An auroral substorm may include large variations in colour, intensity and location, many substructures, and may look completely disordered to the casual observer. Looking at many auroral substorms allows one to see that a pattern exists [e.g., Akasofu, 1968; Carlson and Egeland, 1995; McPherron, 1968]. Figure 17.5 illustrates this pattern schematically.
Figure 17.5: Schematic pattern of an auroral substorm [Akasofu, 1968].
For an observer near the usual auroral oval, the auroral display usually starts with one or more arcs with low intensity (1-10 kR) elongated in the geomagnetic east-west direction (panel A). The first disturbance is a sudden brightening of the most equatorward arc in the pre-midnight sector (panel B). This represents the ``onset'' of the auroral substorm, as also shown in magnetometer data used for the auroral AE and AU activity indices. The brightening extends rapidly westward and poleward (panel C), rapidly forming a broad region around the midnight sector where the aurora first brightened. This region is very dynamic with arcs appearing and disappearing, brightening, folding etc. The auroral brightness may now be several hundred kR. This region continues to expand poleward (panel D) and both westwards and eastwards, corresponding to the expansion phase of the auroral (and magnetospheric) substorm. Eventually the aurora ceases to expand polewards (panel E) and starts to dim and become more homogeneous, ending the expansion phase and starting the recovery phase. After about an hour auroral activity dims at lower latitudes and quiet arcs reappear, being called the recovery phase (panel F). Typically this process takes about 1-2 hours, with about 3 hours between auroral substorms (or breakups) during geomagnetically active periods.
Figure 17.6 illustrates this pattern with a sequence obtained from the Dynamics Explorer 1 spacecraft [Frank and Craven, 1988; Carlson and Egeland, 1995]. Each picture is separated by 8 minutes.
Figure 17.6: An example of an auroral substorm [Frank and Craven, 1988].