1. Introduction
The substorm is, by now, a well-known and well-visited topic. The substorm paradigm got its start when Akasofu [1964] recognized that the episodic brightening and expansion of auroral activity was a universal time phenomenon. Previously, it was believed that through the night the auroral activity went through a local time dependent pattern in which quiet arcs would form in the evening, break up at midnight and irregular patches would form in the morning. Akasofu made a systematic study of worldwide all-sky-camera data collected during the IGY. He found that auroral activation was manifest over the entire darkened auroral zone.
The early sixties was also a time when satellites began making deep space observations and began to map out the characteristics of the Earth’s magnetosphere. One observation that was to have a profound effect on the direction of substorm research was the discovery by Ness [1965] of the neutral sheet in an extended tail to the Earth’s magnetosphere. The sixties was also a period of explosive growth of our understanding of plasma physics through international efforts to achieve controlled fusion. This new understanding was quickly applied to understanding the newly discovered magnetosphere. Coppi et al. [1966] seized on the observation of the extended neutral sheet to propose that it might be unstable to the tearing-mode instability. They further suggested that the tearing mode instability might be responsible for the substorm proposed by Akasofu. Hones [1977] presented rather vivid pictures characterizing topological changes to the Earth’s magnetotail during a substorm cycle. Later, Fairfield et al. [1981] found a strong relationship between the orientation and strength of the IMF and accumulation of energy in the magnetotail and substorm dissipation. They also found that southward IMF leads to plasma sheet thinning.
Important substorm effects were also seen in space nearer to the Earth. Akasofu [1972] found that substorms resulted in intense field-aligned currents in the form of a wedge connecting to the ionosphere from geosynchronous altitudes. Also observed were impulsive energetic particle injections at synchronous altitudes [Akasofu, et al., 1974] and magnetic field dipolarization [Nagai, 1982].
The research attention became primarily focused on substorm manifestations in space. More significantly, the substorm came to be viewed as a result of a large-scale plasma instability, primarily related to the tearing mode and magnetic reconnection. Beginning in the seventies and early eighties, numerical simulation began to play a prominent role in our efforts to understand the substorm. Global scale MHD simulations [Fedder and Lyon, 1987; Walker, et al., 1993] have elucidated the response of the magnetosphere to the changing IMF. These simulations appear to nicely reproduce the changes in the magnetic field topology that Hones [1977] had previously conjectured on the basis of field and plasma flow measurements in the Earth’s magnetotail. Numerous other simulations focused on smaller-scale plasma processes using a variety of MHD, as well as particle and hybrid codes.
There have been some theoretical and modeling attempts [Fedder and Lyon, 1987; Kan, 1998] to relate substorm processes in the magnetosphere to field-aligned currents connecting to the ionosphere and to study possible magnetosphere-ionosphere interactions. However, with few exceptions, like that of Otto and Birk [1993], relatively little attention has been directed to understanding magnetospheric processes that might account for the aurora. This is ironic in view of the fact that the substorm paradigm got its start with analysis of visual aurora observations.
A major purpose of this paper is to investigate a new set of physical processes that may help us understand magnetospheric phenomena that might account for the discrete aurora. This paper presents results of a two-dimensional hybrid code simulation of the midnight meridian plane. The simulation domain extends from the surface of the Earth to x = -36 RE in the antisunward direction and to z=± 11 RE along the polar axes. Full ion kinetics is presented in a global-scale context. The code uses a curvilinear coordinate system that affords high spatial resolution near the Earth yet encompasses the vast regions of the magnetospheric tail. The model is driven from the boundaries by an externally imposed dawn-dusk convection electric field. The varying field topology and varying plasma convection velocities can result in the interpenetrating of plasma beams. These can give rise to a variety of micro-instabilities that cannot be accessed with a purely MHD code, and the conditions for the instabilities are difficult to reproduce in kinetic codes that encompass only a limited region of space. The fine spatial scale of auroral arcs does argue that a microinstability might be involved in their production.
Although the physics contained in the simulations will allow the study of new processes, the two-dimensionality of the model does pose a number of limitations. One is that it cannot reproduce the substorm current wedge [Akasofu, 1972] and associated Region I currents, because these involve variations in the dawn-dusk directions. Although the code includes the Earth’s surface, it is not possible to model the auroral electrojet current, also because this current connects to the upward and downward field-aligned currents, which comprise the substorm current wedge.
The next section describes the simulation code, initial conditions and boundary conditions of the models. Section 3 describes a run in some detail and then describes three other runs with differing initial and boundary conditions in considerably less detail. Interpretation of the results is discussed in Section 4, and concluding remarks are provided in Section 5.