Discussion



The efficiency with which the He+ ions are accelerated compared with H+ and O+ and the consistent association between preferential acceleration and EMIC waves suggest a cyclotron resonant acceleration mechanism. A similar mechanism has been proposed to explain the unusually high 3He abundances seen in impulsive solar flares [Temerin and Roth, 1992; Roth and Temerin, 1997]. In impulsive solar flares, ions are accelerated to several MeV/nucleon, and the 3He/4He abundance ratio, normally ~ 10-4, can exceed 1. A Temerin-Roth-type mechanism in the topside auroral ionosphere could easily account for the enhanced He+ energization seen in these events. In addition, this mechanism would account for the elevated He+ fluxes reported in the DE-1 data [Collin et al., 1988].

Some previous studies of ion heating by EMIC waves [e.g., Horne and Thorne, 1997] have explicitly assumed that this heating occurs when the wave frequency is at or above the ion hybrid resonance frequency f
IH, which plays a similar role in EMIC wave propagation to that of the lower hybrid frequency for whistler mode propagation [Smith and Brice, 1964]. Above this frequency a resonance cone exists and waves are guided along the field line; below this frequency perpendicular propagation is allowed and the wave can be reflected [Rauch and Roux, 1982]. A cutoff of EMIC waves at fcHe+, which has been observed in Freja data [Erlandson et al., 1994], is readily explained by this effect, since when H+ is the majority species, fIH approaches fcHe+ as nHe+ approaches 0. However, reflection at fIH does not always occur in the auroral zone, since H+ EMIC waves generated in the auroral acceleration region have been detected on the ground [Sato and Hayashi, 1985]. While the data in Figure 1 do not rule out reflection at the ion frequency, they require perpendicular wavelengths as short as 200 m at the satellite speed of 4 km/s in order to Doppler shift the observed wave frequencies to the local ion hybrid frequency. The skin depth, a typical minimum perpendicular wavelength for electromagnetic waves below fcH+, is at least several times longer in this region. A more likely scenario, therefore, is that the waves propagate beyond the ion hybrid resonance, as has been predicted in recent ray tracing calculations [Lund and LaBelle, 1997], and heat the He+ near the altitude where the wave frequency matches the local helium cyclotron frequency. At 3600 km, fcHe+ = 56 Hz, near the lowest frequency of the observed EMIC band. As a result, He+ can be accelerated by cyclotron resonance even when the plasma contains a substantial He+ concentration.

It should be stressed that the resonant heating observed here is a secondary heating mechanism; some other acceleration mechanism must heat the protons to the energies observed. In this case there are two possibilities: the broad-band electrostatic waves in the beam region could heat ions which then are convected equatorward as they move upward, or the ions could be heated by electrostatic waves generated below the satellite. The relative importance of the two or more mechanisms is an open question which will be addressed in future investigations.

Next: Acknowledgements
Previous: Description of Data

Title Page | Abstract | Introduction | Description of Data | Discussion | Acknowledgements | References


Up to Eric Lund's Home Page