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.     

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.     

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.     

Relative occurrence rate and geoeffectiveness of large-scale types of the solar wind

We investigate the relative occurrence rate for various types of the solar wind and their geoeffectiveness for magnetic storms with Dst < —50 nT. Both integrated effect for the entire time 1976–2000 and variations during this period of 2.5 cycles of solar activity are studied As raw data for the analysis we have used the catalog of large-scale types of the solar wind for the period 1976-2000 (see ftp://ftp.iki.rssi.ru/omni/) created by us with the use of the OMNI database (http://omni.web.gsgc.nasa.gov) [1] and described in detail in [2]. The average annual numbers of different type of events are as follows: 124 ±81 for the heliospheric current sheet (HCS), 8 ±6 for magnetic clouds (MC), 99 ±38 for Ejecta, 46 ±19 for Sheath before Ejecta, 6 ±5 for Sheath before MC, and 63 ±15 for CIR. The measurements that allowed one to determine a source in the solar wind were available only for 58% of moderate and strong magnetic storms (with index Dst < —50 nT) during the period 1976–2000. Magnetic clouds (MC) are shown to be the most geoeffective (~61%). The CIR events and Ejecta with Sheath region are three times less geoeffective (~20–21 %). Variations of occurrence rate and geoeffectiveness of various types of the solar wind in the solar cycle are discussed.     

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.     

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.     

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 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.     

Long-term variations of asymmetry of the solar magnetic field and geometry of the interplanetary magnetic field

A comparative analysis of variations of the mean solar magnetic field and the interplanetary magnetic field over all history of their measurements is performed. Asymmetry of the solar magnetic field and its manifestation in the heliosphere is investigated. Long-term variations of the solar magnetic field and the heliosphere, which manifest themselves in alternation of dominating magnetic polarities of different sign, are discovered. On the basis of the analysis of cumulative sums of the IMF components, long-term variations of the IMF geometry and of the solar wind spiral angle are found. The cumulative sum of the IMF B _z component perpendicular to the ecliptic plane also shows long-term variations. Time intervals are revealed, in which negative values of the IMF B _z component dominate, and an increased geomagnetic activity is observed.     

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.     

Relationship between the fluxes of relativistic electrons at geosynchronous orbit and the level of ULF activity on the Earth’s surface and in the solar wind during the 23rd solar activity cycle

We present the results of a cross-correlation analysis made on the basis of Spearman’s rank correlation method. The quantities to correlate are daily values of the fluence of energetic electrons at a geosynchronous orbit, intensities of ground and interplanetary ultra-low-frequency (ULF) oscillations in the Pc5 range, and parameters of the solar wind. The period under analysis is the 23rd cycle of solar activity, 1996–2006. Daily (from 6 h to 18 h of LT) magnetic data at two diametrically opposite observatories of the Intermagnet network are taken as ground-based measurements. The fluxes of electrons with energies higher than 2 MeV were measured by the geosynchronous GOES satellites. The data of magnetometers and plasma instruments installed on ACE and WIND spacecraft were used for analysis of the solar wind parameters and of the oscillations of the interplanetary magnetic field (IMF). Some results elucidating the role played by interplanetary ULF waves in the processes of generation of magneospheric oscillations and acceleration of energetic electrons are obtained. Among them are (i) high and stable correlation of ground ULF oscillations with waves in the solar wind; (ii) closer link of mean daily amplitudes of both interplanetary and ground oscillations with ‘tomorrow’ values of the solar wind velocity than with current values; and (iii) correlation of the intensity of ULF waves in the solar wind, normalized to the IMF magnitude, with fluxes of relativistic electrons in the magnetosphere.     

Some features of solar cosmic ray penetration into the Earth’s magnetosphere

We compared fluxes of the 1–100 MeV solar energetic particles (SEP) measured in the interplanetary medium (ACE) and in the magnetosphere (Universitetsky-Tatiana, POES—in polar caps, and GOES-11—at geosynchronous orbit) during several SEP events of 2005–2006. Peak intensities of the SEP fluxes inside and outside the magnetosphere were compared for each event. It is shown that observed inside-outside difference depends mainly on direction of interplanetary magnetic field (IMF), on degree of the SEP anisotropy (pitch-angle distribution) in IMF, and on distance of the dayside magnetopause from the Earth.     

Influence of Solar Wind Parameters on the Level of Geomagnetic Field Fluctuations

We investigated the dependence of the geomagnetic activity index K _p on the velocity and density of the solar wind and the intensity of the interplanetary magnetic field (IMF). A three-layer neural network was used to create the model. The degree of the influence of input parameters on K _p was determined by the value of the mean and root-mean square deviations of the model index values from the real indices. It was found that the largest contribution to the K _p index is provided by the Z component of the IMF, the velocity and density of the solar wind measured with a delay from 0 to 3 h relative to the studied value of K _p, and the previous value of the index itself. For the model with such a set of input parameters, the correlation coefficient between model and real series is ±0.89. The analysis of deviations from the real values of K _p showed that high indices are simulated worse than low indices. In order to solve this problem the data distribution was reduced to a uniform distribution over K _p, and this considerably decreased the standard deviations for large values of K _p.     

Geosynchronous magnetopause crossings on October 29–31, 2003

During the period October 29–31, 2003, geosynchronous magnetopause crossings (GMC) have been identified based on the magnetic data of the GOES series spacecraft and plasma data of the LANL series spacecraft. It is shown that most of the time the size of the dayside magnetosphere was highly decreased under the effect of very high pressure associated with high velocities and densities of the solar wind plasma, as well as high negative values of the B_z component of the interplanetary magnetic field (IMF). For tens of hours the subsolar magnetopause was deep inside the geosynchronous orbit. During the main phase and at the maximum of the strong geomagnetic storms that occurred in the period under consideration, the dayside magnetosphere was characterized by a strong dawn-dusk asymmetry, so that its size in the postnoon sector considerably exceeded the size in the pre-noon sector. The geomagnetic disturbances in the morning on October 30 and 31, 2003 were accompanied by global magnetospheric pulsations with periods of 5–10 min and high amplitude (up to 0.8 R_E).Translated from Kosmicheskie Issledovaniya, Vol. 42, No. 6, 2004, pp. 574–584.Original Russian Text Copyright © 2004 by Dmitriev, Suvorova.     

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.     

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.     

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.     

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.     

The Earth's Magnetosphere Response to Solar Wind Events according to the INTERBALL Project Data

The results of studying the interaction of two types of the solar wind (magnetic clouds and solar wind of extremely low density) with the Earth's magnetosphere are discussed. This study is based of the INTERBALL space project measurements and on the other ground-based and space observations. For moderate variations of the solar wind and interplanetary magnetic field (IMF) parameters, the response of the magnetosphere is similar to its response to similar changes in the absence of magnetic clouds and depends on a previous history of IMF variations. Extremely large density variations on the interplanetary shocks, and on leading and trailing edges of the clouds result in a strong deformation of the magnetosphere, in large-scale motion of the geomagnetic tail, and in the development of magnetic substorms and storms. The important consequences of these processes are: (1) the observation of regions of the magnetosphere and its boundaries at great distances from the average location; (2) density and temperature variations in the outer regions of the magnetosphere; (3) multiple crossings of geomagnetic tail boundaries by a satellite; and (4) bursty fluxes of electrons and ions in the magnetotail, auroral region, and the polar cap. Several polar activations and substorms can develop during a single magnetic cloud arrival; a greater number of these events are accompanied, as a rule, by the development of a stronger magnetic storm. A gradual, but very strong, decrease of the solar wind density on May 10–12, 1999, did not cause noticeable change of geomagnetic indices, though it resulted in considerable expansion of the magnetosphere.     

Radial profile of pressure in a storm ring current as a function of D st

Using satellite data obtained near the equatorial plane during 12 magnetic storms with amplitudes from ?61 down to ?422 nT, the dependences of maximum in L-profile of pressure (L _m) of the ring current (RC) on the current value of D _st are constructed, and their analytical approximations are derived. It is established that function L _m(D _st ) is steeper on the phase of recovery than during the storm’s main phase. The form of the outer edge of experimental radial profiles of RC pressure is studied, and it is demonstrated to correspond to exponential growth of the total energy of RC particles on a given L shell with decreasing L. It is shown that during the storms’ main phase the ratio of plasma and magnetic field pressures at the RC maximum does not practically depend on the storm strength and L _m value. This fact reflects resistance of the Earth’s magnetic field to RC expansion, and testifies that during storms the possibilities of injection to small L are limited for RC particles. During the storms’ recovery phase this ratio quickly increases with increasing L _m, which reflects an increased fraction of plasma in the total pressure balance. It is demonstrated that function L _m(D _st ) is derived for the main phase of storms from the equations of drift motion of RC ions in electrical and magnetic fields, reflecting the dipole character of magnetic field and scale invariance of the pattern of particle convection near the RC maximum. For the recovery phase it is obtained from the Dessler-Parker-Sckopke relationship. The obtained regularities allow one to judge about the radial profile of RC pressure from ground-based magnetic measurements (data on the D _st variation).