Walter J. Heikkila

Definition of the Cusp

The history of the dayside or polar cusp is reviewed briefly, beginning with the work of Chapman and Ferraro, Axford and Hines, and Dungey. Solar wind plasma comes down the cusp to low altitudes; the remarkably broad area over which this penetration occurs prompted the use of the term cleft, apparently at the foot of the entry layer inside the magnetopause. The consensus of discussion at the workshop was that cusp should refer mainly to the narrow feature in the magnetic profile. Cleft would refer to the broader region defined by the solar wind plasma (perhaps somewhat modified by acceleration or retardation processes); distinctive features observable in the ionosphere or on the ground are photo-emission, enhanced plasma density and temperature, plasma convection, electromagnetic waves, and turbulence.

Particle simulation of plasma response to an applied electric field parallel to magnetic field lines

[1] We study response of thermal plasmas to an induction electric field via one-dimensional particle simulations. The induction electric field is assumed to be uniform in space and constant in time. Because of acceleration of electrons and ions in the opposite directions, there arise counter streaming electrons and ions that cause the Buneman instability. Depending on the ratio of the ion temperature T i to the electron temperature T e , responses to the electric field are different. For a case with hot ions (T i >> T e ) the Buneman instability leads to formation of large isolated electrostatic potentials which trap some electrons to move with ions. For a case with colder ions (T i « T e ) the Buneman instability is taken over by excitation of ion acoustic waves, which diffuse the low-energy part of the accelerated electrons to stabilize the instability. However, a substantial part of the electrons are grouped together at the high-energy part, forming a distinct bump in the electron distribution. In the present simulations we have found that the induction electric field can form an electron beam along the magnetic field line. Since the electron beam leaves the region of the induction electric field and moves into an unperturbed plasma, the accelerated electrons can cause a bump-on-tail instability. This can lead to formation of electrostatic solitary waves as frequently observed by the GEOTAIL spacecraft in the plasma sheet boundary layer (PSBL). The persistent observation of the electrostatic solitary waves indicates their association with the induction electric field that results from meandering motion of the current sheet in the magnetotail.

Magnetic reconnection, merging, and viscous interaction in the magnetosphere

Two ideas were advanced for the process of solar wind-magnetospheric interaction in the same year 1961. Dungey suggested that the interplanetary magnetic field (IMF), although weak, might determine the nature of this process by magnetic reconnection as the solar wind plasma flows across the separatrix surface which divides the IMF from the geomagnetic field. Axford and Hines pointed out that the flow inside the magnetopause is in the same sense as the magnetosheath flow and appears to be viscously coupled. Within a few years the dependence of geomagnetic activity on the IMF predicted by Dungeys mechanism was observed, and reconnection began to dominate current theories. One difficulty, that of the implied dissipation at the magnetopause, was troublesome; however, the ISEE-1/2 observations of the predicted high speed flows on several occasions was enough to convince many persons that reconnection ideas were basically correct. Several investigators found some evidence in the ISEE-3 data in the distant magnetotail for the steady-state reconnection line, as demanded by the Dungey model, in the form of a southward sense of the magnetic field through the current sheet. Here, again, there is some hard contrary evidence when the data are analyzed exactly at the cross-tail current sheet: the instantaneous values show a northward sense, even at high values of auroral activity. Coupled with the anti-Sunward plasma flow, this repudiates the steady-state Dungey model. On the other hand, it lends strong support to some kind of viscous effect through the medium of the magnetospheric boundary layer. This is not a semantic problem, as the sense of the electric field (as well as the magnetic field) is opposite for the two cases. The downfall of the reconnection model is its implicit use of frozen-field convection; this problem is obvious when the problem is viewed in three dimensions. Instead, the view is taken that the relevant process must be essentially time-dependent, three-dimensional, and localized. It is proposed that the term merging be used for this generalized timedependent form of reconnection. The merging process (whatever it is) must permit solar wind plasma to cross the magnetopause onto closed field lines of the boundary layer. Once it is there, it provides the viscous-like effect that Axford and Hines had envisaged.

Interpretation of recent AMPTE data at the magnetopause

Phan and Paschmann [1996] have done a superposed epoch analysis of conditions near the dayside magnetopause and have found significant structure within the magnetopause current sheet itself. Among their many important results is that the electron temperature for an outward profile shows cooling of the solar wind plasma for the inner part followed by heating for the outer. Since these two cases are associated with E · J 0, this pivotal result can be interpreted as evidence for a dynamo-load combination. This was hypothesized by Heikkila [1982a] for the localized impulsive penetration of solar wind plasma through the magnetopause current sheet; the process involves an inductive electric field Eind given by Lenzs law around the current perturbation (the electromotive force) and a plasma response through charge separation caused by Eind, a process which is controlled by the normal component of the magnetic field Bn at the magnetopause. A dynamo is not included in the standard definition of reconnection, only the reconnection load. Another key result is a remarkable difference between inbound and outbound crossings of the normal component of plasma velocity υn. This can be understood on the basis of two complementary processes involving (1) a polarization electric field which does not depend on the movement of the magnetopause itself [Lemaire and Roth, 1978] and (2) the inductive electric field due to magnetopause erosion which does. These results have opened a new chapter on solar wind-magnetospheric interaction. They demonstrate that the concepts of frozen-in flow and of magnetic reconnection (as defined) are inappropriate at the magnetopause. Rather, the interaction of the solar wind plasma at the magnetopause depends on a localized pressure pulse whose effects vary greatly on the magnetic field topology, i.e. whether the magnetopause is a tangential or rotational discontinuity. Since the plasma is doing work, this is a form of viscous interaction.

Analysis of the substorm trigger phase using multiple ground-based instrumentation

The authors discuss in detail the observation of an event of auroral activity fading during the trigger, or growth phase of a magnetic storm. This event was observed by all-sky cameras, EISCAT radar and magnetometers, riometers, and pulsation magnetometers, from ground based stations in Finland and Scandanavia. Based on their detailed analysis, they present a possible cause for the observed fading.

Critique of fluid theory of magnetospheric phenomena

It is pointed out that the fluid theory has been successful in magnetospheric problems (such as the shape of the magnetopause) which involve basic considerations such as the conservation of particles, of momentum, and of energy, but that it is inadequate for other problems (such as the energization of auroral particles). Difficulties arise from the fact that it is not always possible to specify ‘a volume of plasma’ because particles do not remain as neighbours. Misuse of the fluid theory has led to a number of fallacies, such as the idea that the causal order of physical events in cosmic electrodynamics is the reverse of that in the familiar laboratory electrodynamics. This mistaken idea comes from a confusion of a mathematical sequence of calculations with the causal order. Also, the importance of the magnetic field as an active element is over-emphasized. Appreciation of the fact that kinetic theory is the more fundamental seems to be widely lacking. A plea is made for a common sense approach to magnetospheric and auroral problems wherein the fluid theory is used whenever it can, but where it is not expected to be adequate for all purposes.

Cause and Effect At the Magnetopause

Recent analyses of spacecraft data, especially AMPTE/IRM data, provide a test of reconnection theory; an analysis for the signature of a local tangential stress balance in a one-dimensional time-stationary rotational discontinuity has left crucial questions unanswered. A key result is that the electron temperature profile inward through the magnetopause current sheet shows heating followed by cooling. Electrons must be one of the carriers of the current; hence this result reflects the sign of E · J in the frame of reference of the magnetopause current carriers. Since the current is directed from dawn to dusk, the inescapable conclusion is that the electric field must reverse within the current sheet. This is direct evidence of a load–dynamo combination; in that dynamo, energy is transferred from the solar wind plasma to the electromagnetic field. A dynamo is not included in the reconnection model which includes only the electrical load; therefore, we argue that the reconnection problem is improperly posed. A second compelling observation is a remarkable difference of the normal component of the plasma velocity between inbound and outbound crossings. For an inbound crossing (outward current meander) this component does reverse, but not quite as assumed in the reconnection model; on the other hand, for outbound crossings of the spacecraft (corresponding to erosion) there is no reversal at all. The normal component is approximately constant at 20 km s-1, anti-Sunward throughout. Since the typical motion of the magnetopause is 10 km s-1 this revealing result shows that solar wind plasma can go across the magnetopause, even onto closed field lines to feed the low latitude boundary layer. This is in stark contrast to the reconnection model where the plasma goes to open field lines. The interaction can be understood by appealing to Poyntings theorem, where E · J describes the net effect on or by the plasma. Time-dependent terms (even in the initial conditions) must be used so that it is possible to draw upon energy which has been stored locally in both electrical and magnetic forms. An extended discussion of observational results from ground-based, rocket, and satellite instruments indicate the impulsive nature of the solar wind–magnetospheric interaction. There is a lot of plasma involved in this interaction, over 1027 ions electrons-1 per second; the anti-Sunward flow takes place in the low latitude boundary layer. There is no flux catastrophe produced by this flow since the frozen-field theorem does not hold for plasma transfer across the magnetopause. The LLBL completely envelops the plasma sheet; the LLBL is the source of its plasma, not the plasma mantle as hypothesized in the reconnection model of the magnetotail. A number of serious errors have occurred in some articles in the literature on reconnection, and we list and discuss the most important of these. In the conclusion it is emphasized that the failure to provide a viable energy source, within the necessary spatial and temporal constraints, is responsible for the failure of reconnection model. This does not mean that the state of interconnection between the geomagnetic field and the interplanetary magnetic field can not change, but it does mean that the advocated process is not relevant to such changes. True reconnection requires that the electric field has a curl so that an electromotive force ∈ = ∮ E · dl = -dΦMdt exists through which energy can be interchanged with stored magnetic energy.

ELEMENTARY IDEAS BEHIND PLASMA PHYSICS

This contribution is in support of Alfvéns use of circuit theory to advance the understanding of complex plasma physical problems, such as magnetic reconnection; these ideas have often been mis-understood. Circuit analysis is not a full description of the physics, being a scalar relationship. However, it is suitable for dealing with energy relationships and cause and effect, and it is fully capable of showing fallacies behind various fashionable ideas, including steady-state reconnection.

Near Earth Current Meander (Necm) Model of Substorms

We propose that the appropriate instability to trigger a substorm is a tailward meander (in the equatorial plane) of the strong current filament that develops during the growth phase. From this single assumption follows the entire sequence of events for a substorm. The main particle acceleration mechanism in the plasma sheet is curvature drift with a dawn-dusk electric field, leading to the production of auroral arcs. Eventually the curvature becomes so high that the ions cannot negotiate the sharp turn at the field-reversal region, locally, at a certain time. The particle motion becomes chaotic, causing a local outward meander of the cross-tail current. An induction electric field is produced by Lenzs law, Eind=−∂A/∂t. An outward meander with Bz>0 will cause E×B flow everywhere out from the disturbance; this reaction is a macroscopic instability which we designate the electromotive instability. The response of the plasma is through charge separation and a scalar potential, Ees=−∇φ. Both types of electric fields have components parallel to B in a realistic magnetic field. For MHD theory to hold the net E∥ must be small; this usually seems to happen (because MHD often does hold), but not always. Part of the response is the formation of field-aligned currents producing the well-known substorm current diversion. This is a direct result of a strong E∥ind (the cause) needed to overcome the mirror force of the current carriers; this enables charge separation to produce an opposing electrostatic field E∥es (the effect). Satellite data confirm the reality of a strong E∥ in the plasma sheet by counter-streaming of electrons and ions, and by the inverse ion time dispersion, up to several 100 keV. The electron precipitation is associated with the westward traveling surge (WTS) and the ion with omega (Ω) bands, respectively. However, with zero curl, Ees cannot modify the emf ε=∮E⋅dl=−dΦM/dt of the inductive electric field Eind (a property of vector fields); the charge separation that produces a reduction of E∥ must enhance the transverse component E⊥. The new plasma flow becomes a switch for access to the free energy of the stressed magnetotail. On the tailward side the dusk-dawn electric field with E⋅J<0 will cause tailward motion of the plasma and a plasmoid may be created; it will move in the direction of least magnetic pressure, tailward. On the earthward side the enhanced dawn-dusk induction electric field with E⋅J>0 will cause injection into the inner plasma sheet, repeatedly observed at moderate energies of 1–50 keV. This same electric field near the emerging X-line will accelerate particles non-adiabatically to moderate energies. With high magnetic moments in a weak magnetic field, electrons (ions) can benefit from gradient and curvature drift to attain high energies (by the ratio of the magnetic field magnitude) in seconds (minutes).

Magnetospheric Plasma Regions and Boundaries

Several magnetospheric regions and their boundaries are discussed in critical terms. Serious questions are raised by recent observations that suggest that the magneto sheath plasma on the day-side penetrates deep into the region of closed magnetic field lines. This deduction, and the evidence that the polar cap field lines interconnect with the interplanetary magnetic field lines, suggest that the magnetopause be defined as a plasma boundary, rather than in terms of open or closed magnetic field lines. The magnetopause is here defined as that surface where there is an abrupt reduction in the phase space density from the values typical of the magneto sheath plasma. On the night side this definition, coupled with the evidence of closed field lines poleward of the auroral oval, requires a modification to Dungey’s model of the magnetosphere in which there is a gap or space between the end of the plasma sheet and the X-line. These considerations imply that the usual model of magnetic merging is not applicable to the magnetosphere. This implication is supported by the failure to detect the energy dissipation at the dayside magnetopause that would be a consequence of the tangential component of the electric field characteristic of the merging process. Careless use of words or phrases such as interconnection and generation is discussed. It is suggested that the resolution of these serious questions can be achieved only by turning to a more primitive description that avoids the shortcomings of the fluid theory.