5:15 PM - 7:15 PM
[PEM13-P20] Characteristics of high-energy electron precipitation before and after substorm onset: superposed-epoch analysis of spectral riometer data in Finland
Keywords:Riometer, Substorm list, Superposed Epoch Analysis: SEA
It has been known that, immediately after substorm onsets, highly energetic electrons precipitate from the magnetosphere and ionize the D-region ionosphere. Such transient, low-altitude ionization associated with substorm onset is detected as an enhancement in cosmic noise absorption (CNA) observed by riometers, called as an "absorption spike." To reveal the characteristics of absorption spikes, we conducted a statistics of CNA before/after substorm onset using multi-frequency riometers capable of broadband observations in the 20–55 MHz frequency range and a substorm list compiled by Forsyth et al. (2015). From the substorm list, we identified 359 substorms, that occurred during 20—24 UT, and the Superposed Epoch Analysis (SEA) was applied to the CNA data from Kilpisjärvi (KIL), Finland (69.07N, 20.75E). Through this approach, we could clarify the average temporal variation of CNA around substorm onset.
First, it is revealed that weak electron precipitation is observed ~1 h before onset, and CNA slightly decreases ~20 min before onset. After onset, CNA increased, reaching its peak within ~20 min, followed by a gradual decay over about three hours. Examining the frequency dependence of CNA, we found that the increase in CNA was stronger in the lower frequency range than in the higher frequency range. Furthermore, when we sorted the data based on the distance between the onset region and the riometer observation site, we found that signature of absorption spikes occurs later at locations farther away from midnight toward the morning sector. In general, substorm is known to occur more frequently around magnetic local midnight (~23 MLT), and there is an offset of ~3h between MLT and UT in Kilpisjarvi (MLT ~ UT + 3). Therefore, substorms that occur after 21 UT (e.g., at 24 UT) are expected to show a time delay due to the larger distance from the onset region. In the presentation, we will show the results of SEA with substorms in all UT time, which will minimize the impact of exceptionally rare substorms that do not occur before midnight. We will also discuss the seasonal dependence of electron precipitation associated with substorm onset and further examine the characteristics of electron precipitation from the magnetosphere during substorms.
First, it is revealed that weak electron precipitation is observed ~1 h before onset, and CNA slightly decreases ~20 min before onset. After onset, CNA increased, reaching its peak within ~20 min, followed by a gradual decay over about three hours. Examining the frequency dependence of CNA, we found that the increase in CNA was stronger in the lower frequency range than in the higher frequency range. Furthermore, when we sorted the data based on the distance between the onset region and the riometer observation site, we found that signature of absorption spikes occurs later at locations farther away from midnight toward the morning sector. In general, substorm is known to occur more frequently around magnetic local midnight (~23 MLT), and there is an offset of ~3h between MLT and UT in Kilpisjarvi (MLT ~ UT + 3). Therefore, substorms that occur after 21 UT (e.g., at 24 UT) are expected to show a time delay due to the larger distance from the onset region. In the presentation, we will show the results of SEA with substorms in all UT time, which will minimize the impact of exceptionally rare substorms that do not occur before midnight. We will also discuss the seasonal dependence of electron precipitation associated with substorm onset and further examine the characteristics of electron precipitation from the magnetosphere during substorms.