Modelling of plasmaspheric flows.

Abstract

Flows on a plasmaspheric flux tube were modelled, using a fully interhemispheric model in which no low speed assumptions are made. The model is time-dependent and hydrodynamic; it is an adaptation, for closed dipolar field lines, of a polar wind model developed by Gombosi et al. (1985). Both one- and two-stream versions were developed. The model simultaneously solves the coupled continuity, momentum, and energy equations of a two-ion (H$\sp+$ and O$\sp+$) quasineutral, currentless plasma. The coupled time-dependent partial differential equations are solved using a combined Godunov scheme/Crank-Nicolson method with dimensional splitting. The model takes into account the effects of ionization, charge exchange, recombination, collisions, heat conduction, and allows for external heat sources. The model flux tube connects external reservoirs, each at an altitude of 200 km; these reservoirs represent photochemically controlled regions of the ionosphere. The neutral species densities and the neutral temperature are found using the MSIS-86 model (Hedin, 1987). The one-stream version of this model has been used to investigate plasmaspheric refilling following density depletions on an L = 2 flux tube, which were simulated by dividing steady state ion densities by an arbitrary factor above 2500 km altitude; to study the effect of equatorially confined H$\sp+$ heating on an L = 4 flux tube; to model diurnal variations on an L = 2 flux tube; and to investigate annual variations in plasmaspheric density. The two-stream version was used to study plasmaspheric refilling on an L = 4 flux tube following a density depletion; the initial H$\sp+$ density profiles were chosen to be similar to those used by Rasmussen and Schunk (1988). The results from the plasmaspheric refilling studies indicate that for an initial period refilling occurs from the equator downward, between downward moving shocks. The study of equatorially confined H$\sp+$ heating showed that heating rates which are consistent with observed magnitudes of warm, trapped, anisotropic plasma would not significantly affect the thermal population. The most striking result from the diurnal variations study was the large downward H$\sp+$ velocity which occurs in the southern end in the early morning. The study of annual variations in plasmaspheric density showed that these are due to variations in O$\sp+$ density in the upper ionosphere. The results found with the model are generally consistent with those found in earlier work; significant differences include temperature variations, as were seen in the refilling studies, and strong transients, as were found in the diurnal variations study.