in space, simply called oblique. All the remaining properties,

819 designations, and distinctions in Figure 1, excepting the standing whistlers, have therefore resulted from spacecraft observations. When 9 _ S 80°, but £ 50°, standing whistlers are observed upstream no from the bow shock, as in the laboratory , at least when 3 « 1 and the Mach number is low; i.e., when M S 3 ' . When 9 _ falls below about 50°, the no character of the shock is altered radically. Empirical results from satel- lite observations have therefore divided the shock Into two broad cate- gories, quasi-perpendicular and quasi-parallel. Quasi-perpendicular shocks have a monotonic, sawtooth, or wave-step magnetic profile, and resemble the perpendicular shock in being sharply defined, although the ramp through which the field rises to its downstream level is thicker than in the very specialized perpendicular case. Quasi-parallei shocks have multigradient magnetic profiles, are thicker yet, and do not feature a clearly-definable boundary between upstream and downstream field or plasma. The entire range of 8 has been surveyed statistically by many shock nb passages with many satellites. The asterisks in Figure 1, however, mark those more restricted parts of the range which have been observed at high- resolution with comprehensive, but still incomplete instrumentation and with two or more spacecraft at once. An enumeration of the qualifications to be attached to the very sim- plified Figure 1 serves as a virtual prescription for the next decade's study of shock structure in space. Before discussing these, the proton velocity distributions and the foreshock need to be mentioned.

820 The proton distributions are represented qualitatively by the small three-dimensional sketches scattered around Figure 1. A simple, relatively cool anisotropic solar wind is shown at lower right, with its long axis parallel to B-w, since on average, Tj/Ti y 2 in many solar streams. A modi- fied proton distribution, currently under study and still incompletely de- fined, is attributed to the solar wind flowing through the foreshock at lower left, where a second group of high velocity return protons, reflected and possibly energized by the shock, is traveling away from it. The "foreshock," to be discussed in greater detail later, is a region attached to, indeed part of, the quasi-parallel structure in which protons (and electrons) re- flected from the shock gradients into the solar wind interact with the in- coming, unshocked plasma to generate magnetic waves of appreciable amplitude (=: 1/2 B-.. peak-to-peak). The long dashed line represents the foreshock boundary, which follows the return-proton, guiding-center path resulting from the vector sum of the parallel component of ejection velocity and the drift velocity given by y_w x B-,,. The flat-topped, heated, bimodal distribution found behind the typical high-8(= 1), high-M(z 5) quasi-perpendicular bow shock is shown at upper right, while the sketch at upper left is Intended to represent the multi- modal, undefined, but predominantly solar wind-like distribution In the quasi-para 1 lei structure, of which fragmentary cross sections have been ob- tained by the plasma detectors of several spacecraft. The question marks in Figure I accompany those characteristics of the depicted bow shock that represent reasonable extrapolations from measurement but are by no means established facts. The incompletely-determined issues are, explicitly, left to right:

821 162. What is a typical instantaneous velocity distribution within the quasi-parallel, or parallel, structure? Are return protons separable as a group? So far no plasma probe in the shock or magnetosheath has had the temporal or angular resolution to develop a reliable spectrum in velocity space. One source of difficulty is the variability of the field direction, because any- anisotropy of the particles would be constantly changing dir- ection in phase with the large amplitude waves in the field, at periods of a few seconds. The best that has been established so far is that there is an average energy distribution in the parallel structure that's distin- 49 guishable from either solar wind or quasi-perpendicular magnetosheath plasma . What, really, is the magnetic structure of the parallel shock? We know the mul tigradient profile becomes 2 R or more thick, but does it really extend far upstream? Or downstream as far as the magnetopause? The principal difficulty here is that the IMF is seldom in a constant di- rection for very long, and the establishment of an extensive wave region must take a finite time. We haven't had, knowingly, many opportunities to observe the parallel, or quasi-parallel, structure in steady state condition, let alone to do so with adequate instrumentation. 5. What is the true three-dimensional proton spectrum behind the quasi- perpendicular structure? The post-shock proton distribution sketched at upper right is shown as essentially isotropic, but the true shape is undetermined. Isotropy is improbable, particularly for the second peak of the bimodal dis- tribution, which is most likely a high energy component in the ambient field direction. The direction is not always well-defined, however, and obser- vational distributions remain to be documented.

822 We may now return to the qualifications attached to Figure 1. To begin with, the sketches represent somewhat mixed conditions, for illustrative purposes. The whole magnetic profile is drawn for essentially laminar conditions, i.e., for Q and M both low, but the post-shock proton distribution, upper right, is most appropriate for a moderate 6(~ 0, supercritical M(~ 3) solar wind. In contrast, the distribution corresponding to the laminar case would show rather little thermalization and no secondary peak , which Is why it hasn't been drawn. Further, when B is very high (£ 8), the monotonic nature of the quasi-perpendicular ramp is destroyed, at least on a fine time scale of seconds . For high Mach numbers, M £ 5, the magnetic ramp is also more irregular than displayed here, but the demarkation between upstream and downstream states is never in doubt as it always is in quasi-parallel cases. Mention of 8 and M provides the basis for the generalized qualification that the shock system exists in a multiparameter plasma in which every com- bination of values affects shock structure on either the macro or micro- scopic scale, or both. In addition, there are numerous physical properties of the plasma that simply do not appear in Figure 1 , all of which are of great interest: electron spectra, T /T., T\\+/Ti + , ct-particle contributions, <E> wave spectra, <B> wave spectra, are some of these. Finally, the most important geometrical and temporal qualifications are not pictured at all: the figure is drawn to represent the shock's structure in a single plane defined by §-,, and V-,,, but all shocks in space are doubly curved in three- dimensional space. Moreover, every parameter characterizing the solar wind varies either continuously or sporadically so any picture of a shock In the

823 solar wind is soon replaced by another. Indeed shocks often interact with each other as, for example when a fast-stream shock, tipped at some angle, sweeps past the earth, colliding and passing through the curved bow shock, first on one side, then on the other. No hint of such a dynamic situation is indicated in the static, two-dimensional, limited-parameter drawing offered here. General Scheme The gross effects of M and 8 on shock structure are summarized in a classification scheme devised by Dobrowolny and Formisano . Some examples 43 29 were illustrated by Greenstadt and Formisano , and some cases have been described in specialized reports. The classification is elaborated in Table 1, with citations of detailed descriptions where available. In addition to the entries in the table, some perpendicular, or very nearly perpendicular, cases have been displayed in detail by Rodriguez and 8l 82 Gurnett ' . Their examples would be summarized according to the same scheme as follows: Parameter Values Plasma Conditions Name of Structure Features 6 * If M. > 3 Warm plasma, PERPENDICULAR, Sharp field grad- supercritical TURBULENT lent followed by mach number appreciable multi- gradient fluctuation- Proton distributions unspecified. For quasi-paraliel geometry, the classifications according to parame- ter have not been clearly differentiated by observation, largely because of the difficulty in knowing whether any given spacecraft passage has encountered

824 a steady-state or a transient view of the shock. 0nly one study has produced generalizable details, as noted. The other cases are isolated incidents, possibly of partially developed structures. Wherever no detailed study has been conducted, the parenthesized citation of the Asilomar report (^i) provides at least an example of the corresponding case. In the table, the dividing value M = 3 is the experimentally-determined value approximating the critical Mach number known in laboratory shock 76 studies . Under q-parallel geometry, and for very high 8, strictly local combinations of parameters may play an important role in defining the pro- cesses taking place in local gradients. Microturfaulence An overall description of the wave properties that have been inferred from a combination of statistical and case-history studies in the bow shock is sketched in Figure 2. Electromagnetic noise power P_, shown here in terms of magnetic field, increases with 8 to levels roughly an order of magnitude or more higher when 8 > 10 than when Q ~ .1. Average electric wave strength <E> is seen at well over an order of magnitude higher at M = 10 than at M«l, and peak fields in critical portions of quasi-perpendicular shocks may reach two orders higher than typical values in laminar shocks. The general outlines of Figure 2 result from numerous observations, published and unpublished, of varying degrees of depth. The asterisks and triangles mark the parameter combinations of particular examples that have been exhibited in one or more of the references in Table 1. Most of these have been studied with care so their places in the scheme are reasonably

825 reliable. The perpendicular cases of Rodriguez and Gurnett^1'^2 are included in the figure (circled asterisks). Clearly, the quasi-parallel division is not only undocumented, but largely undefined for warm and hot solar wind conditions. In dealing with waves, as with particles, energy distributions are important, so the hybrid quantities of Figure 2's vertical axes give an inadequate description of the shock profile. The magnetic power density must be interpreted in terms of its place in the total spectrum, whose form varies from f to f , ' and whose integrated power has been found to be about k x 10~ ergs/cm from 1 to 140 Hz' and about x.** x 10 3 82. ergs/cm from .02 to k.Q kHz . The vertical scale for P0 in Figure 2 applies D at f « 10 Hz. Electrostatic waves are -treated differently. The scale of <E> refers to the peak rms field strength for f £ 3 kHz, because in the shock there is a noise maximum in this frequency range, formed by many dis- crete frequencies ' * . The noise seems to arise largely from ion-acoustic wave generation, some of which is undoubtedly responsible for electron heat- 66 ing in the shock ramp . The role of wave-spectral form in shock characterization can be ap- preciated by viewing Figure 3. In the center, a magnetic profile along shock normal n is drawn for a typical supercritical, essentially perpendicular, bow shock in a warm solar wind, with electron and proton velocity space distri- butions superimposed. At left, sequences of magnetic, and, at right, se- quences of electric, spectral "snapshots" in the appropriate scale along n 82 (guided by Figures 13 and 14 of Rodriguez and Gurnett ) have been reproduced

826 81 from Figures 3 and k of Rodriguez and Gurnett . The magnetic spectra at left show the sharp enhancement of electromagnetic noise corresponding to the main field gradient, but no isolated peak in the measured range of frequencies. The electric wave noise at right is more complicated. Ahead of the main gradient, where there is a "foot" in the solar wind with associated small amplitude precursors, there is also a clear wave resonance at about 10I* Hz marking the electron plasma wave frequency f corresponding to the up- stream density. The resonance is very likely associated with electrons streaming through the solar wind along §_.. at the outer edge of the shock transition layer. Strictly speaking, this feature should be most apparent when B-.. is not exactly perpendicular to n, allowing electrons to stream outward ahead of the shock proper. Once the main (average) gradient in B is encountered, the f noise at 10I* Hz vanishes and the lower frequency end of the spectrum becomes enhanced. Unlike the magnetic noise, however, the electric noise has a shoulder, or even a local maximum, between .1 and 1 kHz, indicating the presence of ion acoustic wave noise at the local proton frequency f .. This characteristic of the electric spectra persists, at lower levels, into the downstream plasma. The differences between B and E spectral signatures motivated the different presentations of their scales in Figure 2.

827 The particle velocity distributions, symbolized by the negative and positive signs in Figure 3, are altered by the shock in several ways. The electrons, hotter in the solar wind than the protons, are partially scattered in the foot of the ramp, but toward the rear of the field gradient, and beyond, they become rather uniformly scattered, giving the flat-topped cross section observed by Montgomery et al. . The protons begin in the • solar wind with T1I/II ~ 2, and are slowed slightly in the foot with some scattering of particles to higher energies. They become partially heated in the ramp and exhibit a secondary, high energy peak, and by the rear of the main gradient have been scattered to higher "temperature" than the electrons, but with a nonmaxwel1ian distribution featuring a high energy tail. The directional properties of the proton velocity distributions, i.e., their degree of anisotropy, are known only in the solar wind, so the sketches are illustrative rather than representational. In and behind the shock ramp, distributions represented by double, or even multiple bubbles, might be appropriate. Returning to Figure 2, the most important message of the figure is that microscale magnetic and electric wave activity are somewhat Independent of each other, but are both low for the bow shock in cold, low M solar wind flow and both high for the bow shock in hot, high M solar wind flow. The variation of PD with B has not revealed any inflections, but the Mach number B dependence of <E> appears to undergo a significant change at M a 3. This Is a reasonable phenomenon to expect since in the absence of particle collisions It is plasma waves that presumably supply the electric fields needed to scatter and heat both electrons and protons in the shock transition, and the critical Mach number at which resistive dissipation by drift current insta- bilities becomes inadequate is at M * 3. Thus when M > 3 either a different,

828 more potent wave mechanism or two separate mechanisms operate to produce higher dissipation than is required when M 'y 3- The investigations of 81 82 Rodriguez and Gurnett ' " revealed that for perpendicular and quasi-perpen- dicular shocks of supercritical magnetosonic Mach number, electric wave noise in the .02-4 kHz band tends to rise with T /T or, alternatively, since T doesn't change very much in the solar wind, tends to fall with increas- ing T . Thus we know T /T plays its part in determining microscopic shock structure, as expected, but it must be noted that the scatter in the cited correlations was large and significant work is still to be done. Certainly the record of influence of this parameter must be extended to include the entire ranges of the other macroscopic parameters. Similar diagrams based on T /T „ T I/T I etc. await the outcome of future investigations, e p' p| pj_ Foreshock One of the most interesting discoveries of space plasma physics has been the shock-generated region of particles, waves, and wave-particle inter- actions that inhabit a vast volume of space outside the bow shock. The existence of such a region where B is parallel to n was broadly predicted *"oW by Kellogg-* , who modeled the regions in terms of whistler waves under station- ary conditions. Experimental data, almost always describing transient con- ditions, have disclosed a somewhat different picture that may include the developmental stages of Kellogg's model, but the latter has yet to be docu- mented. We describe the observational results. Charged particles exist in the foot of the classical shock for various reasons: some have failed to negotiate the shock electric potential; some

829 have been reversed in direction by the enlarged field in the ramp gradient; some have been deflected by plasma waves in the shock transition. Many of these have guiding center velocity components opposite to, and greater in magnitude than, V-w, and can therefore leave the shock "upwind." These spiral around 6... into the solar wind with net motion outward away from the j W shock. As they do, they move, in the shock frame, according to the expres- sion • £ ' (VBSW> HSW * §SW + * x Bsw) , at where fl is the appropriate cyclotron frequency ' . The result is a modi- fied spiral trajectory displaced from B-w in the shock frame, with net guid- ing center velocity V » Vn + V,, where V. • Vn » 0, as shown in Figure k (a). ~r ~|| ~d ~d ~H We designate the upstream region occupied at any instant by some species of particles and/or the waves they excite the "foreshock." In practice, space physicists have formed the habit of representing the particles as leaving the shock with guiding center velocity pV.y along B , while the field itself is carried downstream with the particles, at speed V . The result is the same, as Figure k (a) indicates, but the artificial p\/_w +• V - resolution of V , which arose from observations of waves rather than particles, has lent itself more readily to experimental treatment. We expect this practice to decline as improved particle detection mandates a more direct analysis of the fores hock's material constituents in terms of their actual pitch angle velocities Vi and Vi determined by plasma-physical processes in the shock. For the present, however, we shall continue to deal with V and 9 using experimental results on the value of p.

830 Since the vectors governing the foreshock are B ' and V_w, and the former is variable in time, the action at a given instant takes place in the infinite set of parallel planes containing the solar wind's flow velocity and its magnetic field. One of these planes is illustrated in Figure Mb) for a sunward IMF oriented about 30° above the ecliptic. Electrons are hot, around 1.5 x 105°K , and are the first to be affected at the outermost fringe of the shock . Their M« is high (p z 10) and they are not displaced i 2k very much from B.,, as they progress upstream . The negatives (dashes) in Figure 4(b) signify the region occupied by return electrons. Most return protons are appreciably slower than electrons (p « 2) and V, is a large" part of V . Return protons are therefore displaced considerably from §_,.. Their region is denoted by the pluses in Figure k (b). Figure k(b) displays the complexity of the foreshock to first order: there is an-electron foreshock and a proton foreshock. But there is also much more. There is of course the three-dimensional foreshock consisting of the infinite set of planes intersecting the shock, parallel to the one depicted. There are also electrons and protons upstream with energies, i.e., velocities, differing from those described above, because there is a spec- trum of each return species at any given point of origin in the shock. Further, these spectra may vary from point to point in the shock. In ad- dition, return electrons and protons each generate waves in the oncoming solar wind by their interaction with it, and these waves are carried down- stream with the plasma at doppler-shifted frequencies. The most prominent of these waves appear to be caused by the protons with p « 2. They are detected by magnetometers as quasi-sinusoidal oscillations of amplitude

831 19 k7 about B_/4 superposed on the IMF ' ' and are the principal phenomenon by which the foreshock has been studied and characterized. These waves consti- tute what may be cal led the ULF foreshock. The varieties of particle energies and wave frequencies that have been detected outside the bow shock in the cislunar region have been enumerated kk In a recent summary review . Here, we shall review newer results and des- cribe the current outstanding issues involving the foreshock. Two properties of the foreshock have been the subject of five recent studies. These properties are the variability of p and the interaction of foreshock constituents with the solar wind. A very curious result of early investigation of upstream ULF waves and quasi-parallel structure was the reproducibi 1 ity of the q-perpend icular to q-parallel transformation at the tangent point of the ULF foreshock boundary when the value p » 1.6 was used to compute the angle of the boundary 9X_ • k\ 47 arctan [p sin 9vD/(p cos 9VQ - 1)] * . However, most early observations AD AD were confined to the subsolar region of the appropriate B-X cross section. A dependence of p on position away from the subsolar point has emerged in a statistical study by Diodato et al . that covered most of the upstream region. Diodato et al. found that the overall best value was p » 2 on the sunward (of the earth) side of the shock and that p increased with distance from the subsolar point. In application the Diodato £t_ a_i_. result means that when the passing orientation of B.u places the changeover point from local q-perpend icular to q-parallel structure toward either flank of the shock, the ULF foreshock

832 boundary, which is tangent to the shock at the changeover point, is a little more forward (i.e., GXF is a little less) than it would have been were p constant. Equivalently, the backstream ing protons assumed responsible for the forwardness t ULF waves are not displaced as much from B when they originate near the flank as they are when they originate near the subsolar point. The most straightforward physical inferences are that either a larger proprotion of return particle energy appears as Vi, or return par- ticles are more energetic along the flanks, or both. Figure 5 (a) shows the crude dependence of 9VI- on 0V_ implied by the Diodato et al . study. The Ar AD broken and overlapping nature of the graph results from their study's division of data according to discrete ranges of 9X_ (defined In Figure 5(b)). The second property of the fpreshock that has received experimental attention is its influence on the solar wind. Unfortunately the newest results are in conflict with each other. If return particles interact with the plasma to produce waves, as theory and observation suggest ' "'•"'-3 »* ' then there ought to be some discernible effect of the interaction on the particles themselves. In fact, we know that return electrons change the heat flux in the electron foreshock, reversing its direction and elevating 2k T i/T I slightly by about nine percent . But what of the protons? A study by Feldman et al. set out to examine the thermal properties of the protons in the proton foreshock and discovered no temperature effect but a slightly lower thermal anisotropy. The data used for the study, however, did not include a magnetometer or return proton detector, so the direction of the electron heat flux was used as a guide to the ambient regime applicable to each data point. Since this criterion identifies only

833 the electron foreshock, at most an unknown fraction of the observations came from the proton foreshock as well. What is significant is that an anisotropy distinction was found even though the proton foreshock was mis- identified in some cases. Presumably the effect would be more pronounced in an undiluted sample. 6 28 Auer et a_l. and Fbrmisano and Amata undertook statistical analyses of the solar wind proton properties associated with the ULF foreshock and reported what appeared to be distinguishable differences between N, V, and B2/N2 in the foreshock and in the unperturbed solar wind. However, a T| very extensive statistical study by Diodato and Moreno involving almost 11,000 hourly averages of data from four satellites, using both single and paired spacecraft observations, found no significant differences between the foreshock and the undisturbed solar wind in speed, proton density, or proton temperature. Moreover, Diodato and Moreno demonstrated that an automatic biasing effect of variations in N on shock location could explain the apparent differences reported earlier. For the time being the composition, behavior, and influence of the foreshock's various components remain open subjects of inquiry. In addition to the obvious need for further experimental results along the same lines as those already described, requirements for numerous other measurements and analyses can easily be generated with a few moment's thought. Most impor- tant is the necessity for direct measurements of return particle characteris- tics throughout the foreshock and for their correlation with local wave observations.

834 Besides detailing the upstream plasma interaction processes themselves, future investigation must discover the sources of return particles, their distributions, and the forces governing their creation. We have glibly referred to return particles in the foreshock after touching briefly at the beginning of this section on the role played by the "foot" of the shock in supplying them. But the foot accompanies quasi-perpendicular structure, whereas the foreshock, or foreshocks, exist upstream only from the quasi- parallel structure. Where is the foot, or what takes its place, in the q-parallel structure with its multiple gradients, irregular profile, and quasi-periodic oscillations? Neither theory nor experiment, laboratory or extraterrestrial, has yet supplied answers to this question. Finally, the "steady state" question has yet to be explored even to the extent such exploration is possible fn the constantly varying solar wind. The character of quasi-parallel shock pulsations appears super- ficially to be consistent with that of large-amplitude whistler-like waves, and we know a large portion of the whistler spectrum can propagate outward from the shock along B-... There is no reason at present to doubt that under stationary solar wind conditions a very extensive whistler foreshock should be evident where 6 . is small, as predicted by Kellogg . Further measure- no ment and analysis should succeed in testing this picture at least to a satisfactory steady state approximation. Magnetosheath The entire region between a detached bow shock and the obstacle It shields from a surrounding supersonic flow is not ordinarily thought of when referring to the post shock fluid. In the case of a magnetospheric

835 bow shock, however, such an extension of terms is unavoidable. 0nly in that section of a magnetospheric bow shock where quasi-perpendicular structure prevails can the traditional concept of average fluid properties just behind a shock "front" be maintained as a basis for describing the post shock "jump conditions." Elsewhere, the downstream "turbulence" appears to occur on an amplitude scale comparable to or greater than the nominal jump itself and on a spatial scale that may extend the entire distance to the magnetopause. Speculations on the extent to which the magneto- sheath, and even the magnetosphere, participates in quasi-parallel shock k2 structure have been published before and are being actively investigated 48 with available data resources , so we avoid repetition here and concentrate on issues related to shock mechanics. • From a plasma-physical standpoint, the downstream regime is of inte- rest for several reasons. If we think first in terms of isolating q-perpen- dicular and q-parallel cases locally, there are two reasons immediately: 1. The magnetosheath supplies the post-jump conditions needed to characterize the q-perpendicular shock. Moreover, the conditions must be catalogued for a wide range of each upstream parameter, and there are numerous parameters. 2. The definitions of "preshock" and "post shock" are parameter- dependent in individual cases. For example, the magnetic ramp, or gradient, I.e., the "jump," is not strictly superposed on the density ramp. Especially in the cases of extreme parameter values such as very high betas or Mach numbers, a delayed or expanded particle redistribution mechanism may re- quire appreciable post-shock data collection to deduce the physical processes

836 making up the shock profile. The electrons may be principally responsible for one shock subprofile and the protons for another. One model of proton heating at supercritical Mach numbers, for example , does not call for any prominent "viscous" dissipation instability in the ramp, so that plasma wavemode identification would be fruitless and only substantial particle observation well behind the principal magnetic ramp would allow verifi- cation of this model. In the case of q-parallel structures, it appears that the magnetosheath j_s_ the shock to some extent. We simply do not know at present just where the downstream, or post-shock regime begins. If we expand our inquiry beyond isolation of separate cases, we find two more penetrating reasons why magnetosheaths are interesting to plasma- shock investigators: 3. Col 1isionless shocks in the solar wind are three-dimensional, with magnetic field lines connecting one region of the nominal surface, through the post shock standoff region, to the other "side." If one side Is q-perpendicular in a uniform field, the other side is q-parallel. Hence there is the possibility of communication by field-aligned phenomena pro- pagating from one part of the nonuniform shock to another through the downstream region. The three-dimensional shock favored by nature is an entity by itself with a continuity of parameter changes differentiating it from the two-dimensional cases that might be found locally, reproduced in the laboratory, or approximated in theory. 4. Finally, the most exciting reason of all, is the opportunity to observe transient plasma processes at high resolution. The magnetosphere is the source of phenomena that communicate the presence of an obstacle -29-

837 to the solar wind. The magnetosheath therefore is the location in which the processes originate that continuously form and reform the various shock structures in response to changing solar wind conditions. The dimensions of a typical magnetosheath, being on the order of many planetary radii, transform ephemeral phenomena that might take microseconds or nanoseconds in the laboratory into finite events that can be recorded over intervals of seconds and minutes, with almost arbitrary precision. The reader will have perceived that the straightforward summary of physical results in the preceding sections has been replaced here by an undocumented discussion of justifications for obtaining physical results. This is because spacecraft data have not yet yielded generalizable results of plasma physical application except in a few instances. Results close behind the shock front have been described already in Figures 1 and 3 and Lfc in an earlier review ' . Deeper in the earth's magnetosheath, the most pertinent results are, first, that ion acoustic wave noise continues to modify the solar wind electric wave spectrum well behind the quasi-perpen- 81 dicular magnetic ramp , and, second, that suprathermal protons of energy > 100 keV appear in sharp correlation with large amplitude magnetic field 102 oscillations similar to those of q-parallel pulsations . Improvements in instrumentation and expanded data analysis will be necessary to catalogue magnetosheath properties and relate them to the bow shock.

838 PLANETARY SHOCKS Four planets other than the earth have been found by direct measurements with spacecraft instruments to have identifiable bow shocks. These are Mercury, Venus, Mars, and Jupiter. In addition, there is evidence for a limb shock associated with the moon's wake. Four results regarding planetary shocks are of plasma-physical signi- ficance. 1. The overall classification of shock structures into quasi-perpendi- cular and quasi-parallel categories is generalizable to shocks at least as widely spaced in the solar system as Mercury and Mars and apparently as far as Jupiter^ too. The local characteristics of shocks of both categories, with regard to magnetic field and electron behavior, are familiar from the earth's bow shock. 2. The relatively swift passage of the respective mission spacecraft through their target magnetospheres has provided bilateral views of nonuni- form bow shocks difficult to obtain at earth even with multiple spacecraft observations. Thus the coexistence of q-perpendicular and q-parallel structures 67 68 on opposite faces of the nominal shock has been demonstrated at Mercury ' , JiO kQ Venus , and Mars . The data from Mercury are particularly striking be- cause the shock profile during the third pass of Mariner 10 was virtually a mirror image of the pass two years earlier, with the macrostructures en- countered in reverse order because the interplanetary field orientation during the third pass was complementary to the orientation during the first pass 67,«>,77772

839 3. Each pair of bilateral shock views just cited has, of course, been part of a one-dimensional cross section of a corresponding magnetosphere- solar wind interaction region. These planetary cross sections have there- fore provided pictures of the extensive downstream regions of their res- pective shocks. A prominent, repeated feature has been the penetration of the q-parallel magnetic pulsations deep into their post shock magnetosheaths. Thus these planetary passes have served as principal sources for the in- ference that the q-parallel shock and its downstream sheath are essentially inseparable, and that pulsation macrostructure stretches all the way from the magnetopause to some, as yet undefined, outer pulsation boundary. Mag- netosheath involvement in q-parallel oscillatory structure has yet to be systematically studied in the earth's interaction region. 4. The fourth result from probes to other bodies is that the solar system offers some unusual cases of special interest not to be found in the earth's vicinity. One is the moon, which having no intrinsic field of di- mension comparable to itself, but possessing small local conductive or mag- netic anomalies, offers the opportunity to examine marginal col 1isionless 91 shock formation and maintenance near the scale limits required by theory A second case is Venus, where the obstacle in the solar wind's path is neither a strong magnetosphere nor the absorptive surface of the planet itself (or its atmosphere), but an induced magnetic field arising from cur- rents in the ionosphere. Here, there is a possibility that the solar wind is partially absorbed in the upper atmosphere and that the shock may be so close to the ionosphere that the latter may be part of the downstream region, and shock effects could play a role in the planet's meteorological energy budget. Certainly Venus presents the prospect of displaying some unique 8k plasma-shock phenomenology

840 A third case is Jupiter, whose magnetosphere may be so rotationally asymmetric, so tilted, and so rapidly revolving as to create wholly un- familiar nonunifortuities and asymmetries covering new areas of parameter 26 space not seen elsewhere . Little data of significance to plasma shock study has yet been released, however. There is only one example of detailed study of an alien bow shock system from a plasma physical viewpoint. Fairfield and Behannon analyzed magnetic field data from Mariner 10 and found whistler waves propagating upstream from the shock as observed previously at the earth . They went further, however, and identified some of the waves in the Hermean magnetosheath as ion cyclo- tron waves, a result yet to be achieved at the earth. ASTROGENIC SHOCKS All shock phenomenology described so far pertains to shocks formed by supermagnetoacoustic flow past large, effectively stationary obstacles. The other great class of shocks on space pertains to those formed in front of objects moving at supermagnetoacoustic velocity through relatively slow, although nonstationary, plasma. The "objects" contemplated in this class are not solid bodies or their magnetospheres, but masses of plasma of suf- ficient density and conductivity to appear impenetrable to the background hydromagnetIc gas through which they proceed. We shall call such an object a "driver plasma," or "driver gas," and its shock a "driven shock," although we intend to include In this category the "blast shock" which is very much detached from its driving source.

841 Driven shocks differ from planetary shocks in two major ways; scale and speed. The sources of driven shocks are activity centers on the sun and, by implication, on other stars, and stars themselves. Should these pulse or explode or, In the case of binaries, revolve at high enough orbital velocities, shocks would presumably be generated in the interstellar medium. The scale size of such shocks is therefore of heliospheric or stellar, rather * than magnetospheric, proportions. This means that wave-particle effects that may depend on tangential motion along a shock front have an immensely enlarged arena in which to operate. Thus, multiple charged particle ac- celeration associated with a shock front could result in particles develop- ing cosmic ray energies before being released by local alterations in field or shock geometry to escape into the general interstellar environment. The speed of a driven shock will depend on the speed of the plasma piston that drives it. The latter varies depending on strength of the source and distance from the source, whereas the flow speed of the solar wind is essentially constant with distance from the sun after an initial acceleration in the outer corona. Although the absolute value of solar wind velocity changes as the source region changes, the nominal speed is a few hundred km/sec and usually varies by no more than a factor of about two to one. Piston plasmas and their shocks, in contrast, may race through the solar atmosphere or outward from the sun as fast as a few thousand km/sec, slowing appreciably as they progress outward through the solar system. Since shocks driven from solar flares or by high-speed streams have a wide range of

842 velocities, they may include both slow shocks and shocks of extreme Mach number which makes their observation worthwhile, if dIfficult, at customary data rates. 0nce again, however, particle acceleration is involved; to the extent that energy exchange between a shock and an individual ion contributes to the ion's gain in energy, driven shocks have the potential of accelerating ions to much higher energies than do planetary shocks. The property of accelerating ions gives the subject of driven shocks a value for those concerned with high energy cosmic particles, because there is ample reason to believe that col 1isionless MHD shocks are commonplace in the astrophysical environment. The most primitive example would be the shock expected to exist in front of the heliosphere as a result of its CO motion in the galactic plane^ . Presumably such shocks can be attributed to the myriad of stars later than F5 where the existence of stellar winds is 79 inferred . A more sophisticated example has been envisioned by Siscoe and 92 Heinemann , who have proposed that binary stellar winds might include oppos- ing shocks bracketing the contact discontinuity between appropriately spaced star-pairs. Finally, the velocities of supernova ejecta and remnants may be as high as several thousands of km/sec and are quite likely to drive high Mach number shocks ahead. of them Into the interstellar region . If such examples are multiplied, there are numberless opportunities for particles to be accelerated throughout the universe by shock waves of many origins. Study of the energization mechanism taking place in the solar system should have direct application to such astrophysical sources.

843 Locally, we know that accelerated particles are associated with the bow shock, but neither their origin nor the mechanism of the?r~production has been established. The subject has been discussed separately for elec- 23 59 trons and for protons . More to the point, there has been both analysis and observation of high energy ions in connection with interplanetary shocks, 4 86 87 and this subject is currently under investigation °'0' . it has yet to 0 be learned whether microscopic plasma physical processes play a direct role in the energization of particles by col 1isionless shocks or whether they are involved only indirectly in formation of the shocks themselves. If microinstabi1ities are directly responsible, then the prospect opens that not only shocks, but any plasma interaction region where these phenomena can occur will become more accessible to analysis based on spacecraft in- vestigations.

844 DISCUSSI0N Recent Status of Theory. The experimentally-observed presence of electro- static turbulence which clearly correlates with the more macroscopic MHD structures in the bow shock has resulted in significant theoretical activity, especially in searches for electrostatic microinstabi 1 i ties that could be driven by currents of one kind or another. The outstanding unanswered theoretical question concerning dissipative mechanisms in col 1 isionless shocks is simply: what randomizes the protons across the shock? At first, it was thought that diamagnetic electron drift currents setting up a jump in field AB over a shock thickness AS would produce Langmuir, Buneman or ion-acoustic turbulence, and hence anomalous resistivity, and that ion-acoustic turbulence would cause the observed ion heating. How- ever, quasi linear theory predicted only electron heating in such turbulence, and even though the formula has been successful at estimating magnetic ramp, i.e., drift region, thicknesses * , doubt has been cast upon this as the main dissipative mechanism (ion randomizer). Furthermore, observations of 0GO-5 data on bow shock crossings re- vealed shock structures in which electrostatic turbulence appeared in many magnetic gradients, without randomization of the proton flow. A summary of all the known attempts to resolve this theoretical question by postulation of electrostatic or electromagnetic instabilities and conse- quent turbulence up to 1973 was given by Greenstadt and Fredricks . Their

845 conclusion was that electron heating in bow shock precursors by electrostatic microinstabi1ities is qualitatively rather well understood, while the ion heating mechanism is not. The most reasonable conjecture they implied was that ion-ion sound instabilities are responsible for the observed rapid ion heating across most bow shocks. This latter conjecture has been put on a stronger basis in work by 38 Galeev3 , who reviews again the possible candidates. Galeev proposes that counterstreaming ions produce turbulence due to instabilities with frequen- cies near the lower hybrid resonance. However, he points out that whether this mechanism is the correct one depends on verification by direct measure- ment on future spacecraft missions. Furthermore, as amply documented in earlier sections of this review, there are obvious differences in the microscopic particle dynamics associated with the several morphological categories (perpendicular, quasi-perpendicular, quasi-etc.) of bow shock structures, so that exact identification of the microinstabi1ity responsible for the dissipative process in each case is not an easy task. It should be pointed out that measurements of lower hybrid resonance turbulence in the earth's bow shock structures have not been made in the past due to lack of Instrumentation to cover the extremely low frequency electro- static spectrum (f.u_ - 10-20 Hz). It is hoped that this deficiency will LnK be removed by proper choice of instrumentation on such missions as the ISEE spacecraft in the future. 6k Morse has suggested an alternative model for proton heating in the earth's bow shock. He starts from the premise that plasma fluctuations

846 within the bow shock structure of sufficient amplitude have not yet been ob- served. Morse then puts forth the hypothesis that the motion of ions in the macroscopic field fluctuations in the shock and its downstream region is sufficient to explain the observed broadening (and decrease of the average velocity) of the proton velocity distribution behind bow shocks, without introducing subshocks or wave instabilities as ion heating mecha- nisms. This idea is reminiscent of the ion "orbit-crossing" mechanism for ion randomization discussed by Auer et aj. and Morawetz ' Shock Parameters in the Solar System. There need be little doubt that solar system plasma offers an arena in which shock investigations can proceed and theoretical estimates can be tested experimentally. The values of the principal upstream plasma parameters in which shocks may be found in the solar system between Mercury and Jupiter is displayed in Figure 6. In 6(a), parameters B, Cu_, and Muc and the nominal quasi-perpendicular unit of thick- rib ns ness c/oi . are shown vs solar distance in astronomical units (AU) for a typi- pI — cal solar wind speed of 400 km/sec, taken as constant at all distances in the solar equator. The parameters have been computed using the relations N » 7/r2, T » 7 x l0Vr, T » 1.5 x loVr1'5, and the Parker model for the p e spherical field components: B » 5^2/2r2, B, » 5^2/2r, BQ » 0. The numbers typify the solar wind at 1 AU. Although some question arose for a time spendence^of ,9,54,70,75 12 8^ about the radial dependence of B * , all the dependences used here have - now been documented' The long, solid and dotted curves at top and bottom of 6(a) represent parameters of the solar wind independent of whether any shocks exist there

847 or not. The Mach numbers, in contrast, depend on some assumption regarding shock and solar wind velocities. The long dashed curve shows the magneto- sonic Mach number that would apply to a planetary, or cometary, shock sta- tionary in a 400 Km/sec solar wind at any r, but the circled points mark only the M-values at the permanent planets, identified by their initials just above the horizontal coordinate line. The Q-turbulent/turbulent designation under the line denotes the category of planetary shock struc- ture defined by the M _ and B combinations in the corresponding regions. Ho The two shorter curves illustrate the Mach numbers of two solar flare shocks propagating into the assumed 400 Km/sec wind. They terminate at the lower end, of course, where MM§ « 1. The declining velocity profiles from which the curves were computed were taken from two examples (Figures 15,16) 18 of Dryer . Such shocks would be turbulent above MM- y 3, quasi-turbulent below. • The dotted curve in 6 (a) accounts for the third source of shocks in the solar wind, namely, stream-stream interaction. Fast streams overtaking slower ones in the solar wind can evolve into compressed structures contain- ing sufficient density to serve as "bodies" coursing through the wind, capable of generating shock pairs, one forward and one reverse. Such fast- 74 33 53 96 95 stream shocks have been predicted , inferred " ' t and demonstrated The velocity differential between the fast plasma behind a stream's steep- ening front and the slower plasma ahead of it falls in the range 40 to 120 Km/sec and averages about 60 Km/sec , so the behavior of the magneto- sonic velocity CM- implies that formation of a forward shock where AV > C... (i.e., MM_ > 1) is marginal for a 400 Km/sec background wind, but generally

848 most probable where C drops below 60 Km/sec, i.e., where r > 1.2 AU. Clearly, the Mach numbers of such shocks are unlikely to be supercritical, so the classification of almost all of them would be laminar out to large r-ciistances, where B « 1 would make them quasi-laminar, or quasi-turbulent nearer 1 AU where 8-1. They would be expected to be well defined and of relatively simple structure when observed. Figure 6(b) plots two more parameters determined by the IMF: its longitude angle <J>B - arctan (B./B ), for the sense away from the sun, and f 43 foreshock boundary angle 9xp arctan p sin 9vg/(P c°s 9..- ~ 1) , as defined in Figure 5(b). The 9XF curve in 6(b) is dashed, since it really applies only at the planets, indicated by the circled points. The plotted values were obtained by using p « 2 for the entire range of r . Note that for purposes of illustration, the sense of B has been reversed in 6(c), so 9XB " 4>XB + 180°. The importance of 9X_ in 6(b) lies in its implications regarding the extent and location of the average foreshock at each planet. For the first three planets, the foreshock literally reaches out In front of its bowshock, since its forward boundary makes an acute angle with the X-axis. At Mars and Jupiter, however, and by inference, beyond them, the foreshocks do not reach sunward of the subsolar points of their bow shocks, since Q^ < 90°, but they may occupy a considerable region to the sides of the shock flanks. The trend in 9 _ also implies that, given the usual fluctuations of the IMF direction, there Is a fair probability that the foreshocks of the inner planets will occupy their entire upstream regions a substantial fraction of the time, i.e., when 9..- slews toward 0°. Equivalently, the sunward face of Mercury's shock should be expected to exhibit quasi- parallel structure frequently.

849 One paramount feature of Figure 6 has been saved for final emphasis. At the right of 6(a) and (b), three vertical flags represent the ranges of 6, Mur, and <J>D that have been recorded by spacecraft at 1 AU. According to MS B these, the bow shock system of the earth alone can occur in any combination of parameters, thus providing a rich table from which to select samples for the study of shock structure, both in transient and steady-state conditions. In reality, not all combinations of the structural parameters are equally probable or equally observed. High 6, low M (quasi-turbulent) conditions are unusual for the earth's shock at 1 AU and would presumably be much more probable at Mercury. Laminar cases, on the other hand, are considerably more common than the curves in 6(a) would suggest. On balance, detailed study of the earth's shock system should, with patience, produce a full display of the structural panoply found among the planets. A Contemporary view of the multilocal approach to observations by spacecraft. The constellation of sensors needed for diligent examination of plasma shock systems in space is suggested by the juxtaposed data plots of Figures 7 and 8. The figures show an overall view, at low resolution, of a sequence of encounters with interplanetary and bow shocks during two days of re- peated crossings by both IMP 7 and 8 on opposite sides of the magnetosphere. We call attention to certain highlights of these plots that illustrate major features of the most significant events. The formats of both figures are identical. The four upper panels are from IMP-8, the five lower from IMP-7. The IMP-8 panels contain, beginning

850 at the top, one channel of the Iowa electric wave data, the ambient magnetic field magnitude B, and the ambient field's longitude <£. and latitude A D O in solar ecliptic coordinates. Field measurements were made by the Goddard instrument. The inserts in the B-panel show the relative positions in the ecliptic of the satellites and the field's projected orientation, together with nominal magnetopause and shock cross sections,during the time intervals where the inserts are placed. Details will be described below. The IMP-7 panels, continuing toward the bottom, contain views in four sectors, in the ecliptic, of low energy protons detected by Iowa's LEPEDEA experiment. The apex of each caret at left indicates the direction in which protons must be flowing to be detected in the corresponding sector. These directions coincide, counterclockwise, wit-h the axes of the inserts in the B panel. The last panel at the bottom represents the average energy density from a wide channel of the TRW electric wave experiment on IMP-7. Note that the narrower Iowa channel on IMP-8, top panel, falls at about the center of the TRW channel. Both are chosen as monitors of local ion acoustic noise, which predominates in and around the bow shock. Descriptions of the various instruments can be found in references ^* ' Our principal Interest in Figure 7 is the foreshock, but we note the interplanetary shock of 1530, whose jump in B is accompanied by a sudden increase in average V_w, a spread (thermalization) of its distribution (uppermost LEPEDEA panel), and enhanced plasma wave noise at both space- craft. The triangle at bottom denotes the sudden commencement of a mag- netic storm at the earth's surface.

851 The apparent discontinuity in B at left, around 0115, is of unidenti- fied character and will not be discussed. What Figure 7 shows very clearly is the control of the foreshock by the IMF orientation in three distinct intervals. The key factor was <j)_, D since XD remained for the most part within 30° of the ecliptic. B First interval. Early in the day, until 0800, the IMF projection was directed from sunward to slightly west of the sun-earth line, i.e., 4>0 =: 330-3600, as depicted by the small arrows in the first insert. The B wavy line F in the same insert is the forwardmost proton-foreshock boundary that applied during this early interval. IMP-8 was obviously in the fore- shock then, and we see that electric wave noise was enhanced (top panel) and that B oscillated as if upstream waves were present. Second interval. At about 0800 the IMF rotated into the first quadrant, where it remained for several hours with <J>_ between 0 and 45° (second insert) B and then gradually advanced past 90°. The result at 0800 was to shut IMP-8 out of, and place IMP-7 within, the proton foreshock. We see that at IMP-8 the enhanced electric noise subsided and B became very quiet, while at IMP-7, return protons made their appearance. These are represented by the dark green traces between 103 and 10I* eV in the second proton panel, indi- cating particles of those energies coming to IMP-7 from the shock to the right of it in the second insert. Shortly before 1600 UT, and slightly after the interplanetary shock of 1530, when <(>_ had increased to about 70° (and there was also a swing away from the ecliptic in A ), as pictured by B

852 the dotted arrow In the second insert, the foreshock boundary (dotted wavy line) passed behind IMP-? and the return protons disappeared. They re- appeared briefly at about 1730 coincident with a slight shift for a few minutes in IMF direction. Third interval. After about 17*»5 electric wave noise became enhanced "again at IMP-8 and remained so until nearly the end of the day. The field- • shock geometry for this -interval is shown in the third insert, where it is clearly seen that IMP-8 was outside the expected foreshock boundary. In this case the field projection was actually opposite the small arrows, which have been drawn for visual clarity and to emphasize the direction of possible particle velocities toward the satellite. It cannot now be stated with cer- tainty whether the enlarged E-fleld amplitudes at IMP-8 were related to the bow shock or to changed interplanetary conditions, but we observe that the wavy foreshock boundary refers only to the region containing slow pro- tons with UH s 2 V_w and their associated ULF upstream waves. Faster pro- tons, and electrons, could have been streaming more closely along B to IMP-8 from the bow shock, causing the local VLF wave noise. This possibility is favored by what we know of return electron behavior and by noting that proton velocities corresponding to the high energy end (10* eV) of the green return proton traces earlier at IMP-7 were adequate to convey ions to IMP-8 outside the depicted foreshock boundary in the third insert. This possibility Is also compatible with the electric wave record at IMP-7, in the bottom panel. The noise level never seems to have dropped to a sus- tained background, or quiet, level during the 4th, even though we know IMP-7 was sometimes outside the (slow proton) foreshock boundary, as drawn in the

853 first insert. However, even early in the day, electrons and high-speed protons could have been reaching IMP-7 from the shock. This is the meaning of the dashed lines in the first insert, and !MP-7's location may have kept It on IMF lines intersecting the shock the entire day. It remains only to note the short entry and exit of the bow shock itself by "lMP-7 just before 2300, as determined from the thermalization and deflection of protons and the sharp peak in plasma wave noise. We proceed to July 5th, in Figure 8. For this day, we cannot discuss the foreshock within the scope of the present report because the frequent, wide excursions of X from 0° rendered ecliptic representations wholly B inadequate. 0f greater interest is the sequence of shock encounters which illustrate the wealth of diverse events and conditions that can occur in a single day in space. At the outset, IMP-7 encountered the bow shock and was enclosed by the magnetosheath for about 30 minutes between 0100 and 0130. The high B that prevailed at IMP-8 at the start of the day suggests that there was a low Mach number in the solar wind that brought the bow shock out to IMP-7, for when the field dropped later, IMP-7 remained in the solar wind several hours until it saw the shock closer to its nominal position at 1315. The insert in the IMP-8 B-panel shows the wide range of <J»a during the B hour and a half or so before the 1315 crossing. 0f special interest is the swing of the IMF to the fourth quadrant at about 1245, the result of which was to make B essentially tangent to the bow shock in time to determine a nearly perpendicular crossing geometry at 1315- After that, X_ was close B

854 enough to -90° to keep the shock geometry nearly perpendicular while IMP-7 was in the sheath behind it until about 1615 when the solar wind underwent a sudden change, best indicated in the figure by the discontinuity in B and Xg. IMP-7 emerged gradually from what appears to have been a quasi- parallel shock and then, at 1715, began a two-hour series of irregular and noisy shock measurements the nature of which pur information is insufficient to define. It is clear, however, that the local magnetic geometry was quasi- parallel, and we see that (a) the average velocity of the solar wind was little changed, (b) thermalization was taking place but less Intensely than behind the 1315 crossings, and (c) higher energy particles, up to about 10* eV were being deflected into all sectors; this did not take place behind the 1315 crossing. At 1930 a sudden-commencement interplanetary shock reached the vicinity of the earth and drove the bow shock inward, leaving IMP-7 again In the solar wind. The satellite reentered the bow shock at 2210 under conditions varying between quasi-perpendicular and quasi-parallei and remained in the magnetosheath through the end of the 5th. The IMP-8 data of 5 July featured electric wave activity which was low, with some sporadic enhancement through about 1630, and then increased for the remainder of the day. The cause of the noise amplification cannot be unambiguously identified here. It Is always necessary to be cautious about differentiating between Interplanetary and foreshock effects. In this case, a very careful geometric analysis of the possible connecting field geometry to the bow shock would be necessary to determine the plausibility of return

855 electrons or high-speed protons, as factors in producing the amplified wave noise, so we cannot state now whether the plasma wave pattern on the 5th Is a continuation of the same processes that appeared to be operating on the kth. The content and description of Figures 7 and 8 illustrate that two satellites and four instruments are far from superfluous in reconstructing a sequence of physical processes surrounding the bow shock. Indeed, the absences of IMF data at IMP-7 and LEPEDEA data at IMP-8 from the figures are keenly felt. The former were unavailable because the instrument had previously ceased operating; the latter was available but had not yet been obtained for this report. Either would have served to clarify some of the event identifications. From another point of view, however, the figures demonstrate how comprehensive instrumentation permits analysis to proceed, albeit cautiously, even though nominally essential measurements are.missing. The examples of Figures 7 and 8 also show the way in which space plasma behavior can be examined on a vast scale with devices of comparatively in- finitesimal dimensions and negligible influence on the environment they sample. Finally, the figures expose the data techniques of the present and command the methodology of the future. The four plasma panels, for example, were selected from a still wider display of the solar wind's properties developed at Iowa to show the properties of the multidimensional solar wind in metric and velocity space in a manner rapidly understandable to the data analyst. The verbal description of the events of k-5 July, how-

856 ever, underscores the upcoming need for equally routine display of the re- lationship among spacecraft locus, IMF orientation, and bow shock geometry in order to define the delicate control of the foreshock's components and the bow shock's structure by the IMF. RECOMMENDATIONS Future study of col 1 is Ion less shock phenomena will be greatly aided by spacecraft programs designed for both geocentric and interplanetary orbits, if appropriate instrumentation is carried. In addition to conventional field and particle devices, high resolution, omnidirectional plasma detectors will be of great importance. The earth's bow shock will be the principal source of new, detailed measurement covering almost the whole range of solar wind parameter combina- tions. Details will be needed in both spatial and temporal dimensions. Table 2 displays the major categories of shock phenomena, the requirements for their investigation, the sources for meeting these requirements, and the status of the sources at present. Multiple listings under a single let- ter designation mean that coordinated, grouped requirements must be satis- fied as a unit. The table indicates that, for the most part, suitable space- craft instrumentation has been or is scheduled to be available, but that the status of software support and non-mission activities is in general under- • supported. The most important new recommendation outside the table is that a cluster of four closely-spaced vehicles be planned, giving three-dimensional spatial resolution with gyroradial-order separation, omnidirectional plasma

857 particle detection capability, high sampling rates, and an accompanying solar wind monitor in the sunward plasma, far upstream. Much of future progress, however, is rooted in the past. Figures 7 and 8 of the foregoing section illustrate, scandalously, the first published example of double measurements in the foreshock with the instrumentation shown. A commitment is urgently needed to recapture the investment in plasma physics lying idly in data warehouses throughout the nation. We recommend such a commitment. Further, it is of great importance that theoretical modeling of shock phenomena, most of which is years old, be encouraged to catch up with the rich store of observational information already published, let alone awaiting disclosure in data freightyards and tape libraries. 0ne of the most promising avenues is numerical simulation, and we recommend dedir cation of at least part of the effort of one or more large computer facili- ties to the digital-modeling of shocks in space. Finally, we note that comprehension of physical phenomena are best completed when we can reproduce and manipulate them in the laboratory, and we recommend wholeheartedly that promising experimental work with tenuous, if not col 1isionless, plasma shocks in laboratory apparatus be extended as far as possible into the parameter domains found in space. ACKN0WLEDGMENTS We express with pleasure the help of K. L. Ackerson, K. W. Behannon, R. A. Chevallier, L. A. Frank, A. A. Galeev, J. R. Jokipii, R. P. Lepping, C. T. Russell, and F. L. Scarf in supplying helpful discussion and/or advance copies of unpublished data and manuscripts used to prepare this report.

858 in ^^ • PO tTi *-^- U. -a- rO 4) i -3- ee 1 OJ D -C C O) I _J u I 0> u 4-i O •O 3 •o 0 c — k. I C u 4-i — 4J I ^_J LU o -o 4) U O I0 m 2 O E O — w - O) ml — 0 a. O> IB .— — -lj 4) ~ -in C —I 4-i — in E 4-i > •D O) O. i- 4) — C m C 0> ^ •— I— 0> ui •— J^ MB C C >4- C i4- Ol 4) O I0 QC in 4) c i4- ID in 4-i ui C Ol IB > •— E <c c a . O O — C — C 2 ^^ ^ O) '"0 k. E I0 4V o- 10 I t- a. I0 2 4) I0 — c E 10 23 c t \- •>• 4-i m k. 4) — — — 4) -Q ••• in c c I0 JZ — U U — 4) 4) O 3 Ol E — Ol k. I0 k. 4) 10 4) 4J I/I k. JZ 4-i 5 4> 2 O O) 2 i*- a =s •o I0 •a o tn £ X >4- -o O) Oi — 4) 4-i C ^B * •» k . — k. in 4J in 4) C *2 4j" c 4) J= in C ui E ^ *^ • 1_ 4-i — -o 4) — 4J 4) O) — JZ I/) ^ it— •' C 3 • i4- in "- i 2 mm •— 3 I0 4-i a c 3 •o m 4i in UJ— • * c in • 0. w 2 E IB 4»i 4) 4) c - c 4-i ae 3 4) >• O c k . in E .^ c -a • > o -o o c a . Ol 4-i C O I0 E . C - m 4) c in I0 in 4) 4) ^ in I0 a. O — 4) cs X _ in lu — 4) C 2 E — 4J — 4! -* I0 U Ol 4J O O 4) — O C — 4) IB •O J3 — O >- 3 •a — o a. 2 I0 E Q E o 4) -a 10 a.— 10 in 3 O u — — .a § I0 4J «^ X • k. 4i k. O) 4J — I- JZ k . — in c cn 4-> O) 4) — in 4) > — i— 4) 4-i o cn in k. 2 E 4-i k . — JZ 4) u •— I/I 4-i — 3 4-i J— I0 C — c 3 O v O) I0 4) 4J 4-I > 4-J "O "Q Ul 4J Q. k. 4_i 0> in in 4V 4) _i 4-i I0 U C — — I0 in O c — — 4J £ o 4J C 4) — — JZ — k. x Ji 3 JZ 1 322 * 3 4J 4) C -Q in U IB 3 X _l X Z> 0- m j^_ Q O PO rr\ cn •• -0 4) in P-I -3- -3- CM 00 4 > t OC .0 "k . _^ I0 4-i 0 o f^ 1 >. o — — ln 01 < i — 4) •o -^ >. - c I I i — 0. C ja k. (J — a . I0 3 X k . — 3 i- E 0 u 4) E — J2 I0 X in Q.O 4) in 104-i 10 U 3 — k. E • U 4J 10 J= 3 4V J3 I0 C t- O 4) C — \ — C 4) 4) in in 4) M. a . j=2 4) ^.i — 4) 2 4-i - O J3 — k. O 4-i a. ft .— kB 0) I — 4) — O "04J *J LU 2 3 a. x 4J » 3 l- X E -Q 3 I0 g - 4) in 4J in in a. C c in 4) "i — u- O 0. i in 4) O £ £ — IB k. 4^i £ 10 k. ja U — C E 0 2 O 3 4J O c •o 2 O 3 CO 4) 3 I0 C 4) E C * I— u 4-i k. O in 4) k . i 0 •o in t/l C *—"•' 4-* 4-i 4) C 4-i i u 4-i 4-J k. 4J — — U- O) 4-i - O CM U l in 4) a. Ul — 3 in I0 -o a. c 10 C Ol k. 4) Ml E 4) O > o n -o !«§ c I "io I0 -a E v. 3 — I0 4J « — — 0 — 10 — 3 Ol 1) Ol 3 M Oi in 4) O 4J 4-i O 4-i tn k. c o E • — IB ul I0 °"5.l •o — — •o m 4) E E 3 3 ««- -O — i4- 4) i 2 3 JK U JZ £ 3 — C 4J — X. — O U n- (B •— in ID 4) h- 4) O 3 I0 a. u — "V. >- a. - 4-i _a w a. 4) k. Ol— X 4) 1a k . o C 0 4) -1 4-i CN *-* ! « S 1 I0 §4-i 4-i -a IB 4-i m •a a. u 3 k . J U. 4J Q.U) — i > a. 4J I0 in 4-» in —l C C O 2 4) 4) 3 C — > E O — QUO. 4) JZ — 4) Q. — Ol k. « | ^J •a k . zt i Q •o 2 4) — Ol — o-§ U- I0 C i4- — CQ a . C U l — 4) >. a.-o — — -a k. jz c 3 £ 4> a . *V 3 C O) o -a _ _ ro — 4) 2 a . = k. 4) O • I0 4J in k. cn 4) o I0 >' C > in 0 "4- -o • 4> 4) > — I flj «t- •o — 10 o o -o 4) t*. • 10 4) 4 > — — — — C 3 C I0 O 3 *4 > k. in u 0> i 4) C 4) § c a. u C — 4) O c u — — 4) C 2 « i*- I0 E c IB U — I0 U k. E Ol I0 C 4J — O) «- O 4-> M 4) 18 O) — 3 4> 4) 4> a . — 4) 4) 4J 4) kk 4) E 3 k. k. O 4J <4- 3 • — -a — 4) 4-i 2 — a. O.-Q jZ k. •o — J3 k. 4J — X 4) J3 — U ce u < U — J — CD — LU 4) 4) ^y r- ^^ J3 O 3 ce. •C Ul mmm, 1 i < UJ UJ 00 4) U — ZJ 3 I E 3 •w V) ••• ui aa aa ^ I0 k. •y < x < m cc 0 Z 4-i 5 §5 P I I0 I0 i 1 U l E .8 in in )z I0 U in IO JZ in JZ I0 U s» I0 -C — u •a-x-k . — U — I0 — 4-i in 4J 10 — a.x « 4) a. IB •- X 4) a.x v- 4) JZ — I0 u E — -o •o -o E J3 E JZ J3 - O ul 0. C is J i o 0> e sil k. oi E 4-i -^ I0 o u 3* i O U — X > Q. w in 4) 4) — . A « ^B E 3 fr\ — PO *— f"> — p^ P*^ I0 — V A u I0 V VI V A' » V? l» A? A A I0 > « on ^ CH X CO. X OQ X CQ X

859 Ul § I I • CM V ^•^ i. ^ "« I0 I — 3 . k . l— UJ I0 4J Ul — o 3 Ul U _ in 4-I o x 0 a o — 0 OL — — X Q. 01 4»i 4I ^— . • M A Q Ol***"- •V C ^^ ^»^ ^-v X U •> O- l- T» C 3 • aa CD - ui j< 1o 0 3 • • • 4J ^ z tn c — k . 4J > 4f <£ <£ c 1o -1 I0 O U P— ^ Ul i UJ Ul 4) U in o ui < ja in p— i 0 a. as w UJ UJ Ul 4) Q. a. •- I0 — x a. k 3 in v> I/) k. in X 00 4J 4J 4) ^o 1o 44 •M — * — Q. — - I0 — C • •*•' •— •'• — - i0 *— < •o C 3 VI I/I (/i tfl X Z Z O. M X Ul C i c 0 •o 4) — o U3 yj •o *V u 4J Ul — ,_ . _ "^ 0 i4- a c X 4J o 1O I0 4J I0 a § p^ — I0 0 U E 4) 3 U U I0 U l 0 10 — C 3 10 J I0 C 4) •o 0 0) Q. 3 4J *J 0 (0 — J § k V|- • 3 •w 10 w Ul p- Ul IK O 4-i a. I0 Ul U 0 10 _Q 10 , _ 4) I/I Ul k . C U 4-> U 3 y — in d, '— c I0 — 0 — I0 U l — I0 X H9 a. 0 I U — a. OJ U X I0 — • 0 (jj 4) MB ^ X C Ul U l — • a 0 Ul 44 l- U o Q) 4~» i ouk.ce -a ui 1o O 4) X 4-' P— i4- 0 10 1a a wi ee. 11 1— Z m 4-i 4) 10 " •o *3 C J3 3 p- a i Is "3 "o •o k •- u to in 1/1 in I0 4-i JZ 0 10 i 10 -d "* C 4-i a Ul U k k C i C w I0 C 4-1 O o o u a en O 0 — — U 0 4J a. P- — . Ul i *• a. p- W D v o m g .^ Ul O O Ul I0 — 0 E »— 0 E p— 0 V. !„ 4) PM *~~" P1B 4| 1_ ui 0 > X 0 O Q I0 0 > 0 C I0 I- JZ f c Ul I0 JZ 0 S J 4J 4J I0 § s s t Ul — e 0 1o i_ e c 4-> ui x E k I0 as ui O Q. O 8 Id — Ul X 0 .2 0 w I0 ui a. > — +* 3 •—. 0> c 10 — 3 4V I- Q. (6 0 I/I C k 4V c o > a, *>» x ui 10 0 C g 0 0 4-i k . k . 4) _ a *• — k . -O 10 10 E (9 *0 O C i- in £ c 8 O 1 u ui 10 E 0 10 -a C in I c O O I0 10 O O 0 9 0 3 Ul o «•• <a o« 4-i •o -a O 4V 4) a. C -w +- U IU ~ 0 C -~ 4J 4J CL 0 4J — o a. o u u _0 a . 0 I0 •o O Ul U U C U — 0 0 0 -O Ol 0 w 0 Ul •o 0 0 4. | 10 3 U C 0 l- C 0 4»i 4-i I0 4J — k . a. c 3 k u — — e ia ui — if 4-l 3U •— Ul *0 E — 10 — X -0 J3 -Ok w O -c> O -a i TJ . I0 c 44 Ul — — i JZ — 41 -K: k •J E O U — C I U — 4V M» e 0 ^-"* 4-* «— 4-i 4J p» u 44 ••• "-• •~ Ul 4J o — 4J u u JU k «n p- -a p- c — u Q U l 44 u ^ 4)U — •— —. ui — 0 ... 0 3 C 3 0 3 — J= — X I0 U. h- X *• 01 U "! a a I0 Q. X Ul I k 0 3 3 I 3 a k . i ui XIAU.P-XUI o o • ... X 3 x u a ja u -a 0 u- — 1 vo I0 J3 O "O o* i < 03 u Q < \ 1 J 2J 4V i o io cr rt N U — C I0 •— •— P-- U 0 §3 • Ul U 0 •* 3 k i U 0 — U U 4V 0 Ul T3 3 JZ ui C *- 49 ** • • y u a. 3 Ul k 8 c o k . k . 0 -^ O k • « I^ Ul 0. — a. a. ui k Q Q» a.. 3 in <»1 a vi - 0 VI — — ja •?= • 1o 1o 0 .53 0 10 !/<</'•— ::! O O I0 Ik *J Ul •o ui —• 0 10 — Q. 4V 3 k . •O C O 0 — S. J= Q. o -o a • Ul 0 Ul 0 4J •— Ul U L. jQ .. Q. « — Ul 3 0 I0 4-> Ul U > n 3 I0 — e -a X 0 0 U l ««o~ 0 I. w k ui a. k 3 1o o 0 — o k U 3 o o cr » k 0 0 *j a. k . C 03 E o a "0 a. 3 u — « o- o 3 C 0 k o-0-oQ. 0 i to a c - k *3 *~ C o o- k 0 4-i 0 o 6 U Q. « — — — M- I0 i0 3 •' — E O- «« a. x

860 k . O in •— in — X a. 4V 4) Z — e C 3 2 x wk. o "«' UJ — - * O. 01 Ul I0 in a. k . CO D O io — 4V 4V — 4i IB 00 z Z X a. z « x cry z •x. in o I i^ C 4V 4V 4V i*- i k . I0 0 vi 0 «4- U- t.. I0 O m I0 I0 — k . l- -X C 4) U — I0 •* 4V k. T3 I0 I0 — U I0 4) I0 4) O 4V O in 4) 1- "o C 4) O o w a. •— in u u 4V U 4V Q. O 4V — in S8 — 4) JC M ft m in -* C OJ C 3 U C 4) 4) U in C O 4) O x k . •a 4J E 0 4) I0 U I0 4) U U U I0 •— l I0 E nj 3 a . k . in 4) •0 4V 4) u. u I0 U 4) k. O > I0 J= c Q. in 4) i4- 4V 1Q > I0 O I0 I0 -a c a. O in in k . — 4) I0 I0 U c a.U- I0 U tn J 3 I0 k. U U a . u 0 4V Q) k . U a. u O I0 4) X — u -o i* •— I0 in c C 4) U I0 4) X — u -a — a. u 4) 4V c I0 •03 — — C a. u 4) 4V C I0 4) I0 4) o — .* e U O 4V 4) 4) I0 4) U — C in o 4) i- — J U O .— oi w in x-o 4V in a 4) i- — J — in 41 i in .j O '^ 4V — °3 U — c l in 4V O X IB 4) 10 i- -C 4V 4) -0 ID ^^ X 10 4) u — -a E JC C • r0 m « k. 4V 4V 4) io _ "« c E -c E I0 C C 4) C — 4V O O ••^ O 0 in i_ w -a 4) .— 4V 4V "o 0 — 3 •o — w o 0 0 O) — O i«- TJ — 4V o — i C IiS 4V 3 in w — JC X u. o 3 O I0 C i£ 3 VI 4V 4V 3 i- UJ CO in 3 < a rt „ l k. U l T C 4V 4) in •• o u in «— in — u 4) o Ul X ^— ' 4V CL I0 O X *m U u i k ^» I0 in 4) w C I0 k. C e — 4) o o k. C o I0 O 4) > 4) O I0 4V — I0 Q 4V 4V •— l6 4V m J in I0 M- i4 C 4V 10 E O) C J^ i § U > vi Z « U I0 UJ X — 4) — k. U I0 1 in a. O 0 t» 4V a. •^ in & *s 1u — k . 4V a. 4) U 3 0) in J3 4V 4V o. 4) U 3 • fci 3 U l c X O — I0 4) k. 4V £ •o ~ I0 E E I0 C C U l O Q 4) i •o 4) U 4) U 10 k 4) c O O I0 4V I0 I0 !• 4ri ^ I0 2 U 4V • •o x | 4V 4V 10 Q. C ID I0 •a X a. 4) "io I0 3 u k . o) e a. 4) "I0 E 4V 4V — 4) 4V -o ig k . I0 I0 c — I0 O — . k. U C E 4) U £ I0 ID C a T) •o in i n — 4) I0 JC O VI I0 C I0 in 4) ••• U 4) « -a a-a u w — c 4) — 4-i !• •o *J •• ^~ k . — — U l i 4) I0 — o k . •a 4V C 4V k . •^ 4V (Q U *-1 — ul l_ 4J fgf i in 8 u 4) E x — c — i a "3 I0 in 8 U a. a 3 U 3 O 3 — 3 a. 4) ^ u ••» O c — x — U, • «/> Z > ID Pm &- a. 4) 4) •o o a. •— . . E JZ U 4) 4) . ... • . . 3 U— 0> l- Z vi a -a o -o « VU. at ^B O V . • . 1 . x CO O a -< CD in 4> m i- — 3 QL I 4V o- o 4) 3 > C U in in c in 4i 4) — JC in 4) O in — U O — . fo I0 v_ IQ a. . u a. a.

861 - M a» w TJ 4> I — 10 Q O I0 k. •» "a. k . u a. 4) 4) X X > U 4) 4) (/) 4) I0 C C 3 in a. 3 3 10 ^^ in *~* *~^ £ z z * M Z Z Z •x. Z Z x tfl 4> U l C C 4-> I I i u- in tfl C — o o »•^ M 4) 4) o O m 3 — 0 in — — I0 C TJ >4- U U u k. I0 tfl «— c X4V • • 4»i k. o 4) 1O nj I0 c 4) a. X 4J o — I0 U l U U 4V k. a. a. 4-i Q 4J O — 10 in m — -* C 0) 4-1 C U U l in c I0 — ^£ 4V Si C 3 O I0 4) 4) £ u 3 10 "TJ 4) in. E 4) k . 1o c C 3 C 10 ^3 I0 I0 — l I0 5 ra E .oca. 3 4) ' 4) U Ol 3 — in O tfl E k. a. k. > I0 I0 — in E I0 IO U 4J in I0 I0 a c a . E I0 I0 0 — « 01 in in k . a. o o — U l — I0 I0 tfl C C 5 10 I0 x O TJ tO — 10 10 4V U TJ 3 — X — 4) a. a. 4) C C tfl 4-' O TJ X 4) _ O 4J C 4) 4) O u — .* c 4) MB k. k. k. Ol 4J tfl O X in in 4) _ ~ J U 0 k. k. Ol o U l 0 O « 4> E — k. • k . JB *< Si c — 3 X I0 -0 0 X 4-i x ** io 10 U 4-i 4-i 4) tj 10 — l4- — i4- £ k . §nj m 1o f 3 4J L_. k H Z in k. O *•" TJ 4) 10 4) I0 ._ 4) •»• y h- •*. in o 1 . — O u- ! X «•- O *3 "3 TJ k. TJ k. _ o 4-i a. c o *J O O O "x _9 I0 I0 — y 3 Ol 4-' >— . . M io -a LU _J O 0 U l io -a UI tfl I I0 — Ok. ra X •— U k. I— c 4) 4) O I0 U l Q. 10 -C §in k. 4V 4V I0 in ** c C 4-i — C U k. O c U •u o4>a. io — a . 3 4-> • 1 in Ol > m x ui 10 k . 4-> a. in Ol Ol ••» c k . to k. 4) y 3 Ol MB c c U l f) 4) 4V — k . 1 3 in c in X "o VI C > 4) 10 4) in X "v -a O 10 I c y 4) I0 o k. j 5 o _Q 10 k. .- 4) I0 ^o ^5 c Q 4V 4J — O k. M •3 c Q 1 I0 • C O Z 4) «-« I0 O 4) UJ C U U a. 1 X | a • • .-_ — — Ol 4) 4J 10 I0 1o ~— Io «0 4-« -— 4) *a g k. u 3 in E — 4) u o I0 c 'a 4V I0 I0 I0 I0 C 10 M U TJ 1 — « mm -oo 0 — E E u — tfl in u in in 4) 4J 4) u TJ C k. TJ e 4) X 10 I0 tfl — C 4-J | •M Io 1 k. U — — I 3 O "3 tfl u 1 k. k. a. k. k . k. 1 a a. a. — z — z 1 1 4) 4) 4) 4) X O D • • » • a. x 9 fm O LU -j tfi a .a u -a 4) ^4_ Ol f- 0 UI 0 4) « . . • . • • • " U O UI ** -• CD U a UI § 0 u — £. <a •w j: I0 4) 4) A tfl v» c o o 4»i C c « o 1 z *«

862 1 Ul Ul T U. 1. *-^ — 03 mm <c 4) — O {•1n 1 4) Wl " 1 Ul c o a Mfc Zoo •i Ul _o — > 3 a. 4) c 2 M — 13 to - a. Q. u. </, 4) Wl "O *-• U ^~ Wl 4) 14. 10 10 X 1n c 4-1 10 a. 4 J j£ — — 0 C U Wl c U 10 Wl — 4) U E 4) j 0 C X 4 - • — 10 •o js m 3 U 4) Wl 10 U V. 10 u 4) C 4) 4-< Q. 10 4) — a. 10 — Wl Wl 1u a c 10 10 4) C 4d 10 x o c x: — O "I a 4) — O m ro U U 10 X C ^^ 4) 4) 10 M — Wl 4-* l*S ^•' 1U 1 Wl 4 J L. 1o 4J C 0) J3 h> 4-J X 4> 10 4) 4-1 TJ 4) O u 10 10 E a. 3 4) O ~~ u 3 4 - -o M 4) O 38 O. 4) §a. *3 O 4* H- O 3 — a. c u in -C O V) M 10 Wl 1- O 4J — O 1 4) — >» — 0.5 *—* "••-" M E C 19 C me 4) a c a ID .c u O U L. 4J 0 id «— 4J 4 J X 10 U tl w u u u <- O mm c 4) 4) 4) *J Q. -M c o > — Q. 4) 3 •— 1_ uI In E M e g 4-1 4) O u O Wl 1d 4) 4^ *"~ g X 3 c J3 > 4) 10 4J O C 10 C C O 2 01 O >- ~~ 4) ^B 4«| E U 4J O O *~ X I 4 - 10 u o U 1 U •o o a. — -a ui u L. G 14- 4) ** C -0 4) 4J 4) — . a 3^-» — «d C L. u c 10 3 ftmm — In J O — « 3 C Wl fl 2 O 41 a u -o a In — L. E 4) — tn -* •a u u C 4) U «B M M 4J U U u ^ I U X — 10 — C J 8 OJ - ^° 5Z-S5 M (/> X C 1 01 U 3 i m m ja u T) a! Ib en OC * a — u u O ^c In L» a C M e 10 4) i — M a. In L. « 0 U *-• O c u 1 — a. w a. •

863 FIGURE 1 Conceptualization of collisionless shock macrophenomenology as manifested in the earth's curved bow shock. Unshocked interplanetary field direction Bsw is indicated on the foreground field "platform." Field magnitude is plotted vertically; field direction would be deflected somewhat in the Bsw plane by the quasi-perpendicular shock, but would share the agitation of the magnitude in the quasi- parallel shock in all components. The superposed three-dimensional sketches represent solar wind proton thermal properties as number distributions in velocity space. < t > (mV/m) 100 M * Q - PERPENDICULAR ® PERPENDICULAR A Q - PARALLEL FIGURE 2 Conceptualization of bow shock microphenomenology as represented by electromagnetic noise power density and plasma electric wave noise amplitude in the various shock structures determined by up- stream plasma parameters 0 and M. Individual symbols show the parameter combinations in which details have been documented in specific cases. The clear contours apply to quasi-perpendicular, the shaded (and truncated) contours to quasi-parallel, geometry.

864 25 5.12 SEC ELECTRON PROTON VELOCITY DISTRIBUTIONS FIGURE 3 Sequences of (electro) magnetic wave, left, electric plasma wave, right, and particle events, center, relative to typical quasi-perpendicular, supercritical magnetic profile plotted in the central block. Electron and proton distributions in velocity space are shown in the cold, fast solar wind, foreground, in the foot of the magnetic structure where the electrons are initially heated and the protons retarded and partially scattered, in the center of the principal magnetic gradient, or ramp (shaded) where the electrons are fully scattered and the protons partially heated and scattered to form a bimodal distribution, and fi- nally behind the magnetic front, where electrons and protons are both found heated and scattered into nonmaxwellian distributions. (b) FIGURE 4 Geometry of return proton detection, (a) Particle guiding-centers in the plane of Vsw (ie., X) and Bsw ad- vance along B3W at speed V, while drift- ing perpendicular to Bsw at speed Vd, but resulting velocity Vr is treated as if Yr = Pvsw§sw/Bsw + Ysw- 0>) A CTOSS section of the foreshock is formed in each §sw ~ Ysw plane by electrons and pro- tons streaming away from the shock ac- cording to the diagram in (a).

865 XF 140° XT e p 120 ,/45° TO 756 2.3 / 15° to 45° 1.85 100 / -15 3 to 15°| 1.6 80 .f / -45° to -15° 1.85 60 / -75° to -45° 2.3 40 - 20 (a) 0 1 1 1 1 1 0 20 40 60 80 100 120 e XB (b) FIGURE 5 (a) Dependence of the foreshock's proton wave boundary angle 0XF on field angle 0XB for different positions along the shock, indi- cated by 0XT. The quantities are def1ned in rela- tion to each other in (b). 1000 400 km/sec Q_ Q-TURB TURBULENT 0 1 2 3 4 DEGREES 180 150 120 90 60 30 0123456 r AU r AU W « FIGURE 6 (a) Typical parameters of the solar wind vs heliocentric distance r from the sun to Jupiter; (b) Average field longitude 0B and foreshock boundary angle 0X F vs heliocentric distance.

866 IMP-8, JULY 4, l974 E-FIELD l0 562 Hz V/M UNCORRECTED 270-8l0 Hz E-FIELD V/M (Hz)'"4 10 0000 0400 0800 l200 i600 2000 2400 FIGURE 7 Simultaneous measurements by IMF's 7 and 8 on 4 July 1974. Positions of the satellites relative to the bow shock system are shown in the diagrams superposed on the second panel from the top.

867 IMP-8, JULY 5, 1974 E-FIELD 10 562 Hz V/M UNCORRECTED 10 270-810 Hz E-FIELD V/M (Hz)* 10 -5 0000 U400 0800 1200 1600 2000 2400 FIGURE8 Same as Figure 7 for 5 July 1974.

868 REFERENCES 1. Adlam, J. H. and J. E. Allen, Hydromagnetic disturbances of large ampli- tude in a plasma, Proc. U.N. Interntl. Conf. Peaceful Uses Atomic Energy, 2nd, JL 221, 1958. 2. Anderson, K. A., Energetic electrons of terrestrial origin upstream in the solar wind, J. Geophys. Res., 73., 2387, 1968. 3. Anderson, K. A., Energetic electrons of terrestrial origin behind the bow shock and upstream in the solar wind, J. Geophys. Res., Ik, 95, 19&9. k. Armstrong, T. P. and S. M. Krimigis, Interplanetary acceleration of rela- tivistic electrons observed with IMP 7, J. Geophys. Res., 81, 677, 1976. 5. Auer, P. L., H. Hurwitz, Jr., and R. W. Kilb, Large-amplitude magnetic compression of a collision-free plasma, 2, Development of a thermalized plasma, Phys. Fluids, 5_, 298, 1962. 6. Auer, R.-D., H. Grunwaldt, and H. Rosenbauer, Bow-shock-associated proton heating in the upstream solar wind, J. Geophys. Res., 8J_, 2030, 1976. • 7. Axford, W. I., The interaction between the solar wind and the earth's magnetosphere, J. Geophys. Res., 67, 3791, 1962. 8. Barnes, A., Theory of generation of bow-shock-associated hydromagnetic waves in the upstream interplanetary medium, Cosmic Electrodyn., j_, 90, 1970. 9. Behannon, K. W., Variation of the interplanetary magnetic field with helio- centric distance, submitted to Rev. Geophys. & Space Phys., 1977- 10. Benson, J., J. W. Freeman, H. K. Hills, and R. R. Vondrak, Bow Shock Protons In the Lunar Environment, The Moon, 14, 19, 1975-

869 11. Chevallier, R. A., The interaction of supernovae with the interstellar medium, Ann. Rev. of Astron. & Astrophys. ,"~T5r-in press, 1977- 12. Coleman, P. J., Jr., E. J. Smith, L. Davis, Jr., and D. E. Jones, The radial dependence of the interplanetary magnetic field: 1.0-1.5 All, J. Geophys. Res., 74_, 2826, 1969- 13* Daughney, C. C., L. S. Holmes, and J. W. M. Paul, Measurement of spectrum of turbulence within a col 1isionless shock by collective scattering of light, Phys. Rev. Lett's., 25_, 497, 1970. 14. De Hoffmann, F., and E. Teller, Magnetohydrodynamic shocks, Phys. Rev., 80, 692, 1950. 15* Diodato, L., E. W. Greenstadt, G. Moreno, and V. Formisano, A statistical study of the upstream wave boundary outside the earth's bow shock, J. Geophys. Res., 8j[, 199, 1976. 16. Diodato, L. and G. Moreno, A two-spacecraft study of the foreshock pertur- bations of the solar wind protons, CNR, Frascati, submitted to J. Geophys. Res. 17. Dobrowolny, M. and V. Formisano, The structure of the earth's bow shock, Rev. Nuovo Cim., 3_, 419, 1973. 18. Dryer, M., Interplanetary shock waves generated by solar flares, Space Sci. Rev., _£, *°3, 1974. 19. Fairfield, D. H., Bow shock associated waves observed In the far upstream interplanetary medium, J. Geophys. Res., 74_, 3541, 1969. 20. Fairfield, D. H., Whistler waves observed upstream from col 1Isionless shocks, J. Geophys. Res., 79_, 1368, 1974. 21. Fairfield, D. H. and K. W. Behannon, Bow shock and magnetosheath waves at Mercury, J. Geophys. Res., 8l_, 3897, 1976.

870 22. Fairfield, D. H. and W. C. Feldman, Standing waves at low Mach number laminar bow shocks, J. Geophys. Res., 80, 515, 1975- 23. Feldman, W. C., J. R. Asbridge, and S. J. Bame, Bow shock perturbation of the upstream solar wind proton component, J. Geophys. Res., 79, 2773, 1974. 2k. Feldman, W. C., J. R. Asbridge, S. J. Bame, and M. D. Montgomery, Solar wind heat transport in the vicinity of the earth's bow shock, J. Geophys. Res., 78., 3697, 1973- 25. Fishman, F. J., A. R. Kantrowitz, and'H. E. Petschek, Magnetohydrodynamic shock wave in a collision-free plasma, Rev's. Mod. Phys., 32, 959, I960. 26. Formisano, V., Jupiter's bow shock structure, Lett. Nuovo Cim., 10, 182, 1971i 27. Formisano, V., The physics of the earth's col 1isionless shock wave, pre- sented to the Third International Congress on Waves and Instabilities in Plasmas, Paris, June 1977. 28. Formisano, V., and E. Amata, Solar wind interaction with the earth's mag- netic field. 4. Preshock perturbation of the solar wind, J. Geophys. Res., 8J_, 3907, 1976. 29- Formisano, V. and P. C. Hedgecock, 0n the structure of the turbulent bow shock, J. Geophys. Res., 78_, 6522, 1973- 30. Formisano, V., G. Moreno, F. Palmiotto, and P. C. Hedgecock, Solar wind interaction with the earth's magnetic field. 1. Magnetosheath, J. Geophys. Res., 78_, 3714, 1973. 31. Formisano, V., C. T. Russell, J. D. Means, E. W. Greenstadt, F. L. Scarf, and M. Neugebauer, Collisionless shock waves in space: A very high- beta structure, J. Geophys. Res., 80, 2030, 1975-

871 32. Frank, L. A., K. L. Ackerson, and R. P. Lepplng, 0n hot tenuous plasmas, fireballs, and boundary layers in the earth's magnetotai1, J. Geophys. Res., 8J_, 5859, 1976. 33. Fredricks, R. W., A model for generation of bow-shock-associated upstream waves, J. Geophys. Res., 80, 7, 1975- 34. Fredricks, R. W., G. M. Crook, C. F. Kennel, I. M. Green, F. L. Scarf, P. J. Coleman, Jr., and C. T. Russell, 0GO 5 observations of electro- static turbulence in bow shock magnetic structures, J. Geophys. Res., 75, 3751, 1970. 35. Fredricks, R. W., C. F. Kennel, F. L. Scarf, G. M. Crook, and I. M. Green, Detection of electric field turbulence in the earth's bow shock, Phys. Rev. Lett's., 2J[, 1761, 1968. 36. Fredricks, R. W., F. L. Scarf, and L. A. Frank, Nonthermal electrons and high-frequency waves in the upstream solar wind. 2. Analysis and interpretation, J. Geophys. Res., 76, 6691, 1971. 37. Fredricks, R. W., F. L. Scarf, and I. M. Green, Distributions of electron plasma oscillations upstream from the earth's bow shock, J. Geophys. Res., 77, 1300, 1972. 38^ Galeev, A. A., Col 1islonless shocks, Physics of Solar Planetary Environment, ed. D. J. Williams, American Geophysical Union, k6k, 1976. 39. Gosling, J. T., A. J. Hundhausen, and S. J. Bame, Solar wind stream evo- lution at large heliocentric distances: Experimental demonstration and the test of a model, J. Geophys. Res., 8j_, 2111, 1976.

872 40. Greenstadt, E. W., Dependence of shock structure at Venus and Mars on orientation of the interplanetary magnetic field, Cosmic Electrodyn., 1, 380, 1970. 41. Greenstadt, E. W., Binary index for assessing local bow shock obiiquity, J. Geophys. Res., 77, 5467, 1972. 42. Greenstadt, E. W., Field-determined oscillations in the magnetosheath as possible source of medium-period, daytime micropulsations, Proc. of Conf.- on Solar Terrestrial Relations, Univ. of Calgary, 515, 1972. 43. Greenstadt, E. W., Structure of the terrestrial bow shock, Solar Wind Three, ed. C. T. Russell, Institute of Geophysics & Planetary Physics, UCLA, 440, 1974. 44. Greenstadt, E. W., Phenomenology of the earth's bow shock system. A summary description, Magnetospheric Particles & Fields, ed. B. M. McCormac, Reidel Publishing Co., Dordrecht, Holland, 13, 1976. 45. Greenstadt, E. W., Energies of backstreaming protons in the foreshock, Geophys. Res. Lett's., 3_, 553, 1976. 46. Greenstadt, E. W. and R. W. Fredricks, Plasma instability modes related to the earth's bow shock, Magnetospheric Physics, ed. B. M. McCormac, D. Reidel, Dordrecht, Holland, 355, 1974. 47. Greenstadt, E. W., I. M. Green, G. T. Inouye, D. S. Colburn, J. H. Binsack, and E. F. Lyon, Dual satellite observation of the earth's bow shock. 2. Field-aligned upstream waves, Cosmic Electrodyn., ]_, 279, 1970. 48. Greenstadt, E. W. and J. V. 0lson, PC 3,4 activity and interplanetary field orientation, J. Geophys. Res., 81, 5911, 1976.

873 49. Greenstadt, E. W., C. T. Russell, V. Formisano, P. C. Hedgecock, F. L. Scarf, M. Neugebauer, and R. E. Holzer, Structure of a quasi-parallei, quasi- laminar bow shock, J. Geophys. Res., in press, 1976. 50. Greenstadt, E. W., C. T. Russell, F. L. Scarf, V. Formisano, and M. Neugebauer, Structure of the quasi-perpendicular, laminar bow shock,- J. Geophys. Res., 80, 502, 1975. 51. Greenstadt, E. W., V. Formisano, C. T. Russell, M. Neugebauer, and F. L. Scarf, Ion acoustic stability analysis of the earth's bow shock, sub- mitted to Geophys. Res. Lett's, 1977. 52. Holzer, T. E., Interaction between the solar wind and the interstellar medium, Exploration of the Outer Solar System, ed. E. W. Greenstadt, M. Dryer, and D. S. Intri1igator, AIAA, New York, 21, 1976. 53i Hundhausen, A. J. and J. T. Gosling, Solar wind structure at large helio- dentric distances: An interpretation of Pioneer 10 observations, J. Geophys. Res., 8J_, 1436, 1976. 54. Jokipii, J. R., Radial variation of solar wind parameters, Geophys. Res. Lett's., 3., Hi, 1976. 55. Kellogg, P. J., Flow of plasma around the earth, J. Geophys. Res., 67, 3805, 1962. 56. Kellogg, P. J., Solitary waves in cold, col 1isionless plasma, Phys. Fluids, If 1555, 1964. 57. Kennel, C. F., and F. V. Coroniti, Is Jupiter's magnetosphere like a pulsar's or earth's?, The Magnetosphere of Earth & Jupiter, ed. V. Formisano, D. Reidel, Dordrecht, Holland, 451, 1975. 58. Kennel, C. F. and R. Z. Sagdeev, Col 1isionless shock waves in high 8 plasmas, 1, J. Geophys. Res., 72, 3303, 1967.

874 59. Lin, R. P., C.-l. Meng, and K. A. Anderson, 30- to 100-keV protons upstream from the earth's bow shock, J. Geophys. Res., 79, 489, 1974. 60. Montgomery, 0., Development of hydromagnetic shocks from large amplitude Alfven waves, Phys. Rev. Lett's., 2_, 36, 1959. 61. Montgomery, M. 0., J. R. Asbridge, and S. J. Bame, Vela 4 plasma observa- tions near the earth's bow shock, J. Geophys. Res., 75, 1217, 1970. 62. Morawetz, C. S., Magnetohydrodynamic shock structure without collisions, Phys. Fluids, 4_, 988, 1961. 63. Morawetz, C. S., Modification for magnetohydrodynamic shock structure without collisions, Phys. Fluids, 5_, 1447 64. Morse, D. L., A model for ion thermalization in the earth's bow shock, J. Geophys. Res., 8j_, 6126, 1976. 65. Morse, D. L., W. W. Destler, and P. L. Auer, Non-stationary behavior of collisionless shocks, Phys. Rev. Lett's., 28, 13, 1972. 66. Morse, D. L. and E. W. Greenstadt, Thickness of magnetic structures associated with the earth's bow shock, J. Geophys. Res., 81, 1791, 1976. 67. Ness, N. F., K. W. Behannon, and R. P. Lepping, The magnetic field of Mercury, 1., J. Geophys. Res., 8_£, 2708, 1975- 68. Ness, N. F., K. W. Behannon, and R. P. Lepping, 0bservations of Mercury's Magnetic Field, Icarus, 28_, 479, 1976. 69. Ness, N. F., K. W. Behannon, R. P. Lepping, Y. C. Whang, and K. H. Schatten, Magnetic field observations near Venus: Preliminary results from Mariner 10, Science, 183, 1301, 1974.

875 70. Neugebauer, M., Large-scale and solar-cycle variations of the solar wind, Space Sci. Rev., ]7., 221, 1975. 71. 0gilvie, K. W., J. D. Scudder, R. E. Hartle, G. L. Siscoe, H. S. Bridge, A. J. Lazarus, J. R. Asbridge, S. J. Same, and C. M. Yeates, 0bser- vations of Mercury encounter by the plasma science experiment on Mariner 10, Science, 185, 145, 197^. 72. 0gilvie, K. W., J. D. Scudder, V. M. Vasyliunas, R. E. Hartle, and G. L. Siscoe, 0bservations at the planet Mercury by the plasma electron experiment — Mariner 10, Goddard Space Flight Center Report X-692-76-223, 1976. 73. 0lson, J. V., R. E. Holzer, and E. J. Smith, High-frequency magnetic fluctuations associated with the earth's bow shock, J. Geophys. Res., 7^, 4601, 1969. 74. Parker, E. N., Interplanetary dynamical processes, Interscience, New York, 1963. 75. Parker, G. D. and J. R. Jokipii, The spiral structure of the interplanetary magnetic field, Geophys. Res. Lett's., 3., 561, 1976. 76. Paul, J. W. M., Review of experimental studies of col 1isionless shocks propagating perpendicular to a magnetic field, ESR0 Spec. Publ. 51» Frascati, Italy, 97, 1969. 77. Paul, J. W. M., G. C. Goldenbaum, A. loyoshi, L. S. Holmes, and R. A. Hardcastle, Measurement of electron temperatures produced by col 1isionless shock waves in magnetized plasma, Nature, 216, 363, 1967.

876 78. Paul, J. W. M., L. S. Holmes, M. J. Parkinson, and J. Sheffield, Experi- mental observations on the structure of collisionless shock waves In a magnetized plasma, Nature, 208, 133, 1965- 79. Roberts, P. H., Stellar Winds, Solar Wind Three, ed C. T. Russell, Univ. of Calif., Los Angeles, 231, 1972i. 80. Robson, A. E., Experiments on oblique shock waves, ESR0 Spec. Publ. 51, Frascati, Italy, 159, 1969- 81. Rodriguez, P. and D. A. Gurnett, Electrostatic and electromagnetic turbulence associated with the earth's bow shock, J. Geophys. Res., 80, 19, 1975. 82. Rodriguez, P. and D. A. Gurnett, Correlation of bow shock plasma wave turbulence with solar wind parameters, J. Geophys. Res., 81, 2871, 1976. 83. Rosenberg, R. L. and P. J. Coleman, Jr., The radial dependence of the interplanetary magnetic field: 1.0-.7 AU, Inst. Geophys. & Planet. Phys. Publ. 11196-26, Univ. of California, Los Angeles, 1973. 84. Russell, C. T., The Venus bow shock: Detached or attached?, J. Geophys. Res., in press, 1977. 85. Sagdeev, R. Z., Cooperative phenomena and shock waves, collisionless plasmas, Reviews of Plasma Physics, 4_, ed M. A. Leontovich, Consultants Bureau, New York, 1966. 86. Sarrls, E. T., S. M. Krlmigis, and T. P. Armstrong, 0bservations of a high-energy ion shock spike in interplanetary space, Geophys. Res. Lett's., 3, 133, 1976.

877 8/. Sarris, E. T., and J. A. Van Allen, Effects of interplanetary shock waves on energetic charged particles, J. Geophys. Res., 79, 4157, 1974. 88. Scarf, F. L., R. W. Fredricks, L. A. Frank, and M. Neugebauer, Non- thermal electrons and high-frequency waves In the upstream solar wind, I, 0bservations, J. Geophys. Res., 76, 5162, 1971. 89. Scarf, F. L., R. W. Fredricks, I. M. Green, and G. M. Crook, 0bservations of interplanetary plasma waves, spacecraft noise, and sheath phenomena on IMP 7, J. Geophys. Res., 79_, 73, 1974. 90. Scudder, J. D., D. L. Lind, and K. W. 0gilvie, Electron observations In the solar wind and magnetosheath, J. Geophys. Res., 78, 6535, 1973. 91. Siscoe, G. L. and B. Goldstein, Solar wind interaction with lunar magnetic fields, J. Geophys. Res., 78_, 6741, 1973. 92. Siscoe, G. L. and M. A. Heinemann, Binary stellar winds, Astrophys. & Space Sci., 3]_, 363, 1974. 93. Smith, E. J., L. Davis, Jr., D. E. Jones, P. J. Coleman, Jr., 0. S. Colburn, P. Dyal, and C. P. Sonett, Jupiter's magnetic field, magnetosphere, and interaction with the solar wind: Pioneer 11, Science, 188, 451, 1975. 94. Smith, E. J.t R. H. Holzer, M. G. McLeod, and C. T, Russell, Magnetic noise in the magnetosheath in the frequency range 3-300 Hz, J. Geophys. Res., 72_, 4803, 1967. 95. Smith, E. J. and J. H. Wolfe, 0bservations of interactive regions and co- rotating shocks between one and five AU: Pioneer 10 and 11, Geophys. Res. Lett's., 3, 137, 1976.

878 96. Sonett, C. P. and D. S. Colburn, The SI -SI pair and interplanetary forward-reverse shock ensembles, Planet. Space Sci., 13, 675, 1975- 97. Spitzer, L., Jr., Physics of fully ionized gases, Interscience, Wiley, New York, 1962. 98. Spreiter, J. R., and A. Y. Alksne, Plasma flow around the magnetosphere, Magnetospheric Physics, ed D. J. Williams and G. D. Mead, William Byrd Press, Inc., Richmond, Virginia, 11, 1969- 99. Tidman, D. A., Turbulent shock waves in plasmas, Phys. Fluids, 10, 5^7, 1967- '00. Tidman, D. A., The earth's bow shock wave, J. Geophys. .Res., 72_, 1799, 1967. 101. tidman, D. A. and N. A. Krall, Shock Waves in Col 1isignless Plasmas, John Wiley-lnterscience, New York, 1971. 102. West, H. I. and R. M. Buck, 0bservations of > 100 keV protons in the earth's magnetosheath, J. Geophys. Res., 8l, 569, 1976. 103. Zhigulev, V. N., 0n the phenomenon of magnetic detachment of the flow of a conducting medium, Soviet Phys., Doklady, A., SI4, 1959. • 104. Zhigulev, V. N. and E. A. Romishevskii, Concerning the interaction of currents flowing in a conduction medium with the earth's magnetic field, Soviet Phys., Doklady, k_, 859, I960.

Next: Magnetic Field Reconnection »
Space Plasma Physics: The Study of Solar-System Plasmas Get This Book
×
MyNAP members save 10% online.
Login or Register to save!
Download Free PDF

The space age began exactly 20 years ago with the launch of Sputnik I and Explorer I. The Explorer spacecraft discovered regions of trapped radiation around the earth—the van Allen belts. This was the beginning of the study of particles and fields in space, or space plasma physics. A large part of the effort in the early years of the space program was devoted to the mapping of the magnetosphere, the measurements of time variations in particles and fields, and the exploration of the solar wind.

From these studies a sophisticated empirical knowledge of phenomena in space plasma physics has emerged. with the attainment of this observational maturity in the field, NASA funding for space plasma physics has declined as priorities have shifted to other exploratory ventures. The present study of space plasma physics was requested by NASA to obtain guidance for future directions in the subject.

The Committee on Space Physics of the Space Science Board was charged with the responsibility for soliciting technical review papers on a large number of topics in space plasma physics. These reviews are Volume 2 of the report; they constitute a most valuable resource for those working in the field.

  1. ×

    Welcome to OpenBook!

    You're looking at OpenBook, NAP.edu's online reading room since 1999. Based on feedback from you, our users, we've made some improvements that make it easier than ever to read thousands of publications on our website.

    Do you want to take a quick tour of the OpenBook's features?

    No Thanks Take a Tour »
  2. ×

    Show this book's table of contents, where you can jump to any chapter by name.

    « Back Next »
  3. ×

    ...or use these buttons to go back to the previous chapter or skip to the next one.

    « Back Next »
  4. ×

    Jump up to the previous page or down to the next one. Also, you can type in a page number and press Enter to go directly to that page in the book.

    « Back Next »
  5. ×

    To search the entire text of this book, type in your search term here and press Enter.

    « Back Next »
  6. ×

    Share a link to this book page on your preferred social network or via email.

    « Back Next »
  7. ×

    View our suggested citation for this chapter.

    « Back Next »
  8. ×

    Ready to take your reading offline? Click here to buy this book in print or download it as a free PDF, if available.

    « Back Next »
Stay Connected!