Search

Although solar activity may significantly impact the global environment and socioeconomic systems, the mechanisms for solar eruptions and the subsequent processes have not yet been fully understood. Thus, modern society supported by advanced information systems is at risk from severe space weather disturbances. Project for solar–terrestrial environment prediction (PSTEP) was launched to improve this situation through synergy between basic science research and operational forecast. The PSTEP is a nationwide research collaboration in Japan and was conducted from April 2015 to March 2020, supported by a Grant-in-Aid for Scientific Research on Innovative Areas from the Ministry of Education, Culture, Sports, Science and Technology of Japan. By this project, we sought to answer the fundamental questions concerning the solar–terrestrial environment and aimed to build a next-generation space weather forecast system to prepare for severe space weather disasters. The PSTEP consists of four research groups and proposal-based research units. It has made a significant progress in space weather research and operational forecasts, publishing over 500 refereed journal papers and organizing four international symposiums, various workshops and seminars, and summer school for graduate students at Rikubetsu in 2017. This paper is a summary report of the PSTEP and describes the major research achievements it produced.

To obtain the basic information necessary for planning a research data management (RDM) system in Kyoto University, Japan, we conducted a campus-wide survey of research datasets twice with different topics. Although only a part of campus members responded, some clear tendencies were observed. That is, (1) considerable portion of researchers are not very positive to make their data open because of a confidentiality obligation or for holding their predominance in their research fields, (2) strength of consciousness on open data is considerably different depending on their discipline, (3) the above difference may come not only from inherency of each discipline but also from the requirement of open data by the journals. (4) There exist more than fifty open data repositories, and more than half of them are managed by the institutions established for each specific field of research or by the organization outside of the campus. From the survey, it is suggested that the researchers are still conservative to make their data open, and a current practical necessity to RDM system in the campus seems to be a system for keeping and making the data open within each research group. By preparing a convenient RDM system as well as the requirement of open data by the journals in wider discipline may change the consciousness of the researchers on open data.

Data-DOI, data publication, and data citation will promote "Open Science". Recognizing their importance, solar-terrestrial physics (STP) data centers in Japan have been working to mint DOI to their database since August 2013. We participated from October 2014 in a 1-year pilot program for DOI-minting to science data launched by Japan Link Center, which is one of the DOI registration agencies. In the pilot program, a procedure of the DOI-minting for STP data was established. As a result of close collaboration with Japan Link Center, the first case of data-DOI in Japan (doi:10.17591/55838dbd6c0ad) was created in June 2015. The first case of data citation in Japan was also made. As of March 2018, there are 18 data-DOIs for the STP data. The effort of the DOI-minting to scientific data will be continued for database in our data centers and those in other fields of Earth Science.

The magnetic data obtained by the SWARM (the Earth's Magnetic Field and Environment Explorers) satellites in middle or low latitudes during the initial 2 months after launch were analyzed, when they flew nearly on the same orbit with variable time separation ranging from 5 to 100 s. It was confirmed that the small-scale magnetic fluctuations having period around 10–30 s are the manifestation of spatial structure of small-scale field-aligned currents along the orbits. From the statistical relation between correlation coefficients and two satellite separation in time, the typical time scale of temporal variation of the field-aligned current system is estimated to be around 200 s for meridional component and 340 s for zonal components of the magnetic fluctuations, respectively. Existence of shorter time scale around 30–50 s was also found. These results suggest that the main source of current generation is the acoustic mode of atmospheric gravity waves.

A time-dependent, two-dimensional, nonlinear, non-hydrostatic, compressible, numerical model is used to investigate the horizontal extension of acoustic waves trapped in a thermal duct between the lower thermosphere and the ground or the stratosphere, and evanescent and gravity waves around them. The simulation shows that an impulsive point source on the ground excites acoustic waves ducted in the mesosphere. The ducted acoustic waves have horizontal wavelengths longer than 130 km and frequencies of 3.7–3.8 mHz. This ducted acoustic waves are identified as acoustic resonance modes expected in previous studies. They extend to 300 km distance from the source after 2 h. In the case of the source with a finite width, the acoustic resonance would extend to 300 km out of the edge of the source. The temporal variation of two resonance modes is also declared. In the vicinity of the source, two acoustic resonance modes are formed which have frequencies of 3.7 and 4.4 mHz. The mode of 4.4 mHz is dominant above the mesopause for about 50 min after the impulsive perturbation of the source since the mode is leaky. On the other hand, the mode of 3.7 mHz is well trapped in the thermal duct and persist over 2 h. In addition to the acoustic waves, evanescent gravity waves with horizontal wavelengths shorter than 100 km and with frequencies between 2.4 and 3.6 mHz are excited in the stratosphere. Gravity waves freely propagating are also excited with frequencies lower than 2 mHz and with horizontal wavelengths shorter than 130 km.

The equatorial electrojet (EEJ) is a unique feature of the Earth’s external current systems because it must flow along the dip equator. This provides us with a tool to determine the nature of the variations imposed by competing main field models on the equatorial region. First we show that for certain regions a comparison between scalar geomagnetic measurements that use different models to remove the main field may not be reasonable. Next we found the intrinsic error in the determination of the possible location of the dip equator was ±9.8 km (0.088°) at 108 km altitude for the models shown here. Using scalar measurements from over 14,000 CHAMP satellite passes, the latitude of the maximum of the EEJ field at the satellite altitude was determined by subtracting four different models of the main field. We find that the location can be statistically determined to within ±0.5° of the dip equator (calculated at 108 km altitude) irrespective of longitude, time of the measurement, degree of magnetic activity, and subtracted model. However, variations of the latitude of the maximum EEJ field with longitude are sometimes caused by the actual model and are not always a physical phenomenon. By choosing one model, and assuming it is the best representation of the main field, we have also shown that the accuracy of determination of the position of the EEJ signal is reduced in the morning and evening hours and that a morning and evening shift in the location of the EEJ found using ground measurements is also seen here. There exists a clear annual variation in the position of the EEJ regardless of longitude: it is south of the dip equator in December which is in agreement with the findings of all previous studies.

One-second resolution geomagnetic data from 5 stations located at low-latitudes (i.e., L = 1.2 ∼ 1.6) were used to examine the local time dependence of the dominant frequency of Pi2 pulsations. We analyzed 183 Pi2 events simultaneously recorded at the 5 stations and discussed their possible generation mechanisms. The averaged dominant frequency of the H (horizontal) component is higher on the dawn side than that on the dusk side and shows a peak value in the post-midnight at around 03 LT, which confirms the LT dependence previously suggested from single station data (i.e., non-simultaneous observation). However, some of the events have no LT dependence. For the events which show the LT dependence at low-latitudes, we infer that the cavity resonance mode is a plausible generation mechanism, but we do not rule out the possibility of the plasmaspheric surface wave mode which has also been suggested. For the events having a common frequency between the eastern and western stations, we suggest the cavity resonance mode to be the mechanism. The averaged dominant frequency of the D component does not show any clear LT dependence, and only about 20% of Pi2s have identical frequency for both the H and D components, therefore we suggest that the H and D oscillations of Pi2s are generated from different mechanisms. We also found that the frequency of different parts of a Pi2 pulsation, i.e., the dominant frequency for leading part and trailing part of the pulsation, is different, and the frequency of the trailing part is lower than that of the leading part.

The geomagnetic jerk amplitudes, which are defined as abruptness of changes in the trends of geomagnetic time series, are investigated with geomagnetic monthly means computed from hourly mean values at each local time. A statistical time series model in which the trend component is expressed by a second order spline function with variable knots is constructed for each time series. The optimum parameter values of the model including positions of knots are estimated by the maximum likelihood method, and the optimum number of parameters including the number of knots are determined based on the Akaike Information Criterion (AIC). The jerks are detected objectively and automatically by regarding the optimized positions of knots as the occurrence epochs. This analysis reveals that the spatial distributions of jerk amplitudes essentially do not depend on the local time, which indicates that the jerks cannot be explained by abrupt changes in intensities of latitudinally flowing external currents such as the field-aligned currents. Longitudinally flowing currents, on the other hand, such as the ring current could explain the distributions. The abrupt changes of the ring current intensity are estimated from the distributions of jerk amplitudes in the eastward component in 1969, 1978, and 1991 supposing that an abrupt change in the ring current intensity causes a jerk. However those estimated changes cannot consistently explain the distributions of the jerks in the northward and downward components. Therefore it is plausible that the jerks which occurred in 1969, 1978, and 1991 are not caused by external sources but internal ones. It is also confirmed that the occurrence epochs of jerks in the southern hemisphere are a few years after those of the 1969 and 1978 jerks in the northern hemisphere, and it is also found that the jerk in the southern hemisphere occurred a few years after the occurrence of the 1991 jerk in Europe. Taking these time lags in occurrence epochs into account, it can be said that the 1969, 1978, and 1991 jerks are global phenomena.

Using magnetic field data obtained by the Challenging Minisatellite Payload (CHAMP), we show global and frequent appearance of small-amplitude (1 to 5 nT on the dayside) magnetic fluctuations with period around a few tens of seconds along the satellite orbit in middle and low latitudes. They are different from known phenomena, such as the Pc3 pulsations. The following characteristics are presented and discussed in this paper: (1) The magnetic fluctuations are perpendicular to the geomagnetic main field, and the amplitude of the zonal (east–west) component is larger than that of the meridional component in general. (2) As latitude becomes lower around the dip equator, the period tends to become longer. (3) The amplitudes have clear local time dependence, which is highly correlated to the ionospheric conductivities in local time (LT) 06–18. (4) The amplitude of the fluctuations shows magnetic conjugacy to a certain extent. (5) The amplitude shows no dependence on solar wind parameters nor geomagnetic activity. (6) A seasonal dependence is seen clearly. The amplitudes in the northern summer and winter are larger than those in the equinoxes. In the northern summer, the amplitudes above the Eurasian and South American continents and their conjugate areas are larger. In the northern winter, those above the eastern Pacific Ocean are larger. We suggest that the above characteristics, (1) to (6), can be attributed to the small spatial scale field-aligned currents having a lower atmospheric origin through the ionospheric dynamo process.

The HF radars of the Super Dual Auroral Radar Network (SuperDARN) observe the E x B convective drift, of ionospheric plasma when suitable small-scale field-Migned irregularities (FAI) from which to scatter are present. In order to estimate the distribution of FAIs in the subauroral F region, we have investigated the scattering occurrence percentage using data from six Northern Hemisphere radars during 39 months between July 1995 and September 1998. We have ident, ified a morphological tbature known as the dusk scatter event (DUSE; first reported by Ruohoniemi et al. (1988)) in detail and clarified its relation to the boundaries of the plasma density within the submroral F region. In all mont, hs, the DUSE appears within a fbw hours local time on the eveningside of sunset, where the magnetic latitude is slightly lower than the equatorward edge of the mroral oval, which corresponds to the density-depleted structure known as the midlatitude trough. There exists a clear UT effect in the characteristics of DUSE, such that DUSE is more pronounced tbr UTs corresponding to dusk on the prime magnetic meridian (0 o geomagnetic longitude). Furthermore, an enhancement of scattering occurrence around sunrise within t, he trough, which is termed dawn scatter event, (DASE), was newly identified. The region where the DUSE or DASE occurs has a close relationship with the duskside and dawnside ends of the midlatitude trough in the longitudinal direction (which are termed the sunward edges of the trough in this paper), where sunward directed steep plasma density gradients exist. The Kp dependence of the scattering occurrence was also examined. In disturbed conditions, the DUSE appee(rs at, earlier local times compared with that in quiet conditions, which is consistent with the Kp dependence of the sunward edge of the trough. Three models based on the gradient drift, instability were discussed. It, was tbund that the model which is based on the sunward plasma densit, y gradient at the sunward edges of the midlatitude trough and ambient electric field is most favorable tbr the growth of FAIs in this region and can consistently explain the statistical tbat, ures of DUSE and DASE. These facts indicate that we can est, imat, e t, he longitudinal extent of the midlatitude trough fi'om the appearance of DUSE and DASE observed by SuperDARN.

A new integration technique, symplectic method, is introduced and applied for the tracing of a charged particle motion in a dipole magnetic field. This method is an integral technique for the Hamilton's system by the repetition of canonical transformation, and has been tested in celestial mechanics in recent years. Compared with the standard Runge-Kutta method (4th order), the numerical error accumulation is much smaller for the symplectic method. An effective potential is introduced and used to discuss an abrupt transition between trapping and un-trapping regions.

Sudden perturbations of the ground magnetic fields at solar flares are called geomagnetic Solar Flare Effect (SFE). An SFE is the extra ionization produced by the X-ray emission from a solar flare. We analyzed 10 intense SFE events from 1996 to 1998 using SuperDARN HF radar network and other instruments located in the northern hemisphere. Two typical ionospheric signatures associated with the solar flares are revealed, one is a sudden fade-out of backscatter echoes, and the other is an appearance or variation of field-aligned irregularities. Sudden fade-out is observed only in the sunlit hemisphere, and the appearance or variation of irregularities are observed only near the terminator. In addition, we investigated one event out of ten in detail and found that there exists a discontinuity of the electric field or conductivity around the irregularity. This fact suggests that the variation of conductivity or electric field in the E region could affect the irregularity formation in the F region.

Conjugate Freja satellite and MU radar observations were conducted from 1993 to 1995 to clarify the generation mechanism of mid-latitude electric field fluctuations (MEFs) which were found from DE-2 satellite observations. The detection of MEFs at higher altitudes than 1000 km by the Freja satellite confirms that the MEFs are transmitted along the geomagnetic field line without any significant damping. On one night of such conjugate observations, MEFs were observed in a traveling ionospheric disturbance (TID). On the other nights when the MEFs were observed by Freja, large scale density structures in the ionosphere were observed with the MU radar. Spread-F phenomena were also observed by ionosondes associated with these structures. The large scale modulation of the ionosphere, such as TIDs, might cause small scale modulations which grow through ionospheric instabilities. The linear growth rate is evaluated using a set of equations describing the growth of the mid-latitude ionospheric irregularities. The growth rate is shown to be too small under average conditions; however, when the neutral wind in TIDs is included, it is large enough to generate MEFs.

After initial observations of Auroral Kilometric Radiation (AKR), their associations with substorms had been intensively investigated during 1970's. Herein we extend these studies and propose a new index available for substorm studies, named the AKR index. We define the AKR index by the power flux of AKR normalized to a reference distance from the Earth. The AKR index is found to show good correlations with Kp and Dst indices for weak magnetic disturbances. The AKR index tends to saturate for moderate and severe disturbance.

Interactions between Pc5 pulsations and particles are investigated with the magnetic field data obtained by the DE-1 polar orbiting satellite and the precipitating electron data from the DMSP satellites. We found geomagnetic conjunction events of DE-1 and DMSP in which these satellites respectively observed Pc5 pulsations and particle precipitation approximately on the same geomagnetic field lines. The invariant latitude ranges for the observations of Pc5 and precipitation of 3 keV electrons were almost the same. A statistical study also showed that the region of the central plasma sheet obtained by P. T. NEWELL and C.-I. MENG (J. Geophys. Res., 99,273,1994) falls in the region of the occurrence of transverse Pc5 pulsations. The precipitation of electrons having energies greater than 3 keV was investigated. It is shown that there are differences in the characteristics of 30 keV electron precipitation between the cases in which pulsations were observed and those in which pulsations were not observed. These results suggest that electrons are modulated by transverse Pc5 pulsations.

This paper reports plasma and field changes associated with magnetosheath encounters by GEOTAIL near the center of the tail axis at a down tail distance of X = -94 R E during the main phase of a weak geomagnetic storm on January 25, 1993. The magnetosheath entries of the satellite occurred in conjunction with an enhancement in the total pressure, having a temporal scale between 2 and 18 minutes. It is suggested that temporal and/or local structures in the solar wind such as pressure pulses caused these brief encounters. Assuming the flow direction in the magnetosheath to be parallel to the magnetotail boundary, we estimate the tail magnetic flux and the dimension of the tail for the encounter events. It is shown that solar wind variations, which generate geomagnetic storms, alter significantly the orientation as well as the size of the distant tail, such that the magnetosheath can be observed even near the center of the average tail.

Impulsive co-seismic geomagnetic variations were observed at two stations nearly 100 km apart from the source region of the 1995 Hyogoken-Nanbu earthquake (M= 7.2). The maximum amplitude at the main shock was 0.6∼1.0 nT and the duration was 20∼30 seconds. These variations seem to have commenced at the origin time and before the arrival of the seismic waves to the observation site of 17 seconds later. A superposed epoch analysis for the aftershocks also reveals similar geomagnetic variations though the amplitude of them is one order of magnitude smaller than that of the main shock. A crustal dynamo mechanism is discussed as a possible cause of such co-seismic geomagnetic variations.

There is generally high correlation between the orthogonal magnetic and electric field perturbations on the ionospheric field-aligned current region based on observations by polar orbiting satellites. The ratio between the magnetic and electric field perturbations, ⊿B_z/μ_oE_x, has a scale-length dependence : the ratio ⊿B_z/μ_oE_x decreases with decreasing scale length. B. FORGET et al.(J. Geophys. Res., 96,1843,1991) explained this scale-length dependence using a static model. In this paper, we compare the ratio ⊿B_z/μ_oE_x observed by the DE-2 satellite for various spatial scales with the effective Pedersen conductivity Σ_ calculated by the method of FORGET et al. The results show that in many cases this model can adequately explain the observed scale-length dependence. However, cases exist in which the difference, ⊿B_z/μ_oE_x-Σ_, decreases at about 0.25-1.0s in temporal scales (2-8km in spatial scales). This behavior is explained as being due to Alfven waves.

International audience

We have derived on a test basis a geomagnetic index by using presently available data in the southern polar cap. Observatories used here are Scott Base, Dumont d'Urville, Vostok and Mirny. To derive disturbance fields a base value of each station for each month is first calculated using a moving average method, and this base value is subtracted from hourly data from the station in that month. Then after a coordinate transformation the largest and smallest values are selected from the four stations, respectively in the same way as the AU and AL indices are derived, and we call them the polar cap AU and AL indices. The difference between these values gives the polar cap AE index, and we derived these indices for the H and D components and T, the total force of the geomagnetic field for the years 1966 (near solar activity minimum) and 1980 (near solar activity maximum). The results of the variation of the disturbance fields from the four stations show a clear diurnal and seasonal variation depending on distribution of the stations and the ionospheric conductivities in the southern polar cap. We find the highest correlation between the polar cap AE indices and the AE, AU, and AL indices in the local winter, and the correlation decreases in the local summer due to the increase of the ionospheric conductivity in the polar cap caused by solar UV radiation. The relationship between the IMF B_y component and the polar cap AU index during local summer seems to be quadratic. The results indicate that both positive and negative B_y increase the value of the polar cap AU index due to enhancement of the B_y dependent currents in the polar cap.

On April 9-11, 1983, the ISEE 3 spacecraft was continuously located within the earth's magnetotail for more than 36 hours at downstream distances of X = -76 to -80 R(e). During this span of time, 12 major intervals of substorm activity were observed in the AL index with good ISEE 3 telemetry coverage for 11 of them. In addition, there were two small substorms outside of these intervals, both with complete observations in the distant tail. This unusual ISEE 3 data set provides a unique opportunity to test the predictions of the near-earth neutral line model. In particular, the hypothesis that energy stored in the tail lobes during the growth phase is later dissipated, in part, through the release of one or more plasmoids following expansion phase onset is examined. Clear growth phase enhancements in the lobe magnetic field intensity preceded the onsets of nine of the substorms. Plasmoids, or their lobe signatures, traveling compression regions (TCRs), were observed at ISEE 3 in association with all 11 of the major substorm intervals for which there were ISEE observations, as well as for the two small substorms. No plasmoids or TCRs were observed in the absence of substorm activity. If these ISEE 3 observations are representative, then the release of plasmoids down the tail may be a feature common to all substorms.

[1] Periodic variations of the ionosphere were detected by ground-based GPS total electron content (TEC) measurements after the Sumatra-Andaman earthquake in 2004. The observational data showed that the 4-min periodic TEC variations occurred 1 h after the earthquake and continued for longer than 4 h. At the PHKT station, about 30 cycles of the 4-min periodic TEC variations were observed from 0230 to 0430 UT. The maximum peak-to-peak amplitude of the variations was about 0.6 total electron content unit (TECU; 1 TECU = 1016 el m−2) around 0320 UT. The frequency of these periodic variations was 3.9 mHz. They were detected by the SAMP station in the northern Sumatra and the PHKT and BNKK stations in Thailand with the signal from seven GPS satellites: PRN 8, 11, 13, 19, 23, 27, and 31. They were observed in a limited area from 4°N to 15°N in latitude and from 96°E to 101°E in longitude, although the western boundary was not certain because of the limit of the observational field of view. The amplitude of these TEC variations showed the dependence on the zenith angle of the path between the GPS receiver and satellite. The amplitude had a maximum when the zenith angle was the smallest. This could be caused by the vertical structure of the electron density variations. This also suggested that the electrons were oscillating along the geomagnetic field. The 4-min periodic TEC variations were interpreted to be induced by the long-lasting free oscillation of the atmosphere set up by the earthquake. Their long duration also indicates that they were generated by a nontransient process like resonance.

The formation of a closed ring current from a partial ring current is examined quantitatively by a particle trajectory tracing method. That is, we simulated the storm-time ring current by a model with monochromatic energy and pitch angle distribution using a dipole magnetic field with a time-varying electric field. It is found that the combination of the particle energy and the temporal variation of the cross-tail electric field can generate various patterns of the ring current development. It is also found that the recovery of the ring current is explained by the superposition of the “flow out” effect and the loss process of trapped particles by charge exchange. We estimated the geomagnetic Dst, index from the particle motions and obtained results similar to the actual Dst recovery with two different time constants.

The existence of a sharp break in the power spectrum of AE at about 5 hours is demonstrated. Several possible explanations of the nonlinear response of AE to the IMF Bs driver are briefly discussed, including: variable ionospheric conductivity (increasing with Bs) for the high frequency regime, and several AE saturation mechanisms for the low frequency regime.

We simulate particle motions in a realistic magnetospheric model under an impulsive dawn-dusk electric field and show an oscillation of the ring current intensity. This oscillation is caused by the effect of the asymmetric configuration of magnetospheric magnetic field. It is suggested that this effect explains the second development of the Dst often observed during recovery of storms after an impulsive southward IMF or in the impulse response function of the Dst to the IMF southward component.

Both D and H components of geomagnetic field data from 10 mid-latitude stations are used to examine the storm-time magnetospheric current system by deriving the longitudinally symmetric and asymmetric components of the fields. The asymmetry in D component suggests that net downward and upward field-aligned currents exist in the morning sector and from the evening to the early morning sector, respectively, when the IMF is southward. It also suggests the appearance of a net downward current in the afternoon sector at the developing stage of the storm-time ring current. These suggestions are confirmed by Magsat observation over the ionosphere since the direction of disturbances over the ionosphere is not the opposite but the same as that on the ground. This downward current signature co-exists with that of the morning sector during the southward IMF, suggesting that two different current systems co-exist at the developing phase of the ring current. One is the so-called partial ring current system whereas the other is the current system controlled by the IMF-Bz. A systematic phase shift of asymmetry in local time is seen in the dusk-side H but not in the D component, indicating that the contribution from westward drifting ring current particles can not be neglected as a cause for the H component asymmetry.

[1] To investigate the radiation belt, the Gauss Runge-Kutta method (RKG), one of the symplectic numerical integration methods, is introduced and tested by tracing protons and electrons in geomagnetic dipole field. The tracings of protons having the energies from 10 keV to 10 MeV with pitch angles from 90° to 30° at L ∼ 5 are practiced by fourth- and sixth-order RKG with time step 0.016 of the Larmor period on the equatorial plane. The accuracy of calculations are examined with Cartesian, cylindrical, and polar coordinates. With all coordinates, we confirmed that the calculations are stable for 10,000,000 steps with RKG. On the other hand, the calculations for protons in the same conditions with standard Runge-Kutta methods fail within 10,000 steps. The tracings of relativistic electrons having energies from 10 to 150 MeV with pitch angles from 90° to 30° at L ∼ 5 are also practiced. For the electrons above 50 MeV, the calculations are successful for 10,000,000 steps with the time step 0.1 × RE/c by fourth and sixth RKG with cylindrical coordinates. The calculations by standard Runge-Kutta method are failed in any coordinate system, and the calculations by fourth or sixth RKG with Cartesian or polar coordinates are also failed for all energy range above. The calculations for the electrons below 10 MeV with this time step size are all failed by any methods and with any coordinate systems mentioned above. For the effective calculations, for example, for 10 MeV electron with pitch angle 90° by fourth order RKG, we must use shorter timescale size than 0.2 of the above size.

[1] A typical Pi3 pulsation is examined by magnetic field measurements from multiple satellites and ground stations. Low-latitude ground observations with a wide longitudinal span indicate that the amplitude of the Pi3 pulsation peaks on the dayside and is gradually decreasing toward the nightside. This effect and the fact that the wave phase on the dayside leads that on the nightside, imply that the source of the Pi3 lies on the dayside. Variations in solar wind dynamic pressure observed by the GEOTAIL satellite (just outside of the magnetosphere) are highly correlated with these ground magnetic field variations. We argue in this case that the global Pi3 pulsation is directly driven by impulsive variations in the solar wind dynamic pressure. The Pi3 pulsation observed along the latitudinal magnetometer chain at 0930LT shows significant equatorial enhancement and additional observations along a latitudinal chain at 1630LT show that the phase of the Pi3 pulsation at high latitudes lags behind that at low latitudes. The low-altitude polar orbiting satellite Oersted also observed this pulsation in the dayside inner magnetosphere. The B∥ (northward) component at Oersted is strictly out of phase with the X component observed at the dip equator below the spacecraft path, which indicates that the Pi3 pulsation at the dip equator is caused by oscillation of an ionospheric current. We propose that the Pi3 pulsations at different latitudes are generated by different mechanisms.

[1] Global geomagnetic field data, IMAGE FUV data, and many other in situ observations are presented for the initial 10 min of the magnetic storm starting 29 October 2003 at around 0610 UT. Within 1 min after sudden commencement (SC), two independent strong westward ionospheric electrojets (>2000 nT) at the inner magnetospheric region started simultaneously, one in the evening-midnight sector and the other in the morning sector. Both activities expanded and accompanied auroral expansion. The locations (inner magnetosphere), morphologies (expansion), and intensities (>2000 nT) of both activities fall into substorm expansive phases. Having such simultaneous independent 2000 nT level expansions makes this event unique. The interplanetary magnetic field condition before the SC was not favorable in causing an AL < −2000 nT activity. A timing analysis indicates that these strong westward electrojets were most likely triggered by the interplanetary shock, with the triggering location not farther than the geosynchronous distance. They are also probably maintained by the direct energy pumping from the solar wind because cross-tail current derived from the closely located GOES-10 and Polar did not decrease very much during this period. A local but even stronger geomagnetic (nearly 4000 nT) and auroral activity started only 6 min after the start of SC at postmidnight where and when the above two expanding activities met each other, although the relation between the onset of 4000 nT activity and the preceding expansions is not clear. The suddenness of this third activity (3000 nT change within 2 min) is another unique feature.

[1] On the basis of analysis of east-west geomagnetic disturbance fields at midlatitudes, we investigated the characteristics of temporal variations of field-aligned currents on the nightside during geomagnetic storms. The results indicate that upward field-aligned currents develop on the nightside, especially in the postmidnight, when the interplanetary magnetic field (IMF) is directed southward. This suggests that the upward field-aligned currents are principally associated with the convection electric field. It is also found that the upward field-aligned currents in the postmidnight can be intensified at substorm expansion, while they principally depend on the southward component of the IMF. The intensification of the upward currents is regarded as a different physical process from the formation of a substorm current wedge at a substorm expansion. We discuss some possible effects of the IMF and substorms on the generation of the upward currents on the nightside, especially in the postmidnight.

[1] A long period Pc5 pulsation was observed at Phimai in Thailand, shortly after the origin time of the Sumatra earthquake on December 26, 2004. The localized nature and the period of oscillations suggest that the long period magnetic pulsation was generated by dynamo action in the lower ionosphere, set up by an atmospheric pressure pulse which propagated vertically as an acoustic wave when the ocean floor suddenly moved vertically. It is speculated that a Pc3 type pulsation observed at Tong Hai in China, 10 degrees north of Phimai in latitude, was the result of magnetic field line resonance with a magneto-sonic wave generated from the electric and magnetic fields of the dynamo current caused by the Earthquake.

[1] Orsted is a low-altitude polar orbiting satellite, which is advantageous for investigating the spatial structure of low-frequency events like Pi2 pulsations. Using 1-s vector magnetic field data from April 1999 to May 2002 obtained from Orsted and Kakioka magnetic observatories, we found that the field aligned component (B∥) of Pi2 pulsations at the satellite height is highly correlated with the H component on the ground, but the other two components, i.e., eastward component (BE) and radial component (BR), do not show a clear correlation with the ground observations. The observed results for nightside events provide evidence that nightside Pi2s at low latitude are generated from cavity resonance. Two cases observed by Orsted when it was located on the dawnside or duskside also show cavity resonance properties. However, when Orsted was on the dayside, the oscillations observed by the satellite are out of phase with that observed on the ground, suggesting that the dayside Pi2s are more likely related to some dayside ionospheric current systems rather than that caused by a global cavity resonance mode. The amplitude variation recorded by Orsted shows a peak in the equatorial region, which, for the first time, gives clear observational support for earlier model calculations. We also estimated the screening effect of the ionosphere on MHD waves and suggest that when the cavity resonance mode is valid for generation of Pi2 pulsations at low latitudes, the screening effect is negligible; that is, the compressional waves are seen directly as Pi2 pulsations at the ground.