Mean thermospheric winds observed from Halley, Antarctica

Abstract. Thermospheric winds on a total of 237 nights have been studied for the effects due to geomagnetic activity, solar flux, and season. The observations have been made from 1988 to 1992 by a Fabry-Perot interferometer (FPI) operating at Halley (75.5^°S, 26.6^°W), Antarctica. This is the first statistical study of thermospheric winds near the southern auroral zone. The main factor affecting the wind velocities is the geomagnetic activity. Increases in activity cause an increase in the maximum equatorward wind, and cause the zonal wind in the evening to become more westward. Smaller changes in the mean wind occur with variations in season and solar flux. The small variation with solar flux is more akin to the situation found at mid-latitudes than at high latitudes. Since the geomagnetic latitude of Halley is only 61^°, it suggests that the variability of the wind with solar flux may depend more on geomagnetic than geographic latitude. These observations are in good agreement with the empirical Horizontal Wind Model (HWM90). However, comparisons with predictions of the Vector Spherical Harmonic Model (VSH) show that for low geomagnetic activity the predicted phases of the two components of the wind closely resemble the observations but the modelled amplitudes are too small by a factor of two. At high geomagnetic activity the major differences are that modelled zonal velocity is too westward in the evening and too eastward after 04 UT. The modelled ion densities at the F-region peak are a factor of up to 9 too large, whilst the predicted mean value and diurnal variation of the altitude of the peak are significantly lower than those observed. It is suggested that these differences result from the ion loss rate being too low, and an inaccurate model of the magnetic field.