In a set of 17 low-to mid-latitude crossings of the dayside and near-dayside magnetopause, Imp 6 plasma measurements show 11 cases of decreases in magnetosheath density just outside the boundary which are consistent with plasma depletion owing to magnetic flux tube compression as the field becomes draped against the magnetopause. Pressure anisotropies in the sense p• > p• are a predicted result of the plasma depletion and field compression, and such anisotropies are observed. Application of the mirror instability criterion, which predicts growth of slow mode magnetoacoustic waves for values of p•/p• greater than a critical value, suggests that dayside magnetosheath plasma is usually unstable. One of the seventeen cases shows long-period waveforms in the 100-s density data that are 180 ø out of phase with simultaneous waveforms in the magnetic field strength. We interpret these dat, a as the signature of slow mode magnetoacoustic waves.

Three fast search routines to be used in the encoding phase of vector quantization (VQ) image compression systems are presented. These routines, which are based on geometric considerations, provide the same results as an exhaustive (or full) search. Examples show that the proposed algorithms need only 3-20% of the number of mathematical operations required by a full search and fewer than 50% of the operations required by recently proposed alternatives.

The ecliptic plane components of the pressure tensors for low‐energy (<30 keV) electrons and protons have been examined for more than 400 hours of quiet time plasma sheet data from Imp 6 and Imp 8. Individual high time resolution (≲100 s) measurements show that while at times the tensors may be clearly anisotropic with the pressure parallel to the magnetic field (P∥) being as much as 1.5–2.0 times the pressure perpendicular (P⊥), the anisotropies are not usually of this magnitude and are typically below the value thought necessary to balance the tension of the magnetic field. One‐hour averages of the particle counts have also been calculated. While these averages mask more rapid changes due to variations in the distributions and the magnetic field and can yield only a lower limit to the true anisotropy, examinations of the individual measurements indicate that they do provide a reasonable summary. On this time scale, the protons are nearly isotropic: the ratio of the maximum ecliptic component of the proton pressure tensor to the minimum is typically ≲1.1, and there is no apparent tendency for the distributions to be field aligned. On the same time scale, the electron distributions do show a tendency to be field aligned with P∥/P⊥ ≥ 1.2 approximately 25% of the time. Due to the low energy density of the electrons, however, this anisotropy is not itself sufficient to balance the tension of the magnetic field.

Artificially stimulated VLF emissions (ASE's) are emissions triggered in the magnetosphere by the whistler mode signals from VLF transmitters. These emissions may be separated into two classes, rising and falling, depending on whether the final value of df/dt is positive or negative. Several hundred ASE's triggered by three transmitters (NAA, Maine, 14.7 kHz, 1 MW; Omega, New York, 10.2 kHz, 100 W; Sipie station, Antarctica, 5.5 kHz, 400 W) have been analyzed using the fast Fourier transform with a filter spacing of 25 Hz and an effective filter width of about 45 Hz. The study was limited to the initial frequency-time behavior of ASE's. Averages taken over many events reveal that both rising and falling tones show the same initial behavior. The emissions begin at the frequency of the triggering signal rather than at an offset frequency, as has previously been reported. Both tones initially rise in frequency, falling tones reversing slope at a point 25-300 Hz above the triggering signal. The slope of rising tones, particularly those triggered by NAA, often abruptly levels off in this same tYequency range; as a result, a short (•40 ms) plateau is formed that precedes the final rising phase. The initial frequency offset commonly observed in individual events appears to result from the frequent coincidence with this plateau of a peak in amplitude. Emissions stimulated by all three transmitters show essentially the same features; this finding indicates that their frequency behavior does not depend strongly on transmitter power. The process appears to be asymmetric in frequency; no evidence of initial growth below the triggering frequency has been found.Artificially stimulated VLF emissions (ASE's) represent only one of the many diverse forms of discrete ¾LF emissions. Since they are triggered by signals from ¾LF transmitters, however, they have the potential of becoming a valuable controlled tool for the study of the magnetosphere. The triggering of emissions by ¾LF transmitters (NPG, 18.6 kHz, •2 X 105 W; NAA, 14.7 kHz, • 106W) was first reported by Helliwell et al. [1964]. Kimura [1968] extended the known range of the phenomenon to lower frequencies (10.2 kHz) and lower power (•100 W) with his observation that the New York Omega transmitter also stimulated emissions. In 1973 the frequency range was lowered further when the Stanford transmitter at Siple station, Antarctica, was employed in triggering emissions in the band from 2.5 to 7 kHz [Helliwell and Katsufrakis, 1974]. Investigations subsequent to the original discoveries have revealed several intriguing features of ASE's. First, Morse dashes (150 ms long) trigger ASE's much more frequently than do Morse dots (50 ms) [Helliwell et al., 1964; Helliwell, 1965; Lasch, 1969]. Second, triggering tends to occur most often when the transmitted frequency is within a few percent of one half the minimum gyrofrequency along the whistler mode path [Carpenter, 1968; Carpenter et al., 1969]. Third, the emissions often appear (misleadingly, as will be seen) to begin at a point...

Ionospheric convection at auroral circulation and its mapping into a two-cell eyelatitudes has been examined during a series of tem at low altitudes in the polar ionosphere. long duration experiments with the Chatanika, Direct observation of the high-latitude con-Alaska, incoherent scatter radar. These experi-vection pattern was first made by rocket borne ments have been carefully designed to obtain max-probes [Mozer and Bruston, 1967], balloon elecimum latitudinal coverage (56øA -75øA) while ma-trio field measurements [Mozer and Serlin; 1969] intaining a temporal resolution of 30 min in and barium releases at rocket altitudes [Wescot% order to resolve the effects of individual sub-et al., 1969]. Electric field probes flown on storms on the convection pattern. Design criter-polar orbiting satellites provided a wealth of ia for the experiments are described together new data on the magnitude and configuration of with presentation of observational data acquired the electric fields responsible for high-latitude during 400 hr of radar operations during various plasma convection [e.g., Maynard and Heppner, levels of geophysical disturbance. The data ac-1970]. These observations were combined into emcentuate the repeatability of the gross features pirical models of the convection electric field of the auroral convection and its basic conformi-in the polar cap and auroral region; the models. ty to the two-cell pattern predicted from the agreed with the general two-cell pattern deslarge-scale magnetospheric circulation. For mod-cribed by theory [e.g., Heppner, 1977; Stern, erate to active geomagnetic conditions uniform 1977]. sunward convection with velocities of 800 -12OO Other techniques to measure the auroral conm/s spans the 62•A to 72•A latitude band at both vection were being developed at the time of these dawn and dusk. In quieter circumstances the sun-early electric field observations. Direct obserward convection continues in this region, but has vation of the drifting ionospheric plasma, for smaller speeds and is centered at higher lati-example, has been made by using rocket-borne ion tudes. Little evidence of a 'throat' in the day-drift meters and, more recently, the highly sucside convection pattern is seen at latitudes cessful drift meters flown on the Atmosphere Exbelow 75•A except in very disturbed circum-plorer satellites [e.g., Hanson and Heelis, stances. During several experiments coordinated 1975]. The STARE radar observes the drift of operation of the Chatanika and Millstone Hill, electrostatic waves in the E region and thus the Massachusetts, radars permitted the simultaneous convection electric field [Nielsen and Greenwald, observation of the auroral convection pattern at 1979]. The technique of observing ion drift veltwo different local times. Substorms affect the ocity at high latitudes by measuring the Doppler convection at all local times and appear to gen-shift of incoherent scatter radar signals reerally enhance the rate of convection without flected from the ionospheric plasma, described by se...