March 31, 2001 geomagnetic storm
by Volker Grassmann, DF5AI, January 16, 2003
Number of Aurora QSOs versus the number of radio stations
The QSO database comprises about 2.200 observations corresponding to about 4.400 radio stations. In this number, double counts are already rejected (for example: station A reports QSO with station B and vice versa). However, many operators worked or heard more than one station on March 31, 2001, i.e. the number of individual radio stations is only 1.150 (some of these actually worked more than 120 QSOs). As a consequence, there is no linear relationship between the number of QSOs and the number of active stations: 50 percent of the QSOs is represented by 90 percent of the operators, i.e. another 50 percent of QSOs is contributed by only 10 percent of the stations (see fig. 1). However, it is believed that this result does not reflect the true scenario of March 31, 2001. This project cannot address all radio amateurs having managed dx QSOs on this particular day but the group of active VHF dxers connected to the internet portals and email refelectors, that group has indeed contributed the majority of the data.
Spatiotemporal distribution of the QSO activity
On March 31, 2001, the intensity of Auroral backscattering shows significant fluctuations which are present in the QSO activity too, of course (see fig. 2). In the U.S. and Canada, the first phase of high QSO activity occured between 0500 UT and 0730 UT peaking at 0630 UT. The same maximum was observed in Australia. On the other hand, the duration of the band opening in Australia was limited to two hours only. In the European sector, the phases of high QSO activity show a very similar shape compared to the North American sector, a time lag is however present corresponding to a delay of +1 hours in the morning and -2 hours in the afternoon (in the evening around 20 UT no time lag is present at all), see fig. 2.
From 09 UT to 1230 UT, the intensity of Auroral backscattering was almost zero in all geographical regions. At 13 UT, some Auroral backscattering was available but faded away very soon. Finally, European radio amateurs enjoyed an intense Aurora opening from 1430 UT to 19 UT and two smaller ones in the late evening (20 UT and 23 UT, respectively). The Aurora intensity peaked at 1600 UT providing more than 220 QSOs in a 30 minutes time frame. At the same time, the Aurora came back to the American radio amateurs corresponding to a time delay of about two hours. Surprisingly, the Aurora intensity dropped sharply in the American sector (see 1900 UT) but recovered very soon reaching another broad maximum at 2000 UT. Because the number of QSO data from the U.S and Canada is only a tenth in comparison to the data material from European hams, this sharp minimum reflects missing data, perhaps. On the other hand, this minimun corresponds to a minimum occuring in the European sector (19 UT) and the following maximum in the American sector is equivalent to the maximum observed in Europe (20 UT).
In fig. 2, the three categories, i.e. Australia, Europa and North America, respectively, can only provide rough information on the geographical distribution of Aurora QSOs. Fig. 3 studies the meridional occurence of Aurora backscatter in detail by plotting the geographical longitude of all radio stations versus the QSO time. Note, that each QSO contributes two dot marks, i.e. a dark and a light green marker corresponding to the station in the east and in the west of the propagation path, respectively. Although we cannot identify neither the true position nor the latitudinal distribution of the Aurora backscatter, fig. 3 can provide an instructive display of the global characteristic in Aurora dx communication on March 31, 2001.
The very first Aurora observation results from European radio amateurs around 0030 UT, see fig. 3. Hence, there was already some Aurora acitivity prior to the solar wind shock front which pushed into the magnetosphere at 0053 UT (see the storm history section of this web site). Apparently, it is this short horizontal green line in the North America sector (-90 degree longitude, 0115 UT - 0145 UT) which reflects the impact of the solar wind shock front, i.e. even a massive shock front does not necessarily result in immediate and eruptive dx band openings. In fact, Aurora dx openings are primarily caused by aftereffects which are strongly correlated to the actual direction of the interplanetary magnetic field, see the IMF section of the web site. A striking feature is displayed in the North American sector between 1630 UT and 2300 UT: initially, the Aurora opening was only available at the U.S. west coast (-120 degree, 1630 UT - 18 UT), i.e. the central and eastern part of the U.S. and Canada were both excluded, more or less. However, from 1815 UT to 23 UT the situation was reversed, i.e. no Aurora backscatter was available at the west coast but large spread radio Auroras occured between -60 degree and -100 degree longitude.
See also the discussion of the QSO distances in a separate section of this web site.