Meridional structures of electric potentials relevant to premidnight discrete auroras: A case study from Akebono measurements

The relation between latitudinal structures in particles and electric/magnetic field is studied along the premidnight meridian from the polar cap to low latitudes. Observations were made by the Akebono (EXOS D) satellite at altitudes of about 10,000 km on several successive orbits on February 7, 1990, when the magnetic activity was high ( Kp = ∼4) and a series of substorms took place. Combining the particle and the electric field measurements, height‐latitude structures of electric potentials relevant to premidnight discrete auroras are derived. The UV images of aurora are used to monitor the structure of precipitating electrons at the ionospheric heights. A large‐scale minimum of electric potential, where the convection velocity rotates clockwise, corresponded to a discrete aurora that delineated the poleward limit of auroras recorded by the Akebono UV imager. The poleward boundary of the plasma sheet, identified by the particle measurements, did not coincide with this clockwise rotation, and was located at a few degrees further poleward. A weaker but distinct potential maximum associated with a counterclockwise rotation of the flow was observed in the boundary region of the plasma sheet where accelerated electrons with intense lower‐energy ( E <1keV) populations were present, but their energy flux at the ionospheric heights is too low (<1erg s −1 cm −2 sr −1 ) to produce bright auroras. Upward flowing ion (UFI) conics and sometimes ∼10‐keV ions with a clear dispersion (where energy fell down toward lower latitudes) were observed in this boundary region of the plasma sheet. Thus the upward field‐aligned electric field should exist above the satellite level, and no significant acceleration took place below it. In the region of the potential minimum and the clockwise rotation, accelerated electrons were observed which had a “hole” distribution (characterized by the absence in the low ‐ energy component), and UFI beams were also seen. This indicates that the upward field‐aligned electric field existed both above and below the satellite. Because of this electric field the energy flux of the electrons became high enough to produce the discrete aurora. These observational facts suggest that electric potentials in the premidnight auroral zone are more structured in heights and latitudes during substorms. That is, in the boundary region of the plasma sheet (the region of potential maximum) the field‐aligned acceleration takes place at the higher altitudes. On the lower‐latitude side (the region of potential minimum), the acceleration occurs at both the higher and the lower altitudes. The latter corresponds to the region of maximum auroral intensity. In the boundary region of the plasma sheet, the magnetospheric electrons are accelerated by the higher‐altitude electric field only, and would be mostly reflected back by the magnetic mirror force. On the other hand, the electrons in the lower latitudes are further accelerated at the lower altitudes and could sufficiently account for bright auroral emissions. Thus the most poleward discrete aurora does not necessarily project to the plasma sheet boundary layer (PSBL), and could be situated significantly closer to Earth than previously expected.