ring current associated with the sunward convection of hot plasma sheet plasma described in section 2.3.2.
Figure 2.7 (a) A cut through the magnetosphere in the X = 0 plane, showing the sense of plasma flow (black arrows into and out of the page) in the open, closed, and plasmasphere regions, along with the sense of current flows (green) at the magnetopause, ionosphere (polar cap and return flow regions), R1 and R2 FACs, and partial ring current. (b) The deformation of open (blue) and closed (red) field lines by flows in the solar wind and inner magnetosphere and the frictional forces of the ionosphere, and the sense of field‐aligned current produced by the magnetic shear.
Figure 2.8 (Left) Field‐aligned current pattern associated with vorticity in the convection flow, with black as downward current and grey as upward current. The two concentric rings of FAC are region 1 or R1 (high latitude) and region 2 or R2 (low latitude). The overlap of currents in the premidnight sector is associated with the Harang discontinuity. Upward or downward FAC flows poleward of the dayside R1 FAC depending on the sign of IMF BY, associated with east‐west flow asymmetries produced by magnetic tension forces on newly opened field lines
(from Iijima & Potemra, 1976).
(Right) The Harang discontinuity as seen in FACs and convection
(adapted from Iijima & Potemra, 1976; Koskinen & Pulkkinen, 1995).
The magnitude of the FACs can be found from the divergence of J⊥ by combining equations (2.6) and (2.10):
in which the three terms on the RHS are the divergence of JP due to shears in the convection flow, the divergence of JP due to gradients in the Pedersen conductance, and the divergence of JH due to gradients in Hall conductance. In the limit of uniform conductance, this reduces to Poisson's equation. The distribution of FACs can then be compared with or inferred from the vorticity in the convective flow (e.g., Sofko et al., 1995; McWilliams et al., 1997; Green et al., 2006; Chisham et al., 2009). In the auroral zone, with a Pedersen conductance of 10 S, a flow speed of 500 m s−1 produces an ionospheric current across the flow of 0.25 A m−1. If the convection reversal associated with the low‐latitude duskside flow shear is 2,000 km in length, then 0.5 MA of FAC flows in the R2 current there. The R1 current poleward of this will be of a similar magnitude, possibly enhanced somewhat by a contribution from Pedersen current flowing across the polar cap, as shown in Figure 2.3b (though see Laundal et al., 2018, for a discussion of horizontal current closure). By this estimate, it is expected that when convection is ongoing, approximately 2 MA flows into and out of the ionosphere in each hemisphere in the whole R1/R2 system, increasing during periods of strong driving, in agreement with the observations of Iijima and Potemra (1978).
Upward FACs are carried by downgoing electrons from the magnetosphere, which produce auroral emissions and enhanced conductance as they impact on the atmosphere. The requirement of current continuity can lead to field‐parallel acceleration of the electrons to defeat the mirror force that otherwise traps the magnetospheric plasma above the ionosphere, resulting in the production of discrete auroral forms. Indeed, we can draw a link between the convection pattern, the R1/R2 FAC pattern, and the large‐scale distribution of auroras, which is a powerful tool for studying magnetospheric dynamics (section 2.4.2). Not all auroras are formed in this way, however. For instance, auroras can be produced by direct precipitation from the magnetosheath during magnetopause reconnection and from the plasma sheet during tail reconnection, whereas diffuse auroral emissions are produced by more chaotic perturbations of the trapped plasma in the magnetosphere, such as pitch angle scattering.
Pedersen currents do not in general produce significant magnetic perturbations on the ground, as these are cancelled by the perturbations produced by the FACs themselves (Fukushima, 1976), except in specific situations in which there are significant horizontal gradients in the conductance (Laundal et al., 2015, 2018). However, the Hall currents in the high conductance auroral zones, known as the eastward and westward “auroral electrojets” at dusk and dawn (Fig. 2.3b), produce northward and southward directed perturbations on the ground, which can be used as a measure of convection strength. These perturbations are measured by the upper and lower auroral electrojet indices (AU and AL) (Davis & Sugiura, 1966), which quantify the maximum northward and southward directed perturbations associated with the eastward and westward electrojets, respectively.
An additional upward FAC forms in the midnight sector, which tends to bridge the upward FAC of the dusk R1 current with the upward FAC of the dawn R2 current, as shown in the right‐hand panels of Figure 2.8 (Kamide & Vickrey, 1983; Kunkel et al., 1986). This is thought to be due to a pressure buildup in the premidnight sector (Erikson et al., 1991) as hot protons convect earthward and gradient‐curvature drifts toward dawn (section 2.3.2). This produces the “Harang discontinuity” or reversal, a deformation of the dusk convection cell in the nightside convection pattern (e.g., Koskinen & Pulkkinen, 1995).
2.4 TIME‐DEPENDENT CONVECTION
The discussion so far has concentrated on steady‐state convection when the IMF is directed southward and the dayside and nightside reconnection rates are equal, a phenomenon known as “steady magnetospheric convection” or SMC. However, the IMF orientation and solar wind speed continually change, varying the dayside reconnection rate. The nightside reconnection process is largely decoupled from the dayside and cannot instantaneously adjust itself to changes in the solar wind. This means that magnetospheric dynamics and ionospheric convection are inherently time dependent. Much research in recent years has focused on the time dependence of convection associated with the substorm cycle, or the timescales of response to changes in IMF orientation.
2.4.1 The Expanding/Contracting Polar Cap Model
The consequences of nonsteady and unequal dayside and nightside reconnection for magnetospheric and ionospheric convection were first explored by Russell (1972), Siscoe and Huang (1985), and Holzer et al. (1986), and later developed by Freeman and Southwood (1988), Cowley and Lockwood (1992), and Lockwood and Cowley (1992), resulting