Interplanetary shock waves and magnetospheric substorms

It is shown that substorm activity after a storm sudden commencement (S.S.C.) depends on whether or not an interplanetary shock wave is accompanied by a large increase of the solar wind-magnetosphere energy coupling function ε. It has long been thought that substorm activity associated with an S.S.C. results from sudden conversion of magnetic energy stored in the magnetotail and that this conversion is triggered by the shock wave. However, the present result implies that the magnetospheric substorm is not a sudden conversion of stored magnetic energy, but is a direct consequence of increased efficiency of the solar wind-magnetosphere dynamo.     

Interplanetary magnetic field and magnetospheric substorms

The interplanetary magnetic field (IMF) changes and the associated responses of the magnetosphere on November 1, 1972, are examined. IMF B_z changes consisted of a sudden southward turning, a slow northward turning, and a subsequent steady northward sense. Magnetospheric substorms occurred throughout this period.     

Plasma-sheet dynamics and magnetospheric substorms

We present a conceptual model of the formation of the plasma sheet and of its dynamical behavior in association with magnetospheric substorms. We assume that plasma mantle particles $\text{E}\text{～}\text{×}\text{B}\text{～}$ drift toward the current sheet in the center of the tail where they are accelerated by magnetic-field annihilation to form the plasma sheet. Because of the velocity-dependent access of mantle particles to the current sheet, we argue that the convection electric field and the corresponding rate of field annihilation decrease with increasing radial distance. As a consequence, there exists no steady-state configuration for the plasma sheet, which must instead shrink continuously in thickness until the near-earth portion of the current sheet is disrupted by the formation of a magnetic neutral line. The current-sheet disruption launches a large-amplitude hydromagnetic wave which is largely reflected from the ionosphere. The reflected wave sets the neutral line in motion away from the earth; the neutral line comes to rest at a distance (which we estimate to be a few hundred earth radii) where the incoming mantle particles enter the current sheet at the local Alfvén velocity. At this “Alfvén point” reconnection ceases and the thinning of the plasma sheet begins again. Within this model, the magnetospheric substorm (which is associated with the current-sheet disruption) is a cyclical phenomenon whose frequency is proportional to the rate of convection in the magnetospheric tail.     

Some critical issues on magnetospheric substorms

Several evidences for the directly driven aspect of magnetospheric substorms are presented by reinterpreting what have been thought to be supporting evidences for the unloading process. Further, it is stressed that some of our confusions in substorm studies could be resolved by understanding that the magnetospheric substorm is primarily a directly driven phenomenon, but has a variety of internal processes. A method is suggested tto identify the directly driven and the unloading components. It is also demonstrated that the magnetosphere is intrinsically a non-linear system and that a quantitative study of magnetospheric substorms is not possible without taking into account this non-linearity.     

The statistical magnetic signature of magnetospheric substorms

The characteristic magnetic signatures of magnetospheric substorms both on the ground and in space have been determined from the analysis of ～1800 substorm events. The timing and properties of these events were objectively determined according to explicit mathematical criteria by a computer pattern-recognition program. This program processed daily magnetograms from a mid-latitude network of geomagnetic observatories.Ground data analyzed, using onsets determined in this manner, included the AE indices and individual magnetograms at different local times in the auroral zone and at midlatitudes. Superposed epoch averages of these data confirm the local time magnetic substorm signatures, determined in earlier studies of fewer events, and demonstrate the validity of the computerized onset determination procedure.Superposed epoch averages of the interplanetary magnetic field (IMF) associated with the onsets demonstrates both a distinct southward component prior to the onsets and a dependence of the substorm amplitude on the integrated preceding southward IMF flux. Superposed epoch averages of the tail lobe magnetic field magnitude and vector components demonstrates field magnitude changes and rotations in association with the substorm onsets. These lobe field changes are consistent with the growth-phase model of substorm activity and with variations in the magnetopause flaring angle.     

The reason for magnetospheric substorms and solar flares

It has been proposed that magnetospheric substorms and solar flares are a result of the same mechanism. In our view this mechanism is connected with the escape, or attempted escape, of energized plasma from a region of closed magnetic field lines bounded by a magnetic bottle. In the case of the Earth, it must be plasma that is able to maintain a discrete auroral arc, and we propose that the cross-tail current connected to the arc is filamentary in nature to provide the field-aligned current sheet above the arc. A localized meander of such an intense current filament could be caused by a tearing instability in the neutral sheet. Such a meander will cause an inductive electric field opposing the current change everywhere. In trying to reduce the component of the induction electric field parallel to the magnetic field lines, the plasma must enhance the transverse or cross-tail component; this action leads to eruptive behavior, in agreement with tearing theories. This enhanced induction electric field will cause a discharge along the magnetic neutral line at the apex of the magnetic arches, constituting an impulsive acceleration of all charged particles originally near the neutral line. The products of this phase then undergo betatron acceleration for a second phase. This discharge eventually reduces the electric field along the neutral line, and thereafter the enclosed magnetic flux through the neutral line remains nearly constant. The result is a plasmoid that has definite identity; its buoyancy leads to its escape. The auroral breakup (and solar flare) is the complex plasma response to the changing electromagnetic field.     

Equatorward shift of the cusp during magnetospheric substorms

It is shown that the magnetic field of an enhanced dynamo current in the dayside boundary layer and of the connected circuit can quantitatively account for the equatorward shift of the cusp region which is observed during the expansive phase of magnetospheric substorms.     

A mechanism for the growth phase of magnetospheric substorms

A mechanism is presented whereby the rate of energy dissipation in the magnetosphere is controlled by the particle density in the plasma sheet in the near geomagnetic tail. The mechanism is based on a model in which the plasma sheet is sustained by injection of solar-wind particles into the dayside magnetosphere. The efficiency of the injection is controlled by solarwind parameters, in particular, the north-south component of the interplanetary magnetic field; the maximum injection rate occurs when the interplanetary field is northward. During geomagnetically quiet times, this source balances the loss of particles from the edges of the tail current sheet. If the dayside source rate is reduced (e.g. by a southward-turning interplanetary magnetic field), then the plasma sheet is depleted and the rate of magnetic merging is enhanced in the earthward portion of the tail current sheet. This period of steadily-enhanced merging is associated with the growth phase, i.e. the period of enhanced magnetospheric convection for about one hour preceding the breakup of a polar magnetic or auroral substorm. The breakup can be understood as the result of the collapse of a portion of the tail current sheet following the local depletion of the plasma sheet.     

A modeling of magnetic field variations during magnetospheric substorms

Magnetic field variations in the noon-midnight plane during the magnetospheric substorm are studied in terms of changes of three current systems: the dynamo-driven current on the magnetopause, the cross-tail current and the field-aligned current-auroral electrojet system. The field-aligned current is assumed to be generated as a result of interruption and subsequent diversion of the cross-tail current to the ionosphere. It is concluded that the available observations are consistent with a large increase of the three currents.     

The structure of a connected sequence of magnetospheric substorms

The structure of a sequence of four spatially and temporally connected magnetospheric substorms has been determined through the use of a set of mid-latitude magnetometer station data, an auroral zone magnetometer line data set, and the ATS-5 magnetometer record. In two of the substorms, two parallel westward electrojets were observed to develop during the initial part of the expansion phase. In another, the expansion phase consisted of the development of a westward electrojet westward and equatorward of the pre-existing current system.     

Similarities and differences between magnetospheric substorms and solar flares

Magnetospheric substorms and solar flares seem to follow a pattern where a sudden transition from a slow passive evolution to a fast active evolution occurs. This concept which has proven useful for constructing a theory of the onset of magnetospheric substorms is tentatively applied to current sheets which may be relevant to flares.     

Implications of solar flare dynamics for reconnection in magnetospheric substorms

From observations of two-ribbon solar flares, we present a new line of evidence that magnetic reconnection is of key importance in magnetospheric substorms. We infer that in substorms reconnection of closed field lines in the near-Earth thinned plasma sheet both initiates and is driven by the overall MHD instability that drives the tailward expulsion of the reconnected closed field (0 loops). The general basis for this inference is the longstanding notion that two-ribbon flares and substorms are essentially similar phenomena, driven by similar processes. We give an array of observed similarities that substantiate this view. More specifically, our inference for substorms is drawn from observations of filament eruptions in two-ribbon flares, from which we conclude that the heart of the overall instability consists of reconnection and eruption of the closed magnetic field in and around the filament. We propose that essentially the same overall instability operates in substorms. Our point is not that the magnetic field configuration or the microphysics in substorms is identical to that in two-ribbon flares, but that the overall instability results from essentially the same combination of reconnection and eruption of closed magnetic field.     

Effects of solar wind parameters on the development of magnetospheric substorms

Effects of solar wind parameters on the development of substorms during the events of southward interplanetary magnetic field (IMF) lasting more than one hour were studied. Analysis on 175 events with average magnitude of the southward component of IMF larger than l·5γ as observed in July–December 1965 lead to the following results: (1) The total auroral electrojet (AEJ) current associated with the southward IMF event is approximately proportional to the time integral of the magnitude of the southward component. (2) The azimuthal component of IMF also affects the AEJ development. AEJ about twice as intense were observed when IMF was directed duskward than when IMF was directed dawnward. (3) AEJ intensity is strongly affected by the solar wind velocity during the southward IMF events, the intensity being approximately proportional to the square of the velocity. (4) No indication was found that the angle between the Sun-Earth line and the Earth's dipole axis plays any role on the development of substorms if effects of the solar wind parameters as described above are eliminated.     

Pi 3 magnetic pulsations associated with substorms

Using magnetic data from the North American IMS network at high latitudes, Pi 3 pulsations are analysed for a period of $\text{4}\text{1}\text{2}$ continuously-disturbed days. The data were obtained from 13 stations in the Alaska and Fort Churchill meridional chains and in the east-west chain along the auroral zone. In the past, Pi 3 pulsations associated with substorms have been classified into two sub-categories, Pi p and Ps 6. However, we find that Pi 3's which have longer periods than Pi p and which are different from Ps 6 are more commonly observed than these two special types. Power spectra, coherence and phase differences are compared among the stations. Results show that noticeable differences for latitudinal dependence of period and amplitude exist among midnight, morning and late-evening Pi 3 pulsations. Results for Pi 3 occurring near midnight indicate that the periods at which the power spectral density is a maximum are longest, and the amplitude largest, near the center of the westward auroral electrojet. On the other hand, for Pi 3 pulsations occurring in the morning, the periods at which the power spectral density is a maximum are longest, and the amplitude largest, near the poleward edge of the westward electrojet. Furthermore, for Pi 3 pulsations occurring in the late evening, their periods are longer and their amplitudes larger near both the Harang discontinuity and the poleward edge of the westward electrojet than near its center. Correlations between pairs of adjoining stations are better in the polar cap than at auroral latitudes. It is also found from hodograms that the sense of polarization often varies from one station to another for the same event, and that the time duration in which the same rotational sense is maintained is shorter near midnight than in the morning and late evening. It is suggested that the source regions of the morning and late-evening Pi 3's lie on the electrojet boundaries; that is at the Harang discontinuity (in the evening) and at the poleward edge of the westward electrojet (in the morning and evening). The generation of midnight Pi 3 pulsations, centered at a location within the westward auroral electrojet appears to be associated directly with the generation of that electrojet.     

Neutral line motion due to reconnection in two-ribbon solar flares and magnetospheric substorms

Two kinematic models of line-tied reconnection are considered which describe the motion of a magnetic neutral line (NL) during the main phase of a two-ribbon solar flare and during the recovery phase of a magnetospheric substorm in the geomagnetic tail. The models are kinematic in that they use only the magnetic induction equation, which suffices to determine the position and velocity of the NL as functions of time if the rate of reconnection is prescribed. The solar flare model shows that the observed large decrease in the rate at which “post”-flare loops rise upward from the photosphere during the main phase does not require a corresponding decrease in the rate of reconnection. Instead it is found that a constant rate of reconnection can account for the motion of the loops for almost the entire period during which they are observed. By contrast, application of the same procedures to the recovery phase of the magnetospheric substorm in the tail predicts a slightly increasing speed of NL motion if the rate of reconnection is constant. Furthermore, it is found that the motion of the NL relative to the ambient medium may account for much of the observed asymmetry in the magnetic field in the plasma sheet during recovery. Due to this motion, the plasma sheet thickness may be up to 4 times smaller and the normal magnetic field component up to 2 times weaker in the region tailward of the NL than in the corresponding region earthward of the NL.     

Variations of magnetospheric convection electric fields during substorms as inferred from pc1 hydromagnetic waves

The convection electric field in the vicinity of the plasmapause in the midnight sector during magnetospheric substorms has been obtained on the basis of spectral analysis of Pc1 hydromagnetic (HM) waves observed at the low latitude station, Onagawa (Φ = 28.°3, Λ = 206.°8). Variations of the field are consistent for four independent substorm events studied. The calculation implies that the convection electric field increases westwards up to ~1.0 mV/m during the expansion phase of the substorms, changes polarity near the end of the expansion phase, and then points eastwards during the recovery phase.     

Scaling relations governing magnetospheric energy transfer

The functional dependence on solar wind parameters of the rate of energy transfer from the solar wind into the magnetosphere is subject to constraints imposed by dimensional analysis. The form and extent of the constraints depend on assumption about the energy coupling mechanisms, specifically on the relative importance of electromagnetic coupling (MHD flows effects), ionospheric conductivity effects (through Birkeland currents), and the viscous coupling. The effective viscosity coefficient scales in a well-defined manner with solar wind parameters, and its effect is dimensionally the same as that of more general finite-gyroradius mechanisms. We obtain the general form of the expression for energy transfer which takes all these effects into account and which can then be specialized to specific assumptions about the coupling mechanism. We point out the needed changes in energy transfer formulas previously used in the littrature, which make them conform to the requirements of dimensional analysis. Electromagnetic coupling yields the most restrictive formulas for energy transfer, although a unique expression cannot be obtained either on solely dimensional grounds or from presently available theory. Modifications required by the addition of viscous or finite-gyroradius effects are well defined but small and likely to be difficult to detect in practice. Assumptions of energy transfer by solar wind plasma entry leads to expressions equivalent, as far as dimensional arguments go, to those based on assumptions of electromagnetic or viscous coupling. Ionospheric conductivity effects are likely to be minor since Joule heating in the ionosphere is a relatively small fraction of the magnetospheric energy budget. All energy transfer formulas discussed presuppose a well-defined set of solar wind parameters and hence can be valid only on time scales longer than the solar wind flow time past the magnetosphere, which is also the expected time scale for energy storage (if any) in the magnetotail.     

Relationships between radio aurora,visual aurora and ionospheric currents during a sequence of magnetospheric substorms

In an earlier paper the latitudinal and longitudinal structure of ionospheric current flow during a sequence of magnetospheric substorms was presented (McDiarmid and Harris, 1976). In the present paper the relationships between the electrojets, the radio aurora observed at 48 MHz and the all-sky camera-recorded visual aurora are presented for the same substorm sequence. The previously described morphology of radio aurora during substorms is confirmed and the observed relationships can be explained.     

On the cause of northward magnetic field along the negative X axis during magnetospheric substorms

The magnetic fields produced by a three-dimensional current system, consisting of a flow into the morning part of the auroral oval along tail-like field lines, along the auroral oval and out from the evening part of the oval along tail-like field lines, are computed. It is demonstrated that the major parts of the well-known ‘positive bay’ in low latitudes on the Earth's surface, the positive H variation at the synchronous distance and the positive B_s variation along the magnetotail during magnetospheric substorms can be caused by the proposed current system.