Modeling the time behavior of the D st index during the main phase of magnetic storms generated by various types of solar wind

Results of modeling the time behavior of the D _st index at the main phase of 93 geomagnetic storms (?250 < D _st ≤ ?50 nT) caused by different types of solar wind (SW) streams: magnetic clouds (MC, 10 storms), corotating interaction regions (CIR, 31 storms), the compression region before interplanetary coronal ejections (Sheath before ICME, 21 storms), and “pistons” (Ejecta, 31 storms) are presented. The “Catalog of Large-Scale Solar Wind Phenomena during 1976–2000” (ftp://ftp.iki.rssi.ru/pub/omni/) created on the basis of the OMNI database was the initial data for the analysis. The main phase of magnetic storms is approximated by a linear dependence on the main parameters of the solar wind: integral electric field sumEy, dynamic pressure P _d , and fluctuation level sB in IMF. For all types of SW, the main phase of magnetic storms is better modeled by individual values of the approximation coefficients: the correlation coefficient is high and the standard deviation between the modeled and measured values of D _st is low. The accuracy of the model in question is higher for storms from MC and is lower by a factor of ～2 for the storms from other types of SW. The version of the model with the approximation coefficients averaged over SW type describes worse variations of the measured D _st index: the correlation coefficient is the lowest for the storms caused by MC and the highest for the Sheath- and CIR-induced storms. The model accuracy is the highest for the storms caused by Ejecta and, for the storms caused by Sheath, is a factor of ～1.42 lower. Addition of corrections for the prehistory of the development of the beginning of the main phase of the magnetic storm improves modeling parameters for all types of interplanetary sources of storms: the correlation coefficient varies within the range from r = 0.81 for the storms caused by Ejecta to r = 0.85 for the storms caused by Sheath. The highest accuracy is for the storms caused by MC. It is, by a factor of ～1.5, lower for the Sheath-induced storms.     

The Fields of Magnetospheric Current Systems on the Earth's Surface in an Interval of the Magnetic Storm Studied under the International Program on Solar–Terrestrial Physics

Using the magnetic storm in January 1997 as an example, we examined the possibilities to employ the magnetospheric field T96 [1, 2] and the dynamic paraboloid model PM of the magnetosphere [3] for modeling the D _st variation. We have revealed the necessity to refine the results of normalizing the free parameters of the model T96 according to the solar wind parameters. The contributions to the D _st variation of magnetic fields of basic large-scale magnetospheric current systems (the field DCF on the magnetopause, the field DR of the ring current, and the field DT in the magnetotail) are estimated for different phases of the storm from model calculations. Possible causes of a discrepancy between the results of modeling D _st using the T96 and PM models are discussed. Special emphasis is made on the ratios of contributions into the D _st variation of the fields of the magnetotail and the ring current in the main phase of magnetic storms and on the contributions to D _st of the fields of various current systems at the recovery phase.     

Statistical study of interplanetary condition effect on geomagnetic storms

Based on the archive OMNI data for the period 1976–2000 an analysis has been made of 798 geomagnetic storms with D _st < −50 nT and their interplanetary sources-large-scale types of the solar wind: CIR (145 magnetic storms), Sheath (96), magnetic clouds MC (62), and Ejecta (161). The remaining 334 magnetic storms have no well-defined sources. For the analysis, we applied the double method of superposed epoch analysis in which the instants of the magnetic storm beginning and minimum of D _st index are taken as reference times. The well-known fact that, independent of the interplanetary source type, the magnetic storm begins in 1–2 h after a southward turn of the IMF (B _z < 0) and both the end of the main phase of a storm and the beginning of its recovery phase are observed in 1–2 h after disappearance of the southward component of the IMF is confirmed. Also confirmed is the result obtained previously that the most efficient generation of magnetic storms is observed for Sheath before MC. On the average parameters B _z and E _y slightly vary between the beginning and end of the main phase of storms (minimum of D _st and D _st ^* indices), while D _st and D _st ^* indices decrease monotonically proportionally to integral of B _z and E _y over time. Such a behavior of the indices indicates that the used double method of superposed epoch analysis can be successfully applied in order to study dynamics of the parameters on the main phase of magnetic storms having different duration.     

Energy dependence of the fraction of protons of ionospheric origin in the ring current

The energy dependence of a fraction of ring current protons of ionospheric origin is calculated using the AMPTE/CCE data for a typical strong magnetic storm (max|D _st | ≈ 120 nT). It is shown that this fraction monotonically decreases from ～ 83 to 25–30% with an increase in proton energy from 5 to 315 keV at L = 6–7 (L is the McIlwain parameter) and is 30–40% at energy 40–50 keV corresponding to the maximum of proton energy density at L = 6–7. It is demonstrated that the core of the ring current (L = 3.7–4.7) was enriched by solar protons with E ≈ 10–200 keV during the active phase of the storm (the maximum effect is reached at E ≈ 20–50 keV).     

Statistical investigation of Heliospheric conditions resulting in magnetic storms: 2

This work is a continuation of investigation [1] of the behavior of the solar wind’s and interplanetary magnetic field’s parameters near the onset of geomagnetic storms for various types of solar wind streams. The data of the OMNI base for the 1976–2000 period are used in the analysis. The types of solar wind streams were determined, and the times of beginning (onsets) of magnetic storms were distributed in solar wind types as follows: CIR (121 storms), Sheath (22 storms), MC (113 storms), and “uncertain type” (367 storms). The growth of variations (hourly standard deviations) of the density and IMF magnitude was observed 5–10 hours before the onset only in the Sheath. For the CIR-, Sheath-and MC-induced storms the dependence between the minimum of the IMF B _z-component and the minimum of the D _st -index, as well as the dependence between the electric field E _y of solar wind and the minimum of the D _st -index are steeper than those for the “uncertain” solar wind type. The steepest D _st vs. B _z dependence is observed in the Sheath, and the steepest D _st vs. E _y dependence is observed in the MC.     

Statistical investigation of heliospheric conditions resulting in magnetic storms

Time behavior of the solar wind and interplanetary magnetic field parameters is investigated for 623 magnetic storms of the OMNI database for the period 1976–2000. The analysis is carried out by the superposed epoch technique (the magnetic storm onset time is taken to be the beginning of an epoch) for five various categories of storms induced by various types of solar wind: CIR (121 storms), Sheath (22 storms), MC (113 storms), and “uncertain type” (367 storms). In total, the analysis conducted for “all storms” included 623 storms. The obtained data, on one hand, confirm the results obtained earlier without selecting the intervals according to the solar wind types, and, on the other hand, they indicate the existence of distinctions in the time variation of parameters for various types of solar wind. Though the lowest values of the B _z-component of IMF are observed in the MC, the lowest values of the D _st-index are achieved in the Sheath. Thus, the strongest magnetic storms are induced, on average, during the Sheath rather than during the MC body passage, probably owing to higher pressure in the Sheath. Higher values of nkT, T/T _exp, and β parameters are observed in the CIR and Sheath and lower ones in the MC, which corresponds to the physical essence of these solar wind types.     

Variations of the magnetopause position versus the level of geomagnetic activity (according to data of the INTERBALL-1 Satellite for 1995–1997)

The variations in the deviation of the observed position of the magnetosphere boundary from its mean position predicted by the Shue at al., 1997 (Sh97) model [7] are studied as a function of the substorm activity level (the AE-index value) and magnetic storm intensity (the value of the corrected D _st ^* index). The results obtained make it possible to state that the amplitude of motion of the magnetospheric boundary on the dayside and in the low-latitude tail is small. It is likely that the position of the boundary is either independent of the AE and D _st ^* indices or this dependence is weak. At the same time, the boundary of the high-latitude tail shifts inward on the average by 1.5R _E with an increase of the AE-index in the case of absence of magnetic storms (contraction of the magnetospheric tail). On the contrary, in the presence of magnetic storms, this boundary shifts outward by up to 3R _E with an increase of the AE-index (inflation of the magnetospheric tail). It is also shown that the boundary of the high-latitude tail moves outward with an increase of the D _st ^* index, both at low substorm activity and in periods of high substorm activity. The amplitude of the outward motion of the high-latitude tail of the magnetosphere is by a factor of two higher for moderate magnetic storms with strong substorms than for moderate magnetic storms with weak substorms.     

Statistical Relationships between Solar, Interplanetary, and Geomagnetospheric Disturbances, 1976–2000

Within the framework of the Space Weather program, 25-year data sets for solar X-ray observations, measurements of plasma and magnetic field parameters in the solar wind, and D _st index variations are analyzed to reveal the factors that have had the greatest influence on the development of magnetospheric storms. The correlation between solar flares and magnetic storms practically does not exceed a level of correlation for random processes. In particular, no relation was found between the importance of solar flares and the minimum of the D _st index for storms that could be connected with considered flares by their time delay. The coronal mass ejections (CME; data on these phenomena cover a small part of the interval) result in storms with D _st < –60 nT only in half of the cases. The most geoeffective interplanetary phenomena are the magnetic clouds (MC), which many believe to be interplanetary manifestations of CMEs, and compressions in the region of interaction of slow and fast streams in the solar wind (the so-called Corotating Interaction Region, CIR). They correspond to about two-thirds of all observed magnetic storms. For storms with –100 < D _st < –60 nT, the frequencies of storms from MC and CIR being approximately equal. For strong storms with D _st < – 100 nT, the fraction of storms from MC is considerably higher. The problems of reliable prediction of geomagnetic disturbances from observations of the Sun and conditions in interplanetary space are discussed.     

Study of notches in the Earth’s plasmasphere based on data of the <Emphasis Type="Italic">MAGION-5</Emphasis> satellite

Depleted narrow (localized in longitude) regions (field tubes) in the plasmasphere, recently discovered in He⁺ radiation measurements on the IMAGE spacecraft, were first directly observed by the Magion-5 satellite. The low-density regions (notches) occupy <～ 10–30° in longitude and extend from L ～ 2–3 to the plasmasphere boundary in neighboring plasmasphere regions with larger densities. The Magion-5 data give evidence that in the low-density regions temperature is enhanced as compared to the neighboring denser plasmasphere regions. Formation of notches in the plasmasphere is, apparently, associated with AE intensification during weak magnetic storms, while strong magnetic storms usually result in the overall reduction of plasmasphere dimensions. However, even a strong magnetic storm on April 6–7, 2000 (max K _p = 9-and min D _st ～ ?290 nT), but accompanied by an isolated AE impulse, resulted in a density decrease only in the longitudinally limited post-midnight sector of the plasmasphere.     

Storm deviations of the geomagnetic trap core from dipole configuration deduced from data on relativistic electrons

Satellite data on the position of maximum L _m of the belt of relativistic electrons during strong storms, obtained at low altitudes (∼500 km) and at high altitudes (near the geomagnetic equator plane), are compared (L is the McIlwain parameter). Both at low and high altitudes the maximum of the storm belt of relativistic electrons is formed on the outer edge of the ring current. It is shown that the geomagnetic field can substantially deviate from dipole configuration not only at the geomagnetic trap periphery, but at its core as well (at L ∼ 2.5–3.5), and these deviations are nonlinear. Simultaneous measurements of the fluxes of relativistic electrons at low and high altitudes can serve for estimation of the real shape of magnetic field lines at L < 4 during geomagnetic disturbances.     

Simultaneous Response of the Mid-Latitude Ionosphere to a Sudden Commencement of Strong Magnetic Storms

Using the new technology of global GPS detection of ionospheric disturbances (GLOBDET), it is found that a sharp increase of the time derivative of the magnetic field strength during magnetic storms is accompanied by a simultaneous decrease of the mid-latitude total electron content (TEC) over the entire dayside of the globe. The corresponding negative correlation coefficient is no less than 0.8, and the delay relative to the sudden commencement of magnetic storm is about 3–10 min. The effect is especially clearly pronounced for magnetic storms with sudden commencements (SSC). The analysis is carried out for a set of 90 to 300 GPS stations for 10 days (January 6 and April 23, 1998; April 6, June 8, July 13, 14, and 15, 2000; March 31, April 4 and 11, 2001) with various levels of geomagnetic activity (D _st and K _p varied from –6 to –295 nT and from 0 to 9, respectively). The amplitude of the response in the total electron content for the events considered was 0.1–0.4 × 10¹⁶ m^–2 (which is a deviation of 0.2–2.6% from the TEC background value). The velocity of the disturbance motion from the dayside to the nightside was about 10–20 km/s. The results obtained agree with the data of ionospheric parameter measurements conducted earlier by methods with high temporal resolution.     

Development of auroral electrojets and low-latitude geomagnetic disturbances during strong magnetic storms on November 7–11, 2004

The influence of auroral electojets and solar wind parameters on variations in low-latitude geomagnetic disturbances and D _st during strong magnetic storms on November 7–8, 2004 with D _st ≈ −380 nT and on November 9–10, 2004 with D _st ≈ −300 nT is studied on the basis of global geomagnetic observations. It is found that the impulsive variations of the western electrojet intensity with a duration of Δt ≈ 1–2 h (probably, substorm disturbances) lead to positive low-latitude disturbances of ΔH at Φ′ ≈ 10°–30° and to disturbances of the same durations with an amplitude +ΔH ∼ 30–100 nT at latitudes of the polar cap (Φ′ ≈ 75°–80°). More durable (with Δt ≥ 10 h) convection electrojets whose centers are shifted to latitudes of ∼50°–55° in the process of storm development are the main cause of the increase in negative values of ΔH at low latitudes and D _st . It is shown that meridional dynamics of position of the center of electrojets and the equatorial boundary of the auroral oval is governed by variations (increase or decrease) in the intensity of negative values of the IMF B _z component. It is assumed that in these storms the intensification of the magnetospheric partially ring current closes the circuit to the ionosphere with the help of field-aligned currents at the equatorial boundary of the auroral oval is the main cause of the magnetic field depression at low latitudes.     

Statistical study of interplanetary condition effect on geomagnetic storms: 2. Variations of parameters

We investigate the behavior of mean values of the solar wind’s and interplanetary magnetic field’s (IMF) parameters and their absolute and relative variations during the magnetic storms generated by various types of the solar wind. In this paper, which is a continuation of paper [1], we, on the basis of the OMNI data archive for the period of 1976–2000, have analyzed 798 geomagnetic storms with D _st ≤ −50 nT and their interplanetary sources: corotating interaction regions CIR, compression regions Sheath before the interplanetary CMEs; magnetic clouds MC; “Pistons” Ejecta, and an uncertain type of a source. For the analysis the double superposed epoch analysis method was used, in which the instants of the magnetic storm onset and the minimum of the D _st index were taken as reference times. It is shown that the set of interplanetary sources of magnetic storms can be sub-divided into two basic groups according to their slowly and fast varying characteristics: (1) ICME (MC and Ejecta) and (2) CIR and Sheath. The mean values, the absolute and relative variations in MC and Ejecta for all parameters appeared to be either mean or lower than the mean value (the mean values of the electric field E _y and of the B _z component of IMF are higher in absolute value), while in CIR and Sheath they are higher than the mean value. High values of the relative density variation sN/〈N〉 are observed in MC. At the same time, the high values for relative variations of the velocity, B _z component, and IMF magnitude are observed in Sheath and CIR. No noticeable distinctions in the relationships between considered parameters for moderate and strong magnetic storms were observed.     

Proton precipitation to the equator of the isotropic boundary during the geomagnetic storm on November 20–29, 2003

Using NOAA satellite data, we consider the peculiarities of precipitation dynamics for energetic protons to the equator of the isotropy boundary during a geomagnetic storm. In addition to two well-known types of proton precipitation events, a new third type of precipitation is distinguished, which is observed on the dayside at relatively high latitudes. The assumption is made that the third-type precipitation in the dayside sector is associated with the development of ion-cyclotron instability. Apparently, the transverse anisotropy of energetic protons, which is necessary for the development of instability, is caused by splitting of drift shells. All three types of precipitation have different generation regions and different time dynamics during storms. The maximum precipitation intensity takes place in the evening sector during the main phase of a storm. At the storm’s recovery phase major losses of protons of the ring current are due to precipitation in the day and morning sectors.     

A diffusion model of radial distributions of plasma density from measurements on the Cassini spacecraft in the inner magnetosphere of Saturn

Equatorial radial distributions of plasma density in the 3 < L < 9 region of Saturn’s magnetosphere, obtained from measurements on the Cassini spacecraft, are considered on the basis of diffusion theory. The concentration of particles in the magnetic tubes is found to grow with L. The external source is located at L ? 9. The particles diffuse to Saturn. In the 5 < L < 9 interval the distribution is close to equilibrium. A relation between the diffusion coefficient and the densities of internal sources and losses is obtained in this interval. Prevalence of losses over sources is very probable. Estimates of the diffusion flux and its derivative are given. If the diffusion coefficient is expressed as D _LL = D _o L ³ and the contencentration of particles depends on L according to a power law, the diffusion rate is constant.     

On the belt of energetic electrons at <Emphasis Type="Italic">L</Emphasis> = 2.75 in the Earth’s magnetosphere

In 1964, during flights of the ELECTRON satellites the narrow belts of energetic electrons (E _e ≈ 6MeV) have been discovered in the Earth’s magnetosphere at L ≈ 2.75. The same structures approximately at the same magnetic shells were found in 2004 by the CORONAS-F and SERVIS-1 satellites. A comparison of the results of these experiments is presented. It is shown that the additional narrow belts of energetic electrons occur after intense magnetic storms (D _st > 100 nT), in our cases, having a double-triple structure. The lifetime of these belts is a few months and their disappearance had a gradual character. The obtained results separated in time by 40 years suggest the constancy of the sources of particles of the Earth’s radiation belts and processes occurring in the magnetosphere, which ensures not only existence of the radiation belts, but also the recurrence of various exotic phenomena in the belts similar to the belt of energetic electrons at the inner magnetic shells.     

Some problems of identifying types of large-scale solar wind and their role in the physics of the magnetosphere

This paper discusses the errors in analyzing solar-terrestrial relationships, which result from either disregarding the types of interplanetary drivers in studying the magnetosphere response on their effect or from the incorrect identification of the type of these drivers. In particular, it has been shown that the absence of selection between the Sheath and ICME (the study of so-called CME-induced storms, i.e., magnetic storms generated by CME) leads to errors in the studies of interplanetary conditions of magnetic storm generation, because the statistical analysis has shown that, in the Sheath + ICME sequences, the largest number of storm onsets fell on the Sheath, and the largest number of storms maxima fell at the end of the Sheath and the beginning of the ICME. That is, the situation is observed most frequently when at least the larger part of the main phase of storm generation falls on the Sheath and, in reality, Sheath-induced storms are observed. In addition, we consider several cases in which magnetic storms were generated by corotating interaction regions, whereas the authors attribute them to CME.     

Statistical Relationships between Solar,Interplanetary, and Geomagnetic Disturbances, 1976–2000: 3

In this paper we continue the analysis of the influence of solar and interplanetary events on magnetic storms of the Earth that was started in [9, 10]. Different experimental results on solar-terrestrial physics are analyzed in the study and the effects are determined that arise due to differences in the methods used to analyze the data. The classifications of magnetic storms by the K _p and D _st indices, the solar flare classifications by optical and X-ray observations, and the classifications of different geoeffective interplanetary events are compared and discussed. It is demonstrated that quantitative estimations of the relationships between two types of events often depend on the direction in which the events are compared. In particular, it was demonstrated that the geoeffectiveness of halo CMEs (that is, the percentage of Earth-directed coronal mass ejections that result in geomagnetic storms) is 40–50%. Higher values given in some papers were obtained by another method, in which they were defined as the probability of finding candidates for a source of geomagnetic storms among CMEs, and, strictly speaking, these values are not true estimates of the geoeffectiveness. The latter results are also in contrast with the results of the two-stage tracing of the events: first a storm—an interplanetary disturbance, and then an interplanetary disturbance—a CME.     

Peculiarities of long-wave radio bursts from solar flares preceding strong geomagnetic storms

Radio bursts in the frequency range of 100–1500 kHz, recorded in 1997–2000 on the INTERBALL-1 satellite during the solar flares preceding the strong geomagnetic storms with D _st < ?100 nT, are analyzed in this paper. The observed long-wave III-type radio bursts of solar origin at frequencies of 1460 and 780 kHz were characterized by large values of the flux S _f = 10^?15 ?10^?17 W/m² Hz and duration longer than 10 min. The rapid frequency drift of a modulated radio burst continued up to a frequency of 250 kHz, which testified that the exciting agent (a beam of energetic electrons) propagated from the Sun to the Earth. All such flares were characterized by the appearance of halo coronal mass ejections, observed by the LASCO/SOHO, and by the presence of a southward B_z-component of the IMF, measured on the ACE and WIND spacecraft. In addition, shortly after radio bursts, the INTERBALL-1 satellite has recorded the fluxes of energetic electrons with E > 40 keV.     

Dynamics of polar boundary of the auroral oval derived from the IMAGE satellite data

Based on a new database on positions of the auroral oval boundaries including measurements made by the IMAGE satellite in 2000–2002 with correct determination of the glow boundaries, statistical estimations of the latitudinal position of the polar cap boundary (PCB) are obtained depending on the IMF B _y and B _z , and the PCB evolution during a magnetic storm is analyzed. At zero IMF in the noon (midnight) sector, PCB is located approximately at 80° (76°) CGMLat. The PCB displacement along the noon-midnight meridian is controlled by the IMF B _z , and in the noon (midnight) sector it is equal to 0.45° (0.15°) CGMLat when B _z changes by 1 nT. The PCB displacement along the dawn-dusk meridian depends on the IMF B _y , and it equals 0.1° CGMLat when B _y changes by 1 nT. Accordingly, the north polar cap as a whole is shifted to the dawn (dusk) side at B _y > 0 (B _y <0). After northward turn of the IMF during the storm’s recovery phase, the PCB on the dayside is shifted to the north practically without time delay. The night boundary requires 25 h or more in order to be shifted to the pole to a latitude corresponding to B _z > 0.