Solar wind control of auroral substorm onset locations observed with the 
IMAGE-FUV imagers

J.-C. Gerard, B. Hubert, A. Grard, M. Meurant, S. B. Mende


The FUV imagers on board the IMAGE satellite provide multispectral snapshots of
the polar region every 2 min. The combination of the Wide-Angle Imaging Camera
(WIC) with SI12 (Doppler shifted Lyman-ƒ¿) and SI13 (135.6 nm) spectral imagers
is used to discriminate between the electron and the proton aurora. We describe
a statistical study of the location of 78 substorms observed close to the
2000Ð2001 winter solstice. The latitudinal distribution of the onsets observed
with WIC is asymmetric with a median at 65.6¡ MLAT and a full width at half
maximum (FWHM) of 3.5¡. Their local time distribution is concentrated between
2000 and 0200 MLT with a median at 23.4 ± 0.3 hours MLT and a FWHM of 1.8 hours.
No statistically significant difference is found in the spatial distribution of
the proton and electron onsets. All onsets take place within a region of
preexisting proton precipitation, indicating that substorm initiation occurs in
regions of stretched but dipole-like field lines that cross the equatorial plane
close to the Earth. Latitudes of substorm onsets are located at a variable
distance from the poleward FUV auroral boundary but remain at a nearly constant
distance from the equatorward limit of both proton and electron auroral ovals.
The magnetic latitudes of the onsets are correlated with some of the solar wind
plasma properties measured by the ACE satellite prior to the substorm breakup.
In particular, a clear anticorrelation is found between the onset MLAT and the
1-hour averaged solar wind dynamic pressure. A decrease of the onset latitude is
also observed for larger B intensity values. No dependence of the onset MLT on
the solar wind speed is observed, in contrast to the relationship expected from
the thermal catastrophe model for substorm initiation. Our results are in
agreement with models locating the initial instability in the near magnetosphere
such as the near-Earth current disruption models.

_______________
Journal of Geophysical Research, 109, A03208, doi:10.1029/2003JA010129, 2004