Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.izmiran.rssi.ru/magnetism/Feldstein_2012/PDF/431.pdf
Дата изменения: Wed Jan 11 22:42:08 2012
Дата индексирования: Mon Oct 1 21:31:36 2012
Кодировка:
Поисковые слова: р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п р п

Manuscript prepared for Hist. Geo Space Sci. A with version 5.2 of the LTEX class copernicus2.cls. Date: 4 January 2012

Spiral regularities in magnetic field variations and auroras
Ё Y. I. Feldstein1 , L. I. Gromova1 , M. Forster2 , and A. E. Levitin1
1

2

Institute of Terrestrial Magnetism, Ionosphere, and Radiowave Propagation of the Russian Academy of Sciences (IZMIRAN), 142090 Troitsk, Moscow region, Russia Helmholtz-Zentrum Potsdam, GFZ German Research Centre for Geosciences, 14473 Potsdam, Germany

Abstract. The conception of spiral shaped precipitation regions, where solar corpuscles penetrate the up-

per atmosphere, was introduced into geophysics by C. StЬrmer and K. Birkeland at the beginning of the last century. Later, in the course of the XX-th century, spiral distributions were disclosed and studied in various geophysical phenomena. Most attention was devoted to spiral shapes in the analysis of regularities pertaining to the geomagnetic activity and auroras. We review the historical succession of perceptions about the number and positions of spiral shapes, that characterize the spatial-temporal distribution of magnetic disturbances. We describe the processes in the upper atmosphere, which are responsible for the appearance of spiral patterns. We considered the zones of maximal aurora frequency and of maximal particle precipitation intensity, as offered in the literature, in their connection with the spirals. We discuss the current system model, that is closely related to the spirals and that appears to be the source for geomagnetic field variations during magnetospheric substorms and storms. The currents in ionosphere and magnetosphere constitute together with field-aligned (along the geomagnetic field lines) currents (FACs) a common 3-D current system. At ionospheric heights, the westward and eastward electrojets represent characteristic elements of the current system. The westward electrojet covers the longitudinal range from the morning to the evening hours, while the eastward electrojet ranges from afternoon to near-midnight hours. The polar electrojet is positioned in the dayside sector at cusp latitudes. All these electrojets map along the magnetic field lines to certain plasma structures in the near-Earth space. The first spiral distribution of auroras was found based on observations in Antarctica for the nighttime-evening sector (N-spiral), and later in the nighttime-evening (N-spiral) and morning (M-spiral) sectors both in the Northern and Southern Hemispheres. The N and M spirals drawn in polar coordinates form an oval, along which one observes most often auroras in the zenith together with a westward electrojet. The nature of spiral distributions in geomagnetic field variations was unabmibuously interpreted after the discovery of the spiral's existence in the auroras had been established and this caused a change from the paradigm of the auroral zone to the paradigm of the auroral oval. Zenith forms of auroras are found within the boundaries of the auroral oval. The oval is therefore the region of most frequent precipitations of corpuscular fluxes with auroral energy, where anomalous geophysical phenomena occur most often and with maximum intensity. S. Chapman and L. Harang identified the existence of a discontinuity at auroral zone latitudes ( 67 ) around midnight between the westward and eastward electrojets, that is now known as the Harang discontinuity. After the discovery of the auroral oval and the position of the westward electrojet along the oval, it turned out, that there is no discontinuity at a fixed latitude between the opposite electrojets, but rather a gap, the latitude of which varies smoothly between 67 at midnight and 73 at 20 MLT. In this respect the term "Harang discontinuity" represents no intrinsic phenomenon, because the westward electrojet does not experience any disruption in the midnight sector but continues without breaks from dawn to dusk hours.
Keywords. spirals, auroral oval, auroral forms, high-latitude magnetic variations, field-aligned currents,

equivalent ionospheric currents, dynamo theory

2
1

Y. I. Feldstein et al.: Spiral Regularities
Introduction
52 53

1

2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51

1.1. This review is dedicated to one of the branches of planetary-scale research on geomagnetic field variations and the morphology of discrete auroral forms the spiral distributions, which are used in geophysics for the description of various phenomena. A comprehensive view of the results obtained in the scientific literature is given, in particular the contributions in Russian language. Such information will be most notably interesting and useful for those scientists, who often have problems to get proper access to scientific publications in Russian tongue either due to the language barrier or because of undue hardship to obtain such informations, which are buried in Institute's collections or in particular topical proceedings. The scientific development in the USSR, including geophysics, was detached to a large extent from global science progress for a long time period. Only the participation of the Soviet Union in the International Geophysical Year (1957-1958) and in subsequent international geophysical projects changed the situation fundamentally. There arose the additional possibility for publications in international journals, but many results continued to be known primarily in Russia only and were not cited in the global literature. We hope that the present review attracts attention to researchers in the field of Earth's magnetism and auroras, as they are such phenomena that intrigues mankind's exploratory spirit for many centuries. The following elaborations about the development of scientific knowledge on a specific geophysical problem might also be interesting for historians of science. 1.2. The conception of spirals, along which the precipitation of corpuscular solar fluxes occur, is closely related to the model experiments of Kristian Birkeland (Birkeland, 1908, 1913) and the theoretical calculations of Carl StЬrmer (StЬrmer, 1917a,b, 1955). Birkeland illuminated a magnetised model sphere ("terrella") representing the Earth by bundles of low-energy electrons. The bundles of charged particles were bent towards the night side of the terrella and toward the magnetic poles. The regions of precipitation were made visible through a phosphorecent layer which covered the terrella. Glowing stripes, one in each hemisphere, surrounded in the experiments the magnetic poles of the terrella. The stripe winded up around the near-pole region of the Northern Hemisphere in counter-clockwise direction. The terrella experiments have been repeated and extended by Bruche (1931). 1.3 StЬrmer (1917a,b, 1955) developed the methodology and performed mathematical calculations of the trajectories of positively charged particles (protons and -particles) and of electrons in a dipole magnetic field. Figure 1 shows results of StЬrmer's calculations for the distribution of precipitations of positively charged particles at high latitudes of the Correspondence to: Y. I. Feldstein afield@aol.com Hist. Geo Space Sci.

54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106

Northern Hemisphere. The entire precipitation region has been obtained by a series of calculated orbits. It explains the formation of a spiral shaped region of precipitation in Birkeland's terrella experiment. This "laboratory" spiral has been obtained from calculations, adopting particular values of the -parameter (constant of integration) and of (declination of the corpuscles' source region). In the theory of StЬrmer, can assume arbitrary values, while the sun's declination varies in the course of the year by ±34 with respect to the equatorial plane of the dipole. The "laboratory" spiral in Fig. 1 results from a 1-year long integration time of the calculations. Not the whole spiral in Fig. 1 corresponds to a given UT moment with its particular value, but only a few discrete points along the spiral. 1.4 The variation of along the spiral is non-uniform; there is rather a bunching of points within four sectors, centred around = 0, i.e., close to the equinoctial periods. The higher density of points occurs near 03, 09, 14, and 20 hours local time, when the direction of the corpuscular source region varies most quickly. 1.5 In order to move the theoretical auroral spiral to the 23 colatitude required for agreement with auroral observations during the night time, StЬrmer postulated the existence of a ring current at higher geocentric distances. The magnetic field of this current deflects the path of solar particles, so that they precipitate at lower latitudes, but at the Earth's surface its field strength is small. 1.6 Precipitations of charged particles into the upper atmosphere leads to a series of phenomena. Certainly it could be assumed that the most intense specific geophysical phenomena, caused by these precipitations, occur along the spirals, which are oriented in a determinate way according to the sun's direction. The preciptiation region in the upper atmosphere has the form of a spiral in the coordinates of magnetic latitude versus magnetic local time. At the Northern Hemisphere, the spiral wind up clockwise from the magnetic pole for positively charged particles and in anti-clockwise directions for electrons. 1.7 Based on observations of the geomagnetic field, Birkeland (1908, 1913) concluded that the magnetic disturbances and the closely related aurorae are caused by the precipitation of electrically charged particles from the sun into the upper atmospheric layers. He assumed that the currents, calculated from the geomagnetic perturbations are real free currents, probably consisting of free electrons coming in from space towards the auroral zone and bent back again. The problem of the BirkelandStЬrmer theory consisted in the unsteadiness of solar charged particle fluxes of one polarity, which are dissipated before they can reach the Earth. The energy of protons with a flux velocity of 500 km/ s amount to a few keV . Such a particle velocity within the flux was deduced from the time lag of the beginning of magnetic disturbances on Earth with respect to the flare at the sun. This lag comes along with the ejection of a particle flux, the subsequent development of a magnetic storm and the appearance of aurowww.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities

3

Figure 1. Spiral of positive corpuscles' precipitation around the magnetic axis. Here, stands for a constant of integration and is the

declination of the sun (StЬrmer, 1955, Fig. 177).

107 108 109 110 111 112 113 114 115 116 117 118 119

ras. Protons of such energy precipitate, according to StЬrmer, into a much smaller region than the polar latitudinal distances of auroral features. Electrons with energy of a few eV , corresponding to velocities of 500 km/ s, would precipitate close to the geomagnetic pole. However, their energy appears to be too small to penetrate down into the upper atmosphere to a level of 100 km above the Earth's surface, where discrete auroral forms are usually seen (Harang, 1951). The results of StЬrmer's theoretical calculations appeared to be applicable to the particles of cosmic radiation, whose energy is several orders of magnitude larger than those of the plasma fluxes, which are responsible for the geomagnetic disturbances and auroras. In the following sections, we sum up the basic findings concerning the spiral distributions in its historical progression, discuss the results of various authors, and in Section 6 we consider the relationship between the spirals and models of the magnetospheric-ionospheric current systems. In the Conclusions we list the basic steps in exploring the spiral distributions of geomagnetic variations and auroras and discuss www.hist-geo-space-sci.net

127

their place within the magnetospheric plasma structure.
2 Spiral distributions in regular (S D ) geomagnetic field variations

128 129

130 131 132

2.1 The variations of the magnetic field D, which are observed at the Earth's surface during geomagnetically disturbed days are composed of various sources: D = DC F + DR + DPC + DP + DT + Di (1)

133

134 135 136 137 138 139 140 141 142 143

120 121 122 123 124 125 126

where DC F describes the field of magnetopause currents, DR the field of the ring current in the inner magnetosphere, DPC the field of the current system with the electrojet in the dayside cusp, DT the field of the cross-tail current and its closure magnetopause current, DP the field of elementary polar magnetic disturbances (substorms) with characteristic lifetimes of some tens of minutes to several hours, and Di the field of irregular oscillations of the geomagnetic field. The DP variations are refered to the substorm type variations (DP1) only in the following. DP2 variations are of different Hist. Geo Space Sci.

4

Y. I. Feldstein et al.: Spiral Regularities

Figure 2. Loci of maximum intensity of auroral-zone electric currents (electrojets), as observed on geomagnetic meridian 120 , for different

ranges of storminess (1,2,3,4). Mean picture for H and Z fields, letters M, N, and E correspond to morning, nighttime, and evening spirals (Harang, 1946).

144 145 146 147

nature and are not considered here. The DC F , DR, DPC , and DT variations are significantly smaller than DP and Di at high latitudes on ground level, except for magnetic storm intervals, so that one usually assumes: D = DP + Di (2)

164 165 166 167 168 169

148

149 150 151 152 153 154 155 156 157 158 159 160 161 162 163

Avereaged over the international magnetic disturbance days, one obtains the regular solar-daily variation field S D , which represents the integral effect of the polar disturbances DP. The S D variations are calculated as difference between the hourly averaged values of the magnetic field elements for the international disturbance days and the quiet days. The current system obtained and named by Chapman as solar-daily disturbed variation S D has influenced the study of geomagnetic disturbances to a high degree and became the standard model. This was a major paradigm for a few decades. The S D current system consists of a pair of electrojets along the auroral zone as a circular belt with the centre at the geomagnetic pole. The westward electrojet (WE) in the morning sector and eastward electrojet (EE) in evening sector are usually considered. On average, the WE has its Hist. Geo Space Sci.

170 171 172 173 174 175 176 177 178 179 180 181 182 183 184

maximum intensity at 03.5 MLT, while the EE at 17.5 MLT (Allen and Kroehl, 1975). The observations of the magnetic field variations were primarily of experimental character until the campaign of the Second International Polar Year (IInd IPY, 1932/1933). There were practically no permanent magnetic observatories at high latitudes by that time. The observations of highlatitude geomagnetic field variations by a network of magnetic stations during the IInd IPY period extended the possibilities of the researchers considerably. Based on observational material of the IInd IPY, Harang (1946) in Norway, Meek (1955) in Canada, and Burdo (1960) in the Soviet Union proposed models of polar magnetic disturbance current systems. 2.2 Harang (1946) used measurements of a chain of magnetic observatories along the 120 magnetic meridian from Spitzbergen (Svalbard) over Scandinavia till Potsdam to determine the S D variations from three-component measurements of 11 observatories for four disturbance levels. He obtained so-called isopleths (lines of equal disturbance intensity) of these components. www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230

5
238

2.2.1 Figure 2 shows the position of the maximum isopleths for all four magnetic disturbance levels. The diurnal variation of the isopleths' position is shown in the coordinates of geomagnetic latitude versus geomagnetic local time. The letters M, N, and E indicate segments of maximum isopleths, which researchers interpreted afterwards as morning (M), nighttime (N), and evening (E) spirals. The current flow is westward along the M and N spirals, but eastward along the E spiral. The isopleths run approximately parallel with the geomagnetic latitude for the H and Z components, and the maximal isopleths of the H component coincide with the auroral zone at 67 magnetic latitude. The H component within the auroral zone is positive during the evening hours (maximum at 17 MLT) and negative in the midnight to early morning hours (maximum intensity at 2401 MLT) with a gap between the isopleths of opposite sign. 2.2.2 A characteristic Z component variation is observed at the auroral zone stations TromsЬ ( = 67.1 ), Bossekop ( = 66.6 ), and Petsamo ( = 64.9 ): negative values are north of and positive to the south of the maximum H isopleths during the evening disturbance, and this reverses to the opposite during the nighttime disturbance. Throughout this paper, stands for geomagnetic latitude and indicates corrected geomagnetic latitude. The significance of the distinction between and consists in that to determine the deviation between the auroral zone and the geomagnetic parallel of 67 or, with other words, in the consideration of higher order harmonics of the internal potential sources, additional to the dipolar field (Hultqvist, 1958; Gustafsson, 1970). The curve for S D in Z appears as a double wave. This double wave appears in all four disturbance levels. Assuming that the field of the magnetic disturbances is produced by a linear overhead current, it is evident that maximum values in H and a zero crossing in Z are located just below this current flow. This variation in H and Z in a certain distance from the line current is qualitatively in accordance with the isopleths. Between the evening and the nighttime isopleths at auroral latitudes ( 67 ) exists therefore a longitudinal discontinuity, both in the variations of the magnetic field and in the current system of polar magnetic disturbances. Resulting from the existence of a discontinuity in the auroral zone, the zone itself consist of two separate branches. The discontinuity of the current system in the nighttime auroral zone causes the existence of the double wave structure in the Z component variations. 2.2.3 From the consideration of Figure 2 follows: 1. a shift of the maximum intensity current toward the equator with increasing disturbances; 2. the existence of a discontinuity in MLT for the isopleths at auroral latitudes in the midnight sector (discontinuity between E and N spirals); 3. the placement of the evening isopleths with an eastward current in the auroral zone latitudes over the whole inwww.hist-geo-space-sci.net

terval of 1121 MLT; 4. the shift of the isopleths with westward current (M and N spirals) to higher magnetic latitudes from the midnight auroral zone both toward the morning and the evening hours; 5. in the evening hours there exists a latitudinal gap between the positive and negative isopleths (E and N spirals). The N spiral with the westward current is at higher latitudes than the E spiral. Harang didn't give due attention to the existence of the latitudinal gap but to a discontinuity at a fixed latitudinal distance near midnight. At the time of the publication (1946), there was a generally accepted paradigm about a auroral zone at 67 and about the position of electrojets along this zone (but not along the auroral oval yet unidentified at that time), so that he couldn't make an adequate conclusion about the nature of the gap between the electrojets. 6. Based on the close connection between the magnetic disturbances and the auroras and the existence of a local time discontinuity in the current system between the electrojets, Harang (1946) concluded, that the auroral zone should also consist of two parts with a discontinuity between them. The most intense auroras occur during 17-18 LT and 21 LT with a quiet period during the intermediate interval. 7. the currents producing polar magnetic storms as well the diurnal variations are most simply explained according to the dynamo theory (Harang, 1951). 2.3 Meek (1955), in his analysis of the magnetic field variations, uses data from all high-latitude geomagnetic observatories north of 40 , which participated in the IInd IPY. Based on S D variations, moments of the diurnal positive maximum of the disturbance vectors as well as their negative diurnal minimum were determined. Then a polar plot of magnetic latitude versus local geomagnetic time was made to display the diurnal maximum of the H component decrease for all stations. The points drawn lie along a spiral which expands clockwise (M spiral with westward current). The corresponding plot of the diurnal maximum of the H component increase shows a spiral expanding in the opposite direction (E spiral with eastward current). M and E spirals intersect on 10 a.m. (MLT) at 70 and on 10 p.m. at 60 . It is assumed that the magnetic spirals obtained are StЬrmer's spirals of precipitating particles from a plasma cloud arriving from the sun. The spiral M expanding clockwise will be due to the precipitation of negatively charged particles, while the spiral E expanding anticlockwise due to positively charged particles. 2.3.1 The patterns of spirals with currents in the papers of Meek (1955) and Harang (1946) differ in the number of spirals, their positions, and with respect to their sources: Hist. Geo Space Sci.

239 240 241 242

243 244 245 246 247 248 249 250 251 252 253 254

255 256 257 258 259 260 261 262

263 264 265

266 267 268 269 270 271 272 273 274 275 276 277 278 279 280

231 232

281 282 283 284 285 286

233 234 235

236 237

287 288

6
289 290 291

Y. I. Feldstein et al.: Spiral Regularities 1. two spirals M and E, which form the oval of Meek (1955) versus three spirals M, E, and N at auroral latitudes of Harang (1946); 2. the spirals M and E intersect on 22 MLT at 60 latitude without any disruption according to Meek (1955), while there exist a discontinuity in MLT according to Harang (1946); 3. Meek relates the maximum decrease in the H component of the magnetic field along the M spiral with precipitating electrons, and the maximum increase in the H component along the E spiral with proton precipitations (the sign of the charged corpuscles precipitating along the spirals oppose the sign of those in StЬrmer's calculations). The winding of the spirals is therefore opposite to the assumption in the model of StЬrmer. For the explanation of the configuration of maximum isopleths (spirals), Harang utilises the dynamo theory of magnetic variations, and Meek the precipitating corpuscles; 4. after the midnight intersection of the spirals with westward and eastward currents, the westward current in the evening sector is positioned equatorward of the eastward current according to Meek (1955), while it is poleward of the eastward current in the evening sector according to Harang (1946).
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360

292 293 294 295

296 297 298 299 300 301 302 303 304 305 306

307 308 309 310 311 312

361 362 363 364 365 366 367

313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340

2.3.2 Meek (1955) modifies the opinions of StЬrmer by assuming the corpuscular solar flux as a plasma, but not as an ensemble of individual charged particles. It ensued just from such a modification, that electrons precipitate along the M spirals and protons along the E spirals. The westward electrojet (H < 0) yields from the precipitation of electrons, while the eastward electrojet (H > 0) is related to protons. Assuming that the magnetic disturbance deviations are due to the motion of electric charges near the base of auroral forms, Meek (1955) expected some peculiarities in the distribution of auroras, caused by the spiral distribution of maximum intensities of magnetic variations and by particle precipitations of different charges along the M and E spirals. The auroral luminosity configuration along the spirals allows the interpretation of some peculiarities of their morphology in the auroral zone and equatorward of it according to the descriptions in the literature: the diurnal variation of their occurrence frequency; the motion during the evening and morning hours; the orientaion of the arcs; the appearance of auroras as due to separate precipitations of electrons and protons. 2.4. Based on data from 22 high-latitude observatories in the Northern Hemisphere with > 60 , Burdo (1960) has shown that the diurnal variations of the disturbance intensity of the horizontal magnetic field components (T ) have 2 or 3 maxima in dependence on the latitude. The positions of the maxima in (T ) within plots of MLT versus (corrected geomagnetic latitude) are arranged in 3 groups: morning (M), nighttime (N), and evening (E) points. The time Hist. Geo Space Sci.

368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395

of maximum T changes with latitude for each group separately. These changes are for all three groups practically linear in the orthogonal projection used here. This relation can be written as = 0 + ktm . In polar coordinates for a particular UT moment, this relation transforms into a part of a spiral with r = A + B, where r is the distance (colatitude) of from the pole, is the geomagnetic longitude or the geomagnetic time, and A, B are constants. The T vectors are drawn onto this polar projection as arrows. Such a use of perturbing vectors as "current arrows" at a network of stations gives a direct picture of the field distribution and was often applied by Birkeland. 2.4.1 The arrows cut the spirals under an angle near 90 . Numbers at the arrows' footpoint indicate the values of the vertical component of the disturbance vector Z . Each spiral sub-divides areas with positive and negative Z . The T vector assumes therefore a maximum value along the three spirals at high latitudes with the direction perpendicular to the spirals and at the intersection points the vertical component Z of the disturbance vector changes its sign. Under the assumption that the horizontal current system constitutes the source of the magnetic disturbances, the spirals are the places of maximum current density. The direction of T and the sign of Z to both sides of the spiral determines the direction of the current along the spiral: along the morning and nighttime spiral the current is westward, while along the dayside spiral the current points eastward. 2.4.2 Burdo (1960) proposed that the dynamo-effect plays a notable (or even main) role in the generation of magnetic variations during disturbed time intervals. The currents producing the Sq variations and the polar magnetic disturbances are most simply explained according to the dynamo-theory proposed by Steward (1882). According to this theory the air in the upper atmosphere is ionised and thus electrically conducting. The tidal and other motions of the atmosphere will set up horizontal motion in the layers across the lines of force of the Earth's permanent magnetic field. Due to this electromotive forces will be formed, and electric currents will be caused to flow in directions which agree with those required to produce Sq variations. The precipitation of electrically charged particles which produce the aurorae, strongly increases the ionisation and thus the conductivity of the ionized layers, which leads to cardinal changes of the high-latitude current system. This system of ionospheric currents is - according to the dynamo theory - responsible for the generation of magnetic disturbances, so-called S D variations, as well as for polar magnetic storms. Processes, which are generated in the ionosphere due to the penetration of corpuscular fluxes, play the main role for the generation of magnetic disturbances. These fluxes enter the upper atmosphere at auroral latitudes and a 2-D current system, which generates the magnetic disturbances, spreads out in the ionosphere. 2.5 The most exhaustive development of the dynamo theory, which is applicable to the magnetic field variations at high latitudes, was accomplished in the works of Japanese www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425

7
440 441 442 443 444 445 446 447 448 449

scientists. Nagata and Fukushima (1952) and Fukushima (1953) used the dynamo theory for the interpretation of two specific types of the geomagnetic variations, which were observed during the IInd IPY: a single-vortex current system with the westward electrojet at auroral latitudes in the nighttime sector, or a double-vortex system with westward and eastward electrojets at auroral latitudes during the nighttime and evening hours, respectively. The conductivity is steeply enhanced in the auroral zone (colatitude range of 20 В 25 ) during magnetic disturbances and stays at the higher level of about one order of magnitude compared with the polar region. The neutral wind in the conducting layer depends on colatitude, longitude and the phase angle . The model current system can be brought in accordance with the corresponding calculated magnetic field variations by varying the input parameters of the model. In case of enhanced conductivity in the auroral zone along all longitudes, the model current system results in a double vortex with westward and eastward electrojets. If the conductivity is enhanced only within a limited longitudinal range, then the model current system remains a single-vortex system with an electrojet in those longitudes of enhanced conductivity. The angle controls the direction of the current closure across the polar cap. 2.6 Fukushima and Oguti (1953) considered the dynamo theory with an anisotropic ionospheric conductivity. Under several simplifying assumptions they obtaind an expression for the current function JD in northern and southern auroral zones, in the polar caps, and in the equatorial zone. The current function has the form: J
D 0 1 = 2 K3 G k1

electrojets in the auroral zone (65 < < 70 ) and its distributed currents in the polar cap ( > 73 ). The ionospheric neutral wind system for the model calculation of S D was adopted analogous to the calculations of S q variations. There is a very essential difference between the observed and the modelled S D systems: the current direction in the polar cap differs by an angle between 110 and 150 . This problem is usually avoided in the dynamo theory by assuming different neutral wind systems for the S q and S D variations at ionospheric altitudes (Nagata et al., 1950). 2.8 A series of articles of M. I. Pudovkin is devoted to the use of the dynamo theory for interpreting the specifics of the evolvement of magnetic disturbances at high latitudes, generalized in his thesis on "The morphology and nature of polar magnetic storms" (Pudovkin, 1968). In these articles the authors estimated the changes of the ionospheric conductivity and deduced the neutral wind circulation at high latitudes based on observations of geomagnetic field variations, of plasma densitites, of drift velocities deduced from plasma irregularities in the ionospheric E-layer, and from movements of discrete auroral forms. It is assumed that currents, which cause the magnetic field variations, flow during magnetic disturbances along extended auroral forms (arcs). These currents are carried by the neutral wind away from the arc and their position is determined from magnetic field observations, while the direction and velocity magnitude of the neutral wind in the ionosphere is calculated from the direction and extent of the transport (Pudovkin, 1960; Pudovkin and Evlashin, 1962). The current is usually shifted toward south relative to the arc during positive disturbances (in the evening) and toward north during negative disturbances (nighttime to early morning hours). The velocity of the current shift amounts to 100 m/sec. It is assumed that this shift results from the action of neutral wind at ionospheric heights. The sign of the geomagnetic disturbances in the horizontal components is related to the direction of the causing currents: during positive disturbances the current shift toward the equator and during negative toward the pole (Pudovkin, 1965b). For the determination of the wind velocity it is assumed, that the source of ionization in the atmosphere (the auroral arc) is static or moves only marginally for the disturbance time (Pudovkin and Korotin, 1961; Pudovkin, 1964). The intensity of the magnetic disturbance is estimated from the increase of plasma density as due to the interaction of the corpuscular flux with the atmosphere and the velocity of the neutral wind in the ionosphere (Korotin and Pudovkin, 1961). A considerable contribution comes also from ionospheric irregularities in the E-layer. Owing to the generation of electric polarization fields, the velocity of the electrons in the electrojets increases by an order of magnitude compared to the neutral wind velocity. The Hall currents, which are hereby generated along the auroral zone, appear to be the cause for the observed magnetic field variations (Pudovkin, 1964). Hist. Geo Space Sci.

450 451 452 453 454 455 456 457 458 459 460 461 462

463 464 465 466 467 468 469

A1 U1 () sin + B1 U1 () cos

(3)

470 471

426

... in the Northern polar cap (b - 1) 1 S 1 () cos + b [A2 U1 () + A3 V1 ()] sin + [ B2 U1 () + B3 V1 ()] cos

472 473

J

D

0 1 = 2b K3 G k1

474

(4)

475 476 477 478 479

427

... in the Northern auroral zone
0 where K3 represents the height-integrated ionospheric conductivity in the polar cap during quiet times, G is the Earth's 1 magnetic field strength at the equator, k1 is the coefficient of a spherical harmonic series for the potential field of the ionospheric wind velocity, U1 () = tan /2, V1 () = cot /2, 1 S 1 () = sin(2/4), A1 A3 and B1 B3 are numerical constants, obtained from the boundary conditions, and b, finally, is the enhancement factor of the conductivity in the auroral zone during disturbed intervals in comparison with the quiet time level. 2.7 The model current system reminds of the S D current system, though of a somehow smaller intensity, with two

480 481

428 429 430 431 432 433 434 435 436 437 438 439

482 483 484 485 486 487 488 489 490 491 492 493

www.hist-geo-space-sci.net

8
494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527

Y. I. Feldstein et al.: Spiral Regularities
547 548 549 550 551 552 553 554 555 556

2.9 Pudovkin (1965a) estimated the system of ionospheric neutral winds in the auroral zone from magnetograms and plotted it according to the longitude of the stations. At all stations in the evening hours the neutral wind is directed from north to south, while in the nighttime and early morning hours it is oppositely directed from south to north. The course of the S D variations is controlled by the diurnal variation of the ionospheric neutral wind in such a manner that the currents, which are the cause for the S D variations, are generated due to the dynamo action of the ionospheric neutral winds. The neutral wind system is comparatively stable according to the dynamo theory and experiences only a small variation, but the magnetic disturbances are caused by steep enhancements of the ionospheric conductivity due to the precipitating corpuscular fluxes (Pudovkin, 1965b). Based on the discussion of ionospheric processes, Pudovkin (1964, 1965b) concluded, that the dynamo theory is the best substantiated theory to explain the generation of ionospheric currents that are responsible for the high-latitude magnetic disturbances. These currents are excited due to the dynamo action of ionospheric neutral winds. 2.10 Viewed from the vintage point of the present, it is amazing how broadly and deeply the success of the dynamo theory was in interpreting the quiet-time features of the ionosphere. But the application of the dynamo theory as physical base to interpret the nature of the geomagnetic disturbances at high latitudes encountered distinct difficulties, both in observational as in theoretical respects. We already mentioned above the discrepancies in the results of the classical dynamo theory with regard to the phase and intensity of the observed current systems, both in magnitude and direction (phase) within the polar cap. The modifications of the theory by M. I. Pudovkin were also not without any problems as listed subsequently. 1. The assumption about the static behaviour of the corpuscular source (auroral arc), which is used for the determination of the neutral wind velocity at ionospheric heights, disagrees with present perceptions of the dynamics of discrete auroral forms. In the majority of cases the auroras appear and allocate not at a fixed latitude, but along the auroral oval, which is asymmetric with respect to the geomagnetic pole. In consequence of this there is a shift of the aurora with 30 m/sec from north toward south in the evening and from south toward north in the morning. During substorm creation (growth) phases, the arcs in the eveningnighttime sector are shifted with a velocity up to 250 m/sec southward. This is just the same range of velocities as that of the neutral wind velocities in the modified dynamo theory. 2. The magnetic variations at auroral zone latitudes ( 67 ) in the evening sector are positive and, therefore, in the frame of the dynamo theory, the neutral wind in the Hist. Geo Space Sci.

ionosphere should be directed from north to south. In the same sector at latitudes of the auroral oval ( 70 ), the magnetic variations are negative and consequently the neutral wind should be directed there from south to north. This divergence of the neutral wind orientation within close latitudinal distances at the evening sector suggests the presence of a neutral wind source within this small latitudinal range. But the existence of such a source is not envisioned in the frame of the dynamo theory. 3. The current system in the dynamo theory is a 2-D one, spanning at the ionospheric E-layer. According to present conceptions, based on experimental data, the current system of the magnetic disturbances is 3-D. The ionospheric electrojets are connected by field-aligned currents (FACs) with the magnetosphere. The character and the intensity of the magnetic disturbances at the Earth's surface is determined by a complicated system of ionospheric and magnetospheric currents. 4. Nowadays it it generally accepted, that the magnetic disturbances at high latitudes as well as the currents in the magnetosphere and ionosphere, which cause them, are controlled by the parameters of the interplanetary medium. The dominating parameter here appears to be the orientation and intensity of the interplanetary magnetic field (IMF), in combination with the solar wind velocity. 2.11 New approaches for the determination of the upper atmosphere neutral wind system at high latitudes were established during the "Sputnik era". It became possible to determine the velocity and direction of the neutral wind insitu in the thermosphere. The satellite Dynamics Explorer 2 (DE2) measured the velocity of the neutral wind at altitudes between 300 km and 550 km by use of an FabryPerot interferometer and a wind and temperature spectrometer (Killeen and Roble, 1988). The empirical neutral wind pattern obtained in geomagnetic coordinates is characterised by a strong antisolar flow over the geomagnetic pole bordered by sunward flow in the dawn and dusk sectors. Negative vorticity occurs in the dusk sector while positive vorticity in the dawn sector. Measurements of lower thermospheric neutral winds at altitudes between 100 km and 300 km, where according to the dynamo theory ionospheric currents are generated, have been realized onboard the Upper Atmosphere Research Satellite (UARS). Richmond et al. (2003) used data of the Wind Imaging Interferometer (WINDII), which records the lineof-sight intensity of the green atomic oxygen line airglow emissions. They showed that the neutral wind motion is controlled by the IMF action down to 105 km altitude, while above 125 km the wind patterns shows considerable similarity with the ionospheric convection patterns. The correlation between the IMF Bz component and the diurnal harwww.hist-geo-space-sci.net

557 558 559 560 561 562 563 564 565

566 567 568 569 570 571 572 573

574 575 576 577 578 579

528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543

580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596

544 545 546

597 598 599

Y. I. Feldstein et al.: Spiral Regularities
600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654

9
655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709

monic of the neutral wind is best, when the IMF is averaged over the preceding 1.0В4.5 hours. Zonal wind below 120 km correlates with the IMF By component when the By component is averaged over the preceding 20 hours. Above 120 km, the dusk side clockwise circulation wind cell is prominent and it intensifies for IMF Bz < 0. Around 120 km one observes a dawn side anticlockwise wind cell that responds to IMF Bz variations. During intervals with IMF Bz > 0, the neutral wind modification is mainly confined to the polar cap, while for Bz < 0 it extends to subauroral latitudes. On timescales of 20 hours an IMF By dependent zonal wind generally exists at 120 km altitude and 80 geomagnetic latitude with maximum wind speeds of 60 m/ s. The paper of Richmond et al. (2003) includes a broad range of references on thermospheric neutral wind investigations and their modelling. Cross-track accelerometer measurements at the CHAMP satellite were used to deduce statistical neutral wind patЁ tern at the high-latitude upper atmosphere (Luhr et al., 2007; Ё Forster et al., 2008). These measurements were obtained at altitudes of 400 km and confirmed the existence of a strong antisolar flow with an average velocity of 500 m/s and an azimuth of 168 over the central polar cap. At the evening side a large clockwise circulation cell is formed, resulting in the tendency for westward (sunward) neutral wind velocities in the auroral region of the duskside upper atmosphere. Ё Forster et al. (2008, 2011) analysed the neutral wind observations of CHAMP in dependence on the IMF orientation. The data were sorted into 8 separate IMF sectors with 45 width each. Measurements over a 2-year interval resulted in a good data coverage for each IMF sector. The upper thermospheric neutral wind system is characterized in all sectors by a transpolar neutral wind orientation across the central polar cap and two vortex cells with opposite circulations on the dawn and dusk side, the strength and shape of which depend on the IMF By and Bz components. The amplitude of the cross-polar cap neutral wind flow is largest for negative IMF Bz values ( Bz < 0) and amounts on average to 570 m/s Ё (cf. Forster et al., 2008, Table 1). A couple of characteristic phenomena of the IMF By and Bz dependence on the spatial distribution of the neutral wind in the upper atmosphere are gathered and described in detail. The satellite observations of thermospheric neutral wind and ionospheric plasma convection revealed a fairly close similarity of their patterns. The correlation of the velocity and direction of the neutral wind and plasma convection with the IMF Bz and By components verifies the control of the thermospheric neutral wind regime by the IMF parameters. Both the plasma motion and the neutral wind appear as a result of the electric fields, generated in the outer magnetosphere. They are constituded due to the interaction of the solar wind with the geomagnetic field and then transmitted along the magnetic field lines down into the ionosphere. The plasma in the thermosphere drifts in the electric and magnetic fields. The neutral thermospheric component adopts the www.hist-geo-space-sci.net

plasma motion due to the ion drag forces. The secondariness of the neutral wind results in a time lag of the neutral wind with respect to the convection. The magnitude and direction of the ion motion is determined by the intensity of the IMF components. Due to these relations, the neutral wind at high latitudes also depends on the IMF. The global neutral wind system, that is due to this interaction, differs considerably from that system, which is essential for the generation of magnetic disturbances observed at the Earth's surface, as explained by the existing dynamo theory. 2.12 Patterns of magnetic disturbance spirals, that were obtained from the materials of the IInd IPY by L. Harang, J. H. Meek, and O. A. Burdo, differ both in the quantity of spirals and in their sources (2 spirals of corpuscular precipitation according to J. H. Meek, but 3 spirals in the dynamo theory according to L. Harang and O. A. Burdo). All three studies used yearly averaged data. There are, however, significant seasonal dependences in the variations on the geomagnetic field (Benkova, 1948; Feldstein, 1963a). Feldstein and Zaitzev (1965a) studied the S D variations of three geomagnetic field components using data of 24 magnetic observatories during the IGY of the Northern Hemisphere separately for winter, equinox, and summer periods. The horizontal projection of the disturbance vector T was determined according to the relation T = (H 2 + D2 ) as deviations from the quiet time level with D as the deviation from the East-West component. 2.12.1 During the winter season, the S D variation at > 73 is characterized by one maximum near midday. In the latitude range 73 > > 67 the maxima of T occur in the evening and morning hours, and in the latitude range 67 > > 60 they shift to nighttime and evening hours. Below < 60 the nighttime maximum in T is the only one. Figure 3a shows the patterns of T and Z for the winter season of the IGY (November 1957 February 1958) in coordinates of versus MLT. The arrows indicate the disturbance vector directions in the horizontal plane; the arrow's length is in accord with the amplitude of T . The scale is given in the upper left corner of the Figure. The numbers at the arrow's origin specify the amplitude of Z . The dependence on the maximum moments in the course of the diurnal variation of T over is indicated with thick lines: (1) morning maximum (M), (2) nighttime maximum (N), and (3) evening maximum (E) in Fig. 3a. The sign of the Z-component of the geomagnetic variations changes along these lines. Such a distribution of the vector magnitude and direction implies, that intense equivalent currents are flowing along these lines. Along the lines 1 and 2, forming an oval belt, the current flows westward and decreases the horizontal magnetic component. Along line 3 the current flows eastward, increasing this component. 2.12.2 In the limited latitudinal range of 69 < < 72 , there exist only negative disturbances of the horizontal component during the whole day. This region at auroral latitudes, where we have H < 0 in the winter season during Hist. Geo Space Sci.

10

Y. I. Feldstein et al.: Spiral Regularities

Figure 3. Loci of maximum intensity T and minimum Z fields in LMT versus corrected geomagnetic latitudes in Northern Hemisphere.

The digits 1, 2, and 3 correspond to morning, nighttime, and evening spirals, respectively (Feldstein and Zaitzev, 1965a). a) winter season of IGY (upper panel) and b) summer season of IGY (lower panel).

Hist. Geo Space Sci.

www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757

11
762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811

the whole day, is fundamentally different from the character of the diurnal variation within the auroral zone. There occur beside the disturbances with H < 0 during the nighttime and morning MLT hours generally at the same UT also disturbances H > 0 in the evening MLT hours (Burdo, 1960). The existence of a latitude interval with H < 0 was confirmed by the observations of the drifting station North Pole 6 68 < < 70 during the period from September 1957 to April 1958 (Zhigalova and Ol, 1964). During all hours of the day, they practically observed only negative bays. The frequency of their appearance maximizes at 07 and 21 MLT, which corresponds to the times of the station's passage under the oval current belt (Fig. 3a, lines 1 and 2) in the course of the Earth's rotation. The current belt at Fig. 3a, which consists of the spirals 1 and 2, coincides with the position of the auroral oval (Feldstein, 1963b, 1966), Auroras were observed along the auroral oval during the winter season of the IGY practically continuously at zenith. The appearance of discrete auroral forms implies the precipitation of energetic electrons into the upper atmospheric layers. The electron precipitation with energies of 1 В 5 keV along the auroral oval leads to steep increase of the conductivity at dynamo region altitudes. The incoming and outgoing FACs result in the generation of a meridional electric field, directed equatorward. In this way the conditions for the appearance of the electrojet with a westward directed Hall current are established along the auroral oval. 2.12.3 Fig. 3b shows the patterns of T and Z for the summer season of the IGY in coordinates of versus MLT. The legend here is the same as for Fig. 3a. The dependencies of the maximum moments in the course of the diurnal variation of T are depicted with the thick lines 1(M), 2(N), and 3(E), together with the corresponding sign changes of Z . The M and N spiral with negative values of H forms, as in the winter season, the auroral belt, along which the current flows in westward direction. In contrast to the winter season, T increases sharply in daytime hours at > 75 . The intensity of the disturbances in the dayside sector at the auroral oval is comparable with the nighttimeearly morning disturbances at latitudes of the auroral zone. A noticable increase of the T vector intensity, caused by the positive values of H , occurs during evening MLT hours in the range 60 < < 70 . The patterns of T and Z for the summer and winter seasons during the IGY give no hint on the existence of two circular zones at fixed latitudes with maxima of the magnetic activity. The absence of such circular zones is confirmed by latitudinal cross-sections in T during various MLT hours.

758 759

3

Spirals in the irregular geomagnetic field variations Di (agitation)

812 813 814

760 761

3.1 The high-latitude diurnal geomagnetic variations contain a component, which displays an irregular character. It is usuwww.hist-geo-space-sci.net

815 816

ally investigated by studying the regularities of geomagnetic activities. Mayaud (1956) proposed the term agitation for the indication of geomagnetic disturbances, which are due to the precipitation of corpuscular radiation into the upper atmosphere, leading to irregular magnetic field variations. The intensity of agitation is estimated by means of various indices, which characterize the magnetic activity in different regions of the globe. 3.1.a The Polar Cap (PC) index is the most representative one for the polar cap, both for the Northern (PCN) and the Southern (PCS) Hemisphere (Troshichev et al., 1988, 2006). The PC indices are derived from ground geomagnetic measurements in the Northern (PCN) and Southern (PCS) Polar Caps. The PCN index is based on the data from Thule (Greenland) while the PCS index is based on data from Vostok in Antarctica. The PC index in its present form was first formulated by Troshichev et al. (1988) as a planetary index. PCN index values have been supplied from the Danish Meteorological Institute (DMI) while PCS index values have been supplied from Arctic and Antarctic Research Institute (AARI) (Troshichev et al., 2006). Basically, the PC index with a temporal resolution of 1 min represents the polar cap magnetic variations F PC in nT , associated with the transpolar part of the current system driven by the electric field of the solar windmagnetosphere dynamo. 3.1.b The planetary indices AE, AU, and AL were introduced by Davis and Sugiura (1966) to characterize the UT variation (in nT ) of the eastward (AU), the westward (AL) and the total electrojet intensity (AE). These indices were determined by the World Data Center A for Solar-Terrestrial Physics, NOAA, Boulder, USA (Allen et al., 1974, 1975) from 1966 to 1971. The indices were determined with a temporal resolution of 2.5 min as maximal deviation from the quiet time level for positive (AU) and negative (AL) values of the H-component of the geomagnetic field based on a network of 1011 magnetometer stations. These observatories were distributed at auroral latitudes with a rather good coverage in longitude (Allen and Kroehl, 1975). During the subsequent years up to now, the AE, AU, and AL indices are determined with a 1-min cadence by the World Data Center for Geomagnetism, Kyoto, Japan, based on the data of 12 observatories. 3.1.c For the study of the regularities of agitation developments at high latitudes, one-hour amplitude averages of the horizontal components of H and/or D were largely employed (cf., e.g., Nikolsky, 1951; Burdo, 1960; Lassen, 1963, and references therein). Nikolsky (1947) used the curve-length of the D-variometer records as an every-hour characteristic of magnetic disturbances. 3.1.d Disturbances at mid-latitudes were estimated for particular geomagnetic observatories by means of the quasilogarithmic index K (Bartels et al., 1939). This agreed measure, with values from 0 to 9, estimates the amplitudes of the H and D component variations as deviations from the quiet time level within 3-hourly intervals of UT. The estimation of Hist. Geo Space Sci.

12
817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871

Y. I. Feldstein et al.: Spiral Regularities
872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926

K indices in various observatories for one and the same time interval results generally in different values. The averages of the K values from 11 different observatories within the latitudinal range of 44 < || < 57 , from which only one is localized at the Southern Hemisphere, represents the planetary index Kp. This index was introduced by Bartels in 1949. The Kp values counted according to a particular scale, translate to the equivalent 3-hourly amplitude index ap (in nT ) and their daily averages Ap. The indices Kp, ap, and Ap were derived by the Geophysical InЁ stitute of the Gottingen University from January 1932 until the end of 1996. Since January 1997 the derivation and distribution of the Kp values have been moved to the Adolf Schmidt Geomagnetic Observatory Niemegk, part of the GeoForschungsZentrum (GFZ) Potsdam (now: GFZ German Research Centre for Geosciences, Helmholtz Centre Potsdam). To characterize the value of the daily magnetic activity level, the indices C and Ci are used. The magnetograms of any observatory are quantified for each day (in UT) with a magnetic character figure of the value 0, 1, or 2. This character figure appears also as the C index of this day for the given observatory. Arithmetic mean values of the C index values from all reporting observatories result in the Ci index, which runs from 0.0 for very quiet conditions to 2.0 for very disturbed ones. The planetary index Cp for the daily disturbance characteristic, which varies from 0.0 to 2.5 in steps of 0.1, is calculated from the Ap index. The Cp index is used to discern the international magnetically disturbed and magnetically quiet days. Mayaud (1967) assumed that the Kp index does not sufficiently reflect the planetary character of a magnetic disturbance and proposed its modification. The modified indices Km, am, and Am are based on the K index of 24 mid-latitude stations, including 9 at the Southern Hemisphere. 3.1.e Magnetic disturbances of more than 100 years are described with the aa index (Mayaud, 1972). This index was calculated from the K index of the observatories Melbourne and Greenwich (i.e., one station in each hemisphere), beginning from 1868. Later these stations were replaced by Abinger and Hartland for Greenwich and by Toolangui and Canberra for Melborne. From the aa values, daily means (Aa index) and monthly means are calculated. 3.1.f To describe the variations of the Earth's magnetic field during geomagnetic storms, the planetary Dst index is broadly used. A starting point for the studies concerning Dst derivation can be found at the end of the forties of the last century. More systematic determinations started after the IGY (Sugiura, 1964). For the derivation of the hourly Dst values, one usually uses hourly UT values of the low-latitude magnetic observatories Honolulu (USA), Kakioka (Japan), Hermanus (SAR), and San Juan (USA). These observatories were chosen for the reason that their locations are sufficiently distant from the auroral and equatorial electrojets and that they are distributed in longitudes as evenly as possible. The Hist. Geo Space Sci.

Dst index represents the axially symmetric part of the disturbance magnetic field at the dipole equator on the Earth surface (in nT ). Nowadays the World Data Center for Geomagnetism, Kyoto, Japan, provides with 1-min resolution the so-called SymH index (replacing Dst) as well as an asymmetric index AsymH as difference between the maximum and minimum values of the H component from a network of lowlatitude stations. 3.1.g Up to the nineties of the past century, i.e. practically over 50 years, it was generally assumed, that during geomagnetic storms the Dst index represents the magnetic variation fields of the magnetopause currents DCF and of the ring current DR (see, for instance, the reviews of Gonzalez et al., 1994; Kamide et al., 1998). The contribution of the magnetospheric tail current fields (DT) were assumed to be less than a few nT in the observations at the Earth's surface. Only resulting from the studies of Arykov and Maltsev (1993); Arykov et al. (1996); Alexeev et al. (1996); Maltsev (2004) it became clear, that the DT contribution in the Dst index during magnetic storm periods is comparable to the DCF and DR contributions. 3.1.h The Sa (used in the literature to identify the diurnal change of geomagnetic activity) has a complicated form at high latitudes, which changes with latitude. The consideration of irregular magnetic disturbances revealed, that they follow certain regularities at particular stations, i.e. the magnetic disturbance has a significant diurnal course. The onset time as well as the maximum intensity of the disturbance depends on the station's position with respect to the auroral oval. Below we present, according to our guess, basic results, summarizing the research on magnetic agitation regularities at high latitudes concerning the spiral distributions. We seek to follow in this review the historical progression according to the publication time of new ideas. 3.2. The first study on Sa at high latitudes was performed with observational material of the IInd IPY by Stagg (1935), carried out with data of six arctic , one mid-latitude, and three antarctic stations. It was shown, that the form of Sa is determined by the magnetic latitude and local time. In mid-latitudes the activity maximizes during evening hours of local time. Progressing to higher latitudes, the maximum moves to later hours and during midnight it is observed at 69 . The form of Sa changes at latitudes of highest activity between 70 < < 78 . There two maxima are observed: one at the morning hours, the other during nighttime. At > 78 , the maximum activity reaches morning or even midday hours. Nikolsky (1947) assembled observational results of 16 stations, partly positioned within the formerly inaccessible Eastern regions of the Arctic, and found some differences from Stagg's results: the morning maximum for stations along one and the same geomagnetic parallel occurs at different local times; the equatorial boundary of the active zone is recorded down to a latitude of 62 ; the morning maximum appears in the western hemisphere almost concurrently at 15.516.5 MLT, while its appearance lags by www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
927 928 929 930 931 932 933 934 935 936 937 938

13
980 981 982 983 984 985 986 987 988 989 990 991 992

0.7 hours in the eastern hemisphere with shifted by 1 ; intensity variations with latitude are different during nighttime and morning maxima. Noting the differences of the evening and morning magnetic activities, Nikolsky (1947) assumed, that their behaviour follows different regularities. He didn't even exclude different natures of the corpuscular fluxes, responsible for the two types of magnetic disturbances. 3.3 Benkova (1948) considered 17 Arctic observatories and concluded, that the Sa value of high-latitude stations follow the scheme, that was proposed by Stagg (1935). The description of Sa as expansion of harmonic series according to Sa =
n

939

Cn cos(nt - n )

(5)

993 994

940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979

has shown, that over the range of = 0 to 90 the param- 995 eter 1 varies between -2 and +2 and the sign change of 996 the first term of the expansion occurs at the latitude of the 997 auroral zone. The dependence on 1 from in polar co- 998 ordinates turns out to be a spiral, winding up in clockwise 999 direction through the morning sector. This appears to be the1000 first mention of a spiral distribution in the literature relating1001 to the morning maximum of magnetic activity. 1002 3.4 A.P. Nikolsky addressed the analysis of spiral distribu-1003 tions of the high-latitude magnetic activity in a large number1004 of publications. The most important results are collected in1005 the anthologies of the Arctic and Antarctic Research Insti-1006 tute (AARI) in St. Petersburg (Nikolsky, 1951, 1956, 1960a).1007 It was shown, that the specific characteristic of the Sa con-1008 sists in the appearance of maxima at morning, daytime, and1009 nighttime hours, the relative amplitude of which varies in de-1010 pendence on the stations's distribution. The intensity of the1011 morning and dayside disturbances increases with increasing1012 geomagnetic latitude, the nighttime disturbances are fixed1013 to the local midnight and maximize at latitudes of the au-1014 roral zone. A.P. Nikolsky turned his particular attention to1015 the study of morning and daytime disturbances. Consider-1016 ing observations at 30 places with > 60 , the observational1017 material is very broadly scattered about the indices of activity1018 used. It turned out that the agitation during morning and day-1019 time hours is not connected with the auroral zone. Consider-1020 ing contour maps of concurrent appearance (with respect to1021 UT) of the morning agitation maxima, it became clear, that1022 the isoline represent a system of spiral sections, which begin1023 at the pole of homogeneous magnetization. At the Northern1024 Hemisphere they wind up clockwise. Nikolsky assumed, that1025 these sections of spirals should be seen as projections onto1026 the Earth's surface of those real trajectories, along which the1027 solar corpuscles penetrate into the upper atmosphere. The1028 positions of the most frequent particle precipitations along1029 the spirals are determined by the character and the parame-1030 ters of the particles in the corpuscular fluxes. 1031 3.5 Nikolsky (1960a) identifies the spirals of the morning1032 agitation maximum with the StЬrmer's spirals of penetrating1033 solar corpuscles. Because these spirals wind up clockwise1034 www.hist-geo-space-sci.net

in the Artic, these corpuscles should be protons according to StЬrmer's theory. The appropriateness of Nikolsky's assumption about the congruency of his spirals with the spirals of StЬrmer was challenged by Agy (1960). He paid attention to the fact, that the Nikolsky's spirals of the morning Sa maximum represent positions at the Earth's surface of the simultaneous (in UT) onset moment of maximum activity along the whole spiral. These spirals differ essentially from those of StЬrmer, along which the penetration of corpuscles occurs simultaneously only in a few points. The whole spiral of StЬrmer results from corpuscular precipitations over a longer time interval of the order of a year. 3.6 Based on the calculation results of StЬrmer (1955), Nikolsky (1960a) concludes, that there exist four regions within the morning spirals, where the solar proton trajectories are gathering closer (see Fig. 1). These regions ("A", "B", "C", and "D") occur at 02, 08-09, 14, and 20 hours LT, respectively. Nikolsky suggests that as the Earth rotates diurnally under the spiral, the high point density sections of the spiral on Fig. 1 will trace out four ring "auroral zones" or regions with auroras in the zenith and increased magnetic agitation. In Fig. 1 there are four such high pointdensity sections and therefore, according to Nikolsky, four auroral zones will occur. Fig. 4 from the paper of Nikolsky (1960a) shows the positions of all four regions "A", "B", "C", and "D" at the Earth's surface. The figure shows also the image replication of StЬrmer's spiral (named "S"), the position of which was arbitrarily changed by Nikolsky. The spiral in Fig. 4 is prolonged to low latitudes (observatory Bombay in India). The extension of each of the "auroral zones" to the equator is limited by the position of the high point density section of the spiral in Fig. 4. 3.7 Zone "B" is identified as second (inner) region, where the intensity of magnetic agitation and the occurence of auroras achieves maximum values at 0809 LT. The summary effect of the precipitation zones "A" (02 LT) and D (20 LT) appears as the main auroral zone ("Fritz zone"), which is located at mid-latitudes over the American continent. Such a position of the "Fritz zone" is quite unusual and in case of its observational confirmation it would be a good argument in favour of the planetary distribution of corpuscular precipitation and the related auroras and magnetic agitation, as suggested by A. P. Nikolsky. Unfortunately, it is not possible to verify the appearance of zenith auroral forms in region "C" with observations - due to daylight conditions around 14 LT. The appearance of auroral forms in the zenith discloses, in contrast to the magnetic disturbances, directly the location of precipitating corpuscular fluxes. Feldstein (1963a) verified hence the existence of four zones of maximum corpuscular precipitation useing auroral observations from the IGY period (Annals, 1962). 3.8 The appearance of auroras is usually very rare at midlatitudes and if it appears, then preferably with northern azimuths. Near the zones "A" and "D", in case they are precipitation zones, the auroral forms should appear mainly at Hist. Geo Space Sci.

14

Y. I. Feldstein et al.: Spiral Regularities

Figure 4. Location of solar corpuscular flux precipitation zones "A", "B", "C", and "D" above Earths surface and corresponding maximum values of agitation and appearance frequency of auroras according to Nikolsky (1960a). Letter "S" marks StЬrmers spiral, which is arbitrarily prolonged towards low latitudes.

1035 1036 1037 1038 1039 1040 1041

zenith. The positions of some observatories in the environ-1042 ment of zones "A" and "D" are shown in Fig. 4 and their1043 coordinates are given in Table 1 together with the respec-1044 tive probability for the appearance of auroras in the zenith in1045 per cent of the total number of observations as well as the1046 number of observation intervals with auroras northward and1047 southward of the station. 1048
1049 1050

# 1 2 3 4 5

Station Fritz-Peak Delaware Ithaka Pulman Meanook , 39 40 42 46 54 N 52' 30' 27' 43' 36'

Coordinates , W 105 31' 83 76 31' 117 10' 113 20'

,N 48.7 51.6 53.8 53.5 61.8

Zenith % 2.2 3.6 5.2 11.0 90.0

Number 1051 N S 1052 82 1 1053 152 261054 275 901055 236 941056 888 506057 1
1058 1059

Table 1. Auroral observatories during the IGY with their coordi-

nates (geographic latitude and longitude as well as the corrected geomagnetic latitude ; appearance of auroras in the zenith in per1060 cent of the total and the number of observations northward (N) and1061 southward (S) of the station. 1062

From Table 1 follows, that the probability of zenith auroras at the stations Delaware and Ithaka are small, although these stations are close to the zones "A" and "D". The auroras are seen mainly northward from the observatories, as it is usually expected at mid-latitudes of the Northern Hemisphere, i.e., in the direction toward the auroral zone. The Fritz-Peak station is situated between zones "A" and "D", which suggests observations of auroras both northward and southward. But for the period from August 1st to December 1st, 1957, there was only one case of observations in southward direction. The station Pulman is close to zone "A" in the western part of the continent. The number of auroras in the zenith was small and they were recorded mainly in northward direction. At the station Meanook, which is north of the zones "A" and "D", one would expect by far more aurora toward south. But they appeared more often northward, toward the aurora zone. 3.9 The observations of auroras at the mid-latitude stations at the American continent, situated in the region of corpuscular precipitation that are formed by the mid-latitude parts of zones "A" and "D", show the incorrectness of the planetary www.hist-geo-space-sci.net

Hist. Geo Space Sci.

Y. I. Feldstein et al.: Spiral Regularities
1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117

15 maxima are interpreted as zones of enhanced magnetic activity (agitation): the zone of maximal disturbances at 66 is thought to be coinciding with the main auroral zone and that at 77.5 coincinding with the second, high-latitude zone or region "B" according to Nikolsky (1960a). Such an interpretation of the latitudinal amplitude distribution of the first harmonic, that was made by Mansurov and Mishin (1960), differs from the general view on the zones of maximal magnetic agitation. The latter is usually characterized by the largest daily averaged agitation value, but not by the amplitude variation value in the course of the day. Further, the amplitude values of the first harmonic K1 is determined not only from the activity level, but also from the characteristic Sa value. The minimum of K1 at 73 might be caused by two Sa maxima at this latitude in different local times. The K1 values appear to be minimal during sufficiently strong daily averaged disturbance levels. 3.12 Beside of those studies, which employ statistical methods to derive regularities of magnetic disturbances, a number of scientists investigated the morphology of magnetic storms by comparing the magnetograms of various high-latitude observatories. Such a method was used in the studies of Birkeland (1913). The possibilties for applying this kind of methods extended greatly after the IGY. The data of a planetary-scale network of stations, collected in the World Data Centers, give a comprehensive perception of the global magnetic disturbances at all latitudes and longitudes. Results obtained by this kind of methods were presented, e.g., by Bobrov (1960). 3.13 Whitham and Loomer (1956) addressed in their investigations the question of the existence and position of a second zone of enhanced magnetic activity in the Canadian sector of the Arctic. The magnetic observatory Resolute Bay ґ ( 83 ) is situated, according to Alfven (1955) and Nikolsky (1956), directly under the second zone, while the station Baker Lake 73.7 sits between the first and the second zone of enhanced magnetic activity. The comparison of the magnetic data of these two stations for the years 1953-1954 showed an interconnection between their disturbance states, but no unambiguous evidence for the existence of the second zone of corpuscular precipitation that spans above Resolute Bay. Whitham et al. (1960) analysed the latitudinal variation of the magnetic disturbances based on 16 magnetic stations in Canada with 54 < < 86 for winter, equinox, and summer periods of the IGY. They used the hourly amplitudes of horizontal components (H, X, or Y) as the index of magnetic activity and the levels of activity were determined for all data as well as separately for the disturbed and the quiet days. It was concluded, that there exists a maximum in the latitudinal distribution at 67 for all seasons and for all activity levels, in which during disturbed days the level reaches 200 nT. Confirming the former results of Whitham and Loomer (1956), no anomaly effects were found at the station Resolute Bay, which could indicate a second zone of Hist. Geo Space Sci.

precipitation scheme, which had been proposed by Nikolsky.1118 Along these zones, which stretch into mid-latitudes, zenith1119 auroras are practically missing. Consequently corpuscular1120 precipitations, which would lead to discrete auroral forms,1121 are non-existent there. The existence of corpuscular precipi-1122 tation regions, which extent into mid-latitudes or even to the1123 equator during evening and nighttime hours, that would lead1124 to the appearance of auroras at these latitudes, belongs rather1125 to the genre of scientific fiction than to scientific results. 1126 Referring to some morphological characteristics of mag-1127 netic disturbances at mid- and subauroral latitudes, Hope1128 (1961) suggests, that they may be explained as disturbances1129 transported from high to low latitudes by the ionospheric cur-1130 rent system, probably identical with the S D circulation. In1131 addition we note that one cannot preclude a certain contribu-1132 tion to the disturbances at these latitudes originating from the1133 magnetic fields of the magnetospheric currents systems and1134 from the FACs between ionosphere and magnetosphere. 1135 3.10 Mayaud (1956) studied the magnetic agitation in the1136 polar region by using three-hourly amplitudes, obtained from1137 the K-indices of 49 magnetic observatories at the Northern1138 and Southern Hemispheres with | | > 50 . He separated1139 nighttime and daytime types of agitation, which corresponds1140 to the nighttime (N) and morning (M) maxima of activity ac-1141 cording to the terminology of Nikolsky. Nighttime agitation1142 is characteristic for magnetically disturbed days and maxi-1143 mizes at latitudes of the auroral zone (magnetic inclination1144 of I 77 ). The time moment of reaching the maximum ag-1145 itation in the diurnal variation at the auroral latitudes lagged1146 behind those at neighbouring (northward and southward) lat-1147 itudes. The daytime agitation is most characteristic for the1148 near-polar region with I 85 , whose maximum is observed1149 at local magnetic midday and shifted to 06 MLT at I 81 .1150 The daytime agitation has maximum values in summer, mini-1151 mum values in winter, and stays at a comparatively high level1152 during magnetically quiet days. Daily averaged agitation val-1153 ues showed in the latitudinal dependence both for individual1154 seasons and for the full year on average, that the 3-hourly am-1155 plitudes have one maximum at 65 , but there is no dis-1156 turbance enhancement in the near-polar region, which should1157 be expected in case of a second, high-latitude maximum of1158 the agitation. 1159 3.11 Mansurov and Mishin (1960) investigated the Sa by1160 using data from high-latitude stations in the Northern (19)1161 and Southern (11) Hemisphere. Again the 3-hourly K-index1162 was employed as measure of agitation. It was assumed that1163 the Sa is controlled by local time and UT. That part, which1164 depends on local time, is described as: S 0 (t) = S 0 + K1 cos(t -1165 1 ) + .... They determined the amplitudes and phases of the1166 first harmonics and modelled the dependence of the ampli-1167 tude from geomagnetic latitudes for the winter and the sum-1168 mer season. During the winter, there exists only one max-1169 imum at 67 , but during the summer season there are1170 two maxima at 66 and 77.5 in the latitudinal vari-1171 ation with a deep minimum at 73 in between. These1172 www.hist-geo-space-sci.net

16
1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227

Y. I. Feldstein et al.: Spiral Regularities imal deviation of these elements from the quiet time level. The disturbance of one hour is estimated at most by four ? 15-min intervals. Observations form 27 stations were used, which cover the high-latitudes of the Northern Hemisphere between 58.4 < < 88.1 . The IGY observations allowed at the highest level the use of homogeneous material both for the interval length and time of the observations as well as with respect to the activity index. Diurnal and latitudinal variations of the agitation were analysed separately for magnetically disturbed and magnetically quiet days. The form of the diurnal variation depends on season and to a great extent on the geomagnetic latitude of the observation. 3.16 The Sa has one maximum in the near-polar region during the winter season converging to the midday MLT hours (at stations Thule, = 86.2 , Resolute Bay, = 83.0 , and Godhavn, = 77.5 ). At lower geomagnetic latitudes, this maximum shift to the morning hours and in the Sa appears a second, evening maximum (stations Murchison Bay, = 75.1 , Tikhaya Bay, = 74.7 , Baker Lake, = 73.7 , Cape Cheluskin, = 71.6 , and Point Barrow, = 69.4 . At even lower latitudes, these maxima are difficult to discriminate as they merge to one single maximum approaching midnight hours (stations Dixon Island, = 68.0 , Leirvogur, = 66.8 , Kiruna, = 65.4 , College, = Ё 64.7 , Murmansk, = 64.1 , Sodankyla, = 63.0 , and Meanook = 61.9 . The Sa forms are better controlled by the corrected geomagnetic latitude (Hultqvist, 1958; Gustafsson, 1970) than by geomagnetic latitude. The stations Julianehab ( = 70.8 , = 68.7 ) and Leirvogur ( = 70.2 , = 66.8 ), which are located at relatively high geomagnetic latitudes, have only one near-midnight Sa maximum, which is characteristic for the auroral zone, but the stations Point Barrow ( = 68.3 , = 69.4 ) and Cape Chelyuskin ( = 66.2 , = 71.6 ) at relatively low latitudes have clearly two maxima (nighttime and morning), which is characteristic for stations at latitudes adjoining the near-polar region. The latitudinal anomalies in the Sa forms disappear, when replacing the geomagnetic latitude with the corrected geomagnetic latitude . When moving from the auroral zone toward higher latitudes, the morning disturbances shift to later hours, and the nighttime disturbances to earlier MLT hours. Drawn in polar coordinates versus MLT, the maxima in Sa are ordered along segments of spirals. The change in local time for varying latitudes s stronger for the morning maximum than for the nighttime maximum. There are no longitudinal differences in the time of maximum appearance for the Sa. This is an evidence for the unimportance of the UT moment for the positions of maxima in Sa. 3.17 In the summer season, the Sa behaviour is subject to substantial changes. From the geomagnetic pole down to 70 , Sa is characterised now by one maximum in the prenoon MLT hours. At lower latitudes, one observes two maxima of Sa: one in the post-midnight to early morning hours and one in the afternoon to evening MLT hours. The Sa maximum in the afternoon to evening hours turn up as the www.hist-geo-space-sci.net

enhanced magnetic activity in the proximity of the station.1228 Some activity enhancements at the station Alert ( 85.7 )1229 close to the geomagnetic pole were assumed to be due to a1230 limited localized region of enhanced activity. 1231 3.14 Some progress in the analysis of regularities in the1232 spatial-temporal distribution of the magnetic activity at high1233 latitudes came along with the research efforts of Burdo1234 (1960). Observational material of 30 locations were used,1235 unfortunately non-uniform in interval length, times of ob-1236 servation and the activity index used. He devoted particu-1237 lar attention in the study of the Sa forms, because the abso-1238 lute amplitude of the magnetic disturbance depends essen-1239 tially from the season and the index used. Seasonal varia-1240 tions could not be deduced, since the observations at 16 (out1241 of 30) locations were shorter than one month. The latitudi-1242 nal dependence on the maximum entry times in Sa appears1243 to be more regularly when using the corrected geomagnetic1244 latitude , which takes into account the position of the sta-1245 tion with respect to the auroral zone. This means according1246 to Burdo (1960), that the magnetic activity at any time of1247 the day is controlled by processes in the auroral zone. The1248 activity is most intense at corrected geomagnetic latitudes of1249 64 < < 66 between 0103 MLT. This maximum splits up1250 with increasing latitude and becomes effectively two maxima1251 one at morning and the other at nighttime. The distance be-1252 tween the two maxima increases with further shift toward the1253 pole. A third (evening) maximum at 64 < < 66 comes in1254 at 1719 MLT. With increasing , the morning and evening1255 maxima convene more and more and converge finally into1256 one at 78 - 80 , which is found at 1012 MLT. In ob-1257 servatories with the same distance to the auroral zone, but1258 at different longitudes, the maximum occurs approximately1259 at one and the same MLT. This fact serves as proof that the1260 universal time has practically no influence on the form of Sa.1261 Synchronous hourly values of the disturbance vector T hor1262 at the stations Fort Rae ( 69.7 ) and TromsЬ ( 67 )1263 were compared for 1516 UT, when these stations are un-1264 der the intense current regions of the morning and evening1265 vortices, respectively. It turned out that the disturbances at1266 both stations appear at the same time, i.e., with the increase1267 of the disturbance intensity at one station it also increased at1268 the other. This disproves the conclusion of Nikolsky (1951)1269 about the independence on disturbances during the morning1270 and evening hours. The disturbances thus appear simultane-1271 ously at the morning and evening side of the Earth, compris-1272 ing the whole high-latitude region with maximum intensities1273 close to the magnetic spirals. 1274 3.15 The regularities of magnetic disturbances at high1275 latitudes were studied by Feldstein (1963a) using data of a1276 global set of stations from the IGY period. Hourly Q-index1277 values of magnetic activity were used, which allow the study1278 of diurnal and latitudinal variations of the activity (Bartels1279 and Fukushima, 1956). This index characterizes the ampli-1280 tude variations of the horizontal geomagnetic field compo-1281 nent for 15-min intervals and accounts further for the max-1282 Hist. Geo Space Sci.

Y. I. Feldstein et al.: Spiral Regularities
1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330

17 63.0 ), Cape Wellen ( = 62.5 ), Meanook ( = 61.9 ), and Lerwick ( = 59.4 ). The three distinct maxima in Sa are obviously the characteristic pecularity of the summer season during days of intense magnetic disturbances. Apart from the dependence on the magnetic activity level, during the winter season, as a rule, there exist not more than two maxima. The evening maximum does not appear related to the strong weakening of the disturbance intensity with H > 0 during these hours. Such disturbances exist, but they do not appear as maximum in Sa at the background of the subsequent more intense disturbances with H < 0. The intensity of the disturbances along the winter and summer spirals is different. During winter, the nighttime and morning disturbances maximize at latitudes of the auroral zone and become smaller toward higher latitudes. This decrease is very small till 75 , but at higher latitudes the disturbance decreases sharply. During summer, the disturbances along the evening and morning spirals increases monotonically from mid-latitudes to the auroral zone and beyond that till 72 , but for the morning maximum it remains at a high level until the near-polar region at 84 . 3.19 What is the nature of the spiral distributions in the time moments of maximum values in Sa? There exist essentially two assumptions in the public literature: firstly, precipitation of corpuscular fluxes takes place along all spirals or at least along some of it (Nikolsky, 1956; Feldstein, 1963a) or, secondly, the spirals are the places of convergent current lines of the equivalent current system (Harang, 1946; Burdo, 1960). In the first case, the morning and nighttime spirals of the winter season should coincide with the position of auroras in the zenith, i.e., with the auroral oval, as the position of auroras points at the places of precipitating electrons with auroral energy into the upper atmospheric layers. The magnetic disturbance can be the result of processes, which take place far from the location of their records. In Fig. 6 (from Akasofu, 1968) shows the M, N, and E spirals, obtained by Feldstein (1963a), together with the auroral oval according to Feldstein (1963b) in polar coordinates versus MLT. This compilation shows, that during magnetically disturbed days the nighttime spiral N and during morning hours the M spiral are situated within the auroral oval, i.e., within the region of precipitating electrons with auroral energy. The N and M spirals, obtained from observations of the magnetic disturbances, are therefore positioned within the auroral oval of precipitating electrons. The evening spiral E adjoins to the equatorial side of the auroral oval. This is the region of the eastward electrojet and of precipitating protons with auroral energies and soft electrons (Starkov and Feldstein, 1970).
4 Spirals in the distributions of auroras

Figure 5. Dependencies of local geomagnetic time of Sa maximum occurrence on corrected geomagnetic latitude for magneto-1332 disturbed days. a- winter, b summer 1 morning (M) maximum of1333 agitation; 2 night (N) maximum of agitation; 3 evening (E) maxi-1334 1335 mum of agitation (Feldstein, 1963a).
1336 1337 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318

1331

characteristic pecularity of the summer season as it is miss-1338 ing in winter. Its appearance changes the form of Sa fun-1339 damentally during the transition from winter to summer at1340 60 В 70 . 1341 The analysis of Sa based on the Q-index of magnetic ac-1342 tivity over all days of the winter and the summer season dur-1343 ing the IGY proves the existence of morning, nighttime and1344 evening agitation maxima at > 60 . Thereby each season1345 has its own specified maxima: during summer in the morning1346 and evening and during winter in the nighttime and morning1347 hours. In the near-polar region one observes one single max-1348 imum during daytime hours. 1349 3.18 Fig. 5 shows the moments of maxima in Sa in polar1350 coordinates versus MLT as observed during magnetically1351 disturbed days of the winter (a) and the summer (b) season.1352 The basic characteristics of Sa are kept during all days in1353 the winter season (Fig. 5a). The diurnal activity variation1354 has two maxima during morning and nighttime hours. The1355 morning maximum appears at later morning hours with in-1356 creasing , while the nighttime maximum appears earlier in1357 the evening hours. The nighttime maximum, however, shifts1358 faster toward earlier evening hours with increasing than1359 for observations, which comprise the full day, and at 801360 this difference in MLT attains four hours. The nighttime1361 maximum appears at all stations up to the highest latitudes.1362 The morning maximum clearly traces from the near-polar re-1363 gion to the latitudes of the auroral zone. The segments of1364 the spiral, that correspond to the nighttime and morning Sa1365 extremes, establish a shape, which reminds an oval, whose1366 symmetry axis traverses the meridian 0214 MLT. 1367 During magnetically disturbed days of the summer season (Fig. 5b), there are three clearly established maxima of Sa activity at many stations. Three maxima appear crys-1368 tal clear at the stations Arctica 2 ( = 77.5 ), Tikhaya Bay ( = 74.7 ), Cape Chelyuskin ( = 71.6 ), Dixon Is-1369 Ё land ( = 68.0 ), Murmansk ( = 64.1 ), Sodankyla ( =1370 www.hist-geo-space-sci.net

4.1 The daily variation of the occurrence frequency of visible auroras is an important aspect of the auroral activity. The Hist. Geo Space Sci.

18

Y. I. Feldstein et al.: Spiral Regularities

Figure 6. The M, N and E spirals determined by Feldstein (1963a) and the auroral oval by Feldstein (1963b) in dipole latitude and time coordinates (from Akasofu, 1968, Fig. 9).

1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387

simplicity of its determination and the informative value of1388 this parameter for comparisons of the auroras with other geo-1389 physical phenomena of the electromagnetic complex stimu-1390 lated the appearance of a multitude of publications in this1391 field of research. Below we consider only a few of them,1392 which generalize the observations and give general impres-1393 sions about acting physical regularities. 1394 4.2 There is a tendency, that intense active discrete auroral1395 forms appear most often 1 hour prior to magnetic midnight1396 at latitudes of the auroral zone, as it was described by Ve-1397 gard (1912) according to the materials of the First Interna-1398 tional Polar Year (18821883). Additional to the nighttime1399 maximum, some high-latitude stations indicated a secondary1400 morning maximum (Tromholt, 1882; Isaev, 1940; Pushkov1401 et al., 1937). The accurate visual observations of auroras1402 performed by Tromholt at the station Godthab ( 73.3 )1403 in Greenland attracted particular attention. The diurnal vari-1404 Hist. Geo Space Sci.

ation of the aurora occurrence frequency is characterized by two maxima, one during late evening, the other during morning hours, and a minimum shortly after midnight. The morning maximum at 0507 MLT was afterwards interpreted as evidence for the existence of a second auroral zone in the polar cap, where the occurrence frequency of auroras and of intense magnetic disturbances maximizes (Nikolsky, 1960b). The maximum during morning hours appears clearly in the daily variations of visual data sets at the stations Kinga Fjord Ё ( 77.8 ) in Canada (Neumayer and Borgen, 1886) and Tikhaya Bay ( 74.7 ) at Franz Joseph Land (Pushkov et al., 1937; Isaev and Pushkov, 1958). This maximum is absent in the records of some other stations. Analysing a large amount of observational data, Hulburt (1931) comes to the conclusion that the probability of the occurrence of a secondary morning maximum is as high as its absence. Full-day visual aurora observations at Greenland are not www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459

19 arises as a part of the spiral in the range 64 < < 71 in the evening sector. The publication of Malville (1959) is the first one, where the existence of a spiral regularity in the spatialtemporal distributions of aurorae is mentioned. Later it was extended both in latitudinal range and by including the distribution of the morning spiral. 4.5 During the IGY, a network of all-sky cameras operated at high latitudes, which produced photos of the sky in 1-min cadence with exposure times of 20 sec or 5 sec. The results of the perusal of the all-sky films were recorded in special tables - the so-called ascaplots, which allowed to localize the aurora on the sky, to exclude intervals of illumination times and of bad meteorological conditions. The ascaplots were assembled according to the instructions of Stoffregen (1959). The full sky was divided along the geomagnetic meridian into three zones: zenith (from 30 elevation north to 30 elevation above the south horizon), North (from 10 to 30 elevation above the north horizon), south (from 10 to 30 elevation above the south horizon). The extent of each stripe amounts to 3 of geomagnetic latitude. The report about the ascaplots of a global network of stations is given in Annals (1962). 4.6 Feldstein (1960) determined the diurnal variation of auroral occurrence frequency in the zenith for 24 stations of the Northern Hemisphere in the latitude range 55 < < 85 . Fig. 7 shows the characteristic shapes of the diurnal variation for various latitudes. In the auroral zone and equatorward of it clearly appears a maximum. It occurs at 03 MLT in Verkhoyansk ( = 56.6 ) and around midnight MLT for 65 В 66 (Murmansk, = 64.1 and Wrangel Island, = 64.7 ). At higher latitudes between 67 < < 78 , there exist two maxima - one in the evening hours, the other in the morning (Arctica I at 72 and Piramida, Spitzbergen at 77 ). With increasing latitude, the morning maximum occurs later in the morning and the evening maximum occurs earlier. Already at 80 , the auroras don't emerge at midnight as at auroral latitudes, but around midday. In the polar cap (Arctica II at 82 , the auroras are rare, but if so, they appear mainly during daytime hours. The maximum around midnight at 73 В 75 appears in visual observations of the full sky during the I st IPY at Greenland and Spitzbergen as intense auroras with a southern azimuth at latitudes of the nighttime auroral zone. The appearance times of the "nighttime" and "morning" maxima in MLT coordinates are distributed along two segment of straight lines. Transferring this relationship into polar coordinates (Fig. 8), then it results in a pattern, which reminds of an oval with minimum distance to the geomagnetic pole on dayside hours and maximum distance at night. The oval consists of two spirals with opposite sense. Fig. 8 represents the first full display of spiral regularities in the auroras of the Northern Hemisphere. A similar double spiral distribution for the Southern Hemisphere was obtained by Feldstein and Solomatina (1961). They are considered as the starting point for introducing the term "auroral oval" in Hist. Geo Space Sci.

possible even during winter season, because of the daylight1460 period from 0717 local time. This complicates the precise1461 determination of the morning maximum and the frequency1462 of its occurrence. Observations by Carlheim-Gyllenskiold1463 (1887) at the station Cap Thordsen, Spitzbergen ( 74.3 ),1464 where practically a full-day monitoring is possible, con-1465 firmed the existence of two maxima - one in the morning1466 and the other during pre-midnight hours MLT. 1467 4.3 Based on hourly observations of the years 195419551468 from 50 polar stations at geomagnetic latitudes 57 < <1469 80 , Feldstein (1958) concluded: 1) the nighttime maximum1470 attains maximal values at auroral zone latitudes ( 65 );1471 2) the shape of diurnal variations changes gradually with in-1472 creasing latitude, and there appears a second maximum in1473 the morning hours at ( > 69 ) with a smaller intensity of1474 the absolute value than the nighttime one. In the near-polar1475 region, the diurnal variation is sparsely developed and dis-1476 crete forms appear likewise rarely both during nighttime and1477 morning hours. 1478 Lassen (1959a,b, 1961) investigated the diurnal variation1479 of auroral occurrences at 5 Greenland stations for the years1480 19481950. It was concluded that the different types of di-1481 urnal distribution can be arranged according to the follow-1482 ing scheme: 1) at latitudes of the main auroral zone, there1483 is one maximum of auroral occurrence around midnight;1484 2) at latitudes from the main to the inner zone of auroras1485 (8 - 10 poleward from the main zone), there appears a sec-1486 ond, smaller morning maximum; 3) within the inner auroral1487 zone, one observes clearly two maxima - one around mid-1488 night and the other at 06 MLT; 4) within the polar cap, the1489 morning maximum dominates about the nighttime one. Sub-1490 sequently Lassen (1963) analyzed the results of visual obser-1491 vations at many high-latitude stations for the interval of the1492 I st IPY (1882/83) and up to the begin of the IGY with respect1493 to the daily variation of the frequency of auroral occurrences1494 and obtained the following results: 1) stations in the latitudi-1495 nal range 64 < < 69 show a maximum near midnight; 2)1496 between 70 < < 76 an evening and a morning maximum1497 of equal size are observed; 3) north of 77 В 78 , the1498 morning maximum is dominating, it occurs late in the morn-1499 ing. Lassen (1963) could not find a regular displacement of1500 the evening and morning maxima. 1501 4.4 Visual observations in the above-mentioned investiga-1502 tions considered auroral forms that are visible over the en-1503 tire sky. Assuming an elevation of 100 km for the lower1504 borders of the auroras, such observations comprise 17 of1505 latitude, from the northern to the southern horizon. Conse-1506 quently, only a limited latitudinal discrimination was pos-1507 sible. By determining the geomagnetic times of the maxi-1508 mum occurrence probability for overhead auroras in 1 wide1509 latitudinal ranges, Malville (1959) demonstrated at the sta-1510 tions Ellsworth ( 67 ) and South Pole ( 77 ), Antarc-1511 tica, that the maximum time occurs systematically earlier1512 in the evening as the pole is approached from the auroral1513 zone. This dependence in polar coordinates versus MLT1514 www.hist-geo-space-sci.net

20

Y. I. Feldstein et al.: Spiral Regularities

Figure 8. Dependence of occurrence times of MLT "night" and

"morning" maxima in diurnal changes of probability of auroras appearance in zenith on geomagnetic latitude in polar coordinates (Feldstein, 1960). 1 night maximum; 2 morning maximum.

1526 1527 1528 1529 1530 1531 1532

Figure 7. Diurnal variations of auroras appearance in zenith (Uni-

1533

versal Time). 1 Verkhoyansk; 2 Murmansk; 3 Wrangel Island; 4 Arctic I (North Pole 6); 5 Piramida (Spitzbergen); 6 Arctic1535 II (North Pole 7). Figures along the axis of ordinates signify the1536 frequency of appearance of auroras at zenith (in percent). The ar-1537 row depicts local midnight, double arrow depicts local geomagnetic1538 midnight (Feldstein, 1960). 1539
1540 1541 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525

1534

solar-terrestrial physics, signifying the region, where ener-1542 getic charged particles penetrate into the upper atmosphere.1543 On the one hand the auroral oval is the region of the most1544 frequent appearance of auroral forms in the zenith at a given1545 geomagnetic latitude, but it signifies also the actual zone of1546 auroral lights for the given moment in time. The oval is1547 asymmetric with respect to the geomagnetic pole due to the1548 deformation of the geomagnetic field by the solar wind. The1549 discussions about the conception of the auroral oval contin-1550 ued in the science community till the 70-ies of the past cen-1551 tury as long as it didn't find its affirmation in the "Sputnik1552 Hist. Geo Space Sci.

era" of active space exploration. 4.7 Lassen (1963) used aurora observations at four Greenland stations during the IGY to determine the diurnal variation of frequencies of occurence in the zenith as well as northward and southward of the zenith. At the stations Godhavn ( = 77.5 ), Kap Tobin ( = 72.7 ) and Julianehab ( = 68.7 ), auroral observations were not possible during 10 daytime hours because of the daylight. A continuous data set was achieved by the use of ascaplots from nine additional high-latitude stations in the Arctic and Antarkic. Four of them allowed auroral observations throughout the whole day. The moments of maxima in the diurnal variation are allocated close to two segments of straight lines. In polar coordinates the maxima rearrange along two spirals of opposite sense. These results coincide with the similar double spiral distribution shown by Feldstein (1960). Such a coincidence of the results of two independent investigations can be expected, considering the accordance in the selected data sets. 4.8 Magnetic disturbances and auroras constitute different aspects of the same electromagnetic complex of phenomena that take place at high altitudes. The magnetic variations, on the one hand, which are recorded at the Earth's surface, represent the integral effect of the corpuscular fluxes and of the current systems in the ionosphere, covering a considerable part of Earth, but the positioning of the auroras points directly to the location of precipitating corpuscular fluxes, that cause the magnetic disturbance. www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563

21
5.2 Discontinuity in the ionospheric current of the auroral belt

The discussion of the interrelationship between the auro-1605 ras of different types and the plasma domains in the mag-1606 netosphere and their structure is beyond the scope of the present review. Detailed considerations about the present1607 understanding of problems in auroral physics is contained1608 in the collective monograph on "Auroral Plasma Physics"1609 (Paschmann et al., 2002). Results about the projection of1610 the boundaries of various aurora types from ionospheric level1611 into the magnetosphere and the relation of their emissions1612 with the magnetospheric plasma domains are sumed up by1613 Feldstein et al. (1994) and Galperin and Feldstein (1996). 1614
1615 1616

1564 1565

5

Interpretations of the spiral regularities in mag1617 netic field variations and auroras
1618

1566 1567

5.1

Magnetic spirals, their number and correlation with 1620 the ionospheric current system

1619

1621 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604

The number of spirals appeared to be disputed, based on the analysis of S D variations of the geomagnetic field as dis-1623 cussed in the previous sections; according to Meek there are1624 two (M and E spirals), while according to HarangBurdo1625 Feldstein ("H.B.F.") we have to deal with three of them1626 (M, N, and E spirals). Referred to Meek (1955), there could1627 be only two spirals, namely one that represents the max-1628 imimum decrease of the horizontal component (H < 0, M1629 spiral) or the other that represents the maximum increase1630 (H > 0, E spiral). A third spiral was not identified, as along1631 it also applies H < 0, but with a smaller amplitude than in1632 the deepest minimum. According to H.B.F., the spirals1633 match both extremal values in H < 0 and a sign change in1634 Z . The number of extremal values in the diurnal variation1635 of H changes in dependence on magnetic latitude from one1636 to three (the M, N, and E spirals). Three characteristic peaks1637 were also found in the diurnal variation of the Sa activity1638 (Chapman and Bartels, 1940; Nikolsky, 1951, 1956; Burdo,1639 1960; Feldstein, 1963a). Also Akasofu (2002) distinguishs1640 1641 three spirals M, N, and E. 1642 The magnetic spirals specify regions, where concentrated isolines of the ionospheric equivalent current system S D de-1643 scribe the magnetic field variations at high-latitudes. Two1644 spirals form the current oval: for Meek (1955) it consists of1645 the westward (M spiral) and eastward directed currents (E1646 spiral), while for H.B.F. only of westward directed (M and1647 N spirals). The current direction in the spirals is not unam-1648 biguously related to the direction of its winding: in the M1649 and N spirals with westward currents, the winding is in dif-1650 ferent directions; in the N and E spirals, which turn in one1651 and the same direction, the direction of the current is op-1652 posite to each other. The winding of the spiral is also not1653 determined by the sign of the charged particles that precipi-1654 tate into the upper atmosphere. Particles of the same sign are1655 precipitating along the M and N spirals (electrons of auroral1656 energy, that are forming the oval), but the winding direction1657 of these spirals differs. 1658
1622

5.2.1 The equivalent current system, which characterizes the intensity and spatial distribution of geomagnetic disturbances at high latitudes, was introduced by Chapman (1935). It was a major paradigm for a few decades. Its characteristic pecularity consists in two concentrated current flows in the auroral zone, which were named by Chapman as electrojets: the eastward electrojet (EE) appears in the evening sector and the westward electrojet (WE) on the morning side. Between these electrojets exist discontinuities during a few hours around midnight and midday. 5.2.2 Harang (1946, 1951) confirmed with his investigations, that the electrojets are located within the auroral zone at 67 from evening to morning hours. This was to a great extent due to the paradigm (ruling at that time) about the positioning of a particle precipitation region at these latitudes. The Harang study presented no drastic revision of the conceptions regarding the structure of the high-latitude current system. As before, the electrojets were associated with the auroral zone. According to Harang a discontinuity between the WE and EE is located at latitudes of the aurora zone but not along the auroral oval (yet unidentified at that time). The morphology of the night time discontinuity between the electrojets was shown in detail in Harang (1946, Fig. 2). It does not come to an agreement with his statement about the discontinuity in the localisation at aurora zone latitudes. The electrojets overlap each other in the dusk sector (in the sense that the EE and WE coexist at the same local time but at different latitudes ) and the westward current is located there poleward of the eastward. Between the electrojets there exists no discontinuity, according to his data, but rather a gap, whose latitude varies between evening and nighttime hours. The westward electrojet is not delimited by the midnight meridian, but extends toward the evening sector, positioned at higher latitudes with respect to the eastward electrojet. Due to this, the electrojets don't stay strictly within the auroral zone at 67 from the evening to the morning side, but comprise also slightly higher latitudes. For the western electrojet during nighttime any discontinuity is absent; the jet rather continues steadily in LT from morning hours over the nighttime to evening hours. 5.2.3 According to Burdo (1960) the latitude of the discontinuity between the electrojets increases from 65 at midnight to 75 at 20 MLT. Those processes that lead to the generation of currents, which are responsible for the observed geomagnetic variations, are both for Harang and Burdo the precipitation of corpuscular fluxes at auroral zone latitudes (the existence of an auroral oval was not known at the time of their publications). Currents at higher latitudes start-up due to the dynamo effect. The existence of dynamo electric fields is also necessary, as they are generated at ionospheric altitudes owing to the neutral wind motion within these layers. The dynamo current system at ionospheric altiHist. Geo Space Sci.

www.hist-geo-space-sci.net

22
1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696 1697 1698 1699 1700 1701 1702 1703 1704 1705 1706 1707

Y. I. Feldstein et al.: Spiral Regularities in the ionosphere and the discontinuity between them are caused by the precipitation of auroral particles and therefore the enhanced ionospheric conductivity at latitudes of the auroral oval as well as by the neutral wind system in the ionosphere. The dynamo action is responsible for the appearance of a 2-D current system at ionospheric heights; FeldsteinAkasofu (FA) no discontinuity, but rather a gap between the eastward and westward electrojets. The morphology of the current gap is analogue to the current discontinuity of HB, but the physical reason for the gap appearance is different. In actuality, there occurs no discontinuity of the electrojet, because in the westward electrojet any discontinuity is missing. The electrojet does not break apart, but continues from nighttime to evening hours, shifting toward higher latitudes. Resulting from that, between the electrojets appears a gap. This gap reflects the boundary of large-scale plasma structures in the nightside magnetosphere. The westward electrojet along the auroral oval is connected with the central plasma sheet in the magnetospheric tail, ґ while the eastward electrojet with the Alfven layer in the inner magnetosphere (Feldstein et al., 2006). The electrojets appear as elements of a 3-D current system, which extends throughout the Earth's ionosphere and magnetosphere (Lyatsky et al., 1974; Akasofu, 2004; Marghitu et al., 2009).
6 Current systems of polar magnetic disturbances

tudes is two-dimensional and controlled by processes in the1713 upper atmosphere close to Earth. 1714 5.2.4 The discovery of the auroral oval brought about a1715 new stage for the interpretation of the discontinuity between1716 the electrojets. This discovery led to a fundamental change in1717 the conception of the planetary morphology concerning the1718 electro-magnetic complex in the near-Earth space (Akasofu1719 and Chapman, 1972). The active auroral forms and hence the most frequent and most intense corpuscular precipita-1720 tions into the upper atmosphere are distributed along the au-1721 roral oval and not along the auroral zone. It was shown, that1722 the westward electrojet runs along the auroral oval from the1723 evening to the morning MLT hours (Feldstein, 1963b; Feld-1724 stein and Zaitzev, 1965b; Akasofu et al., 1965; Akasofu and1725 Meng, 1967a,b,c; Starkov and Feldstein, 1970). 1726 The location and configuration of the auroral oval is inti-1727 mately connected with the asymmetric shape of the Earth's1728 magnetosphere in the Sun-Earth direction (O`Brien, 1963;1729 Frank et al., 1964; Feldstein, 1966), with the plasma struc-1730 ture of the magnetosphere (Vasyliunas, 1970; Frank, 1971;1731 Winningham et al., 1975; Feldstein and Galperin, 1985), and1732 in particular with the plasma sheet in the magnetospheric1733 tail. The eastward electrojet runs in the evening sector at1734 latitudes of diffuse luminosity, adjoining the auroral oval at1735 the equatorial side. This luminosity is caused by the precipi-1736 tation of low energy electrons from the inner magnetosphere1737 with energies up to 10 eV, which are convected from the1738 plasma sheet in the Earth's magnetotail (Nishida, 1966; Feld-1739 stein and Galperin, 1985), and by protons with energies of 10 keV, which are formed in the magnetosphere in the partial ring current of westward direction (Grafe et al., 1997;1740 Liemohn et al., 2001). Its short-circuit by FACs through the ionosphere causes the appearance of the eastward electrojet1741 in the ionosphere. This way the emergence of the auroral oval1742 conception combined the existence of the gap between the1743 eastward and westward electrojet with the magnetospheric1744 1745 plasma structure. 1746 5.2.5 In the course of 30 years (from 1935 till 1965), both the morphology and the physical conception of the night-1747 time discontinuity in the equivalent ionospheric current sys-1748 tem changed successively, Heppner (1972) proposed to call1749 it the Harang discontinuity. This term has been consolidated1750 since that time in the scientific literature as designation of1751 the discontinuity between the electrojets. Note that the exis-1752 tence of the discontinuity was found by Harang in 1946 (see,1753 e.g., Zou et al., 2009). Apparently, when discussing the dis-1754 continuity between the electrojets as a natural phenomenon,1755 debated intensively for a long time, one has to give credit to1756 1757 various cutting-edge research efforts: ChapmanHarang (CH) discontinuity at auroral zone 1759 latitudes during near midnight hours;
1760 1758

1708 1709

1710 1711 1712

HarangBurdo (HB) for increasing latitude of the dis-1761 continuity in the evening sector, it shifts from mid-1762 night to evening hours. The generation of electrojets1763 Hist. Geo Space Sci.

6.1 The description of geomagnetic field variations by means of equivalent currents at ionospheric level and later also in its 3-D extension within the Earth's magnetosphere, has a long history. The sequential progression of perceptions in this field of geophysics was accomplished by the contributions of such "titans" of science like Kristian Birkeland, Sydney ґ Chapman, Hannes Alfven, Takesi Nagata, and many other outstanding scientists. It is beyond the scope of this review to consider this history, we rather have to confine ourself to a short demonstration of the present-day views about the spatial-temporal characteristics of the current system in the near-Earth space (Akasofu, 2002; Feldstein et al., 2006). 6.2. Modelling of equivalent ionospheric currents, based on magnetometer data on the Earths surface, can be realized with different methods. The direction and intensity of the linear current can be found by rotating the horizontal magnetic disturbance vector clockwise by 90 (Kamide and Akasofu, 1974; Baumjohann et al., 1980). Based on data from 70 magnetic observatories, including six meridional chains of magnetometers, Kamide et al. (1982) obtained the spatial distribution of ionospheric current densities for four UT crosssections during a substorm on 19 March 1978. An overview of various techniques used for simulations of the equivalent www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
1764 1765 1766 1767 1768 1769 1770 1771 1772 1773 1774 1775 1776 1777 1778 1779 1780 1781 1782 1783 1784 1785 1786 1787 1788 1789 1790 1791 1792 1793 1794 1795 1796 1797 1798 1799 1800

23 the DP intervals. Such patterns are usually used for generalized representations of perturbed geomagnetic field vector distributions on the Earths surface. Fig. 9a-e shows schematically various current system patterns, that were reported in the literature by different authors. They comprise the classical two-vortex current system (Chapman, 1935; Fukushima, 1953) (Fig. 9a), the single-vortex system with a WE along the auroral oval (AO) and closure currents through the polar cap and at mid-latitudes (Fig. 9b) reported by Feldstein (1963b) and Akasofu et al. (1965), and the two-vortex system with a WE within the boundaries of the AO and with an EE in the evening sector at 65 (Feldstein and Zaitzev, 1965b) (Fig. 9c). It was concluded that during substorms the WE extends to all longitudes along the AO and its intensity decreases from midnight to noon hours. The EE as a separate current system, rather than a return current from the WE, is located at 65 in the evening sector. 6.6. A modification of these patterns (Fig. 9d,e) for substorm and storm intervals was realised by Feldstein et al. (2006). The space-time distribution of currents and their structure at high latitudes during a polar magnetic substorm with an intensity of AL -800 nT is shown in Fig. 9d. The locations of the WE (red strip), EE (green strip) and PE (polar electrojet: dark blue strip) are shown. 6.7. The narrow strip (2 -3 width in latitude) of the PE is located at the latitudes 78 < < 80 under quiet magnetic conditions. These are typical for the ionospheric projection of cusp latitudes around noon. At substorm maximum both PE and cusp shift towards 73 В 75 (Fig. 9d). Both intensity and direction of the PE are controlled by the IMF By component (eastward for By > 0 or westward for By < 0) [Feldstein, 1976 and references therein]. A number of pioneering investigations, devoted to PE and its connection with IMF, are quoted below. Svalgaard (1968) was the first who indicated the existence of two characteristic types of magnetic field variations in the polar regions. The disturbances occur simultaneously in the north and south polar caps and are controlled by the IMF sector polarity. The dependence on the magnetic field variations in the polar regions on the IMF sector polarity was independently obtained by Mansurov (1969). Friis-Christensen et al. (1972) and Sumaruk and Feldstein (1973) discovered independently and simultaneously, that those magnetic field variations are not controlled by the IMF sector polarity, but by the direction of the IMF azimuthal By component. Friis-Christensen and Wilhjelm (1975) and Feldstein et al. (1975) used a regression method for extracting the By -controlled fraction of the high-latitude magnetic field variations and identified the corresponding equivalent DPC( By ) current systems. The detailed description of this type of magnetic field variations and corresponding DPC( By ) current systems can be found in the review (Feldstein, 1976, and references therein). The PE is a characteristic feature of DPC( By ) current system. The PE intensity during the summer season for the IMF By 6 nT is 1.8 в 105 A. Hist. Geo Space Sci.

currents from ground-based magnetometer data was given by1813 Untiedt and Baumjohann (1993). 1814 6.3. Kotikov et al. (1991) developed a practical inversion1815 scheme to infer the fine structure of the auroral electrojet1816 by utilizing a series of linear ionospheric currents (50 al-1817 together) of different intensities located at 100 km altitude.1818 The current distribution was adjusted to fit measurements1819 on the Earth's surface. This numerical method for estima-1820 tion of equivalent ionospheric currents, using magnetic field1821 observations along meridian chains of ground-based vector1822 magnetometers, allows not only for the determination of the1823 latitudinal distribution of the ionospheric current intensity,1824 but also for the separation of the contributions to the ob-1825 served geomagnetic variations of the fields from external1826 (ionospheric and magnetospheric) and internal (induced by1827 telluric currents) sources. Popov and Feldstein (1996) and1828 Popov et al. (2001) suggested a refinement of the Kotikov1829 method by approximating the auroral electrojets with a se-1830 ries of narrow current strips of finite width. The strips with1831 currents of different intensities were distributed along a geo-1832 magnetic meridian at 115 km altitude over the range of lat-1833 itudes covered by the ground magnetometer stations. Both1834 the accuracy of the method and its spatial resolution were1835 1836 considered in detail by Popov et al. (2001). 6.4. Feldstein et al. (2006) applied the refined method of1837 Popov and Feldstein (1996) to some substorms and a mag-1838 netic storm in order to obtain the location and distribution of1839 eastward and westward electrojet intensities as a function of1840 latitude. For that study were used data from three meridian1841 magnetometer chains: the IMAGE chain along the 110 CG1842 (corrected geomagnetic) longitude meridian; the GWC chain1843 along the 40 CG longitude meridian; the CANOPUS chain1844 1845 along the 330 CG longitude meridian. The observed magnetic field variations in the vertical Z1846 (downward) and horizontal H (northward) components at any1847 point "l" along the geomagnetic meridian at the Earth's sur-1848 1849 face due to a single current strip is given by
1850

He xt (l) = Ze xt (l) =
1801 1802 1803 1804 1805 1806 1807 1808 1809 1810 1811 1812

ji arctan 2 ji h2 + ln 2 4 h+

xi + d xi - d - arctan h h 2 ( xi + d) ( xi - d)2

(6)1851
1852 1853

(7)1854
1855

where ji is the current density in the ith strip, d, h, and xi are1856 the half-width, the altitude, and the distance from the obser-1857 vation point to the ground projection of the centre of the ith1858 strip, respectively. Using these expressions for each of the K1859 magnetometers in the chain we obtain 2K equations to deter-1860 mine the current densities in N strips (N=100). If N>2K, the1861 problem is underdetermined and the solution is not unique.1862 In order to constrain the solution the regularization method1863 developed by Tikhonov and Arsenin (1977) is used. 1864 6.5. Latitudinal current density distributions for every1865 chain with a 10-min cadence were the basis for the creation1866 of equivalent ionospheric current patterns in the course of1867 www.hist-geo-space-sci.net

24

Y. I. Feldstein et al.: Spiral Regularities

Figure 9. Schematic view of the electrojet's space-time distribution at ionospheric (Fukushima,1953); b) single-vortex system with westward electrojet along the auroral vortex system with westward electrojet along the auroral oval and eastward electrojet according to (Feldstein and Zaitsev, 1965b); d) the magnetospheric substorm with AL AL -1200 nT , Dst -150 nT (d and e by Feldstein et al., 2006).

altitudes: a) classical current system according to oval according to (Akasofu et al., 1965); c) doublein the evening sector at latitudes of the auroral zone -800 nT ; e) the magnetic storm main phase with

1868 1869 1870

6.8. During substorm intervals the characteristics of the1883 eastward/westward currents as far as their intensities and lo-1884 cations are concerned, can be summarized as follows: 1885
1886

1871 1872 1873 1874 1875 1876 1877 1878 1879 1880

westward currents are most intense after midnight hours1887 at auroral latitudes of 65 < < 70 , and are shifted1888 to cusp latitudes ( 73 ) in the morning and evening1889 sectors. Its latitudinal width is 6 during midnight-1890 dawn hours. As seen in Fig. 9d the WE does not cover all MLT hours (contrary to Figs. 9b,c) during the sub-1891 storm, but is seen only within evening-night-morning1892 hours. During pre-noon and late afternoon hours the1893 WE adjoins the cusp, i.e., it is magnetically mapped to1894 the magnetopause; 1895
1896

hours, become most intense in the evening MLT, and reach auroral latitudes at nighttime hours. The EE comprises the evening sector from early evening to midnight hours and is located at lower latitudes when approaching midnight. During midnight the EE latitude is the same as that of the WE equatorward edge. Both the EE width and its intensity reach maximum values during dusk hours; eastward currents are located just equatorward of the westward currents for evening hours where currents in opposite directions overlap in latitude. During evening hours, when overlapping occurs, the maximum eastward current intensity is higher than the westward current intensity at the same local time; the EE shown in Fig. 9d is not a closure current for www.hist-geo-space-sci.net

1881 1882

eastward currents in the evening sector start from cusp latitudes ( 73 ) during the early afternoon MLT1897 Hist. Geo Space Sci.

Y. I. Feldstein et al.: Spiral Regularities
1898 1899 1900 1901 1902 1903

25 strong poleward-directed electric fields driving fast plasma convection bursts at subauroral latitudes in the evening local time sector and called them polarization jets (PJ). The new name SAPS (Sub-Auroral Polarization Stream) is generally accepted nowadays for this phenomenon (Foster and Burke, 2002). In Fig. 9e the location of the bursts are indicated by a yellow line, the longitudinal prolongation of which is adapted from Foster and Vo (2002). Galperin (2002) proposed a model that explains the formation of PJ or SAPS during magnetic substorms and storms in the evening-tonighttime sector of the discontinuity region between the auroral electrojets. A detailed consideration of the interrelation between auroral luminosity, FAC and PE during magnetic disturbances in the dayside sector was given by Sandholt et al. (2004). 6.12. To show the structure of magnetospheric plasma domains and the 3D current systems, a cross-section of the Earth's magnetosphere is displayed in Fig. 10. It is shown there with cuts in the midday-midnight meridional plane and in the equatorial plane, using ground-based observations of different auroral luminescence types and auroral precipitation (Galperin and Feldstein, 1991). The majority of plasma domains seen in the figure are directly related to largescale current systems in the magnetosphere. Such FAC systems exist permanently in the magnetosphere (Maltsev and Ostapenko, 2004). 6.13. The 3D structure of currents in the near-Earth space is enclosed by the magnetopause. The currents screening the magnetic field of the inner magnetosphere from penetrating into the solar wind are located on the magnetopause. These eastward Chapman-Ferraro (CF) currents screen the dipole field. The magnetopause screening currents for the ring current (RC) fields are in the same direction, but their intensity is an order of magnitude weaker. The tail current (TC) in the central plasma sheet (CPS) is in the dawn-dusk direction. The closure of the TC is attributed to currents on the magnetopause which exist not only on the night side, as well established, but on the day side as well. In Fig. 10 the TC in the equatorial plane of the magnetosphere is indicated by two vectors. At midnight one of them is located in the innermost part of the current sheet, the other along its boundary. Their continuation on the magnetopause can be seen. However, the first remains in the tail and the second reaches the day side of the magnetopause where the directions of the CF and TC are opposite as seen in Fig. 10. Since CF currents are always more intense than TC closure currents the resulting current on the day side is always eastward. 6.14. The field aligned current flowing into and out of the ionosphere in the vicinity of the PE are located on the cusp surface. In Fig. 10 PE-FACs are indicated by two green lines (not vectors) along the magnetic field. The PE-FAC direction is not shown since it is controlled by the IMF By component: under By > 0 ( By < 0) the current flows into (out of) the ionosphere along the cusp inner surface and out of (into) it along its outer surface. The ionospheric closure of the inflowing Hist. Geo Space Sci.

the WE at higher latitudes, contrary to Fig. 9b, and is1948 not located in the auroral zone ( 65 ), contrary to1949 Fig. 9c. During early evening hours the EE adjoins the1950 ionospheric projection of the cusp (CU), i.e. the EE is1951 magnetically mapped to the magnetopause during this1952 MLT interval. 1953 It is clear from Fig. 9d that the WE is located along M and1955 N spiral segments, but it does not cover the MLT hours near1956 midday. Such peculiarity in the WE structure is determined1957 by the character of its connection with the central plasma1958 sheet in the magnetospheric tail. The EE is located immedi-1959 ately equatorwards of the WE, along the E-spiral. 1960 6.9. In Fig. 9d the location of the projection of the plasma-1961 pause (PP) to ionospheric altitudes is indicated by the quasi-1962 circle (blue). There is a latitudinal gap (of a few degrees)1963 between the PP and EE that decreases when reaching night1964 hours. During substorms this latitude gap is comparable to1965 the EE width. 1966 6.10. Fig. 9e shows the distribution of equivalent currents1967 and their structure during the storm main phase with activity1968 indices of AL -1200 nT and Dst = -150 nT . The follow-1969 ing characteristics of the electrojet dynamics are apparent: 1970 during evening hours both the EE and WE shift equa-1971 torward and the intensity of the electrojets increases; 1972
1973 1954

1904 1905 1906 1907 1908 1909 1910 1911 1912 1913 1914 1915 1916 1917 1918 1919

1920 1921

1922 1923 1924 1925 1926

the eastward and westward current intensities in the1974 evening sector imply that the EE cannot be the conse-1975 quence of the WE closing through lower latitudes; it is1976 likely that these electrojets are signatures of different1977 geophysical phenomena; 1978 at the peak of the storm main boundary of the CU (and hence < 67 and the width is 2 В 3 ing and early evening hours the join the PE; phase the equatorward 1980 the PE) shifts to 65 < 1981 . During the late morn1982 westward currents ad1983 1984 1979

1927 1928 1929 1930 1931

1932 1933 1934

at the main phase maximum the EE during near-noon1985 hours adjoins the CU at 65 , the WE is absent in1986 1987 the day-time sector;
1988

1935 1936 1937 1938 1939 1940

the WE asymmetry with regard to the noon-midnight1989 meridian is valid for storm intervals (similarly to the1990 substorm intervals). However, the asymmetry pattern1991 changes essentially; in the night sector the WE pole-1992 ward boundary is located at higher latitudes than the PE1993 in the noon sector; 1994 the WE in the midnight and early morning sectors has a1995 current density of 1.6 k A/km, and the integrated cur-1996 1997 rent is 2 В 3 M A.
1998

1941 1942 1943

1944 1945 1946 1947

6.11. During the main storm phase the bursts of the ion1999 drift velocity attain as much as 4 km/s at subauroral lati-2000 tudes (outside of the plasmapause) in the region of the iono-2001 spheric trough. Galperin et al. (1974) was the first to report2002 www.hist-geo-space-sci.net

26

Y. I. Feldstein et al.: Spiral Regularities

Figure 10. The 3-D system of electric currents in the magnetosphere during magnetic disturbances (Feldstein et al., 2006). Descriptions of

the current system and the magnetospheric plasma structures are given in the text.

2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026

and outflowing PE-FAC is by Pedersen current. Its direction2027 in the ionosphere is identical with the direction of the so-2028 lar wind electric field component V в By , i.e. under By > 02029 ( By < 0) the electric field in the cusp is poleward (equator-2030 ward) at ionospheric altitudes. In the cusp the Hall current2031 in the form of the PE, spreading in the ionosphere (out of the2032 cusp) is generated by this electric field. The PE is eastward2033 (westward) under By > 0 ( By < 0). 2034 6.15. The Region 1 FAC in the dusk sector is usually be-2035 lieved to be mapped magnetically from the ionosphere to the2036 low latitude boundary layer (LLBL), i.e., to the periphery of2037 the magnetosphere, in the vicinity of its boundary with the2038 solar wind. Such a pattern is valid for Region 1 FAC during2039 day-time hours only and is shown by a current arrow, resting2040 against the LLBL. During the dusk to pre-midnight hours,2041 where Region 1 FAC is located at AO latitudes, FACs inflow2042 to the CPS, i.e., into the deep magnetosphere. In Fig. 10 the2043 second current arrow of the Region 1 FAC crosses the equa-2044 torial cross-section of the magnetosphere in the dusk sec-2045 tor of the CPS behind the TC vector that depicts the current2046 along the TC boundary. 2047 6.16. The Region 2 FAC flows into the ionosphere from2048 ґ the Alfven layer periphery, where the partial ring current2049 (PRC) is located. In Fig. 10, the Region 2 FACs are indi-2050 Hist. Geo Space Sci.

cated by three vectors for day, dusk, and nighttime hours. It is generally believed, that the Region 2 FACs are located in the inner magnetosphere and part of a single current system with the EE and PRC. As seen in Fig. 10, the Region 2 FACs are flowing into the ionosphere during evening hours and are located in the magnetosphere at L 4. In the early afternoon sector, where the EE adjoins the PE, the Region 2 FAC in the equatorial plane of the magnetosphere is near the LLBL. 6.17. The auroral electrojets present a continuation of the FACs at ionospheric altitudes. Eastward/westward electrojets are connected with magnetospheric plasma domains (the ґ Alfven layer / the CPS) via FACs. In known models of the EE, the Pedersen current in the ionosphere along with the Region 2 FACs constitute BostrЬm's Case I current system Ё (Bostrom, 1964). At the same time, the Region 2 and Region 1 FACs in the evening sector constitute BostrЬms Case 2 current system. The northward electric field between these FACs leads to Hall currents along the EE. Hence, the EE has a joining effect of Pedersen and Hall currents in the ionosphere. 6.18. As mentioned above, a characteristic morphological feature of auroral electrojets exists in the evening sector, namely in their overlapping. This is due to an additional Region 3 FAC current flowing into the ionosphere from the www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
2051 2052 2053 2054 2055 2056 2057 2058 2059 2060 2061 2062

27 7. The spiral distributions are not related to the dynamo action at ionospheric heights. The existence of the spirals and their distribution is determined by large-scale structures of plasma domains at the night side of Earth. The M- and N-spirals are connected with the inner (nearEarth) part of the central plasma sheet in the magnetospheric tail. The E-spiral relates, however, to the region between the central plasma sheet and the plasmapause, where the partial ring current is situated in the evening sector and where the low-energy plasma drifts in sunward direction out of the central plasma sheet. 8. There is no discontinuity between the westward and eastward auroral electrojet at a fixed latitude around midnight, but rather a gap, the latitude of which increases smoothly from nighttime to evening hours.
Acknowledgements. The authors express their gratitude to V. G. Vorobjev, and V. L. Zverev for reading the manuscript and for their critical comments. Support was obtained by RFBR grant No. 11-050036 to IZMIRAN. Work at GFZ Potsdam (M.F.) was supported by Deutsche Forschungsgemeinschaft (DFG).

PSBL during intense magnetic disturbances. A triple FAC2099 structure arises. In addition to the Region 1 FAC and Re-2100 gion 2 FAC, the Region 3 FAC (flowing into the ionosphere)2101 located on the poleward boundary of the auroral precipita-2102 tion region appears. As a result, the necessary prerequi-2103 sites for overlapping auroral electrojets are created. A south-2104 ward electric field, favourable for the appearance of west-2105 ward ionospheric Hall currents near the PC boundary, ap-2106 pears between the Region 3 FAC and Region 1 FAC. As a2107 result a spatial overlapping of electrojets takes place. Also2108 worth noting is that the WE represents the resultant effect of2109 Pedersen and Hall currents in the ionosphere.
2110

2063

7

Conclusions

2111 2112

2064 2065 2066 2067 2068 2069 2070 2071

Spiral distributions (spirals) were used in the first stage of in-2113 vestigations for explaning the positioning of charged particle 2114 fluxes, precipitating into the upper atmosphere. Later they2115 found their place (and became an intrinsic part) in global2116 spatial-temporal patterns of some high-latitude geophysical2117 phenomena. Below we list the basic peculiarities, that char-2118 acterize the spirals in magnetic disturbances and auroras. With respect to geomagnetic disturbances:
2119

References
Agy, V.: Spiral patterns in geophysics, J. Atmos. Terr. Phys., 19, 136140, 1960. Akasofu, S.-I.: Polar and magnetospheric substorms, D. Reidel, Dordrecht, The Netherland, 316p., 1968. Akasofu, S.-I.: Exploring the secrets of the aurora, Kluwer Academic Publishers, Dordrecht/Boston/London, 235p., 2002. Akasofu, S.-I.: Several 'controversial' issues on substorms, Space Sci. Rev., 113, 140, 2004. Akasofu, S.-I. and Chapman, S.: SolarTerrestrial Physics, Oxford University Press, London, 1972. Akasofu, S.-I. and Meng, C.-I.: Intense negative bays inside the auroral zoneI. The evening sector, J. Atmos. Terr. Phys., 29, 965970, 1967a. Akasofu, S.-I. and Meng, C.-I.: Auroral activity in the evening sector, J. Atmos. Terr. Phys., 29, 10151019, 1967b. Akasofu, S.-I. and Meng, C.-I.: Polar magnetic substorm in the evening sector, J. Atmos. Terr. Phys., 29, 10271035, 1967c. Akasofu, S.-I., Chapman, S., and Meng, C.-I.: The polar electrojet, J. Atmos. Terr. Phys., 27, 12751305, 1965. Alexeev, I. I., Belenkaya, E. S., Kalegaev, V. V., Feldstein, Y. I., and Grafe, A.: Magnetic storms and magnetotail currents, J. Geophys. Res., 101, 77377747, 1996. ґ Alfven, H.: On the electric field theory of magnetic field and auroras, Tellus, 7, 5064, 1955. Allen, J. H. and Kroehl, H. W.: Spatial and temporal distributions of magnetic effects of auroral electrojets as derived from AE indices, J. Geophys. Res., 80, 36673677, 1975. Allen, J. H., Abston, C. C., and Morris, L. D.: Auroral electrojet magnetic activity indices AE(10) for 1966, Tech. Rep. Report UAG-37, NOAA, Boulder, Colo., 1974. Allen, J. H., Abston, C. C., and Morris, L. D.: Auroral electrojet magnetic activity indices AE(10) for 1971, Tech. Rep. Report UAG-39, NOAA, Boulder, Colo., 1975.

2072 2073 2074 2075 2076

1. There are three types of spirals at high latitudes independent of seasons, named as M (morning), N (night-2120 time), and E (evening) spirals. The first two (M and2121 N) appear distinctly and visibly during all seasons; the2122 2123 third one (E) is characteristic for the summer season. 2. Maximum intensities of ionospheric currents are ob-2125 served along these spirals - the westward electrojet2126 2127 along the M- and N-spirals, and the eastward electro2128 jet along the E-spiral. 2129

2124

2077 2078 2079 2080

2081 2082 2083

3. Spirals with westward ionospheric currents are situated 2131 along the auroral oval from the late morning, over the2132 nighttime till the early evening hours. 2133 4. The windings of the spirals in clockwise or anticlock2135 wise direction are not determined by the current flow2136 direction in the spiral nor by the sign of the charged2137 particles, which precipitate into the upper atmosphere. 2138 With respect to auroras:
2139 2140 2134

2130

2084 2085 2086 2087

2088

2089 2090 2091 2092 2093

5. The visual observation of auroras is confined at high lat-2141 2142 itudes primarily to the winter season for the most part 2143 of the day. Only M- and N-spirals are identified, which2144 constitute the auroral oval. Most often there appear dis-2145 crete auroral forms in the zenith along the auroral oval. 2146 6. Discrete auroral forms along the M- and N-spirals are2148 due to precipitating electrons with energies up to a few2149 keV . In the vicinity of the E-spiral, one observes diffuse2150 auroras, which are due both to low-energy electrons and2151 2152 protons with energies of 10 keV and higher. www.hist-geo-space-sci.net
2147

2094 2095 2096 2097 2098

Hist. Geo Space Sci.

28
2153 2154 2155 2156 2157 2158 2159 2160 2161 2162 2163 2164 2165 2166 2167 2168 2169 2170 2171 2172 2173 2174 2175 2176 2177 2178 2179 2180 2181 2182 2183 2184 2185 2186 2187 2188 2189 2190 2191 2192 2193 2194 2195 2196 2197 2198 2199 2200 2201 2202 2203 2204 2205 2206 2207 2208 2209 2210 2211

Y. I. Feldstein et al.: Spiral Regularities
imuths, in: IGY Transactions "Investigations of the aurorae", vol. N4, pp. 6177, Publ. House Academy Sci. USSR, 1960. Feldstein, Y. I.: Pattern of the magnetic activity in the high latitudes of the northern hemisphere, in: Geomagnetism, vol. N5, Publ. House Academy Sci. USSR, 64p., 1963a. Feldstein, Y. I.: The morphology of auroras and geomagnetism, in: IGY Transactions "Auroras and Airglow", vol. N10, pp. 121 125, Publ. House Academy Sci. USSR, 1963b. Feldstein, Y. I.: Peculiarities in the auroral distribution and magnetic disturbance distribution in high latitudes caused by the asymmetrical form of the magnetosphere, Planet. Space Sci., 14, 121130, 1966. Feldstein, Y. I.: Magnetic field variations in the polar region during magnetically quiet periods and interplanetary magnetic fields, Space Sci. Rev., 18, 777861, 1976. Feldstein, Y. I. and Galperin, Y. I.: The auroral luminosity structure in the high-latitude upper atmosphere, its dynamics and relationship to the large-scale structure of the Earth's magnetosphere, Rev. Geophys., 23, 217275, 1985. Feldstein, Y. I. and Solomatina, E. K.: Polar auroras in southern hemisphere, Geomagn. Aeron. (Engl. translation), 1, 534539, 1961. Feldstein, Y. I. and Zaitzev, A. N.: Disturbed solar-diurnal variations at high latitudes during the IGY period, Geomagn. Aeron. (Engl. translation), 5, 477486, 1965a. Feldstein, Y. I. and Zaitzev, A. N.: The current system of SDvariations in high latitudes for the winter season during the IGY, Geomagn. Aeron. (Engl. translation), 5, 11231128, 1965b. Feldstein, Y. I., Sumaruk, P., and Shevnina, N. F.: To the diagnostics of the azimuthal component of the interplanetary magnetic field, C. R. Acad. Sci. USSR, 222, 833836, 1975. Feldstein, Y. I., Elphinstone, R. D., Hern, D. J., Murphree, J. S., and Cogger, L. L.: Mapping of the statistical auroral distribution into the magnetosphere, Can. J. Phys., 72, 266269, 1994. Feldstein, Y. I., Popov, V. A., Cumnock, J. A., Prigancova, A., Blomberg, L. G., Kozyra, J. U., Tsurutani, B. T., Gromova, L. I., and Levitin, A. E.: Auroral electrojets and boundaries of plasma domains in the magnetosphere during magnetically disturbed intervals, Ann. Geophys., 24, 22432276, 2006. Ё Ё Forster, M., Rentz, S., Kohler, W., Liu, H., and Haaland, S. E.: IMF dependence of high-latitude thermospheric wind pattern derived from CHAMP cross-track measurements, Ann. Geophys., 26, 15811595, 2008. Ё Forster, M., Haaland, S. E., and Doornbos, E.: Thermospheric vorticity at high geomagnetic latitudes from CHAMP data and its IMF dependence, Ann. Geophys., 29, 181186, 2011. Foster, J. C. and Burke, W. J.: SAPS: A new categorization for subauroral electric fields, EOS, 83, 393394, 2002. Foster, J. C. and Vo, H. B.: Average characteristics and activity dependence of the subauroral polarization stream, J. Geophys. Res., 107, 1475, doi:10.1029/2002JA009409, 2002. Frank, L. A.: Relationship of the plasma sheet, ring current, trapping boundary and plasmapause near the magnetic equator at local midnight, J. Geophys. Res., 76, 22652275, 1971. Frank, L. A., Van Allen, J. A., and Craven, J. D.: Large diurnal variation of geomagnatically trapped and precipitated electrons observed at low altitudes, J. Geophys. Res., 69, 31553167, 1964. Friis-Christensen, E. and Wilhjelm, J.: Polar currents for different directions of the interplanetary magnetic field in the Y - Z plane,

Annals: Annals of the International Geophysical Year (IGY): As-2212 caplots: half-hourly all-sky camera plots from 114 stations for2213 the period 1957-1958, vol. 20, parts 1 and 2, Perganon Press,2214 Oxford, International Council of Scientific Unions. Special Com-2215 2216 mittee for the IGY edn., 1962. Arykov, A. and Maltsev, Y. P.: Contribution of various sources to2217 the geomagnetic storm field, Geomagn. Aeron. (Engl. transla-2218 2219 tion), 33, 6774, 1993. Arykov, A. A., Belova, E. G., Gvozdevsky, B. B., Maltsev, Y. P., and2220 Safargaleev, V. V.: Geomagnetic storm resulting from enhance-2221 ment of the high-latitude magnetic flux, Geomagn. Aeron. (Engl.2222 2223 translation), 36, 307313, 1996. Ё Bartels, J. and Fukushima, N.: Ein Q-index fur erdmagnetische2224 Ё Ё Activitat in viertelstundlichen Intervalen, Abhandlungen Akad.2225 Ё 2226 Wiss., Gottingen, N2, 37p., 1956. Bartels, J., Heck, N. H., and Johnston, H. F.: The three-hour-2227 range index measuring geomagnetic activity, Terr. Magn. Atmos.2228 2229 Elect., 44, 411454, 1939. Baumjohann, W., Untiedt, J., and Greenwald, R.: Joint two-2230 dimensional observations of ground magnetic and ionospheric2231 electric fields associated with auroral zone currents. 1. Three-2232 dimensional current flows associated with a substorm-intensified2233 2234 eastward electrojet, J. Geophys. Res., 83, 19631978, 1980. Benkova, N. P.: Diurnal variation of the magnetic activity (in Rus-2235 sian), Transactions of the Institute of Terrestrial Magnetism,2236 2237 3(13), 1540, 1948. Birkeland, K.: The Norwegian Aurorae Polaris Expedition of 19022238 -1903: On the cause of magnetic storms and the origin of terres-2239 trial magnetism, vol. 1, H. Aschehoug & Co., Christiania, First2240 2241 Section, 1908. Birkeland, K.: The Norwegian Aurorae Polaris Expedition of 19022242 -1903: On the cause of magnetic storms and the origin of terres-2243 trial magnetism, vol. 2, H. Aschehoug & Co., Christiania, Sec-2244 2245 ond Section, 1913. Bobrov, M.: Types of irregular geomagnetic disturbances and2246 mechanism of the influence of solar corpuscular streams on the2247 2248 outer atmosphere, Astron. Zhurnal, 37, 410424, 1960. Ё Bostrom, R.: A model of the auroral electrojets, J. Geophys. Res.,2249 2250 69, 49834994, 1964. Bruche, E.: Some new theoretical and experimental results on the2251 2252 aurorae polaris, Terr. Magn. Atmos. Elect., 36, 4162, 1931. Burdo, O. A.: Relation of regular and irregular variations of the2253 geomagnetic field, high latitudes, in: Transactions of the Arctic2254 and Antarctic Institute, vol. 223, pp. 2145, AARI, Leningrad,2255 2256 USSR, 1960. Carlheim-Gyllenskiold, G.: Exploration internationale des re-2257 gions polaires,18821883, Observations faites au Cap Thordsen,2258 Spitzbergen, par lexpedition suedoise, T. I, II, Stockholm, 19,2259 2260 1887. Chapman, S.: The electric current systems of magnetic storms, Terr.2261 2262 Magn. Atmos. Elect., 40, 349370, 1935. Chapman, S. and Bartels, J.: Geomagnetism, 1, Geomagnetic and2263 2264 related phenomena, Clarendon Press, Oxford, 542p., 1940. Davis, T. N. and Sugiura, M.: Auroral electrojet activity index AE2265 and its universal time variations, J. Geophys. Res., 71, 785801,2266 2267 1966. Feldstein, Y. I.: Geographic distribution of aurorae in the westward2268 sector of Soviet Arctic, Problems of Arctic, N4, 4549, 1958. 2269 Feldstein, Y. I.: Geographic distribution of aurorae and arcs az-2270

Hist. Geo Space Sci.

www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
2271 2272 2273 2274 2275 2276 2277 2278 2279 2280 2281 2282 2283 2284 2285 2286 2287 2288 2289 2290 2291 2292 2293 2294 2295 2296 2297 2298 2299 2300 2301 2302 2303 2304 2305 2306 2307 2308 2309 2310 2311 2312 2313 2314 2315 2316 2317 2318 2319 2320 2321 2322 2323 2324 2325 2326 2327 2328 2329 2330 J. Geophys. Res., 80, 12481260, 1975. Friis-Christensen, E., Laasen, K., Wilhjelm, J., Wilcox, J. M., Gon-2331 zales, W., and Colburn, D. S.: Critical Component of the In-2332 terplanetary Magnetic Field Responsible for Large Geomagnetic2333 Effects in the Polar Cap, J. Geophys. Res., 77, 33713376, 1972.2334 Fukushima, N.: Polar magnetic storms and geomagnetic bays, vol.2335 8, Part V of Series II, pp. 293412, Geophysical Institute, Faculty2336 2337 of Science, Tokyo University, Tokyo, Japan, 1953. Fukushima, N. and Oguti, T.: Polar magnetic storms and geomag-2338 netic bays, Appendix 1, A theory of SD field, in: Rep. Ionosph.2339 Res. in Japan, vol. 7, No. 4, pp. 137145, Geophysical Institute,2340 2341 Faculty of Science, Tokyo University, Tokyo, Japan, 1953. Galperin, Y. and Feldstein, Y. I.: Auroral luminosity and its rela-2342 tionship to magnetospharic plasma domains, pp. 207222, Cam-2343 bridge University Press, ed. by C.-I. Meng, M. J. Rycroft, and L.2344 2345 A. Frank, 1991. Galperin, Y. I.: Polarization Jet: characteristics and a model, Ann.2346 2347 Geophys., 20, 391404, 2002. Galperin, Y. I. and Feldstein, Y. I.: Mapping of the precipitation2348 regions to the plasma sheet, J. Geomag. Geoelectr., 48, 857875,2349 2350 1996. Galperin, Y. I., Ponomarev, V. N., and Zosimova, A. G.: Plasma2351 convection in the polar ionosphere, Ann. Geophys., 30, 17,2352 2353 1974. Gonzalez, W. D., Joselyn, J. A., Kamide, Y., Kroehl, H. W., Ros-2354 toker, G., Tsurutani, B. T., and Vasyliunas, V. M.: What is a2355 geomagnetic storm?, J. Geophys. Res., 99, 57715792, 1994. 2356 Grafe, A., Bespalov, P. A., Trakhtengerts, V. Y., and Demekhov,2357 A. G.: Afternoon mid-latitude current system and low- latitude2358 geomagnetic field asymmetry during geomagnetic storm, Ann.2359 2360 Geophys., 15, 15371547, 1997. Gustafsson, G.: A revised corrected geomagnetic coordinate sys-2361 2362 tem, Arkiv for Geofysik, 5, 773, 1970. Harang, L.: The mean field of disturbance of polar geomagnetic2363 2364 storms, Terr. Magn. Atmos. Elect., 51, 353380, 1946. Harang, L.: The Aurorae, vol. 1, Chapman & Hall, Ltd., London,2365 2366 The International Astrophysical Series edn., 163p., 1951. Heppner, J.: The Harang discontinuity in auroral belt ionospheric2367 2368 current, Geofysiske Publikasjoner, 29, 105120, 1972. Hope, E. R.: Low-latitude and high-latitude geomagnetic agitation,2369 2370 J. Geophys. Res., 66, 747776, 1961. Hulburt, E. O.: On the diurnal variation of the aurora polaris, Terr.2371 2372 Magn. Atmos. Elect., 36, 2340, 1931. Hultqvist, B.: The geomagnetic field lines in higher approxima-2373 2374 tions, Arkiv Geofysik, 3, 6377, 1958. Isaev, S. I.: About diurnal periodicity of auroras inside the zone of2375 2376 maximal frequency, Problems of the Arctic, 9, 4145, 1940. Isaev, S. I. and Pushkov, N. V.: The Aurorae, Publ. House Academy2377 2378 Sci. USSR, 112p., 1958. Kamide, Y. and Akasofu, S.-I.: Latitudinal cross section of the auro-2379 ral electrojet and its relation to the interplanetary magnetic field2380 2381 polarity, J. Geophys. Res., 79, 37553771, 1974. Kamide, Y., Ahn, B.-H., Akasofu, S.-I., Baumjohann, W., Friis-2382 Christensen, E., Kroehl, H. W., Maurer, H., Richmond, A. D.,2383 Rostoker, G., Spiro, R. W., Walker, J. R., and Zaitzev, A. N.:2384 Global distribution of ionospheric and field-aligned currents dur-2385 ing substorms as determined from six meridian chains of magne-2386 tometers: initial results, J. Geophys. Res., 87, 82288240, 1982.2387 Kamide, Y., Baumjohann, W., Daglis, I. A., Gonzalez, W. D.,2388

29
Grande, M., Joselyn, J. A., McPherron, R. L., Phillips, J. L., Reeves, E. G. D., Rostoker, G., Sharma, A. S., Singer, H. J., Tsurutani, B. T., and Vasyliunas, V. M.: Current understanding of magnetic storms: Storm-substorm relationships, J. Geophys. Res., 103, 17,70517,728, 1998. Killeen, T. L. and Roble, R. G.: Thermosphere Dynamics: Contributions from the First 5 Years of the Dynamics Explorer Program, Rev. Geophys., 26, 329367, 1988. Korotin, A. B. and Pudovkin, M. I.: About possible mechanism of magnetic disturbances origin, in: Transactions "Spectral, electro-photometrical and radiolocation investigations of auroras", vol. N6, pp. 3742, Publ. House Academy Sci. USSR, 1961. Kotikov, A. L., Bolotynskaya, B. D., Gizler, V. A., Troshichev, O. A., Pashin, A. B., and Tagirov, V. R.: Structure of auroral zone phenomena from the data of meridional chains of stations: magnetic disturbances in the night-time auroral zone and auroras, J. Atmos. Terr. Phys., 53, 265277, 1991. Lassen, K.: Existence of an inner auroral zone, Nature, 184, 1375 1377, 1959a. Lassen, K.: Local aurorae in the morning hours at Godhavn, in: Meddelelser, vol. N 24, Det Danske Meteor. Inst., Charlottenlund, 1959b. Lassen, K.: Day-time aurorae observed at Godhavn 1954-1956, in: Meddelelser, vol. Medd. N 15, Det Danske Meteor. Inst., Charlottenlund, 1961. Lassen, K.: Geographical distribution and temporal variations of polar aurorae, in: Meddelelser, vol. 16, Det Danske Meteor. Inst. Meddelelser, Charlottenlund, 1963. Liemohn, M. W., Kozyra, J. U., Thomsen, M. F., Roeder, J. L., Lu, G., Borovsky, J. E., and Cayton, T. E.: Dominent role of the asymmetric ring current in producing the stormtime Dst*, J. Geophys. Res., 106, 10,88310,904, 2001. Ё Ё Luhr, H., Rentz, S., Ritter, P., Liu, H., and Hausler, K.: Average thermospheric wind pattern over the polar regions, as observed by CHAMP, Ann. Geophys., 25, 10931101, 2007. Lyatsky, W., Maltsev, Y. P., and Leontyev, S. V.: Three-dimensional current system in the different phases of a substorm, Planet. Space Sci., 22, 12311247, 1974. Maltsev, Y. P.: Points of controversy in the study of magnetic storms, Space Sci. Rev., 110, 227267, 2004. Maltsev, Y. P. and Ostapenko, A. A.: Field-aligned currents in the ionosphere and magnetosphere, in: Proceedings of the International Symposium in memory of Yuri Galperin on 'Auroral Phenomena and Solar-Terrestrial Relations', edited by Zelenyi, L. M., Geller, M. A., and Allen, J. H., pp. 147151, IKI, Moscow, 4-7 February 2003, 2004. Malville, J. M.: Antarctic auroral observations, Ellsworth Station, 1957, J. Geophys. Res., 64, 13891393, 1959. Mansurov, S. M.: New evidence of the relationship between magnetic field in space and on the Earth, Geomagn. Aeron. (Engl. translation), 9, 768773, 1969. Mansurov, S. M. and Mishin, V. M.: Diurnal behaviour of magnetic activity in the polar region, in: Transactions "Earths electromagnetic field disturbances", pp. 4552, USSR Academy of Science Publishing House, 1960. Marghitu, O., Karlsson, T., Klecker, B., Haerendel, G., and McFadden, J.: Auroral arcs and oval electrodynamics in the Harang region, J. Geophys. Res., 114, A03214, doi:1029/2008JA013630,

www.hist-geo-space-sci.net

Hist. Geo Space Sci.

30
2389 2390 2391 2392 2393 2394 2395 2396 2397 2398 2399 2400 2401 2402 2403 2404 2405 2406 2407 2408 2409 2410 2411 2412 2413 2414 2415 2416 2417 2418 2419 2420 2421 2422 2423 2424 2425 2426 2427 2428 2429 2430 2431 2432 2433 2434 2435 2436 2437 2438 2439 2440 2441 2442 2443 2444 2445 2446 2447 2448 2009. Mayaud, P.: Activite magnetique dans les regions polaires, Rap-2449 ports scientifiques des expeditions polaires francaises, Annales2450 ґ 2451 de geophysique, 12, 84101, 1956. Mayaud, P. N.: Calcut preliminaire dindices Km, Kn, Ks, ou Am,2452 An et As, measures de lctivite magnetique a l chelle et dans les2453 ґ hemispheres Nord et Sud, Annales de geophysique, 23, 585617,2454 2455 1967. Mayaud, P. N.: The aa indices: a 100 year series characterizing the2456 2457 magnetic activity, J. Geophys. Res., 77, 68706874, 1972. Meek, J. H.: The location and shape of the auroral zone, J. Atmos.2458 2459 Terr. Phys., 6, 313321, 1955. Nagata, T. and Fukushima, N.: Constitution of polar magnetic2460 2461 storms, Rep. Ionos. Res. Japan, 6, 8596, 1952. Nagata, T., Fukushima, N., and Sugiura, M.: Electro-dynamical be-2462 haviour of the ionosphere region viewed from geomagnetic vari-2463 2464 ations, J. Geomag. Geoelectr., 2, 3544, 1950. Ё Neumayer, G. and Borgen, C.: Die Internationale Polarforschung,2465 1882 -1883, Die Beobachtungs-Ergebnisse der Deutschen Sta-2466 2467 tionen, vol. 1, Kingua Fjord, Asher, Berlin, 1886. Nikolsky, A. P.: Dual laws of the course of magnetic disturbances2468 and the nature of mean regular variations, Terr. Magn. Atmos.2469 2470 Elect., 52, 147173, 1947. Nikolsky, A. P.: Magnetic disturbances in the Arctic, vol. 36, AARI,2471 2472 Leningrad, USSR, 135p., 1951. Nikolsky, A. P.: On the second zone of magnetic disturbances inten-2473 sity increased in near-polar region, in: Transactions of the Arc-2474 tic and Antarctic Institute, vol. 83, pp. 583, AARI, Leningrad,2475 2476 USSR, 1956. Nikolsky, A. P.: On planetary distribution of magneto-ionospheric2477 disturbances, in: Transactions of the Arctic and Antarctic Insti-2478 2479 tute, vol. 223, pp. 320, AARI, Leningrad, USSR, 1960a. Nikolsky, A. P.: On the issue of geographical distribution of auroras2480 in Arctic, in: Transactions "Investigation of aurorae", vol. 4, pp.2481 1419, USSR Academy of Science Publishing House, Moscow,2482 2483 1960b. Nishida, A.: Formation of the plasmapause, or magnetospheric2484 plasma knee, by the combined action of magnetospheric con-2485 vection and plasma escape from the tail, J. Geophys. Res., 71,2486 2487 56695679, 1966. O`Brien, B. J.: A large diurnal variation of the geomagnatically2488 2489 trapped radiation, J. Geophys. Res., 68, 989997, 1963. Paschmann, G., Haaland, S., and Treumann, R., eds.: Auro-2490 ral Plasma Physics, vol. 15 of Space Sciences Series of ISSI2491 (SSSI), Kluwer Academic Publishers, Dordrecht, Boston, Lon-2492 don, reprinted from Space Science Reviews Volume 103, No.2493 2494 1-4, 2002. Popov, V. and Feldstein, Y. I.: About a new interpretation of "Ha-2495 rang discontinuity", Geomagn. Aeron. (Engl. translation), 36,2496 2497 4351, 1996. Popov, V. A., Papitashvili, V. O., and Watermann, J. F.: Modeling2498 of equivalent ionospheric currents from meridian magnetometer2499 2500 chain data, Earth, Planets and Space, 53, 129137, 2001. Pudovkin, M. I.: Origins of bay disturbances, in: Series Geophys.,2501 2502 vol. No. 3, pp. 484489, Izvest. Acad. Nauk, 1960. Pudovkin, M. I.: Magnetic bay disturbance (in Russian), in: Trans-2503 actions "Investigations on geophysical phenomena of the elec-2504 tromagnetic complex at high latitudes", pp. 322, Publ. House2505 2506 Nauka, USSR, 1964.

Y. I. Feldstein et al.: Spiral Regularities
Pudovkin, M. I.: Diurnal variation of the ionospheric drift velocity in the auroral zone and the SD variations (in Russian), in: Transactions "Investigations of auroras, geomagnetic disturbances and the ionosphere at high latitudes", pp. 310, Publ. House Nauka, USSR, 1965a. Pudovkin, M. I.: Some remarks on the dynamo theory of geomagnetic disturbances (in Russian), in: Transactions "High-latitude Investigations in the field of geomagnetism and aeronomy", pp. 1926, Publ. House Nauka, USSR, 1965b. Pudovkin, M. I.: Morphology and nature of polar magnetic storms (in Russian), Ph.D. thesis, Leningrad University, dSc Dissertation, 1968. Pudovkin, M. I. and Evlashin, L. S.: Spatial relationship of auroras with electric currents in the ionosphere (in Russian), Geomagn. Aeron. (Engl. translation), 2, 669673, 1962. Pudovkin, M. I. and Korotin, A. B.: Magnetic bay disturbances and their relation to ionospheric drifts (in Russian), Geomagn. Aeron. (Engl. translation), 1, 408412, 1961. Pushkov, N. V., Brunkovskaya, N. S., and Abramova, N. V.: Comparison of magnetic and auroral activities according to observations at Tikhaya Bay in 19321933 (in Russian), Transactions "Earth magnetism and Electricity", 4, 7175, 1937. ` Richmond, A. D., Lathuillere, C., and Vennerstroem, S.: Winds in the high-latitude lower thermosphere: Dependence on the interplanetary magnetic field, J. Geophys. Res., 108, 1066, doi: 10.1029/2002JA009493, 2003. Sandholt, P. E., Farrugia, C. J., and Denig, W. F.: Detailed dayside auroral morphology as a function of local time for southern IMF orientation: implications for solar wind-magnetosphere coupling, Ann. Geophys., 22, 35373560, 2004. Stagg, J. M.: The diurnal variation of magnetic disturbances in high latitudes, Proc. R. Soc., A, 152, 277298, 1935. Starkov, G. V. and Feldstein, Y. I.: Instantaneous distribution of auroras and polar geomagnetic disturbances, in: Transactions "Auroras", vol. N19, pp. 3241, USSR Academy of Science Publishing House, 1970. Steward, B.: Terrestrial magnetism, vol. 16, pp. 159184, Gordon and Breach, New York, 9th ed., 1882. Stoffregen, W.: Instructions for scaling auroral ascaplots and auroral indices, in: Medd., Repot N5, Uppsala Ionosfarobservatorium, 1959. StЬrmer, C.: Corpuscular theory of the aurorae borealis, Terr. Magn. Atmos. Elect., 22, 2334, 1917a. StЬrmer, C.: Corpuscular theory of the aurorae borealis. (concluded), Terr. Magn. Atmos. Elect., 22, 97112, 1917b. StЬrmer, C.: The Polar Aurora, Clarendon Press, 402p., 1955. Sugiura, M.: Hourly values of equatorial Dst for the IGY, in: Annals of the International Geophysical Year (I.G.Y.), edited by of Scientific Unions. Special Committee for the IGY, I. C., vol. 35, pp. 9???, Perganon Press, Oxford, 1964. Sumaruk, P. V. and Feldstein, Y. I.: Sector structure of the interplanetary magnetic field and the magnetic variations in the near-polar region (in Russian), Kosm. Issled., 11, 155160, 1973. Svalgaard, L.: Sector structure of the interplanetary magnetic field and daily variation of the geomagnetic field at high latitudes, in: Geophys. Papers, R-6, Danish Meteorol. Inst., 1968. Tikhonov, A. N. and Arsenin, V. I.: Solutions of ill-posed problems, Scripta series in mathematics, Halsted Press, New York, 258p., 1977.

Hist. Geo Space Sci.

www.hist-geo-space-sci.net

Y. I. Feldstein et al.: Spiral Regularities
2507 2508 2509 2510 2511 2512 2513 2514 2515 2516 2517 2518 2519 2520 2521 2522 2523 2524 2525 2526 2527 2528 2529 2530 2531 2532 2533 2534 2535 2536 2537 2538 2539 2540 2541

31

Tromholt, S.: Sur les periodes de laurore boreale, Dan. Meteor. Inst. Copenhagen, 1882. Troshichev, O. A., Andrezen, V. G., Vennerstrom, V. G., and FriisChristensen, E.: Magnetic activity in the polar cap A new index, Planet. Space Sci., 36, 10951102, 1988. Troshichev, O. A., Janzhura, A. S., and Stauning, P.: Unified PCN and PCS indices: method of calculation, physical sense and dependence on the IMF azimuthal and northward components, J. Geophys. Res., 111, A05208, doi:10.1029/2005JA011402, 2006. Untiedt, J. and Baumjohann, W.: Studies of polar current systems using the IMS Scandinavian magnetometer array, Space Sci. Rev., 63, 245390, 1993. Vasyliunas, V. M.: Low energy particle fluxes in the geomagnetic tail, pp. 2547, Gordon and Breach, New York, 1970. Vegard, L.: On the properties of the rays producing aurorae borealis, Philos. Mag., 23, 211237, 1912. Whitham, K. and Loomer, E. J.: A comparison of magnetic disturbance at Resolute Bay and Baker Lake, Tellus, 8, 276278, 1956. Whitham, K., Loomer, E. J., and Niblett, E. R.: The latitudinal distribution of magnetic activity in Canada, J. Geophys. Res., 65, 39613974, 1960. Winningham, J. D., Yasuhara, F., Akasofu, S.-I., and Heikkila, W. J.: The latitudinal morphology of 10-eV to 10-keV electron fluxes during magnetically quiet and disturbed times in the 21000300 MLT sector, J. Geophys. Res., 80, 31483171, 1975. Zhigalova, N. N. and Ol, A. I.: High-latitude bay disturbances of the geomagnetic field (in Russian), Problems of Arctic and Antarctic, 15, 6973, 1964. Zou, S., Lyons, L. R., Wang, C.-P., Boudouridis, A., Ruohoniemi, J. M., Anderson, P. C., Dyson, P. L., and Devlin, J. C.: On the coupling between the Harang reversal evolution and substorm dynamics: A synthesis of SuperDARN, DMSP, and IMAGE observations, J. Geophys. Res., 114, A01205, doi:10.1029/ 2008JA013449, 2009.

www.hist-geo-space-sci.net

Hist. Geo Space Sci.