Perhaps one of the most interesting outcomes of the simulation was the appearance of field-aligned currents flowing into and out of the high-latitude ionosphere. We have observed two, and possibly three, types of currents. One type might be referred to as a growth phase current’. This is the simple up-down pair seen at t = 14.4 and 21.6. These currents are fairly broad and are likely due to the polarization set up by the breaking of the earthward plasma convection at the inner edge of the plasma sheet. Downward-moving electrons could carry the upward current. Because the electron current sheet is reasonably broad the current could be carried by low energy electrons, which would produce very little optical emission. Upward electrons or downward moving ions would carry the downward current. Downward moving ions would probably be energetic and would likely produce some optical emission.

Figure 16 shows a photometer record from Poker Flat, Alaska during a substorm growth and expansive phase sequence. The top trace shows the Hb emission line of atomic hydrogen, which is an indicator of proton precipitation. In all-sky camera photographs this feature shows up as a diffuse arc that moves equatorward during the substorm growth phase. We would identify the equatorward of the field-aligned current pairs in the t = 21.6 frame of Figure 9 with this diffuse proton precipitation. All-sky camera photographs for that period show some auroral activity in the form of discrete arcs propagating into the meridian of the photometer from activity to the east of Poker in the 0900 to 1000 time frame.

A substorm expansion occurs at or slightly before 1000, and all-sky camera photographs show discrete auroral forms propagating from the east into the field of view of the photometers. We would identify the filamentary field-aligned currents seen in the t = 28.8 to 50.4 frames of Figure 9 with the expansive phase beginning at 1000. The hybrid code, which treats electrons as a fluid, is, of course, not able to simulate directly the electron acceleration process. Also, the grid used is incapable of resolving individual auroral forms. However, if in the real world the field aligned current filaments are generated with thickness down to the electron inertial length, then electrons would have to be accelerated in order for them to move fast enough to carry the current. We also note that the figures show both upward and downward currents. These could result in alternating bands of proton and electron precipitation. Reference to the Hb trace of Figure 16 indicates that proton precipitation does indeed participate in the poleward expansion. Figure 17 shows another indicator of the existence of upward and downward filamentary currents. The figure shows both discrete bright and black bands superimposed on a diffuse background precipitation. Kimball and Hallinan [1998] report that the black bands also exhibit a curl structure. The vorticity associated with black bands is opposite that of the bright auroral forms. This suggests that the electric polarity of the black bands is opposite the polarity of the discrete auroral forms. Kimball and Hallinan also report that the black aurora is found in regions of mixed electron and proton precipitation.

We have seen that the filamentary currents appear to originate near the equatorial plane just inside the expanding dipolarization front. The fact that the disturbance appears to propagate along field lines seems to provide a simple explanation of the poleward progression of auroral forms during the expansive phase: The poleward progression is a result of the outward expansion of the dipolarization region. Also since we have continued convection, auroral forms should be seen as drifting equatorward. In the development of the simulation, a number of examples were run in which the density of the fluid plasma was considerably higher than that used in the simulations presented here. This higher density slowed the propagation of Alfvén waves. The result was that equatorward convective drift speed nearly matched the rate of poleward/outward progression of the current filaments, so no poleward expansion was seen at the Earth’s surface.

The third type of current may be seen in the t = 36.0 frame of Figure 9 as the diffuse and most equatorward of the field aligned currents. These currents are also present in the t = 28.8 and 43.2 (not shown) and 50.5 frames, but at greatly reduced intensity in comparison to the other currents. These currents are upward, meaning that downward-moving electrons may carry them. We do not have a ready explanation for these currents, but it is tempting to associate them with the mid-latitude SAR arcs first described by Barbier [1960]. These arcs are primarily in the 6300 Å line of atomic oxygen, and they are attributed to the downward flux of hot electrons seen during magnetic storm conditions [Walker and Rees, 1968]. The suggestion is that the downward flux might be due to the type of upward currents seen in the simulations. However, the simulation suggests that these probably have a substorm time scale, whereas the observed SAR arcs have generally been associated with the main phase of larger scale magnetic storms with a longer time scale.

Contents

References

Section 5