The detailed analysis of an isolated dispersionless substorm is performed on the basis of field and particle data collected in situ by the geostationary satellite GEOS 2 and of data from ground-based instruments installed close to the GEOS 2 magnetic footprint. These data give evidence for (1) quasi-periodic variations of the magnetic field configuration, which is alternatively taillike and dipolelike, (2) in-phase oscillations of the flux of energetic electrons, which is high when the configuration is dipolelike and vice versa, (3) a gradient in the flux of energetic ions, which is, on the average, earthward but undergoes large fluctuations around this average direction, and (4) large transient fluctuations of the quasi-dc electric field, which reverses its direction from eastward to westward. It is shown that these results are consistent with the development of an instability which leads to a westward propagating "wave." The source of the instability is the differential drift of energetic electrons and ions in a highly stressed magnetic field configuration (in a high /3 plasma). Evidence is given for a system of localized field-aligned currents flowing alternately earthward and equatorward at the leading and trailing edges of the westward propagating wave. This current system resulting from the temporal development of the instability produces the so-called Pi 2 pulsations, at the ionospheric level. The closure of this current system in the equatorial region leads to a current antiparallel to the tail current, and therefore to its reduction or cancellation. This reduction/ cancellation of the tail current restores the dipole magnetic field (dipolarization) and generates a large westward directed induced electric field (injection). Hence, dipolarization and injection are the consequences of the instability. Finally, it is suggested that the westward traveling surge observed simultaneously by all-sky cameras, close to the magnetic field of GEOS 2, is the image of the instability in the equatorial region transmitted to the upper atmosphere by precipitating electrons. 12 18 06 AFTER EXPANSION 12 Fig. 1. Schematics adapted from Akasofu [1977]. (a) The auroral arcs as they appear at breakup, at the northern boundary of the diffuse auroras, at a magnetic latitude typically between 65 ø and 70 ø. The shaded areas indicate the regions where diffuse auroras are observed. (b) After breakup, the discrete auroral forms expand to the west, to the east, and to the north. At the end of the expansion phase, the discrete arcs cover a region which typically extends from 650-70 ø to 750-80 ø magnetic latitude. diffuse reverse current has been identified west and more clearly east of the surge [e.g., Kozelova and Lyatskiy, 1984]. According to Baumjohann et al. [1981], the ratio between the height-integrated Hall and Pedersen conductivities increases in the westward traveling surge region, due to precipitation of relatively high energy electrons (10-20 keV). This kind of precipitation was directly observed, for example, by Kremser et...

Because of the high time resolution of many of its scientific instruments, the Viking satellite has detected the presence of intense wave bursts in the cusp/cleft regions, at altitudes between 1 and 2 Earth radii. These bursts are closely related to the acceleration processes of charged particles and are characterized by a strong plasma turbulence illustrated by sporadic and intense broadband electrostatic noise detected in the 4-to 700okHz frequency range of the high-frequency wave receiver. The duration of such bursts is small, typically a few hundreds of milliseconds, corresponding to a spatial width less than 1 km at the satellite altitude. The low-frequency part of this noise is always the most intense, with electric field amplitudes of a few tens of millivolts per meter. This noise is correlated with 0.1o to lokeV electron beams and with small-scale field-aligned currents and is always associated with low-frequency (--• 1 Hz) electric field fluctuations. Perpendicular acceleration of ambient plasma ions occurs in connection with the bursts. The impulsive broadband waves are detected at the equatorward edge of the polar cusp and more generally in the entire cleft region. The central part of the cusp is characterized by an increase of the frequency cutoff of the hiss emissions, related to the increase of the electron density. In addition, electrostatic electron cyclotron waves are also recorded. Terrestrial kilometric radiation is often detected in the cusp/cleft regions usually in close correlation with particle acceleration processes. The Viking satellite tends to have been mostly above the generation regions of these electromagnetic waves in the dayside auroral oval, which makes it difficult to localize precisely their source. 1. by upward flowing ionospheric electrons and locally produced ion conics [Lundin, 1988]. The cleft regions are characterized by discontinuous magnetosheath plasma injection, and the electron fluxes are considerably more energetic and structured than those found in the central part of the cusp. The high-altitude auroral plasma is collisionless, so that waves produced by plasma instabilities provide the necessary interactions to transfer energy and momentum between particles. Recently, the importance of low-frequency (--•1 Hz) electric field fluctuations for energization of both ions and electrons in the dayside auroral regions has been extensively studied [Andre et al., 1988; Hultqvist et al., 1988].The purpose of this paper is to presenthigh-frequency wave data in the 4-to 700-kHz range, recorded by the V4H experiment on Viking in the cusp/cleft regions, and to discuss the relationship between those waves and the complex plasma processes previously described;Indeed, there have been relatively few published obse.•ations of highfrequency plasma waves in this part of the•magnetosphere [Gurnett et al., 1983; Lin et al., 1984], Starting with the lowest-frequency. waves, the main wave modes occurring in the investigated frequency range are as follows:Lower hybrid waves. These are particu...

We present results from the particle measurements made on board the Viking satellite. For the period of interest the Viking orbits covered at high latitudes the whole dayside sector. Data from the Viking V-3 particle experiment acquired during the Polar Region Outer Magnetospheric International Study period (March 15 to June 15, 1986) have been used to study the extension of the cusp and cleft in magnetic local time and invariant latitude, and furthermore, their dependence on solar wind and interplanetary magnetic field parameters. The study is limited to the MLT range from 0900 to 1500 and to invariant latitudes (ILAT) from 74 ø to 82 ø. This region is divided into bins of size (15 min MLT x 0.2 ø ILAT). We concentrated on the region where magnetosheath solar wind plasma penetrates more direcfiy into the magnetosphere and is measured at Viking altitudes (•-13,000 km). This region is called the "cusp proper" (Lundin et al., 1988; Newell and Meng, 1989), to be distinguished from a broader region denoted the cleft, where more energetic particles are observed. Statistically, we find the cusp proper to extend from invariant latitudes of 75 ø to 82 ø and magnetic local times from 0930 to 1400 MLT. The width in ILAT is found to be on average •-2 ø and in MLT = 2 hours. It is shown that a clear correlation exists between the densities in the cusp proper calculated from the Viking V-3 experiment in the cusp proper and those in the solar wind calculated from IMP 8 measurements. It is also shown that the position of the cusp proper in MLT depends on the sense of the By component of the interplanetary magnetic field (IMF By), giving a well-defined displacement of the region of maximum occurrence toward earlier MLTs for IMF By <0 and a less defined displacement toward later MLTs for IMF By >0. INTRODUCllON Ever since the discovery of the cusp by tteikkila and Winningham [1971], and Frank [1971], the cusp has remained an important, challenging, and interesting magnetospheric region to study. According to the definition of the cusp, magnetosheath solar wind particles in this region are expected to have direct access to the magnetosphere. Some of these particles may precipitate into the ionosphere, whereas others will mirror and drift tailward into the plasma mantle. The cusp has been identified from low-altitude satellite/rocket measurements (height < 1000 km), e.g., by Reiff et al. [1977], Meng [1982], and Newell and Meng [1988], these latter two using measurements from DMSP satellites, from measurements with the highly eccentric satellite HEOS at the magnetopause (entry layer region) by Haerendel and Paschmann [1975], Formisano and Bavassano Cattaneo [1978], as well as measurements at mid-altitude (3-5 Re), from DE 1 by Burch et al. [1982]. The observations in the cusp are often described in terms of the model used by Rosenbauer et al. [1975] to explain the observed spatial distribution of the plasma mantle or entry layer plasma. Magnetosheath ions having a wide energy distribution simultaneously gain access to the cusp alon...

Measurements of positive ions (0.1-10 keV) from the polar-orbiting Viking satellite have been obtained in the cusp/cleft region and have been tabulated in MLT versus invariant latitude plots to get a synoptic picture of the occurrence of upgoing auroral positive ions. A distinction was made between ion distributions with peak fluxes along B (ion beam) and those exhibiting flux maxima that are not field-aligned (conics). Both beams and conics are shown to be common auroral phenomena, whose frequencies of occurrence in MLT, invariant latitude, and altitude were studied. During the period of study (March-June 1986) the ion beams were more frequent (about a factor of 2) in the dusk sector than in the dawn sector. The ion conics, however, were more frequent (about 2-3 times) in the dawn sector than in the dusk sector. This effect seemed to increase with magnetic activity but was mostly unchanged with the sign of the interplanetary magnetic field By component. An investigation was also made of the invariant latitude dependence for beams and conics, where the dawn sector beams have a tendency to be located toward higher invariant latitudes for both positive and negative By components. The ion beams were observed primarily above 5000 km with a frequency of occurrence increasing with altitude up to the satellite apogee at about 13,500 km. The ion conics were observed from above 7000 km to be steadily increasing in altitude. seen in Figures 3-5 for UFIs, ion beams, and ion conics.A similar location displacement has also been observed by studying the location for the cusp at different signs of the By component. This cusp study has been reported elsewhere by 13 o oo ß **%*ø ß ß o ß ß ß ß ß I I I I I 810 I % 76 78

No abstract