The amplitude of the E‐W component Ew of the convection electric field in the nightside magnetosphere has been inferred from the observed cross‐L motions of whistler ducts within the plasmasphere. Several ducts distributed over 1–2 RE in L space and over ±15° around the longitude of the Eights, Antarctica, whistler station have been tracked simultaneously. The method appears capable of resolving fluctuations in Ew with period T ∼ 15 min and rms amplitude as low as 0.05 mv/m. For variations with T > 1 hour the method has a sensitivity of the order of 0.01 mv/m. Three case studies are presented, two of which illustrate convection activity associated with relatively isolated substorms. In these two cases Ew reversed from westward to eastward for a period following the decay of substorm bay activity. In the third case the substorm bay activity was prolonged, and Ew remained westward and at enhanced levels until local dawn. Evidence was found that, at least in a limited longitudinal sector, perturbing substorm Ew fields can penetrate deep within the plasmasphere. In two of the case studies comparisons of Ew and the interplanetary magnetic‐field θ component show evidence of a possible relation based on brief (≤ 1 hour) southward excursions but not on long preceding southward events. The growth of Ew can take the form of an initial brief (<15 min) positive surge followed by a larger surge that is simultaneous with the most active phase of the substorm. Certain of the pronounced increases in Ew were found to be coincident with activation or spreading of electrojet or auroral activity. In one instance low‐amplitude (<0.1 mv/m) presubstorm fluctuations in Ew with periods of the order of 30 min were found to correlate closely with ground‐observed midlatitude fluctuations in the magnetic H component. Calculated values of E‐field power spectral density from the tracking of two long‐lived (∼6 hours) whistler paths reveal considerable fine structure. The falloff with frequency roughly as f−2 agrees approximately with results from balloons, but the calculated spectral amplitudes appear lower than the balloon results by a factor of ∼4. The amplitudes from whistlers appear to be within the range identified by other workers as sufficient to drive radial diffusion in the radiation belts. The present research agrees with balloon measurements on the general presence of a westward field during substorms, but there is apparent disagreement on a number of details, including the post‐substorm reversal in Ew.

Information on the location of microburst source regions is limited. One measurement at L ≈ 8.5 placed the source within 4 RE of the ionosphere. Measurements at 5≲ L ≲6, though less conclusive, suggested that source regions may be located either near the equatorial plane or at higher magnetic latitudes along the field line. This paper reports simultaneous observations of bremsstrahlung X rays and VLF radiowave emissions that reveal a detailed correlation between electron microbursts precipitated in one hemisphere and chorus elements of rising frequency recorded at the conjugate point. The measurements were made at Roberval, Canada, and Siple Station, Antarctica (L ≈ 4.1), during magnetic substorms on July 9 and 15, 1975. The relationship between electron energy (50 ≲ E ≲ 200 keV) and wave frequency ( ≲ f ≲4 kHz), and the measured time difference (0.01 s ≤Δt ≤0.13 s) between detection of the electrons and waves at ionospheric conjugate points are consistent with near‐equatorial cyclotron resonance interactions occurring outside the plasmasphere. In both cases, the observations could be accounted for if a diffusive‐equilibrium distribution of electron density along the field line was assumed. The so‐called ‘collisionless’ (or R−4) model of electron density was not in accord with the observations. Some evidence is found for a separation of the wave growth and electron scattering regions. Evidence is also found indicating that the process of electron scattering requires a finite time, up to ∼80 ms under the conditions of these observations. The present results suggest that microburst generation regions are located within 20° of the equator on subauroral field lines.

No abstract

‘Fast hisslers,’ or very brief bursts of auroral hiss dispersed in the whistler mode, have been found to occur in substorms at Byrd station, Antarctica. Broad band VLF data recorded on tape during breakup phases of substorms were monitored aurally. In two of 19 substorm breakup phases observed at Byrd station, fast hisslers were heard. The data were subsequently spectrum analyzed to permit measurement of whistler mode dispersion, from which the altitudes of their origin were estimated. Some of the fast hisslers exhibited a ‘nose’ or nonextremal frequency of earliest arrival. This feature permitted an estimate of altitude of origin that is nearly independent of the field line electron density distribution model. Altitudes of origin of 1800–30,000 km are deduced. Some fast hisslers have been observed at phases other than the breakup phase. Fast Fourier transform spectra show that fast hisslers are indistinguishable from dispersed short pulses of band‐limited white noise. Incoherent Cerenkov radiation from precipitating auroral electrons is of insufficient intensity to account for this phenomenon.

Auroral electrojet parameters measured by the Chatanika incoherent scatter radar have been compared with VHF backscatter observed in a comparable spatial area by the 50‐MHz auroral radar located at Anchorage. We find that the D region absorption, occurring in concert with the morning (westward) electrojet, can significantly decrease the observed scatter amplitude. If the amplitude is corrected for absorption effects, we find that the scatter varies approximately linearly with either eastward or westward current density with the same slope for both periods. A surprising result is that at times, relatively large north‐south current densities do not give rise to detectable backscatter. The Chatanika electric field measurements indicate that during the morning period, substantial backscatter amplitudes occur for southward fields of only 10 mV/m. In contrast, during the evening period the backscatter amplitude is zero or very low until the northward field is at least 25 mV/m. This indicates that most of the auroral backscatter that we observe in the evening period occurs under conditions when the two‐stream plasma instability may be operative. However, in the morning period the backscattering irregularities are strongly generated under conditions associated with either gradient drift or two‐stream instabilities.