Evolution of Auroral Acceleration Types Inferred from Two-Satellite Coincidences

We investigated 1668 cases where one DMSP satellite observed an electron acceleration event and a second satellite subsequently observed an electron acceleration at the same location. The spatial coincidence required was tight, with a maximum separation of 0.1 MLAT and 0.15 h MLT. Time separations of 0-10 min were considered in 1-min bins. Auroral acceleration was flagged as either monoenergetic, broadband, or a mixture of both. Within the first temporal bin (0-1 min), the second satellite had a high probability of observing the same type of aurora as the first, establishing consistency. When the first satellite observed monoenergetic aurora, the second satellite also observed monoenergetic aurora (about 80% of the time), and this continued to up to about 10 minutes of UT separation. In most of the other 20% of the cases, the second satellite a mixture of monoenergetic and broadband acceleration. Thus monoenergetic aurora does not seem to typically evolve over a few minutes. However when the first satellite encounter was with broadband acceleration, the second encounter was highly time dependent, with broadband dominating the second encounter only in the first time bin, 0-1 min. Between 1-5 minutes, the probability of observing a mixture of auroral types jumped, and after 6 minutes, the auroral acceleration was nearly as likely to be monoenergetic as broadband. Finally, if the first satellite encountered a mixture of acceleration, the second encounter was progressively more likely to be monoenergetic aurora as time increased. These results are consistent with the idea that broadband aurora may be inherently a transient, and often progresses to monoenergetic aurora, while the latter is quasi-steady.