The nightside auroral zone is full of exotic, exciting and often unexplained phenomena. The aim of this Section is to introduce you to several of these phenomena, most of which are relevant elsewhere in space physics or in astrophysics, and not to explain them.
Large field-aligned currents are common in the auroral zone, as shown in Figure 17.9 [Iijima and Potemra, 1978], thereby connecting the ionosphere and the magnetosphere.
Figure 17.9: Pattern of field-aligned (Birkeland) currents for relatively quiet conditions
[Iijima and Potemra, 1978].
These currents are usually inferred from magnetometer data using Ampere's Law
. Note that the currents flow down on one
side of the auroral oval and then up on the other side, requiring current to flow around and across the
auroral oval as discussed above in the context of the auroral electrojets. Figure 17.10 shows that the
flux of downgoing electrons typically peaks in the current regions of the auroral zone and is small over
the polar cap. The figure also shows that electric fields exist pointing into and out of the auroral oval.
Figure 17.10: Observations from the S3-2 satellite during a dawn-to-dusk South Pole pass on 19 September 1976
[Harel et al., 1981; Wolf, 1995]: electron energy flux in the downswards direction, the transverse
magnetic field deflection, and the electric field.
An important result that is still not understood in detail is the existence of relatively steady, large (several to tens of kV), field-aligned potential drops in the downward and upgoing current regions. Why is this surprising? Because the plasma is essentially collisionless and so electron motion along the magnetic field would normally be expected to short out any potential drops , i.e., field lines are usually expected to be equipotentials. Figure 17.11 shows typical results from the FAST spacecraft [Ergun et al., 1998a].
Figure 17.11: High resolution FAST observations of the near-midnight auroral zone [Ergun et al., 1998a].
The dashed line separates the downward and upward current regions. (a) DC electric field perpendicular
to 
and along the satellite path, (b) Electric field in a 4 kHz bandwidth. (c) East-west
component of . (d)-(e) Electric field spectrograms of plasma waves. (f) Electron energy
spectrogram. (g) Electron pitch-angles, with 180 degrees upgoing and 0 and 360 degrees downgoing.
(h)-(i) Similar to (f) and (g) but for ions.
Note that the electrons are downgoing and close to mono-energetic, indicative of acceleration by a potential. Similarly, the ions are almost mono-energetic after about 49:50 UT, indicative of acceleration by a potential below the spacecraft. Figure 17.12 compares the potential derived by integrating the observed parallel electric fields with the ion beam energy, demonstrating quantitatively that the acceleration was by a parallel potential that endured for tens of seconds [Ergun et al., 1998a].
Figure 17.12: (a) DC electric fields perpendicular to and (b) upgoing ion energy flux
spectrogram with the parallel potential inferred from the observed electric field
superimposed [Ergun et al., 1998a]. The data are consistent with acceleration by a parallel
potential.
Numerous ideas have been proposed to explain the preservation of these potential drops in
regions with field-aligned currents: electrostatic double layers, anomalous resistivity
(which increases the effective collision frequency and impedes particle transport to short out the
potential), and magnetic mirroring of electrons and ions with different characteristics. An
alternative, very recent suggestion is that nonlinear, solitary wave structures support most of the
potential drops in the auroral field lines [Ergun et al., 1998b]. Figure 17.13 shows two
such structures: they have spatial scales of order several Debye lengths, carry potentials of
order 100 V each, move at speeds of order 3000 km s 
, and often come in long trains
capable of maintaining potentials of order several kilovolts.
Figure 17.13: A fast solitary wave at 0.5 
s resolution, moving anti-Earthward at
about 2000 km s [Ergun et al., 1998b].
These structures are consistent with those predicted for electron hole modes, corresponding to clumps of positive charge moving at speeds commensurate with the downgoing electrons [Ergun et al., 1998c].