Search

Magnetopause shadowing and wave-particle interactions are recognized as the two primary mechanisms for losses of electrons from the outer radiation belt. We investigate these mechanisms, sing satellite observations both in interplanetary space and within the magnetosphere and particle drift modeling. Two interplanetary shocks sheaths impinged upon the magnetopause causing a relativistic electron flux dropout. The magnetic cloud (C) and interplanetary structure sunward of the MC had primarily northward magnetic field, perhaps leading to a concomitant lack of substorm activity and a 10 day long quiescent period. The arrival of two shocks caused an unusual electron flux dropout. Test-particle simulations have shown 2 to 5 MeV energy, equatorially mirroring electrons with initial values of L 5.5can be lost to the magnetosheath via magnetopause shadowing alone. For electron losses at lower L-shells, coherent chorus wave-driven pitch angle scattering and ULF wave-driven radial transport have been shownto be viable mechanisms.

The Bastille Day magnetic cloud in July 2000 occurred with NEAR in conjunction with the Earth at a radial distance of 1.76 AU and 1.9° from the Earth-Sun line. Propagation time from ACE at 0.99 AU to NEAR indicates the cloud did not decelerate significantly between the Earth and 1.76 AU. Using a non-force-free, kinematic flux rope model we find the rope contained 130 TWb of magnetic flux, was oriented with clock and cone angles of 50° and 83°, and had a radius of 0.25 AU at ACE. At NEAR its radius had expanded to 0.43 AU. Simultaneous modeling of ACE and NEAR data indicate the axial and poloidal magnetic fields vary as R−1.4 and R−1.2 where R is heliocentric distance. Magnetosheath thicknesses of 0.14 AU and 0.23 AU indicate the rope cross section is elongated normal to the cloud axis and the radial direction.

During the magnetic cloud event of January 10–11, 1997, several density enhancements in the solar wind were observed while other solar wind parameters were relatively constant. By using this fortuitous opportunity, the effects of the solar wind density on the intensity of the auroral electrojets are examined. It is found that a positive relationship exists between the solar wind density and the intensity of the auroral electrojets. The relationship is found to be much weaker for northward interplanetary magnetic field (IMF) than for southward IMF. It is also shown that when the IMF is directed southward, the density effect is dominant in the westward electro jet, while it is most effective in the eastward electro jet when the IMF is directed northward.

Pioneer Venus observations from 1979–1988 have revealed the solar cycle evolution of magnetic clouds in the inner heliosphere. Non-force-free magnetic rope models, fit to 44 magnetic clouds, allow us to determine the structures of these clouds and compare their properties with potential field maps of their solar source regions. The leading magnetic polarity in these clouds is correlated with the polarity of the sun's global field. The axis of symmetry of the clouds varies with the inclination of the magnetic streamer belt. At 0.72 AU, the median radius of a magnetic flux rope is 0.10 AU. The median rope contains 11 TWb and carries 0.5 GAmps of parallel electric current. The occurrence frequency of different handedness of the ropes is equally left and right throughout the solar cycle and shows no correlation with other modeled parameters.

Riometers monitor the changes in ionospheric conduc- tivity by measuring the absorption of very high frequency radio noise of galactic origin passing through the ionosphere. In this Letter the absorption of radio signals by a thin layer of ionospheric plasma, produced by ionization due to energetic precipitating elec- trons, is modeled by taking into account strong turbulent heating caused by instabilities. The precipitating electron population is obtained from a global MHD simulation of the magnetosphere, along with the electric fields which excite the Farley-Buneman instability and lead to turbulent electron heating. A comparison, the first of its kind, of the data from polar and sub-auroral riome- ters for the magnetic cloud event of January 10, 1997 shows good agreement. The ionospheric conductance modified by turbulent electron heating can be used to improve the magnetosphere- ionosphere coupling in the current global MHD models.

Using data from the MASS high-resolution solar wind spectrometer on the WIND spacecraft, six enhanced ³He2+/4He2+ periods were identified from January 1995 to May 1998. The ratios observed in these events are four to ten times higher than previously reported average solar wind values. All enhanced ³He2+ periods were associated with the passage of CME-ejecta. The observed time periods were either within but towards the end of the CME-associated magnetic clouds or trailing the magnetic cloud by about 24 hours. All the enhanced ³He2+ periods have a mixture of low (cold) iron charge states along with high (hot) oxygen charge states. The unusual charge state composition of these enhanced ³He2+ events suggests the plasma came from a relatively cold and dense solar region, or from a rapidly expanding plasma parcel in the corona. We suggest that the enhanced ³He2+ plasma originated from the prominence core embedded within the CME.

The 3D Plasma and Energetic Particle (3DP) instrument on the WIND spacecraft detected extremely cold solar wind electrons in the January 10–11, 1997 magnetic cloud, ∼4 times lower temperature than any previously reported measurement. Detailed fits to the electron distributions show that the core electron temperature, Tec, generally ranged from ∼1 to 4.5 eV through the cloud, dropping to a low of ∼0.7 eV in an unusually dense (n>∼150 cm−3) region in the trailing portion of the cloud. Due to the extremely low average electron temperature Te and high density, the ions and electrons are collisionally coupled as confirmed by Te=Tp, the first such observation in the solar wind. For most of the cloud the halo density is very low (neh

Using an observed relation between speeds of CMEs near the Sun and in the solar wind, we determine an “effective” acceleration acting on the CMEs. We found a linear relation between this effective acceleration and the initial speed of the CMEs. The acceleration is similar to that of the slow solar wind in magnitude. The average solar wind speed naturally divides CMEs into fast and slow ones. Based on the relation between the acceleration and initial speed, we derive an empirical model to predict the arrival of CMEs at 1 AU.

Small-scale magnetic flux ropes have been discovered in the solar wind at 1 AU in observations from the IMP 8 and WIND spacecraft. These small magnetic structures (diameter of 270 RE, on average) have some similar properties to magnetic clouds (diameters of 0.2–0.3 AU or about 6000–8000 RE), which are well known large-scale magnetic flux ropes, but have durations of 10s of minutes as opposed to many hours or days for most magnetic clouds. The presence of these small helical field structures suggests that solar wind flux ropes may have a wide-range of scale sizes, or possibly have a bimodal size distribution, and are perhaps more common than previously estimated. Similarities and differences with magnetic clouds will be discussed. We suggest that these small scale magnetic flux ropes are signatures of magnetic reconnection in the solar wind as opposed to in the solar corona.

A sudden drastic change in the solar wind dynamic pressure is one proposed mechanism for the excitation of cavity mode resonances in the plasmasphere. To study the possibility of cavity mode occurrence following a 180 cm−3 solar wind density pulse on January 11, 1997, a numerical MHD simulation was run on a dipole grid. The simulated spectra recorded by a virtual satellite are compared with those produced from the Polar magnetic time signals during the interval 0228–0304 UT. In both sets of spectra, power is concentrated between 2 and 5 mHz; observed FLR-like spectral peaks in the range 6–20 mHz are in rough agreement with simulated plasmaspheric FLRs between 7–21 mHz. In the simulation, the fundamental plasmaspheric cavity mode is at 3.5 mHz; phase analysis of observed 3.5 mHz electric and magnetic field signals from the interval 0230–0237 UT supports the possibility of a heavily damped standing wave at this frequency.

November 1997 saw a series of very intense and unusual solar particle events that have been well studied [Cohen et al., 1999; Lario et al., 1998; Mason et al., 1999; Mazur et al., 1999; Mobius et al., 1999]. However, the composition of the solar wind has not previously been analyzed. We report the elemental and charge-state composition of the solar wind during these events. We find unusually broad and high charge-state distributions for O, Si, and Fe. Fe/O was enhanced by a factor >3–5 relative to the ambient solar wind in the magnetic cloud and in the following material. We speculate whether the observed compositional similarities in the solar wind and energetic particles might, in part, be explained by a common reservoir out of which both kinds of particles were accelerated.

Solar wind suprathermal electron velocity distributions measured by the SWEPAM experiment aboard ACE were studied to determine the internal magnetic structure of the coronal mass ejection (CME) that passed 1 AU on 4–5 February, 1998. Pitch-angle distributions range from symmetric counter-streaming to complex distributions that include conics. Identification of suprathermal electron conies in interplanetary CMEs (ICMEs) is reported here for the first time. The origin of these conies is highly uncertain. We suggest, however, that the conics may originate in transient electron heating associated with magnetic reconnection within the magnetic legs of the CMEs. Such reconnection generates large, complex magnetic loops that have enhanced anti-solar electric fields at their solar edge to maintain charge neutrality. This field must be sufficiently strong to bind electrons with high pitch angles to the bottom of the loops as they propagate away from the Sun.

Measurements of the polytropic index γ inside a magnetic cloud showed that there are two non-equal tubes inside the cloud [Fainberg et al., 1996; Osherovich et al., 1997]. For both tubes, γ < 1, but each tube has its own polytrope. We test equilibrium solutions which are a superposition of solutions with cylindrical and helical symmetry [Krat and Osherovich, 1978] as a new paradigm for a multi-tube model. Comparison of magnetic and gas pressure profiles for these bounded MHD states with observations suggests that complex magnetic clouds can be viewed as multiple helices embedded in a cylindrically symmetric flux rope.

Elemental, isotopic and charge state abundances provide valuable information about the source and acceleration mechanism of Coronal Mass Ejections (CMEs). Even though the kinetic properties of the plasma might be subject to changes because of dynamic effects occurring during the expansion of the CME, the composition of the solar wind remains unchanged after it leaves the low corona. Data from the Solar Wind Ion Composition Spectrometer (SWICS) on ACE are used to study the elemental and charge state composition ofHe, O, C, N, and Fe as well as the isotopic ratio of He during the very large CME of May 2-3, 1998. We find in this CME anomalously large enrichment of 3 He ++ / 4 He ++ , He/O and Fe/O. During the 28 hour long cloud portion of the CME unusually cold material ( 4 He + and very low charge state heavy ions) was observed together with hot (high charge state ions) and normal solar wind plasma.

A coronal mass ejection and magnetic cloud containing an unusually large enhancement of He+ was observed in the solar wind by the plasma and magnetic field instruments on the Advanced Composition Explorer (ACE) spacecraft on May 2–4, 1998. The He+/He++ ratio during this event exceeded 0.5% for a period of more than 24 hours, and reached values as high as 100%. The high He+/He++ ratio indicates the presence of prominence material, and in fact a disappearing filament and prominence were observed at the Sun in association with this event. The prolonged observation of He+ indicates that prominence material extended through much of this CME, the first such observation in a CME in the solar wind.

This paper reports the first scientific results from the Solar Wind Electron Proton Alpha Monitor (SWEPAM) instrument on board the Advanced Composition Explorer (ACE) spacecraft. We analyzed a coronal mass ejection (CME) observed in the solar wind using data from early February, 1998. This event displayed several of the common signatures of CMEs, such as counterstreaming halo electrons and depressed ion and electron temperatures, as well as some unusual features. During a portion of the CME traversal, SWEPAM measured a very large helium to proton abundance ratio. Other heavy ions, with a set of ionization states consistent with normal (1 to 2x10(exp 6) K) coronal temperatures, were proportionately enhanced at this time. These observations suggest a source for at least some of the CME material, where heavy ions are initially concentrated relative to hydrogen and then accelerated up into the solar wind, independent of their mass and first ionization potential.

In a previous work we have shown that interplanetary ejecta, which sometimes appear behind transient shocks and sometimes appear alone, are related to coronal mass ejections (CMEs) associated with some kind of near-surface solar events (Hα flares or prominence eruptions). In this paper we study the characteristics of these particular CMEs by taking the list of interplanetary shocks with and without ejecta from the Solwind/Helios combined study of CMEs and transient interplanetary shocks. From the same list and the Solwind CME Catalogue we obtained the characteristics of their associated CMEs and compared them. We find that the mean velocity of ejecta-associated CMEs is higher than that of CMEs associated with shocks without ejecta, and belong to different structural classes. We also find that the CMEs of ejecta cases are almost always associated with near surface solar activity, while CMEs of no-ejecta cases are rarely accompanied by this kind of activity. Also, the duration of the associated near-surface events is much longer for CMEs related to interplanetary ejecta.

On January 10 and 11, 1997, a coronal mass ejection (CME) encountered the Earth's magnetosphere and produced a moderate geomagnetic storm (Dst = −84 nT). During this event, the GEOTAIL spacecraft was skimming the dayside magnetopause, and observed a number of magnetopause crossings in response to the varying solar wind dynamic pressure. Around 1200 UT on January 10, GEOTAIL was traveling in the vicinity of the magnetopause at 1500 LT. Energetic oxygen ion bursts were repeatedly observed by the HEP/LD spectrometer on board GEOTAIL. This was a period of strong southward IMF related to the magnetic cloud structure of the CME. Two of these events with clear energy dispersion effects lasted from 1050 to 1113 UT, on January 10. This can be explained by invoking a quasi-steady layered structure of the energetic oxygen ions in the magnetosheath with the higher energy particles further away from the magnetopause than the lower energy ones. The observed energy dispersion is interpreted as a result of a crossing through this spatial layer due to the arrival of a filamentory pressure pulse in the solar wind. It is suggested that the existence of the structured oxygen layer is the result of particle leakage through the dayside magnetopause into magnetosheath.

The change m in the magnetic energy stored m in the Earth`s magnetotail as a function of the solar wind, BIF conditions are investigated using an empirical magnetic field model. The results are used to calculate the two normal modes contained m in the low-dimensional global model called WINDMI for the solar wind driven magnetosphere-ionosphere system. The coupling of the magnetosphere-ionosphere (MI) through the nightside region 1 current loop transfers power to the ionosphere through two modes: a fast (period of minutes) oscillation and a slow (period of one hour) geotail cavity mode. The solar wind drives both modes m in the substorm dynamics.

A magnetic cloud embedded in a corotating interaction region was observed by Wind 244 Re upstream of Earth on May 13, 1995. This cloud was unusual as it had at least two internal structures and the profile of the magnetic field strength was far from that predicted by a simple force-free flux rope model, although the field direction fit well. About 90 min later, this cloud was detected by the IMP 8 and Geotail satellites in front of Earth's bow shock. Geotail observed a sunward movement of the bow shock 8 min after the passage of the front boundary. An earthward movement of the bow shock was observed about 3 hours later. From the normal vector of the bow shock front, we deduce that the shape of the expanding bow shock was distorted from an axially symmetric geometry. The shape was consistent with distortion due to the observed inclination of the front boundary of the magnetic cloud.

In January 1997 a large fleet of NASA and US military satellites provided the most complete observations to date of the changes in >2 MeV electrons during a geomagnetic storm. Observations at geosynchronous orbit revealed a somewhat unusual two-peaked enhancement in relativistic electron fluxes [ Reeves et al., 1998]. In the heart of the radiation belts at L ≈ 4, however, there was a single enhancement followed by a gradual decay. Radial profiles from the POLAR and GPS satellites revealed three distinct phases. (1) In the acceleration phase electron fluxes increased simultaneously at L ≈ 4–6. (2) During the passage of the cloud the radiation belts were shifted radially outward and then relaxed earthward. (3) For several days after the passage of the cloud the radial gradient of the fluxes flattened, increasing the fluxes at higher L-shells. These observations provide evidence that the acceleration of relativistic electrons takes place within the radiation belts and is rapid. Both magnetospheric compression and radial diffusion can cause a redistribution of electron fluxes within the magnetosphere that make the event profiles appear quite different when viewed at different L-shells.

The polytropic index γ is a key parameter in the modeling of interplanetary magnetic clouds. In contrast to other models, the self-similar model of Osherovich et al [1993a] suggested γ < 1 in clouds. The first measurements [Osherovich et al, 1993b] showed that γ ≈ γe = 0.4 - 0.5 in magnetic clouds demonstrating the dominant role of electrons over protons (Te/Tp ≈ 6-7 at 1 AU). For the January 10-11, 1997 cloud observed by WIND, we found that the polytropric relation Te = FNeγe−1 is valid with γe = 0.52 ± 0.04 and F = 1.97 105 ± 0.12 105 (cgs units) over the number density range 1 < Ne < 10 cm−3. The constant F is half the value found for other clouds studied at 1 AU, while γe is essentially the same. We believe that the small value of F should be attributed to a later stage of the self-similar evolution of this cloud as compared to other clouds studied previously at 1 AU.

A magnetic cloud associated with the 1997 January 6 halo coronal mass ejection (CME) impacted the magnetosphere on January 10. With the WIND solar wind observations we make predictions 1–3 hours ahead of the magnetospheric response (Dst) to the halo-CME event. The recently developed neural network models with different solar wind input parameters are applied in this study. A major geomagnetic storm triggered by the cloud is accurately predicted by the models with inputs: (a) P1/2, V Bz, (b) B, N, V, Bs, and (c) B, By, Bz, N, V. Further, the three models can predict the effect of the density enhancement at the rear of the cloud which might be associated with the disappearing filament observed on January 6, 1997.

We present an initial report on the first direct, concurrent measurements of solar wind oxygen (O) charge state variations inside and outside Earth's magnetosphere. On 10-11 Jan 1997 a magnetic cloud, with a spatial scale of at least ∼100 RE engulfed the magnetosphere. At ∼1200 UT on 10 Jan 1997, the Wind/SWICS ion charge-state spectrometer, at ∼90 RE upstream of Earth, detected an increase in the solar wind O charge state ratio Γo = (O+7 + O+8) / O+6 at ∼0.5-30 keV/e which persisted until ∼2400 UT on 10 Jan 1997. After ∼1110 UT on 10 Jan 1997, Geotail was in and near the mid-afternoon to dusk magnetosphere. The ∼9.4-210 keV/e Γo value from the Geotail/STICS spectrometer showed a longer-term, gradual increase over ∼8–10 hr. Then, after ∼0600 UT on 11 Jan 1997, Wind observed a Γo decrease at a high speed solar wind stream onset. About 6 hr later Γo decreased promptly at Geotail during a plasma sheet substorm injection. Both instances are long durations compared to nominal 3-hr substorm time-scales.

Strong enhancements of the cosmic ray anisotropy were observed before and during the January 1997 CME/magnetic cloud. From a multi-station analysis of neutron monitor data, we conclude that “B × ∇n” drift is a primary source of CME-related anisotropies for 5 GeV cosmic rays. Evolution of the cosmic ray density and density gradients is closely linked to magnetic properties of the ejecta, and provides information on the magnetic cloud and related features as they approach and pass Earth. Strong enhancement of the field-aligned anisotropy was observed primarily during the 9 hours prior to shock arrival. If typical, this phenomenon should prove useful for space weather forecasting.

The Ap index is a daily average of the range of magnetic disturbance in each three-hour UT interval. It is the only global magnetic index routinely forecast. Predictions of Ap are subjective estimates made on the basis of past behavior of Ap and events at the Sun. The predicted variance in present forecasts is only 30%. Here we develop a set of linear filters for Ap forecasting. We show that auto regressive filters with constant coefficients do as well as human forecasters. Filters with time varying coefficients predict 37% of the variance. If we assume Ap is driven by solar wind velocity through an auto regressive moving average we predict 47% of the variance. However, this filter requires today's average solar wind velocity to predict today's Ap. Only with remote sensing of the solar wind a day in advance would it be possible to implement this model. We apply our filters to data for January 1997 determining whether effects of the January 6–11 magnetic cloud can be predicted. The filters do not predict CME related activity, but they do predict activity later in January obtaining a monthly average prediction efficiency of 59% that should be compared SEC results of 40%.

Nearly ten years of continuous magnetic field observations by the Pioneer Venus spacecraft allows us to study the correlation between the structure of magnetic clouds in the inner heliosphere and the phase of the solar cycle. Fifty-six magnetic clouds have been identified in the PVO data at .7AU during 1979–1988. As this period spans nearly two solar maxima and one solar minimum we can study the evolution of the structure of these magnetic clouds through varying solar activity and under various orientations of the coronal streamer belt. Until shortly after the 1979 solar maximum the majority of clouds had an initially southward magnetic field which turned northward as the cloud was traversed, while in the period leading up to the 1988 solar maximum the majority had a northward field that turned southward. In the declining phase of solar activity magnetic clouds continued to occur, but only a minority can be classified as having south-to-north and north-to-south rotations. The majority of these clouds occurred with the field remaining entirely north or south relative to the solar equator. These results confirm observations using Helios and ISEE data indicating that the structure of magnetic clouds varies in response to changes in the magnetic structure of the source region. By interpreting these observations to imply that the leading magnetic field in magnetic clouds is controlled by the polarity of the sun's global field and that the inclination of the coronal streamer belt controls the axis of symmetry of the clouds, we can predict preferred magnetic cloud structure and orientation during varying phases of the solar cycle. The helicity of the observations does not seem to be ordered by the solar cycle.

The arrival of the magnetic cloud on January 10, 1997, 01:05 UT, was preceded by a 12-hour interval of increased fluctuations in the IMF orientation, which drove boundary oscillations in the magnetosphere. The GEOTAIL spacecraft, initially located near the dawnside magnetopause, experienced multiple crossings of the magnetopause: at the geosynchronous distance, GOES 9 observed weak perturbations associated with these oscillations. During this time, the POLAR spacecraft encountered the dense high-latitude boundary layer, followed by several rapid encounters with the tenuous lobe. The boundary between the mantle and the lobe was observed to be thin as the crossings lasted only 30 seconds. The source of these boundary oscillations is magnetosheath pressure variations produced by rapid IMF directional changes.

We simulate the ring current evolution during the magnetic storm caused by Earth passage of the January 1997 magnetic cloud. Compared to previous studies, we include for the first time energy diffusion caused by wave-particle interactions. The modeled Dst index agrees reasonably well with the measured one, corrected for magnetopause currents and currents induced in the solid Earth. We compare H+distributions calculated from our model with those measured by the HYDRA instrument on the POLAR spacecraft and find that: a) the agreement between theory and data at large Lshells (L>5.5) is very good; b) although the agreement at low Lshells is improved when scattering by EMIC waves is included, the result is not entirely satisfactory, suggesting that either transport in a more realistic magnetospheric electric field or additional loss processes should be considered.

Ion drift measurements from DMSP F13 provide the first opportunity to study the evolution of high latitude potentials in both hemispheres as a magnetic cloud passed Earth. Early in the event with IMF B Z strongly southward, the polar cap expanded and the potential measured across it reached 174 kV, then decreased as the IMF rotated toward B Y < 0. During the initial northward IMF phase convection in the polar cap was dominated by large lobe cells, clockwise in the southern and counterclockwise in the northern hemisphere. With purely northward IMF, convection patterns had four cells and were irregular on the day and night sides of the magnetic dawn-dusk meridian. Potential distributions measured with a southward or westward IMF agree with predictions of the Weimer [1996] model, suggesting that some global MHD simulations overestimate dayside merging rates.

Data are presented from two charge plate sensors on the Loral Telstar4 geosynchronous communications spacecraft during an interplanetary magnetic cloud event that intersected Earth on January 10, 1997. The energy input into Earth's magnetosphere from the solar wind during the event produced enhanced plasma populations that caused the charge plates to attain values more than a factor of two larger than at any other time during the month. The time variation of the charging during the event is found to be similar to the variation in conjugate area ionospheric current systems measured by magnetometers near the footpoints of the magnetic flux tube that passes through the spacecraft.

VLF chorus emissions occur near the dawn meridian in fast response (

The polarity of a magnetic cloud refers to its changing magnetic field direction. It is classified as S-N polarity when the magnetic field rotates from southward to northward and N-S polarity when the field is initially northward and rotates southward. A study of 29 magnetic cloud events has found that 40–45% of magnetic clouds, independent of polarity, are followed by a fast solar wind stream which compresses the tail end of the cloud. The compression results in an increase in the solar wind plasma density and in 64% of the cases an increase in the magnetic field strength towards the latter part of the cloud. Such tail end compression can have a significant effect upon geomagnetic storm intensity if the magnetic cloud is of N-S polarity. This is because only in the N-S polarity case does the compression coincide with the southward IMF portion of the cloud. To test the “geoeffectiveness” of N-S versus S-N magnetic clouds three selected magnetic cloud events, two of S-N polarity and one of N-S polarity, are investigated in terms of their geomagnetic response through measured and estimated Dst values. It is found that there is an increased geoeffectiveness of N-S polarity clouds due to both an increased solar wind dynamic pressure and a compressed southward field associated with a following fast solar wind stream.

We study features of the January 1997 magnetic cloud and their effect on the optical aurora at 0630–0930 magnetic local time (MLT). WIND data suggest a thin plasma depletion layer (PDL) preceding the cloud, at whose outer edge the magnetic field rotates southwards, the proton temperature has a local maximum, and the dynamic pressure drops by a factor of 2. At the cloud's front boundary there is a further dynamic pressure drop, another localized temperature rise, and a ∼30° field shear. Prior to arrival of the southward rotation at Earth, the ∼08 MLT aurora is dominated by forms with intense 557.7 nm emission, presumably of boundary plasma sheet origin, located south of zenith (∼75° MLAT) and moving eastward. Minutes after the southward rotation reaches Earth, this emission is replaced by an auroral form which encroaches into the field-of-view at ∼ 73°MLAT, expands southward to ∼70° MLAT and westward, approaching ∼0800 MLT at 0600 UT. Its latitudinal positions satisfy a known relation between cusp latitude and interplanetary BZ, and its morphological and spectral properties are similar to those previously associated with an energy-latitude dispersion signature attributed to plasma transfer at an open low latitude boundary layer. We argue that this auroral precipitation at such far off-noon MLTs may be due to enhanced reconnection occasioned by the strongly southward field in the PDL/cloud. A POLAR pass across the open/closed field line boundary at 66° invariant latitude is consistent with these inferences.

On January 11, 1997, at 03:40:00 UT, while Polar was traveling up the dusk flank toward apogee, two ion instruments, TIDE and TIMAS, detected upflowing H+ with an energy/pitch-angle dispersion resembling an ionospheric reflection of freshly injected solar wind ions. In the same region of space, TIDE and TIMAS observed cold beams of O+ and H+ traveling down the field line with equal bulk velocities. We interpret these ion signatures as concurrent observations of mirrored solar wind ions and downward accelerated ionospheric ions. By 03:42:00, an energy/pitch-angle dispersion of downward moving ions at very low energies was clearly evident in the TIDE data. This additional signature is interpreted as an indication of reconnection on the same field line in the southern hemisphere. We explain this unique combination of plasma distributions in terms of high-latitude reconnection and magnetic field line convection during northward-IMF conditions associated with the January 1997 magnetic cloud event.

During the January 1997 coronal mass ejection/magnetic cloud event, the POLAR spacecraft experienced three successive perigee passes with dramatically different interplanetary magnetic field (IMF) and solar wind conditions on January 9, 10 and 11. The magnetic field observations during these three successive perigee passes are used to examine the response of the low-altitude polar and auroral magnetosphere to different IMF and solar wind conditions. We find that field-aligned currents are controlled mainly by the southward IMF component. They are greatly enhanced during strongly southward IMF conditions on January 10, but the solar wind dynamic pressure has little effect on their strength during the passage of the high-density solar wind filament on January 11. Thus, drag on the magnetosphere is principally due to reconnection at the magnetopause, with little due to viscous effects. Effects of the ring current and the magnetopause current can be identified in the magnetic field data in the low-altitude magnetosphere. During the magnetic storm of January 10, the ring current exhibits strong dawn-dusk asymmetry. It is much stronger in the dusk sector than in the dawn sector.

Dynamic and intense polar auroras were observed over Eureka, Canada (88.9°N magnetic) through January 10 and 11 following the solar coronal mass ejection (CME) of January 6, 1997. A dusk sector arc appeared by 0200 UT on January 10 following the arrival of the magnetic shock at 0110 UT; aurora filled the Eureka field of view between 0225 and 0300 UT. Spectral easurements with allsky cameras, meridian scanning photometers and a CCD spectrograph showed this aurora to be excited by electrons of E0 ≈ 3 keV. After an interval of F-layer patches, faint F-layer arcs (E0 ≈ 300 eV) reappeared by 1900 UT and continued until 0700 UT on January 11, intensifying between 0200 and 0700 UT. The initial auroral display was remarkable in its spectral characteristics, while the prolonged F-layer aurora which followed was similar to the aurora observed during the October 1995 magnetic cloud event.

The ionospheric storm evolution process was monitored during the January 10, 1997 magnetic cloud event, through measurements of the inonospheric total electron content (TEC) from 150 GPS stations. The first significant response of the inonospheric TEC to the geomagnetic storm was at 0300 UT as an auroral/subauroral enhancement around the Alaskan evening sector.

The results of a 3D MHD simulation of the January 10-11, 1997 geomagnetic storm are presented. The simulation results agree well with ground-based and geosynchronous observations. The 28 hours modeled by the simulation include the magnetic cloud responsible for the storm, the shock preceding the cloud, and the dense plasma filament following it. The simulation shows that during the period of southward (MF ionospheric activity was strongly correlated to the solar wind density. The arrival of the plasma filament during northward IMF pushed the dayside magnetopause well within geosynchronous orbit, but generated little ionospheric activity. It appears that n,w as well as the orientation of Bsw plays a role in controlling the intensity of ionospheric and magnetospheric activity.

On May 27, 1996 two all-sky cameras in Antarctica observed the aurora during the passage of a magnetic cloud. Compression of the magnetosphere by the solar wind dynamic pressure pulse in front of the magnetic cloud caused a short, strong, and localized blue auroral emission which moved towards noon in the late morning sector. The emission was caused by electrons of 8 keV energy. The high dynamic pressure caused a diffuse large-scale (at least 3 h MLT and 5° latitude) blue emission on closed field lines. The high energy electron precipitation disappeared after the enhanced solar wind dynamic pressure region had passed the sub-solar magnetopause.

We compare the substructures of the 1997 February 07 coronal mass ejection (CME) observed near the Sun with a corresponding event in the interplanetary medium to determine the origin of magnetic clouds (MCs). We find that the eruptive prominence core of the CME observed near the Sun may not directly become a magnetic cloud as suggested by some authors and that it might instead become the ”pressure pulse” following the magnetic cloud. We substantiate our conclusions using time of arrival, size and composition estimates of the CME-MC substructures obtained from ground based, SOHO and WIND observations.

Geotail was at the magnetopause near the sub-solar point when the magnetic cloud struck the earth's magnetosphere on January 10, 1997. Reeonnection between the southward IMF and the magnetospheric field lines occurred at the dayside magnetopause as observed by Geotail. The signatures of the reconnection were, i) accelerated plasma flow from the reconnection site which reached up to 500km/s, ii) ion heating through the reconnection process, and iii) a velocity filter effect seen in the ion distribution function in the magnetosheath boundary layer. This reconnection may be related to the large substorm onset observed at geosynchronous orbit.

Magnetic field measurements from the POLAR satellite are presented for three dusk to dawn apogee passes during the January 1997 magnetic cloud event. The variability in the residual fields from one orbit to the next are investigated in terms of the changing solar wind and magnetospheric current systems. The observed fields are compared to the Tsyganenko 96 (T96_01) model which allows dynamic solar wind and Dst inputs. The model fields are found to generally agree with the observed fields with the greatest discrepancy occurring during periods of large dynamic pressure. An investigation of the contributions to the T96_01 model by the various solar wind and Dst inputs provides insight into the solar wind control of the high altitude magnetosphere above the polar cap. It is found that the ring current is responsible for the largest proportion of the observed magnetic field residuals while the interplanetary magnetic field and solar wind dynamic pressure control a smaller proportion primarily producing the shorter time scale variations in the observed fields.

Relativistic electrons in the outer radiation belt associated with the January 1997 magnetic cloud event were observed by the HIST instrument on POLAR at kinetic energies from 0.7 to 7 MeV and L shells from 3 to 9. The electron enhancement occurred on a time scale of hours or less throughout the outer radiation belt, except for a more gradual rise in the higher energy electrons at the lower L values indicative of local acceleration and inward radial diffusion. At the higher L values, variations on a time scale of several days following the initial injection on January 10 are consistent with data from geosynchronous orbit and may be an adiabatic response.

We present the results of a global magnetospheric simulation of the initial period of the January 10 - 11, 1997 magnetic cloud event. Distinct magnetospheric boundaries derived from the model are mapped down to the ionosphere and compared to UVI images of the auroral oval from 1:00 to 4:30 UT. The convection reversal boundary, which coincides with the maximum Region I currents, tends to generally match the UVI equatorward boundary. This boundary is almost always poleward of the boundary separating Region 1 and Region 2 currents. The separatrix between open and closed magnetic field lines as mapped in the model matches the poleward boundary of the UVI images very well during quiet periods. During dynamic periods, however, when the separatrix can move several degrees in latitude in some sectors, the poleward boundary of the Region 1 currents matches the UVI images better.

The magnetic cloud and geomagnetic storm on January 10-11, 1997 were associated with a halo-type Coronal Mass Ejection (CME) observed by the SOHO/LASCO coronagraphs near the sun on January 6. We summarize the solar activity related to this CME and the subsequent storm at Earth. This solar activity was remarkably weak and unimpressive. If the wide CME had not been observed, the storm would not have been forecast. Thus this case represents an extreme example of so-called problem magnetic storms that lack obvious surface signatures of eruptive solar activity. It supports the view that CMEs involve the destabilization of large-scale coronal structures which may or may not have associated surface activity, and that CMEs, not the surface activity, are the key causal link between solar eruptions and space weather at Earth.

We determine the geometry of the fluctuations of the magnetic field at frequencies just above the proton gyrofrequency for the January 10, 1997 CME and the magnetic cloud within. The transverse magnetic fluctuations represent a greater fraction of the magnetic energy than is the case in the typical undisturbed solar wind. The break in the power spectrum that is associated with the the onset of magnetic dissipation falls within the frequency range of interest. The fluctuation geometry is markedly different above and below the spectral break frequency. The inertial range geometry remains unchanged in the cloud with only ∼30% of the energy associated with field-aligned wave vectors. The dissipation range wave vectors are highly oblique to the mean magnetic field B with up to 96% of the energy associated with oblique wave vectors.

On Jan. 10–11, 1997 a wide magnetic cloud reached the Earth triggering intense geomagnetic activity. Observations performed at low and very high latitude show that the same features appear simultaneously in correspondence to different changes in the solar wind conditions. In particular, highly polarized modes are simultaneously observed at the same discrete frequencies after the passage of the high density solar wind region following the cloud. SI's and ULF waves polarization are also examined in a wide latitudinal and longitudinal extent.