Numerical simulation of magnetospheric convection including the effect of field‐aligned currents and electron precipitation

Abstract

Early results of a new self‐consistent fluid model are presented for steady convection of the plasma from the geomagnetic tail through the Earth's inner magnetosphere below 10 Earth radii (RE), including its coupling with the ionosphere. This model computes the transport of both the ion and electron fluids and constitutes an important improvement of the fluid numerical model of Fontaine et al. (1985) (referred to as Paper 1), which simulated the convection of electrons only. The coupling with the ionospheric convection is effected by the precipitation of magnetospheric electrons, which enhances ionospheric auroral conductivities, and by the flow of region 2 field‐aligned currents, which connect the magnetospheric and ionospheric circuits. This last coupling was not considered in Paper 1, because region 2 field‐aligned currents are generated mainly by pressure gradients of the ion plasma that develop progressively during its motion. We also take into account the ion precipitation, which affects the generation of region 2 field‐aligned currents, but we neglect their relatively small effect on the ionospheric conductivities. We devoted much effort to produce a self‐consistent description of the transport of the ionospheric and magnetospheric plasmas, and their couplings. As a first step, in order to distinguish basic physical processes from effects of geometrical origin, we solved the equations for the case of a dipolar magnetic field as in Paper 1. As a consequence, we do not expect our results to reproduce very precisely the available observations but rather to provide only correct orders of magnitude. The numerical solution scheme of the hyperbolic equations governing the transport of both magnetospheric ions and electrons, of the elliptic equation of the ionospheric electrostatic potential, and of their coupling due to electron precipitation and region 2 field‐aligned currents, is basically the same as Paper 1. It makes use of the finite element method. The numerical model is run to simulate the evolution of the system from an initial state in which the inner magnetosphere is free of plasma, to a steady state situation in which the plasma has penetrated into the inner magnetosphere from the tail. Stable magnetic conditions corresponding to a Kp between 2 and 3 are considered to allow a comparison with observations. The formation of a belt of electron and ion precipitation in the auroral zone is globally reproduced by the model, but the electron fluxes are underestimated and the auroral ion zone is located equatorward with regard to the DMSP satellite statistical observations. The field‐aligned current region 2 computed by the model is comparable to the available statistical observations, apart from an underestimation of the amplitude. The convection structure is also found comparable to observations, thus improving the results of Paper 1. The amplitude and the direction of the major component of the electric field (the northward component being at least 4 times larger than the eastward component) is consistent w...

Role of collisionless heat flux in magnetospheric convection

“…The results are equivalent to those of Peymirat and Fontaine [1994] if heat flux is neglected; they are, however, developed in a much simplified form that is useful in simulations and in describing the physics of the convection process. They can be stated succinctly: In the absence of heat flux, the number per unit flux is conserved along equatorial streamlines of each species of the plasma as long as the species is isentropic and lossless, but not otherwise; and, again absent heat flux, the entropy per particle in lossless convection is conserved.…”

Polar cap convection patterns inferred from EISCAT observations

“…We solve this second order differential equation over the polar cap by the finite element method with two boundary conditions, in a similar way to Peymirat and Fontaine (1994). More precisely, the domain covers all local times, and the latitudes between two circular boundaries at a constant invariant latitude.…”

“…The equatorward boundary eq is set at 70°invariant latitude, and the conditions along eq are Dirichlet conditions, where the potential takes the value predicted by the statistical models of the convection potential inferred from EISCAT data (Senior et al, 1990). Finally, to solve the elliptic equation of the polar cap potential, we performed a variational formulation which directly includes these boundary conditions: it ensures the uniqueness of the computed solution and has the advantage to satisfy rigorously the boundary conditions (see Peymirat and Fontaine, 1994, for details on the method). The solution of Eq.…”

An extended TRANSCAR model including ionospheric convection: simulation of EISCAT observations using inputs from AMIE

“…One further step towards a self-consistent model is under progress. The aim is to couple TRANSCAR with the Ionosphere-Magnetosphere Model (Peymirat and Fontaine, 1994), which can provide these magnetospheric key parameters.…”