The recent development of several new observational techniques as well as of advanced computer simulation codes has contributed significantly to our understanding of dynamics of the three-dimensional current system during magnetospheric substorms. This paper attempts to review the main results of the last decade of research in such diverse fields as electric fields and currents in the high-latitude ionosphere and field-aligned currents and their relationship to the large-scale distribution of auroras and auroral precipitation. It also contains discussions on some efforts in synthesizing the vast amount of the observations to construct an empirical model which connects the ionospheric currents with field-aligned currents. While our understanding has been greatly improved during the last decade, there is much that is as yet unsettled. For example, we have reached only a first approximation model of the three-dimensional current system which is not inconsistent with integrated, ground-based and space observations of electric and magnetic fields. We have just begun to unfold the cause of the field-aligned currents both in the magnetosphere and ionosphere. Dynamical behaviour of the magnetosphere-ionosphere coupling relating to substorm variability can be an important topic during the coming years.

The Relationship between Field-Aligned Currents and the Auroral Electrojets: A Review

The high-latitude spatial distributions of average signal intensities in 12 frequency channels between 4 Hz and 512 kHz as measured by the ac electric field spectrometers on the DE-2 spacecraft are analyzed for 18 mo of measurements. In MLT-INL (magnetic local time-invariant latitude) there are three distinct distributions that can be identified with 4-512 Hz signals from spatial irregularities and Alfven waves, 256-Hz to 4.1-kHz signals from ELF hiss, and 4.1-64 kHz signals from VLF auroral hiss, respectively. Overlap between ELF hiss and spatial irregularity signals occurs in the 256-512 Hz band. VLF hiss signals extend downward in frequency into the 1.0-4.1 kHz band and upward into the frequency range 128-512 kHz. The distinctly different spatial distribution patterns for the three bands, 4-256 Hz, 512-1204 Hz, and 4.1-64 kHz, indicate a lack of any causal relationships between VLF hiss, ELF hiss, and lower-frequency signals from spatial irregularities and Alfven waves.

High-latitude distributions of plasma waves and spatial irregularities from DE 2 alternating current electric field observations

The existence of an opening on the dayside of the earth's magnetic field through which solar plasma could penetrate was noted by Chapman and Ferraro in 1931, and although observational evidence of its auroral and ionospheric effects accumulated steadily over the years, the dayside cleft (or cusp) was not accepted as a real magnetospheric feature until 1971. The history of this development is described, and some reasons for the long delay in recognizing this profound ionospheric influence are given. The entry of magnetosheath plasma to ionospheric levels provides an essentially permanent dayside energy source for the F region that generates a rather constant 1–2 kR of 6300-A atomic oxygen emission, a weak hydrogen emission, and some other emissions. It also produces ionospheric irregularities and enhanced electron temperatures that can be detected by a variety of methods, all of this in a region normally centered at noon and 78° in invariant coordinates, with a latitude extent of a few degrees and a longitude extent of a few hours. The cleft ionization is an important source for the polar cap ionosphere, and the detailed nature of its effect depends upon the interplanetary field and on magnetospheric influences on the polar cap convection pattern. Some questions for further study are given in conclusion.

Dayside cleft aurora and its ionospheric effects

The Lagopedo experiments were the first in situ ion composition measurements of a chemically modified ionosphere. An artifical ionospheric hole was produced by the injection of H/sub 2/O and CO/sub 2/ into the O/sup +/ dominant F/sub 2/ region. These released chamicals increased the natural ionospheric depletion rate by approximately two orders of magnitude by changing the rate of molecular ion formation. As the rocket payload passed through the event cloud O/sup +/ was observed at a reduced density, and H/sub 2/O/sup +/, H/sub 3/O/sup +/, and O/sub 2//sup +/ were observed in enhanced amounts that varied with time after and distance from the event. H/sub 3/O/sup +/ was dominant at times immediately after the event, when the neutral water density was high, but rapidly decreased to 20-40% of the H/sub 2/O/sup +/. A computer model simulated the postevent O/sup +/ and O/sub 2//sup +/ distributions well, but it had mixed success with the H/sub 2/O/sup +/ and H/sub 3/O/sup +/ distributions. Measureable changes in ion composition levels were detected up to 40 km from the release point.

Chemically depleted F 2 ion composition: Measurements and theory

An analysis of measurements of the 4278-A, 5200-A, 5577-A, and 6300-A emissions obtained from a rocket launch through the postnoon dayside cleft into the polar cap ionosphere is reported. The 4278-A and 5577-A emissions show a clearly defined decrease in intensity at the poleward edge of the dayside particle precipitation, while the 5200-A emission extends well into the polar cap with only a slight decrease in intensity. The 6300-A emission displays a behavior intermediate between these extremes. The optical data are analysed by using a dynamical model of dayside cleft auroral processes incorporating transport by ion and neutral winds. It is concluded that neutral winds are responsible for modifying the spatial distribution of the 5200-A and 6300-A emission, while the ion drift component is negligible. In addition, it is found that (1) 80% of the 4278-A emission intensity is due to resonance fluorescence in the sunlit cleft ionosphere, (2) the reaction O2(c¹Σu−) + O → O2 + O(¹S) is consistent with the 5577-A data if the yield of O(¹S) from the dissociative recombination of O2+ is about 2%, and (3) the rate of the reaction N2+ + O → NO+ + N is apparently higher in the ionosphere than indicated by the laboratory measurements.

An analysis of the spatial distribution of dayside cleft optical emissions

As part of Operation Periquito, two sounding rockets were launched into regions of particle precipitation at invariant latitude 76°N and magnetic local time 13 hours from Cape Parry, Northwest Territories, Canada, in November 1975 On one of the flights, which did not cross any well-defined auroral arcs, the electron spectra were always soft, the thermal ion drift velocity was variable but generally westward at 500 m/s, and energetic particles (>70 keV) were always well above cosmic ray background, indicating a region of closed magnetic field lines The other flight crossed two auroral arcs Electrons with soft (mean energy of 200 eV) power law spectra were observed on the equatorward and poleward sides, whereas intense 1-keV precipitation was measured within the arcs The mean drift velocity of thermal ions was seen to change from 1500 m/s westward, parallel to the boundary, on the equatorward side to ∼200 m/s southward on the poleward side of the arcs Energetic (E > 70 keV) particle measurements showed that the high-latitude boundary was well poleward of the auroral arcs (Heikkila and Winningham's ‘cleft’ precipitation and ion convection velocity reversals), which indicates that the entire region was on closed magnetic field lines Comparison of these features with those from midnight aurora shows that the two are qualitatively similar, strongly suggesting that dayside and nightside arcs are different aspects of the same process No support was seen for the idea that direct penetration of magnetosheath particles is responsible for the dayside aurora

B. A. Whalen

Papers

Rocket‐borne measurements of particles and ion convection in dayside aurora