During 0230-0315 UT, auroral roar is detected at all five stations (67-79) but the peak intensity is observed at Churchill (69). At this time the auroral zone inferred from the magnetometers covers the latitude range 60-77; in other words, the strongest auroral roar lies nearly 8 equatorward of the poleward edge of the auroral oval.
A substorm occurs at 0315 UT, and during the recovery phase the latitude of most intense auroral roar decreases from 71 to 69, in parallel with but equatorward of the poleward electrojet boundary, which shifts from 77 at 0330 UT to 71 at 0415 UT. Briefly, at 0425 UT the auroral roar is most intense at Gillam (67) which is equatorward of the low-latitude electrojet boundary. During 0415-0500 UT the auroral emissions shift poleward from 69 to 74, again coincident with a similar poleward shift in the poleward electrojet boundary, except that in this interval the poleward electrojet boundary and the latitude of strongest emissions nearly coincide.
From 0500 to 0600 UT the latitude of peak auroral roar intensity alternates between Churchill (69) and Arviat (71), and the poleward electrojet boundary remains far to the north, near 76. Between 0600 and 0700 UT the peak emissions are concentrated at Churchill, except for a brief time near 0610 UT when the most intense emissions occur at Baker Lake. During this time the poleward electrojet boundary generally shifts equatorward but still lies poleward of the latitude of most intense auroral roar. After the substorm onset at 0700 UT the boundary shifts around but generally moves poleward from 69 at 0700 UT to 77 after 0800 UT. During this time the emissions also shift poleward, occurring at Baker Lake and Taloyoak during the substorm recovery. During this final part of the record (0730-0900 UT) the more usual pattern reasserts itself; that is, the auroral roar occurs at or poleward of the poleward electrojet boundary.
The first substorm during the time period of this event occurs at 0315 UT when the distant propagating natural and anthropogenic signals abruptly cease in all five stations. The electrojet boundary moves rapidly poleward, indicating an expanding auroral oval, and the magnetic perturbations due to the electrojet currents increase moderately from 119 to 275 nT at 0310-0325 UT. During the recovery phase the electrojet boundary generally moves equatorward as the emissions reappear at the northern stations first: Baker Lake (0330 UT), Arviat (0345 UT), Churchill (0405 UT), and finally at Gillam (0420 UT). The impulsive signals in the Arviat PSFR data are not related to the aurora and are caused by distant atmospherics.
During the second substorm commencing at 0703 UT, MF-burst emissions are recorded at Arviat, Baker Lake, and Taloyoak just prior to the dramatic loss of signals at all five stations, which is frequently observed during substorm onsets LaBelle:94. The intensity of the electrojet currents experiences a more dramatic increase than the previous substorm from 138 to 397 nT at 0640-0720 UT. Again, during the recovery phase the signals clearly reappear first at the more northern stations; the gap in emissions at Taloyoak is only a few minutes.