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.
CONTRIBUTION TO NATIONAL ACADEMY OF SCIENCES SPACE PLASMA PHYSICS STUDY. UNDERSTANDING PLASMA INSTABILITIES IN SPACE: IONOSPHERIC RESEARCH AND COMMUNICATIONS APPLICATIONS F. W. Perkins National Center for Atmospheric Research* Instabilities are ubiquitous in space plasmas because they ofttimes pro- vide the dissipation and transport which are demanded by the general large-scale features of plasma flow in the ionosphere and magnetosphere. Indeed, dissipation and transport by instabilities is a feature which space plasms share with other large, nonlinear systems; turbulent hydro- dynamic flows, geophysical fluid dynamics, astrophysical hydrodynamics and controlled fusion research. One of the major scientific challenges of the present era is to develop techniques to understand these large, nonlinear systems, especially the role of instabilities. The goal of this chapter is to set forth an approach to space physics which will not only provide an overall understanding of space plasmas, but also, in this process, take up problems of true intellectual significance in just how instabilities provide transport and dissipation in large scale flows. We shall point out how two-dimensional ionospheric plasma instabilities have both important similarities and differences with turbulence in neutral fluids. The interplay between the similarities and differences serves to increase our understanding of both the plasma and neutral fluid instabilities. *The National Center for Atmospheric Research is sponsored by the National Science Foundation. 1155
1156 It is also evident that a good understanding of the ionosphere is required if a model for the climatological effects of solar activity is ever to be devised, for it is in the ionosphere that the energy of solar disturbances affects neutral atmospheie flows and chemistry. Lastly, we shall discuss how ionospheric instabilities affect various communica- tions schemes. 1. A Computational Framework for Understanding Space Plasmas The first issue to be addressed is what are the goals of Space Plasma Physics? There is almost general agreement, I believe, that the era of space exploration has past, except for planetary studies. Already, 2 an impressive empirical knowledge has been obtained which is adequate to describe the environment in which future space missions will fly , although the complicated interaction between a given environment and spacecraft electronics, etc. is still poorly understood. A new goal is required, and in my view, the new goal of Space Plasma Physics should be understanding. What form does one expect that understanding will take? I take the point of view that true understanding is expressed by theory which al- lows us to predict the consequences of various applied forces. But in nonlinear systems, one should not expect understanding to take the form of simple equations, like the Schrodinger Equation. A more likely pros- pect is that understanding of nonlinear, dynamical systems will be em- bodied in a sequence of sophisticated computer codes, each of which deals with phenomena on particular spatial and temporal scales, with the re- sults of small-scale codes parameterized in the large-scale codes. Hence
1157 the nonlinear effects of instabilities will appear as viscosities, re- sistivities, etc. in large-scale codes. By turning on and off the param- eterization of an instability in a large-scale code, one can study how the sources of free energy develop which drive the instability in the first place. This brings us to another advantage of the computational approach to understanding space plasmasâidentifying the driving mechanisms for plasma instabilities. In general, the growth time for instabilities is very short compared to the time-scales for evolution of the large-scale flow. Hence the instabilities are always observed in a nonlinear steady state, and linear stability calculations based on measured data, if done correctly, should always yield zero growth rate. Most of the stability calculations found in the space physics literature are based on measured data, and hence should not show strong destabilizing terms. But a computational program, which embodies only limited physics, can evolve into regimes where small-scale processes, not described in the program, go unstable. Thus one can identify what aspect of the large-scale motion drives small-scale instabilities, and how strong these driving terms are. An excellent example of this type of approach is the recent 35 work of Liewer and Krall on collisionless shocks. Experiments, of course, play a major role in this approach to under- standing space plasmas. First, they provide the initial data with which to start computational programs. The limited data combined with the limited capability of computers guarantee that one can not "put the magnetosphere on a computer" and expect an accurate detailed description
1158 of the magnetospherlc response. But one can hope for a much more quan- titative cause-and-effect understanding. The second role of experiments is to normalize the computational programs to actual observations. In nonlinear systems, we can expect many sub-grid scale processes to be go- ing on which must be parameterized, and it is only through comparison with observations that the success of the parameterization process can be determined. Conversely, attempts to normalize computational programs to observations will channel the observational efforts on the key issues where increased experimental sophistication will directly help achieve our goal of understanding. Lastly, experiments should continue to have the ability to discover unexpected process, because we can certainly, expect that in building a computational understanding there will be im- portant processes overlooked in the first attempts. The recent discovery 40 of electrostatic shocks is a case in point. Calling for more theory and computational models has almost reached a motherhood status in space physics recently. But, all too often, theoretical programs can devote much time to relatively straightforward research and bypass the difficult problems of true intellectual signifi- cance. The bulk of this chapter is devoted to describing how ionospheric research has in the past, and can in the future, progress towards a true understanding, taking up the difficult problems as they become germane, and developing the detailed ionospheric models which will be required in future climatological and magnetospheric studies. To summarize the above paragraphs: Space Plasma Physics should set a goal of understanding and the most likely form this understanding will take is a set of nonlinear dynamic computational programs. Experi- ments provide three essential contributions: identifying the basic physi- cal processes, providing initial data, and normalizing computational results. Experimental programs which entail the broadest set of measure-
1159 ment techniques should be emphasized, each geared towards a key plasma physics problem. 2. The Impact of Space Plasma Physics The scientific challenge provided by the complicated nonlinear space plasma system is one of the major reasons for undertaking studies with ambitious goals. Methods developed for treating plasma turbulence, non- linear viscosities and resistivities, magnetic field merging, etc. will find application in many other fields of research and engineering. Achieving this impact, however, implies an aggressive program in space plasma theory with cross-ties to the other fields such as hydrodynamics, meteorology, oceanography, controlled fusion, etc. Currently, these other fields are providing the leadership in developing new theoretical methods and concepts. I Communications systems which propagate through space plasmas are 61 used increasingly for civilian and military purposes. Let us take a broad view of communications to include all forms of electromagnetic transmission of information: standard messages relayed via satellite, TV, satellite telemetry and other data, navigational signals, space track Q and ABM radars, and even raw energy from a satellite power station. Since most of these systems are already in use, the influence of space plasmas is felt on a day-to-day basis. We shall amplify this discussion later. A recent and discouraging development is that space might become an arena of "conventional" warfare because of the large number of space- 58 based military systems. Hence high-energy systems may be found in space, and if their performance depends on interaction with the space environ- ment, a good theoretical understanding of space plasmas is called for to assess their chances of success. At the least, some sort of environmental impact will occur in systems tests.
1160 The role of the ionosphere as the site of the interaction between solar activity energy and the neutral atmosphere meteorology has already been stressed. 3. Understanding of the Ionosphere Let us turn to the question: How far has our understanding of the ionosphere and ionospheric plasma instabilities in particular advanced? The answer to this question will be given within the framework set forth in the preceding paragraphs. Evidently there is a strong theoretical bias in this approach and it does the disservice of minimizing the impor- tance of experimental results. Hence, it should be clearly stated that the theoretical papers referred below all give careful attention to the experimental data which, in many cases, provided the theoretical puzzles in the first place. The references cited in the theoretical papers pro- vide a good guide to the experimental literature. The ionosphere is perhaps the best subject for the computational ap- proach to understanding space plasmas, principally because the magnetic field is very strong and magnetic perturbations can be safely ignored in ionospheric dynamics. Our framework for understanding the ionosphere contains four parts: (1) studies of ionospheric equilibria, (2) the stability at these equilibria, (3) nonlinear calculations of the satur- ation of instabilities, and (4) parameterization of the results of in- stability calculations in the large-scale equilibria studies. The reader should appreciate that all four parts are required in fields such as meteorology, communications channel modeling, and controlled fusion which demand a predictive capability for theory, not merely explanations. Let us examine how far understanding has progressed in each of these four areas.
1161 3a. Ionospheric Equilibria. Of all the topics within space plasma physics, the "equilibria" of the ionosphere has perhaps the most advanced understanding. In this context, equilibria means states that change only on global dimensions in space and diurnal periods in time. One- 5,28,37,46,50,59,60 dimensional computational models which calculate the evolution of the ionosphere as a function of height for a particular geographical coordinate have become quite sophisticated. Photoelectron transport, 59 slowing down, airglow excitation, energetic electron precipitation, and molecular and atomic reaction rates are all included. Nonetheless, cer- tain external forces must be prescribed to have a well-posed model: neutral winds which can support (or push down) the F-region as well as OT create sporadic E-layers, the solar EUV flux which provides the ioniza- tion, magnetospheric fluxes of plasma, heat, precipitating electrons, and electric current and the electric fields which these fluxes generate. In high latitude ionospheres, joule heating by electric currents plays an Q important role. Resonantly scattered solar Lyman-ot radiation is an essen- tial element in nighttime ionosphere models. Within reasonable adjustments of these external parameters, the calculated ionospheric electron density profiles agree well with observed ones. ' But often achieving good agreement forces one into adopting ty Q en solar EUV fluxes that are twice the measured values, 'perhaps an indication of our uncertainty of their variation during the course of the solar cycle. These models are quite good at examining the response of the ionosphere to variations in the global neutral wind patterns and O Q electric fields. Three-dimensional models, which are essential to achieving an understanding of the polar ionosphere where horizontal con- 33 vection is important, have only reached a state of dimensional analysis. But one-dimensional models which allow the possibility of plasma outflow, 6 called the polar wind, are sophisticated.
1162 3b. Stability of the Ionosphere. Although a careful reader can readily identify many areas in which improvements of the equilibria models is desirable, the thrust of this chapter is to ask: to what ex- tent have these models been examined for stability with respect to small- scale perturbations? Let us begin our answer by enumerating what types of equilibria are employed in the current models. We find (1) mechanical equilibrium, (2) thermal equilibrium, (3) chemical equilibrium, and (4) electrical equilibrium (i.e. the divergence of the current is zero). It appears as though there has been very little systematic effort to investigate the stability of the many ionospheric equilibria models in spite of experimental evidence for small-scale structures. Instead, most instability work has concentrated on a few very short-scale (1-10 meter) instabilities discovered by radar backscatter measurements. Struc- ture on the kilometer scale length which is studied by satellites,Ii4,16,17,30,54,55 auroral photographs and radiowave propagation studies (e.g., radio star scintillation) has received less theoretical attention, except for the equatorial ionosphere. Let us set forth some more specific suggestions on what types of instabilities might occur in the ionosphere and indicate whether or not any stability analyses have been performed. Turning first to the ques- tion of mechanical equilibrium of the ionosphere, electric fields and neutral winds play an important role in controlling the height of the F-layer and in producing sporadic-E layers. Near the equator, the geom- etry is appropriate for a classic E x B/Rayleigh-Taylor instability on the bottomside ionosphere and stability analysis clearly predict insta- bility. This instability is invoked as the cause of equatorial spread-F. ' Mid-latitude nighttime F-regions are also unstable to support by neutral
1163 / 0 winds although the growth rates are small unless the layer is high. So far no stability analysis of sporadic-E layers has been performed in spite of history of experimental data which indicate small-scale struc- 27 tures. Sporadic-E layers are interesting because they involve a sub- stantial variation in the chemical equilibrium from the background iono- 20 sphere. The Farley-Buneman acoustic instability has long been recog- nized as occurring in both the equatorial and auroral electrojet; its cause is the high-current required by the electrical equilibrium. Ion-cyclotron 31 , , instabilities which can be generated by parallel currents or ion beams have n n recently been detected by rocket experiments. The source of driving free energy is not yet clear, however. Auroral arc models, which have very fast chemistry time scales as well as important electric currents, have yet to be 20 22 examined for instabilities other than the Farley ' instability. The gradient- 26 drift instability has been proposed to explain diffuse radio aurora but a stability analysis which deals satisfactorily with the high-latitude geom- etry remains to be carried out. The supersonic flow of plasma through a neutral gas which occurs in high-latitude ionospheres can generate 41 microinstabilities if ion-neutral collisions are sufficiently frequent. The auroral zone F-region might well suffer this instability, but no experimental tests have yet been carried out. To my knowledge, there are no attempts to assess the thermal stability of a plasma where joule heating by either perpendicular ion currents or parallel electron cur- rents plays an important role. For example, one can speculate that the / O1 very small-scale instabilities ' observed in the equatorial F-region could be caused by a thermal instability produced by field-aligned cur- rents flowing between hemispheres. Overall, most of the instability calculations have been motivated
1164 by observations, and not by the question of whether equilibria models are stable. Even so, there remains the recently discovered shocks and an im- pressive amount of ionospheric structure, especially at high-latitudes and short OQ scales, that are still unexplained even in a linear stability analysis. For completeness one should also mention that man-made perturbations of the ionosphere such as barium clouds, satellite electron beam gener- ators, radiowave modification, nuclear explosions, etc. have initiated many instabilities in the ionosphere. We shall take these up in a special discussion below. 3c. Nonlinear Saturation of Instabilities; If the stability anal- ysis are lagging equilibrium models and observations, then nonlinear studies of the identified instabilities is still more seriously behind. Essentially only equatorial Farley and gradient-drift instabilities have been studied. Perhaps the most successful nonlinear studies to date have been the application of the computational simulation method first developed 66 for barium clouds, to gradient-drift and Rayleigh-Taylor instabilities in the equatorial E- and F-regions.38,56 jn the E-region, the computational result shewed how daughter instabilities with vertically-oriented wave vectors were created by the nonlinear state of the primary instability which had an approximately horizontal wave vector. The results provide an understand- ing of radar and rocket data. The F-region studies^ show how large- scale density irregularities can occur and rise rapidly into the initially stable region in the post-sunset equatorial F-region. Again good agree- ment with experiment is reported. The Farley acoustic instability has been studied only by analytic methods even though its two-dimensional nature suggests that computation- al simulations would be straightforward. The most advanced analytic *") / 1^ O theories ' combine a number of previous approaches employing quasi-
1165 linear effects or strong turbulence models, and nonuniform geometry. While good agreement with experiment is claimed, it seems to me that a uniform medium instability should have a uniform medium saturation mechanism and that computational simulation is a straightforward way to determine it. From the point of view of fundamental studies, the ionosphere pro- vides a unique opportunity to study strong two-dimensional turbulence. The equations governing the evolution of two-dimensional structures bear 66 a close resemblence to hydrodynamic equations. In these analogies, field-line-integrated plasma density or conductivity takes on the role of vorticity while the electric potential serves as the stream function. The ionospheric turbulence shares with two-dimensional fluid turbulence the property that all moments of the vorticity are conserved, but the elliptic equation for the stream function is sufficiently different that steady-state vortices can not occur in ionospheric turbulence. The ionosphere has the advantage over neutral fluids in that turbulence remains two-dimensional so that the interplay between observations, computer studies, ' and analytic closure theories can be quite close. 3d. Influence of Instabilities on Large-Scale Phenomena. Here there 23 are only a few scattered results. A paper by Fedder ~ surveys what effects anomalous resistivity parallel to a magnetic field might have on auroral arc models. Several resistivity models were required, however, since the form of this resistivity is not yet clear. A nonlinear resistivity arising CO from the Farley-acoustic instability was calculated by Rogister and Jamin^- but has not yet been utilized in global studies of the electrojet current
1166 distribution. By and large, nonlinear studies of various instabilities have concentrated on properties of the instability itself, such as wavenumber spectrum, density fluctuation levels, etc. The studies generally do not produce effective resistivity formulas, turbulent diffusivities, enhanced joule heating rates, tec. which can be fed into equilibria models. Any understanding of how ionospheric instabilities effect the overall structure of the ionosphere must await future work. For example, auroral zone heating exerts a major influence on the meridional circulation of the neutral thermo- sphere. If enhanced joule heating resulting from plasma instabilities occurs, then our model of the overall auroral zone heating could be in error.
1167 3e. Increasing Our Understanding of Ionospheric Instabilities. Clearly a good theoretical representation of the equilibria which has withstood comparisons with experiment is the starting point for stability calculations. Both auroral arcs, where the electrical equilibrium is not clearly understoood, and, polar ionospheres where horizontal convec- tion is a key element require improved equilibrium models. Satellite measurements experiments most likely will continue to provide the best description of magnetospheric current sources in the ionosphere, but one must be careful in deciding whether the observations uniquely define high-latitude current systems, or are merely consistent with a proposed model. Indeed, both ionospheric and magnetospheric physics could greatly benefit from a measurement program which did uniquely define the spatial pattern of electric field and field-aligned current systems. Experi- ments to date have chiefly been concerned about whether field-aligned current exists.3,32 Irregularities in the polar and auroral F-regions are investigated principally by in situ satellite measurements. The Space Shuttle will £ive us the first opportunity to perform satellite-to-ground or satellite-to-satellite radar scattering experiments. Such observations could find out whether hitherto unobservable short-scale field-aligned instabilities exist in the auroral F-region as they do at the equator. Radar incoherent scatter studies of the auroral ionosphere will receive an important and needed boost with the operation of the European incoherent scatter facility, which will complement the very productive Chatanika facility in Alaska.
1168 Further, even radio wave scattering studies have concentrated on density fluctuations which vary rather slowly with time. The possibility of scattering with frequency shifts close to the lower hybrid frequency as the recent Ott and Farley instability predicts has been overlooked. Evidently, experiments will continue to be the essential element in providing magnetospheric boundary conditions for ionosphere studies. But I hope that the discussion above had made clear that our ability to understand ionospheric response to changing these boundary conditions is just in its infancy. At this stage, an increased effort in nonlinear computational models is the single most important step required to in- crease our understanding of ionospheric plasmas. And, as our increase in understanding of ionospheric physics is promulgated through the com- munity of magnetospheric scientists, perhaps the concept of the iono- sphere being just a passive linear resistor will be replaced by better models of its nonlinear response to magnetospheric current sources. 4. The Equatorial Ionosphere - A Major Success A careful reader of the preceding section will have noted that the equatorial ionosphere enjoys the greatest understanding. The purpose of this paragraph is to bring this material together for special emphasis. The unique geometry of the equatorial ionosphere means that it must carry strong currents in the E-region, and have the density stratified in a direction orthogonal to the magnetic field.38,-;6 Currently developed methods of plasma stability analysis are particularly suitable to this geometrjr ' and the two-dimensional nature of the ensuing plasma tur-' bulence permits accurate computational studies of Rayleigh-Taylor type
1169 instabilities.30'56 Hence a rather detailed understanding of equatorial spread-F has emerged from the computational simulations and has been 10 verified by measurements. These results are significant not only in their contribution to ionospheric physics but also in that they repre- sent a unique comparison between nonlinear Rayleigh-Taylor instability theory and experiment. Nonetheless, there is still room for improvement: 34,51,52,64 the analytic nonlinear theories of the current-driven Farley- instability should be compared with truly definitive computational studies, comparable to those used for the Rayleigh-Taylor instability. 5. Active Ionospheric Experiments A number of active experiments have been carried out in the ionosphere with unique and valuable scientific ramifications. Large barium releases have generated major perturbations in the ionospheric conductivity. The 15,24,57,66 computational models developed to describe the evolution and striation of these barium clouds have lead directly to the computational models which have been so successful in modelling equatorial spread-F. The plasma instabilities which ensued when the ionosphere was illuminated by high power radio waves provided one of the first experi- mental demonstrations of plasma heating via nonlinear processes as well as motivating important theoretical work on the anomalous absorption caused by these instabilities. The self-focusing instability of 62 43 high-power radio waves was also observed experimentally and theory suggests that ionospheric striation via self-focusing may be an important 9 environmental impact of a satellite power station. Electron beams launched by satellite borne electron guns provide interesting puzzles in the plasma physics of space-craft neutralization. '
1170 So far there has been insufficient analysis to turn these interesting observations into new concepts in plasma physics. The latest theoreti- cal work on neutralization is apparently that of Linson.^" Altogether, a variety of interesting and new scientific concepts have emerged from active experiments in the ionosphere. On the other hand, the contributions which particular active experiments have made, while valuable and unique, appear to be one-shot affairs, and not a basis for con- tinuing research. But the general technique of a sequence of new and well- conceived active experiments will teach us more about the ionosphere and magnetosphere than most passive observation programs. The reason is that a deeper and more thorough understanding "of space plasmas is required to predict the outcome of active experiments, rather than explain observations. Hence, if active ionospheric experiments are to continue to be of general intellectual significance, qualitatively new experiments must be invented. 1 Q For example, the Space Shuttle could be used for hypersonic gas releases. 6. Communications Applications of Ionospheric Instabilities The unexpected influence of ionospheric structures on transionospheric satellite communications"-^links has lead to much research activity, a recent conference,*â¢-* and an even more recent review article.â¢" Both measurements ,»' and nonlinear plasma instability calculations-^»->^ show that the wavenumber spectrum of ionospheric irregularities has a power-law form. From such ir- regularity models, the effects on various properties of communications chan- 11,47-49 nels have been calculated. The key role for plasma instability theory is to calculate the wavenumber spectrum of the irregularities so that one can obtain an understanding of how ionosphere effects on communications scale as a function of frequency as well as why and where instabilities occur.
1171 7. Summary Although observations have shown that the ionosphere is subject to several plasma instabilities, I hope that this chapter makes clear that a comprehensive study of ionospheric instabilities, their reaction back on the sources of free energy, and their effects on communication systems is just beginning. Even the equilibria states of the ionosphere have been modelled only by one-dimensional codes, and these require neutral winds, solar UV flux, and global electric fields as input parameters. Three-dimensional equilibria models, required for polar regions, have not yet been created. A theoreti- cal program to systematically study what instabilities can occur in the various equilibria states has not really commenced, and the nonlinear study of ionospheric instabilities is even further behind. Hence we can conclude that a truly profound understanding of the ionosphere as a dynamical system with turbulent transport processes awaits substantially more research. Much of this research will be of "true intellectual significance" in that it must necessarily take up basic questions of turbulence. The complementarity between ionospheric and hydrodynamic turbulence holds the promise of being able to achieve a deeper understanding of both of these phenomena. Acnowledgement: Discussions with R. G. Roble made substantive contributions to this chapter. I am also indebted to M. C. Kelly and S. L. Ossakow whose thoughtful reviews resulted in many improvements in the manuscript.
1172 References 1. J. R. Aarons, R. Allen, and T. Elkins, Frequency dependence of radio-star scintillation, J;. Geophys. Res. 72, 2891 (1967). 2. S. I. Akasofu and S. Chapman, Solar-Terrestrial Physics (Oxford, 3,972). 3. H. R. Anderson and R. R. Vondrak, Observations of Birkeland current at auroral latitudes, Rev. Geophys. and Space Phys. 13, 143 (1975). 4. B. B. Balsley, G. Haerendel, and R. A. Greenwald, Equatorial spread- F: recent observations and a new interpretation, J^. Geophys. Res. 7_7, 5625 (1972). 5. P. M. Banks, C. R. Chappel, A. F. Nagy, A new model for the inter- action of auroral electrons with the atmosphere: spectral degra- dation, backscatter, optical emissions and ionization, J_. Geophys. Res. 7_9, 1459 (1974). " 6. P. M. Banks, R. W. Schunk and W. J. Raitt, Temperature and density structure of thermal proton flows, _J. Geophys. Res. 79, 4691 (1974). 7. S. Basu and S. Basu, Correlated measurements of scintillations and in situ F-region irregularities from OGO-6, Geophys. Res. Lett. 3_, 68 (1976). 8. H. F. Bates, Atmospheric expansion through joule heating by horizon- tal electric fields, Planet. Space Science 21, 2073 (1973). 9. W. C. Brown, Satellite power stations: a new source of energy?, IEEE Spectrum 10, 38 (1973). 10. D. G. Cartwright and P. J. Kellogg, Observations of radiation from an electron beam artifically injected into the ionosphere, ^J. Geophys. Res. ^79, 1439 (1974). 11. E. Costa and M. C. Kelley, Calculations of equatorial scintillations at VHF and Gigahertz frequencies based on a new model of the dis- turbed ionosphere, Geophys. Res. Lett. J3, 677 (1976). 12. R. K. Crane, Spectra of ionospheric scintillation, J^. Geophys. Res. 81, 2041 (1976). ' ~ 13. R. K. Crane, Ionospheric scintillation, Proc. IEEE 65, 180 (1977). 14. R. E. Dickinson, E. C. Ridley, and R. G. Roble, Meridional circu- lation in the thermosphere I. Equinox condition, J^. Atmos. Sci. 32, 1737 (1975). 15. J. H. Doles, III, N. J. Zabusky and F. W. Perkins, Deformation and striation of plasma clouds in the ionosphere 3, numerical circulations of a multi-level model with recombination chemistry. J_. Geophys. Res. 81, 5987 (1976).
1173 16. P. L. Dyson, J. P. McClure and W. B. Hanson, jn situ measurements of the spectral characteristics of F-region ionospheric irregular- ities, J;. Geophys. Res. 7_9, 1497 (1974). 17. P. L. Dyson and J. Winningham, Topside ionospheric spread-F and particle precipitation in the day-side magnetospheric clefts, J. Geophys. Res. 79^, 5219 (1974). 18. D. S. Evans, High-velocity gas releases as a method for perturbing the upper atmosphere, J_. Geophys. Res. 79, 3882 (1974). 19. J. V. Evans, A review of F-region dynamics, Rev. Geophys. Space Phys. 1.3, 887 (1975). 20. D. T. Farley, Two-stream plasma instability as a source of irregular- ities in the ionosphere, Phys. Rev. Lett. 10, 279 (1963). 21. D. T. Farley, B. B. Balsley, R. F. Woodman and J. P. McClure, Equa- torial spread-F: implications of VHF radar observations. J_. Geophys. Res. 7j), 7199 (1970). 22. D. T. Farley and B. G. Fejer, The effect of the gradient drift term on type 1 electrojet irregularities, J_. Geophys. Res. 80, 3087 (1975). 23. J. A. Fedder, Effects of anomalous resistivity on auroral Birkeland current systems, Ann. Geophys. 32, 175 (1976). 24. S. R. Goldman, S. L. Ossakow and D. L. Book, On the nonlinear motion of a small barium cloud in the ionosphere, J_. Geophys. Res. 79, 1471 (1974). 25. J. M. Goodman, Ed., Effect of the ionosphere on space systems and communications, U.S. Government Printing Office, 1975. 26. R. A. Greenwald, Diffuse radar aurora and the gradient drift in- stability, J;. Geophys. Res. 79, 4807 (1974). 27. A. DasGupta and L. Kersley, Summer daytime scintillation and sporadic-E, j;. Atmos. Terr. Phys. 38, 615 (1976). 28. K. J. Jones, Wind, electric field and composition perturbations of the mid-latitude F-region during magnetic storms, ^J. Atmosph. and Terrest. Phys. 35, 1515 (1973). 29. M. C. Kelley, E. A. Bering, and F. S. Mozer, Evidence that the electrostatic ion-cyclotron instability is saturated by ion heating, Phys. Fluids 18. 1590 (1975). 30. M. C. Kelley, G. Haerendel, H. Kappler, A. Valenzuelo, B. B. Balsley, D. A. Canter, W. L. Ecklund, C. W. Carlson, B. Hausler, and R. Trobert, Evidence for a Rayleigh-Taylor type instability and upwelling of depleted density regions during equatorial spread-F, Geophys. Res. Lett. 3, 448 (1976).
1174 31. J. M. Kindel and C. F. Kennel, Topside current instabilities, £. Geo- phys. Res. J76, 3055 (1971). 32. D. M. Klumpar, J. R. Burrows, and M. D. Wilson, Simultaneous observa- tion of field aligned currents and particle fluxes in the post mid- night sector, Geophys. Res. Lett. ^, 395 (1976). 33. W. C. Knudsen, Megnetospheric convection and the high-latitude F2 ionosphere, J_. Geophys. Res. 79, 1046 (1974). 34. K. Lee, C. F. Kennel and F. V. Coroniti, On the marginally stable saturation spectrum of unstable type 1 equatorial electrojet irre- gularities, J;. Geophys. Res. 79, 249 (1974). 35. F. C. Liewer and N. A. Krall, Anomalous penetration of a magnetic pulse into a plasma, Phys. Rev. Lett. 30, 1242 ((1973). 36. L. M. Linson, Current-voltage characteristics of an electron-emitting satellite in the ionosphere, J. Geophys. Res. 74, 2368 (1969). 37. M. A. MacLeod, T. J. Keneshea, R.'S. Narcisi, Numerical modeling of a metallic sporadic-E layer, Radio Science 10, 371 (1975). 38. B. E. McDonald, T. P. Coffey, S. Ossakow, and R. N. Sudan, Prelimin- ary report of numerical simulation of type 2 irregularities in the equatorial electrojet, J. Geophys. Res. 79, 2551 (1974). 39. H. G. Moller, Backscatter results from Lindau II - the movement of curtains of intense irregularities in the polar F-layer, ^J. Atmos. Terr. Phys. 36, 1487 (1974). 40. F. S. Mozer, C. W. Carlson, M. K. Hudson, R. B. Torbert, B. Parady, J. Yatteau, and M. C. Kelley, Observations of paired electrostatic shocks in the polar magnetosphere, Phys. Rev. Lett. 38, 292 (1977). 41. E. Ott and D. T. Farley, Microinstabilities and the production of short-wavelength irregularities in the auroral F-region, J_. Geophys. Res. _80, 4599 (1975). 42. F. Perkins, Spread-F and ionospheric currents, J^. Geophys. Res. 78, 218 (1973). 43. F. W. Perkins, and E. J. Valeo, Thermal self-focusing of electromag- netic waves in plasmas, Phys. Rev. Lett. 32, 1234 (1974). 44. F. W. Perkins, Ion-streaming instabilities: electromagnetic and electrostatic, Phys. Fluids 19. 1012 (1976). 45. A comprehensive account of radio wave heating experiments is found in the special issue of Radio Science, Nov. 1974. 46. M. H. Rees and R. G. Roble, Observations and theory of the forma- tion of stable auroral red arcs, Rev. Geophys. Space Phys. 13, 201 (1975).
1175 47. C. L. Rino and E. J. Fremouw, Statistics for ionospherically dif- fracted VHF/UHF signals, Radio Science 8, 223 (1973). 48. C. L. Rino, R. C. Livingston, and M. E. Whitney, Some new results on the statistics of radio wave scintillation 1. Empirical evi- dence for Gaussian statistics, J_. Geophys. Res. 81, 2051 (1976). 49. C. L. Rino, Some new results on the statistics of radiowave scin- tillation 2, scattering from a random ensemble of locally homogen- eous patches, £. Geophys. Res. 81, 2059 (1976). 50. R. G. Roble, The calculated and observed dirunal variation of the ionosphere over Millstone Hill on 23-24 March 1970, Planet. Space Sci. 23, 1017 (1975). 51. A. Register, Nonlinear theory of 'type 1' irregularities in the equatorial electrojet, _J. Geophys. Res. 76, 7754 (1971). 52. A. Register and E. Jamin, Two-dimensional nonlinear processes asso- ciated with 'type 1' irregularities in the equatorial ionosphere, J. Geophys. Res. JJO, 1820 (1975). 53. T. D. Roglien and J. Weinstock, Theory of the nonlinear spectrum of the gradient drift instability in the equatorial electrojet. j;. Geophys. Res. 79_, 4733 (1974). 54. C. L. Rufenach, Power-law wavenumber spectrum deduced from iono- spheric scintillation observations, ^J. Geophys. Res. 77, 4761 (1972). 55. R. C. Sagalyn, M. Smiddy and M. Ahmed, High-latitude irregularities in the topside ionosphere based on Isis 1 thermal ion probe data, J_. Geophys. Res. ^9, 4252 (1974). 56. A. J. Scannapieco, and S. L. Ossakow, Nonlinear equatorial spread-F Geophys. Res. Lett. ^3, 451 (1976). 57. A. J. Scannapieco, S. L. Ossakow, S. R. Goldman, and J. M. Pierre, Plasma cloud late time striation spectra,â £. Geophys. Res. 81, 6037 (1976). 58. Soviet Killer Satellites: U.S. ponders a response, Science 193, 865 (1976). 59. D. J. Strickland, D. L. Book, T. P. Coffey, and J. A. Fedder, Trans- port equation techniques for deposition of auroral electrons, -J. Geophys. Res. _81, 2755 (1976). 60. D. F. Strobel, T. R. Young, R. R. Meier, T. P. Coffey, and A. W. Ali, The nighttime ionosphere: E-region and lower F-region. J^ Geophys. Res. _79, 3171 (1974). 61. R. R. Taur, Simultaneous 1.5 and 4-GHz ionospheric scintillation measurements, Radio Science, 11, 1029 (1976).
1176 62. G. D. Thome and F. W. Perkins, Production of ionospheric striations by thermal self-focusing, Phys. Rev. Lett. 32, 1238 (1974). 63. W. F. Utlaut, Ionospheric scintillation: a potential limitation to satellite communications - important unknown scintillation factors, AIAA 12th Aerospace Sciences Meeting, paper 74-56 (1974). 64. J. Weinstock and A. Sleeper, Nonlinear saturation of 'type 1' irre- gularities in the equatorial electro jet, J_. Geophys. Res. 77, 3621 (1972). 65. J. R. Winckler, R. L. Arnoldy, and R. A. Hendrikson, Echo 2: a study of electron beams injected into the high latitude ionosphere from a large sounding rocket, J. Geophys. Res. 80, 2083 (1975). 66. N. J. Zabusky, J. H. Doles, and F. W. Perkins, Deformation and striation of plasma clouds in the ionosphere, 2, numerical simula- tion of a nonlinear two-dimensional model. .J. Geophys. Res. 78, 711 (1973).
