Possible role of magnetosphere-ionosphere coupling in auroral arc generation
Three models for the magnetosphere-ionosphere coupling feedback instability are considered. The ®rst model is based on demagnetization of hot ions in the plasma sheet. The instability takes place in the global magnetosphere-ionosphere system when magnetospheric electrons drift through a spatial gradient of hot magnetospheric ion population. Such a situation exists on the inner and outer edges of the plasma sheet where relatively cold magnetospheric electrons move earthward through a radial
... rough a radial gradient of hot ions. This leads to the formation of ®eld-aligned currents. The eect of upward ®eld-aligned current on particle precipitation and the magnitude of ionospheric conductivity leads to the instability of this earthward convection and to its division into convection streams oriented at some angle with respect to the initial convection direction. The growth rate of the instability is maximum for structures with sizes less than the ion Larmor radius in the equatorial plane. This may lead to formation of auroral arcs with widths about 10 km. This instability explains many features of such arcs, including their conjugacy in opposite hemispheres. However, it cannot explain the very high growth rates of some auroral arcs and very narrow arcs. For such arcs another type of instability must be considered. In the other two models the instability arises because of the generation of Alfven waves from growing arc-like structures in the ionospheric conductivity. One model is based on the modulation of precipitating electrons by ®eld-aligned currents of the upward moving Alfven wave. The other model takes into consideration the re¯ection of Alfven waves from a maximum in the Alfven velocity at an altitude of about 3000 km. The growth of structures in both models takes place when the ionization function associated with upward ®eld-aligned current is shifted from the edges of enhanced conductivity structures toward their centers. Such a shift arises because the structures move at a velocity dierent from the E Â B drift. Although both models may work, the growth rate for the model, based on the modulation of the precipitating accelerated electrons, is signi®cantly larger than that of the model based on the Alfven wave re¯ection. This mechanism is suitable for generation of auroral arcs with widths of about 1 km and less. The growth rate of the instability can be as large as 1 s À1 , and this mechanism enables us to justify the development of auroral arcs only in one ionosphere. It is hardly suitable for excitation of wide and conjugate auroral arcs, but it may be responsible for the formation of small-scale structures inside a wide arc.