Introduction

Determination of the sequence of events that lead to substorm onset has been a critical problem and the subject of much debate since the discovery of the substorm more than 40 years ago. Recent analysis using THEMIS all-sky imagers (ASIs) [Nishimura et al., JGR, 2010] showed a distinct and repeatable sequence of events leading to onset which is initiated by a poleward boundary intensification (PBI) followed by an north-south (N-S) arc moving equatorward towards the onset latitude. In the present paper, we separate the pre-onset time sequence into three classifications. The most common of these was shown and described in Nishimura et al. This most common type of sequence, including an occasional variation of this type, and two other less common types are discussed here.

Observations

A sequence of selected images from four ASIs during a substorm auroral onset that occurred at 6:13:39 UT on 15 February 2008 near the Churchill meridian is shown in Figure 1. After multiple PBIs, a PBI occurred at 06:09:24 UT, ~5 min prior to auroral onset, and is identified as a bright auroral form near the center of the RANK imager field-of-view (FOV) in Figure 1a. The auroral form expanded both eastward and westward, and then an NS arc extending from the eastern edge of the PBI drifted equatorward and westward, forming an arc rotating clockwise (Figure 1b). The N-S arc contacted the growth phase arc in the FOV of the GILL imager. Finally, auroral onset started on the pre-existing growth phase arc to the west of the NS arc at 06:13:39 UT, immediately after the NS aurora contacted the growth phase arc. The auroral onset and small poleward and westward expansion can be seen clearly in Figure 1d (images taken 4.2 min after onset). The above time sequence clearly demonstrates the connection between auroral onset and the pre-onset aurora starting at the poleward boundary.

Figure 1. THEMIS ASI data during an auroral onset on 15 February 2008. ASIs used are RANK, FSMI, and GILL. White lines are isocontours of magnetic latitude (every 10? in solid lines) and longitude (every 15?). The blue line in each panel is the magnetic midnight meridian. The onset occurred at 06:13:39 UT.

Click here to enlarge the image.

Figure 2 shows ASI data from four imagers located in the central region of Northern America during an auroral onset on 4 February 2008, and illustrate a relatively uncommon type of event where an unusually narrow auroral oval develops near the onset longitude. As seen in Figure 2a prior to the onset, the auroral oval was narrow in latitude near the magnetic midnight meridian with the poleward and equatorward boundaries at ~71 deg and ~68 deg MLAT, respectively. A PBI was detected at ~72 deg MLAT in the FSMI imager 2.6 min prior to the auroral onset (seen clearly in Figure 2b). The onset occurred at 08:37:25 UT soon after the PBI expanded equatorward and contacted the growth phase arc (Figure 2c). The oval width decreased down to ~2 deg at onset.

Figure 2. THEMIS ASI data during an auroral onset on 4 February 2008. The format is same as Fig. 1. ASIs used are FSIM, FSMI, GILL and RANK. The onset occurred at 08:37:25 UT.

Click here to enlarge the image.

The two events discussed above are associated with PBIs and enhanced auroral brightness moving toward the onset locations. A small number of events, however, do not have such precursor auroral forms but instead show a sudden change in auroral motion near the onset location just before onset. Such an example is shown in Figure 3. The pre-onset feature in this event is characterized as equatorward bending of the equatorward portion of the Harang aurora, as seen in Figure 3b. The arc then started to fade and a new arc appeared just equatorward (Figure 3c). The pre-existing Harang arc merged into the new arc, and the new arc developed as a pseudo-breakup, which did not show a remarkable poleward expansion as seen in Figure 3d. The sequence suggests that an abrupt change in the plasma convection pattern and the current system may also initiate onset instability; a direct connection to flows from the polar cap boundary is not evident from the auroral observations.

Figure 3. THEMIS ASI data for an auroral onset on 11 February 2008 auroral onset. The format is same as Fig. 1. ASIs used are FSIM and FSMI. Onset occurred at 07:56:48 UT.

Click here to enlarge the image.

Conclusion

This study presents classifications of the time sequence of events leading to substorm onset using the THEMIS ASI array. The event sequence for most of the events shows a type illustrated in Figure 4a: The initial feature of pre-onset auroral activity is a PBI followed by an NS arc extending from the PBI then moves equatorward, roughly following the preexisting flow pattern around the Harang flow shear. The NS arc contacts a growth phase arc within the equatorward portion of the auroral oval. For about half of the events, onset occurs at the contact location.

A minor population of onset events shows a narrow auroral oval width which decreases to a few degrees in latitude just prior to onset. The sequence is summarized in Figure 4b. The poleward boundary near the magnetic midnight moves equatorward, and the auroral oval near midnight becomes much narrower than the dawn and dusk regions. PBIs with a finite latitudinal width directly reach onset location.

A small fraction of auroral onsets are not associated with any PBI or enhanced brightness moving toward onset location. As illustrated in Figure 4c, the pre-onset feature in this category is characterized as equatorward bending of the equatorward portion of the Harang aurora. The pre-existing arc then generally starts to fade and a new arc appears equatorward and develops as an auroral brightening. It suggests that an abrupt change in the plasma convection pattern associated with enhanced plasma flows leads to initiation of onset instability.

Figure 4. Schematic illustration of motion of pre-onset auroral forms and their relation to nightside ionospheric convection. (a) The time sequence for PBI and NS arc related, (b) poleward boundary contact, and (c) Harang aurora deformation onset events. Pink stars, NS-oriented pink line, and azimuthally-extended wavy lines indicate PBIs, NS-oriented arc and onset arcs, respectively. Blue arrows illustrate the plasma flow pattern inferred from pre-onset auroral motion. Numbers 1–5 show time evolution of pre-onset aurora (see text). Yellow and gray areas correspond to proton and electron precipitations.

Click here to enlarge the image.

Source

Nishimura, Y. et al (2010) Categorization of the time sequence of events leading to substorm onset based on THEMIS all-sky imager observations, IAGA Conference Proceedings, in press.

Please send comments/suggestions to

Emmanuel Masongsong / emasongsong@igpp.ucla.edu