Dynamics of the electrojet during intense magnetic disturbances

Hall current variations in different time sectors during six magnetic storms of the summer seasons in 2003 and 2005 are examined in detail: three storms in the day-night meridional sector and three storms in the dawn-dusk sector. We investigate the sequence of the phenomena, their structure, positions and the density of the polar (PE) and the auroral (AE) Hall electrojets using scalar magnetic field measurements obtained from the CHAMP satellite in accordance with the 5 study of Ritter et al. (2004a). Particular attention is devoted to the spatial-temporal behaviour of the PE at ionospheric altitudes during daytime hours both under geomagnetically quiet and under magnetic storm conditions. We analyze the correlations of the PE and AE with various activity indices like SYM/H and ASYM/H, that stand for large-scale current systems in the magnetosphere, AL for ionospheric currents, and the IndN coupling function for the state of the solar wind. We 10 obtain regression relations of the magnetic latitude MLat and the electrojet current density I with those indices and with the interplanetary By and Bz magnetic field components. For the geomagnetic storms during summer seasons investigated here, we obtain the following typical characteristics for the electrojets’ dynamics: 1. The PE appears at magnetic latitudes (MLat) and local times (MLT) of the cusp position. 15 2. This occurs in the daytime sector at MLat∼73◦-80◦ with a westward or an eastward direction, depending on the orientation of the IMF By component. Changes of current flow direction in the PE can occur repeatedly during the storm, but only due to changes of the IMF By orientation. 3. The current density in the PE increases with the intensity of the IMF By component from 20 1 Ann. Geophys. Discuss., https://doi.org/10.5194/angeo-2018-31 Manuscript under review for journal Ann. Geophys. Discussion started: 25 April 2018 c © Author(s) 2018. CC BY 4.0 License.


Introduction
An intense study of the polar electrojet (PE) at the high-latitude daytime ionosphere was initiated by the works of Svalgaard (1968) and Mansurov (1969).They had demonstrated that its characteristic magnetic field variation depends on the sector structure of the interplanetary magnetic field (IMF) 45 that is much alike the average magnetic field of the solar photosphere.The IMF lines up in a spiral structure near the ecliptic plane with IMF Bx>0 and By<0 (toward the sun) or Bx<0 and By>0 (away from the sun), where one prevalent direction is kept usually for several days.During summer the sector structure is accompanied by intense variations of the geomagnetic Z component within the polar cap at Φ ∼ 86 • and of the H component at Φ ∼ 78 • .An increase or decrease of the magnetic 50 field components with respect to the quiet time is determined by the activity level and the IMF sector The variations of the magnetic field at the Earth's surface at high latitudes, which were derived with the method of regression analysis, allowed to determine the IMF By control of the spatial-90 temporal distributions of the electric field potential at ionospheric altitudes as well as the ionospheric and field-aligned currents (FACs) (Friis-Christensen et al., 1985;Feldstein and Levitin, 1986).The electric field potential for an inhomogeneous ionospheric conductivity is obtained by solving a second-order partial differential equation.Friis-Christensen et al. (1985) used magnetic observations of the summer seasons in 1972 and 1973, while Feldstein and Levitin (1986) obtained it for 95 summer 1968.The potential differences at cusp latitudes in the daytime sector are ∼20 kV for IMF By∼ ±6 nT.Leontyev and Lyatsky (1974) postulated a penetration of the solar wind electric field into the magnetosphere at daytime cusp latitudes.This electric field is generated by the potential difference between the northern and southern boundaries of the magnetotail.Under the assumption of high 100 conductivities along the magnetic field lines, the electric field exists only at open field lines, which have their footprints in the polar caps, and will be short-circuited along closed field lines.The model allowed to estimate the effectivity of the solar wind electric field penetration into the magnetosphere to ∼10%.Olsen (1996) used MAGSAT magnetic field data in a height range of 350<h<550 km to determine 105 the strength and location of the auroral electrojets at 115 km altitude.He showed for the first time the possibility to estimate the horizontal ionospheric currents from scalar magnetic measurements only.The ionospheric currents were modelled by hundreds of infinite linear currents perpendicular to the orbital plane of the spacecraft with discretisation intervals of 111 km.The problem of ionospheric current estimation is underdetermined and its solution is not unique.In order to constrain the 110 solution, a regularization method is used.The compilation of modelled and measured variations of the magnetic field along the satellite orbit on December 04,1979,17:00 UT, demonstrated the good agreement for the field-aligned component, but a significant discrepancy for the field-perpendicular one.The discrepancy is mainly caused by magnetic fields of the FACs.The integral amplitude of the ionospheric currents during the interval November 28 till December 10, 1979, yielded a correlation 115 of r = 0.88 with the AE-index.
The IMF By orientation influences not only the PE, but also the movements of the auroral forms at cusp latitudes (Sandholt et al., 2002).Simultaneously with permanently poleward moving discrete auroral forms at the equatorward boundary of the cusp, which are controlled by the IMF Bz component, there exist east-west moving auroral forms.This azimuthal movement is controlled 120 by IMF By, such that for By>0 the discrete forms move westward and for By<0 eastward.The movement of the auroral forms is in opposite direction to the PE current flow direction.This can be expected, because the discrete auroral forms and the channels of enhanced ionospheric conductivity are both due to precipitating electrons into the upper atmosphere.A detailed consideration of the interrelation between auroral luminosity, auroral particle precipitation, and the PE during magnetic disturbances was given by Sandholt et al. (2004).As shown there, the strong convection channel is located on the dawn side of the polar cap for IMF By>0, and on the dusk side for By<0 conditions.
The electron precipitation in the regime of the convection channel in the morning sector consists of a band (∼500 km) of structured precipitation.The PE is located on the high-latitude boundary of the structured luminosity region in the vicinity of the strong flow channel of magnetospheric convection 130 close to the bright auroral arc.For By>0, this channel is located in the morning sector on the polar cap boundary with FAC out of the ionosphere, and FAC into the ionosphere equatorward of the polar cap boundary.Ritter et al. (2004b) investigated variations in the location and density of the auroral electrojets, which were independently determined both from ground-based (IMAGE magnetometer network) 135 and satellite (CHAMP) measurements.For the estimation of the Hall current from CHAMP data, a current model consisting of a series of 160 current lines were placed at an altitude of 110 km and separated by 1 • in latitude.The magnetic field of the line currents were related to the current strength I according to the Biot-Savart law.The density of each of the 160 line currents were derived from an inversion of the observed field residuals using a least-square fitting approach.They determined 140 the geomagnetic latitude and current densities of the eastward (EE) and westward electrojets (WE) in the evening, nighttime, and morning sectors.Two-or one-dimensional ionospheric Hall current systems were independently determined from variations of the horizontal magnetic field, measured by the IMAGE ground-based magnetometer network.Comparisons of satellite with ground-based measurements of ionospheric currents at au-145 roral latitudes have been done for satellite passages during magnetic storms as, e.g., that of 5-6 November 2001, during substorms, and according to statistical data.The ratio of the current densities from IMAGE and CHAMP was provided for a latitudinal range of 60 • -77 • as well as mean values of the current densities, variations of the correlation coefficients, and coefficients of the regression equations.The ratio of the current densities and of the correlation coefficient as determined 150 from currents above and below the ionosphere, is close to unity.Such a correspondence between the results of two different model approaches constitute the base for the following statements: a) the estimation of the position and density of the auroral electrojets can be carried out with observations above the ionospheric current layer by means of low-Earth orbiting (LEO) satellites; b) the currents can be estimated from scalar magnetic field measurements; c) the result of the calculations 155 are the parameters of the Hall current at an altitude of ∼110 km (where the maximum value of the ionospheric Hall conductivity occurs).This method of Hall current estimation from satellites was proposed for the first time by Olsen (1996).Its detailed justification has been validated quantitatively by ground-based observations by Ritter et al. (2004b).
Based on magnetometer data of the IMAGE and EISCAT networks, Feldstein et al. (1997) showed  Ritter et al. (2004a).

175
In this study we investigate not only the auroral electrojet, but also the polar electrojet characteristics during six intense magnetic summer storms.In this introduction section we have briefly described the historic progression of method used to determine the ionospheric currents from space observations with LEO satellites and from ground magnetic field data.In section 2 we present an overview of the CHAMP data used as well as the indices, which characterize the electro-magnetic 180 conditions in the near-Earth space during the geomagnetic storms under study.Section 3 provides a short description of the method for the determination of the Hall currents from CHAMP scalar magnetic records.In section 4 we consider the latitudinal variation of the density and position of the electrojets during different phases of the magnetic storm on 29-30 May 2003.Particular attention is drawn to the polar electrojet (PE).The discussion of the control of the current direction in the 185 electrojets, its density and latitudinal position by various indices, which characterize the disturbance level and the effectivity of the interaction of the interplanetary medium on the magnetospheric processes follows in the subsections 5.1 and 5.2 for the polar electrojet (PE) and the auroral electrojets (AE), respectively.The Conclusion's section 6 summarizes the main results of the study with respect to the Hall current variations during the various storm phases (subsection 6.1), the polar electrojet 190 (6.2), and the auroral electrojet within four different time sectors (6.3).

Data
The CHAllenging Minisatellite Payload (CHAMP) spacecraft (Reigber et al., 2002)   obtained with the Overhauser Magnetometer (OVM) at the boom tip with a resolution of 0.1 nT.In order to isolate the magnetic effect of ionospheric currents in the satellite data, the contributions from all other sources have been removed from the scalar field readings as described in the study of 200 Ritter et al. (2004a).
The CHAMP orbital intervals during various storm periods used for this study are listed in Table 1.
The quantity, locations, and intensity of the peaks along the latitudinal current density distribution varies over the course of the storm development.For the description of the storm development, we utilise various solar and geomagnetic indices.

205
First, we employ the auroral electrojet index (AE), which is derived from geomagnetic variations in the horizontal component observed at selected (10-13) observatories along the auroral zone in the Northern Hemisphere (http://wdc.kugi.kyoto-u.ac.jp/aedir/index.html).The lower envelope (AL) of the superposed plots of all the data from these stations as functions of UT is used in this study.Finally, Newell et al. (2007) proposed a new solar wind coupling function Index N (IndN) for the 215 correlation analysis in the solar-terrestrial physics: Here, v describes the solar wind speed or, more precisely, the transport velocity of IMF field lines that approach the magnetopause, B T is the magnitude of the IMF, and the IMF clock angle θ c is defined by θ c = arctan(B y /B z ).This function describes best the interaction between the solar wind and 3 Method

225
The Hall current flows at high latitudes are derived from CHAMP scalar magnetometer records along the satellite orbits according to the method that was proposed by Ritter et al. (2004b).These calculations make use of a current model consisting of a series of infinite current stripes at an altitude of 110 km, the magnetic field of which corresponds to the measured values.The model does not take into account the contributions from field-aligned and Pedersen currents, measured at CHAMP alti-  The density of ionisation in the near noon hours at latitudes of 75 • < Φ <80 • decreases from summer to winter season by about an order of magnitude (Feldstein et al., 1975a).The PE current density amounts during winter to ∼0.1 A/m, which makes it difficult to be measured adequately by magnetometers onboard of satellites.Because of that we investigate in this study summer storms only: three storms with CHAMP orbits in the midday-midnight plane and three in the dawn-dusk 240 plane (listed in Table 1).Five of the storms are described in the appendix / supplementary material to this paper.
The storm phases are identified in this study according to the SYM/H index, which describes together with the ASYM/H index the large-scale variations of the geomagnetic field with a 1-min cadence.In essence, they represent mean values of the magnetic field deviation from the quiet 245 time level for a longitudinally distributed chain of mid-latitude observatories.The current at the magnetopause (DCF) and the currents within the magnetosphere as the ring current (DR), which is symmetric with respect to the geomagnetic axis, and the tail current (DT), which closes via the dayside magnetopause, determine the intensity and the development of the magnetic disturbances.
These currents carry the main contributions to the SYM/H values during magnetic storm intervals 250 (Maltsev, 2004;Alexeev et al., 1996).
The density of the ring current varies with longitude.This variability is identified as the partial ring current (PRC), denoted by the ASYM/H index, and determined as the difference between the maximum and minimum magnetic field values from a longitudinal chain of mid-latitude observatories.The PRC current system is a 3-D one that is confined to a limited azimuthal range of the ring 255 current in the magnetosphere, FACs between the magnetosphere and ionosphere at the border of the PRC, and an EE in the evening sector at ionospheric heights.The latitude position is controlled by MLT and toward the near noon sector it shifts toward the ionospheric footpoint of the cusp region (Feldstein et al., 2006).The FAC of the PRC maps from ionospheric heights to the near-cusp magnetopause region, where the Hall currents flow, which are controlled by the IMF By component.The orbit of the CHAMP satellite in its ascending branch was on the dayside (∼14-16 MLT), while its descending branch was in the nighttime sector (∼02-04 MLT).
The beginning of the main magnetic storm phase was identified during orbit 16233 at 22:24 UT In the nighttime sector, the current is westward directed (WE) in the majority of cases at auroral latitudes.Fig. 2 shows the direction, MLat, and density of the Hall currents along the orbit for dayside   MLT range around midday imply that the current observed is the PE.In this case its orientation is controlled by the IMF B y component and for an eastward PE the B y component should be positive (Friis-Christensen et al., 1972;Sumaruk and Feldstein, 1973;Feldstein, 1976).Indeed, according to Fig. 1, B y = 9 nT during the period of this orbit.During the two subsequent orbits, the Hall current changes its direction to westward at MLat∼80 • .If this westward current prove to be the exists at latitudes of the auroral oval and some weakly spreaded eastward currents.Two orbits constitute an exception -one prior to (16231) and another during the beginning of the main storm phase ( 16233).These orbits pertain to the period of intense substorms.Within the polar cap up to the geomagnetic pole, there exist quite intense (up to 0.9 A/m) eastward currents.

335
These currents might contain irregularities, which are caused by the appearance of a peak of eastward currents in the latitudinal profile.The monotonicity of the eastward current variations within the polar cap during most orbits provides some reason to assume, that these currents result from the closure of an intense WE current, which occurs at latitudes of the auroral oval in the nighttime sector.Below we are going to analyse the current directions, their densities, and MLat positions for various MLT sectors with regard to solar wind parameters and some indices of the planetary magnetic activity (SYM/H, ASYM/H, AL, IndN).We use activity indices, which characterize the occurrence 405 and dynamics of large-scale plasma domains in Earth's magnetosphere that are responsible for the existence of concrete variations of the geomagnetic field at Earth's surface.

Polar electrojets
It is well-known from geomagnetic activity researches that the intense magnetic disturbances at the high-latitude projection of the magnetospheric cusp are not related to the occurrence and dynamics   This is assumed to be due to the interaction between the magnetosphere and the supersonic plasma 445 flow (solar wind) with a "frozen-in" magnetic field (IMF).The PE currents are generated at magnetic latitudes of the cusp due to reconnection processes between the IMF and the geomagnetic field (Jørgensen et al., 1972;Wilhjelm and Friis-Christensen, 1971).The reconnection of magnetic fields brings about a north-south electric field and an east-west Hall current at cusp latitudes in the ionosphere.A possible generation mechanism for the PE current system has been suggested by Leontyev and Lyatsky (1974), including the structure of the PE current system.Leontyev and Lyatsky (1974) postulate the penetration of the electric field E z = V x × B y , where V x is the solar wind velocity past the magnetosphere.This electric field will cause a potential difference U between the northern and southern boundaries of the magnetotail: where D m is the size of the magnetosphere along the z-axis.On the assumption of high conductivity Here, we present estimations of the efficiency of the electric field penetration into the cusp region for two concrete orbits of the CHAMP satellite over the daytime sector.The solar wind velocity at this time was V x =640.7 km/s, the plasma density in the solar wind n p =21.4 cm −3 , and IMF B y =-6 nT.The electric field in the solar wind amounts to E z = V x × B y =3.8 mV/m.The dynamic plasma pressure at the subsolar point P sw = 0.88P dyn sw =12.9 nPa, while the distance of the subsolar point at the magnetopause is ∼7 R E .The potential difference 475 along the z-axis between the northern and southern tail boundaries amounts to U m =338 kV, and in the cusp of one hemisphere hence U m =169 kV.According to the CHAMP data, the mean density of the current is ∼0.3 A/m over ∼8.5 • , the integral current in the cusp therefore ∼295 kA, and the potential difference in the cusp ∼42 kV.The efficiency of the electric field penetration from the solar wind into the ionosphere is thus about ∼25%.

480
During orbit 29018 of 24 Aug 2005, 16:42 UT, with V x =630.5 km/s, n p =18.6 cm −3 , and IMF B y =-20.9 nT results in E z =12.6 mV/m, P sw =10.8 nPa, and distance of the subsolar point at the magnetopause is ∼7.3 R E .The potential difference in the cusp of one hemisphere can thus be estimated to U m =585 kV.According to the CHAMP data, the mean density of the current is ∼0.4 A/m over ∼7.5 • , the integral current in the cusp therefore ∼348 kA, and the potential 485 difference in the cusp ∼49 kV.The efficiency of the electric field penetration from the solar wind into the ionosphere is thus about ∼9%.Therefore, for a quite variable electric field voltage applied to the magnetosphere from the magnetized solar wind flow (from 585 kV to 169 kV), the efficiency of its penetration to the ionosphere varies between 9% and 25%.
The SYM/H index varied during the intervals of CHAMP overflights above the polar electrojets 490 considered in this study from -10 nT to -170 nT.As shown in Fig. 3c, there is no correlation between SYM/H and the MLat positions (r = −0.13)nor the PE current density (r = −0.01).The absence of any correlation is as expected, because the current systems of the DCF, DR, and DT are located completely within the magnetosphere.In case of absent FACs, they cannot serve as sources for Hall currents in the ionosphere that is responsible for the existence of PE.Fig. 3e shows the correlation between the MLAT position and the density of the PE with the AL index of geomagnetic activity.This index describes the reduction of the horizontal component of the geomagnetic field at auroral latitudes on Earth's surface during disturbances with respect to quiettime conditions, using a longitudinal chain of magnetic observatories.The AL index appears to be a sensitive tracer for processes in the central plasma sheet of the magnetospheric tail.These processes 505 are created by injection of energetic particles, their accumulation, and the dissipation of their energy during storm times and is accompanied by changes of the boundary positions of large-scale plasma structures.They appear to have relatively small influence on the density and MLat position of the PE (with r = −0.38 and r = 0.46, respectively).However, there is a distinctive tendency for the shift of the PE from ∼78 • to ∼74 • with an increase of the AL index up to -900 nT.

510
As shown in Fig. 3f, there is a correlation of IndN with MLat in the daytime sector (r=-0.52).This is obvious, because both components By and Bz are included in the definition of IndN.With increasing IndN, the latitude of the current decreases.The correlation coefficient of IndN with the Hall current density (I) is r=0.3.

Auroral electrojets 515
The most intense Hall currents at ionospheric heights, which are responsible for the electrojets, are located at auroral latitudes in the nighttime hours.It is even there, where intense auroras occur most often in the zenith (Chapman and Bartels, 1940;Harang, 1951).These electrojets were named auroral electrojets (AE).A huge number of studies has been published on their morphology, their connections with the solar wind parameters and the plasma domains in Earth's magnetosphere, as 520 well as on their internal processes.The AE are present during all hours of the day.The number of electrojets, their internal current structure, and the interconnection with the individual magnetospheric plasma domains depends both on the activity level and on the MLT position of the observation (Feldstein et al., 2006).Therefore, we consider below the results of the Hall current observations of the CHAMP satellite separately for each of the following four MLT sectors: daytime, nighttime, 525 evening, and morning hours.Table 3 provides the correlation coefficients r, the coefficients A and B of the linear regression equations of the type X = A + B * Y , which were obtained by the least-squares method with correlation coefficients r > 0.46, and the mean-square deviation σ from the regression line.

Conclusions
In this paper we investigated the density and spatial-temporal distribution (versus magnetic latitude MLat and MLT) of Hall currents at high latitudes.The currents were determined from measurements of total magnetic field data, sampled by magnetometers on board the CHAMP satellite at ionospheric 600 altitudes of ∼430 km (Ritter et al., 2004a).In this study we used these current estimations to explore the dynamics of the polar and auroral electrojets during a selection of six magnetic storms (see Table 1).We identified their distinctive features and the correlations with activity indices that are usually used to characterize large-scale current systems in the magnetosphere.The main findings obtained are listed below.The PE currents and their MLat positions are characterized by the following peculiarities: -The PE appears at magnetic latitudes and local times of the cusp.

635
-The direction of the current in the PE is controlled by the IMF By (azimuthal) component: for By>0 the current is eastward, for By<0 the current is westward directed.
-The current density in the PE increases with the intensity of the IMF By component from I∼0.4 A/m for By∼0 nT up to I∼1.0 A/m for By∼23 nT.
-The MLat position of the PE does not depend on the orientation and the strength of the IMF 640 By component.
-Assuming that the penetration of the solar wind electric field into the cusp causes the generation of the PE, we estimate the efficiency of such a penetration.Based on two CHAMP orbits across the dayside sector of the high-latitude ionosphere, we estimate the potential difference over the cusp with 169 kV and 585 kV.The efficiency of the electric field penetration into the 645 cusp would then amount to 25% and 9%, respectively.
-There is no connection between MLat and the current density I in the PE with the magnetospheric ring current DR (index SYM/H).-The currents in the central plasma sheet appear to have a weak influence on the current density and the MLat position of the cusp.
-We realized a correlation between MLat and the IndN solar wind coupling function.
-Significant correlation coefficients with r > 0.49 are obtained only for MLat, correlations with the current density I are, however, very small for all indices.
-The EE shifts ∼6 • more equatorward compared to the WE.-The current density in the EE is larger than in the WE.The WE is missing for a certain confined MLT interval of the evening sector, in case of an existing EE (Feldstein et al., 2006).
That means that the EE in the evening sector cannot be a low-latitude branch-off from the WE 685 current.
-The current density in the WE correlates with the activity indices ASYM/H, AL, and IndN, with a maximum correlation coefficient of r ∼0.76 for the IndN.-With increasing activity, the WE shifts equatorward.The lowest observed MLat for the WE is ∼58 • .

705
The existing morphological differences between the EE and the WE probably testify differences of the physical sources, which are responsible for the existence of the EE and WE.One possible option is the interpretation of the EE in the evening and daytime sectors as continuation of the magnetospheric partial ring current (PRC) through the ionosphere via a system of FACs.The WE, which of the WE in the evening hours, which is connected via FACs with the central plasma sheet in the magnetospheric tail in the nighttime and morning sectors.with a maximum density of ∼0.9 A/m.Their direction is controlled by the IMF By component, i.e., they are in accordance with the PE.
For the midnight sector: -As a rule, the Hall currents are westward directed during nighttime.In the concrete observa- In the daytime sector, a EE exist during the creation phase at MLat ∼67 . Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.
160that the electrojets shift equatorward during the main phase of strong magnetic storms.For DST∼ −300 nT, the EE in the evening and the WE in the nighttime and early morning hours shifts to Ann.Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.∼ 54 • -∼ 55 • .Feldstein and Galperin (1999) studied the correlation between EE and WE with the structure of plasma precipitations of 30 eV-30 keV according to DMSP F08, F10, and F11 satellite observations during the magnetic storms of 10-11 May 1992, 05-07 February 1994, and 165 21-22 February 1994.The EE displaces in the region of diffuse aurora, equatorward of the discrete auroral forms, and projects along magnetic field lines into the inner magnetosphere between the plasmasphere and the central plasma sheet of the magnetospheric tail.The WE is located at the auroral oval and projects along magnetic field lines toward the central plasma sheet in the tail.Wang et al. (2008) made use of the Hall current estimations for the intense magnetic storms of 170 31 March to 01 April 2001 and 17-21 April 2002 to investigate the position and current densities of auroral electrojets (WE and EE) as well as the relations of the electrojets to the Dst index and the IMF Bz component.The currents were determined from scalar magnetic field measurements of the CHAMP satellite (orbit in the meridional plane of 15-03 MLT and 16-04 MLT) according to the method that was proposed by was launched on 15 July 2000 into a circular, near-polar orbit with an inclination of 87.3 • .From its initial orbital height at ∼ 460 km, it has decayed to ∼400 km in 2003 and ∼350 km after 5 years.The orbital plane 195 precesses to earlier local times at a rate of about one hour per 11 days so that the orbit covers all local times within about 131 days.The data used in this study are scalar magnetic field measurements

220
the magnetosphere over a wide variety of magnetospheric activity.IndN has a strong correlation with other indices that characterize both the plasma and the IMF in the solar wind as well as the Ann.Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.processes in the magnetosphere.By means of a statistical study of the electrojet characteristics, the new function IndN was used together with the classical indices SYM/H, ASYM/H, and AL.
230tudes.The comparison with ground-based geomagnetic variations of the horizontal component that considers only the contributions from the ionospheric Hall current field, because the contributions from the field-aligned and the Pedersen currents cancel there each other, showed the applicability with high reliability of the modelling assumptions byRitter et al. (2004b) even for the estimation of the Hall currents. 235

(
left column) and nightside (right column) sectors as obtained from scalar measurements of the geomagnetic variations corresponding to the modelled current variations ofRitter et al. (2004b).It is obvious that the quantity, locations, and intensity of the peaks along the latitudinal current density 285 distribution vary in the course of the storm development.4.1 Observations related to SYM/H variations The latitudinal variation of the position and density of the EE is shown in Fig. 2.During the orbits 16229 and 16230, one singular peak of eastward current was observed, which occurred at MLat = 63.4 • and MLat = 64.0• with intensities of 1.0 A/m and 0.6 A/m, respectively.This means 290 that the EE peak current diminishes in density with increasing disturbances according to the SYM/H Ann.Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.

Fig. 1 .
Fig. 1.One-minute values of the ASYM/H, SYM/H, and AL indices and of the By and Bz components of the IMF for the storm of 29-30 May 2003 (analysed interval from 16:00 UT on 29 May to 10:00 UT on 30 May 2003, orbits 16229-16240).The vertical dashed lines indicate the UT time moments of each satellite orbit over the northern polar cap.The orbit numbers are splitted into two parts: the two digits above the UT-axis of each frame denote the last two digits of the orbit numbers of the CHAMP passes, while the first three digits are indicated at the lower left side.

Fig. 2 .
Fig. 2. Direction and density values of the Hall current along the satellite orbit at the dayside (left column, 14-16 MLT, corresponding to the ascending section of the orbit) and nightside sectors (right column, 02-04 MLT, descending orbit section).Positive currents denotes eastward current for the descending orbit section, and, accordingly, westward current for the ascending section.

325
PE, then its appearance should be connected with a change of the IMF By component.Indeed, the currents are accompanied with a change of sign of the IMF B y component, corresponding to -17.5 nT (MLat = 80.7 • ) during orbit 16237 and -5.2 nT (MLat =80.6 • ) during orbit 16238.During the orbits 16239 and 16240 the IMF B y component turns to positive values again and weak eastward directed currents appear accordingly at MLat∼80 • .Ann. Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.

340-
At the beginning of the substorm interval (orbits 16229 and 16230) with the intensification of SYM/H, the WE peak shifts to lower latitudes and the current density diminishes.The most intense peaks of the nighttime WE are obtained during the substorm interval prior to the main phase start with and retain values of 2.7 A/m at MLat = 64.4• (orbit 16231), 1.7 A/m at MLat = 60.9 • (orbit 16232), and during the beginning of the main phase with 1.79 A/m at MLat = 58.0• (orbit 345 16233).Later in the maximum of the main storm phase, the WE peak current density diminishes to 1.07 A/m at MLat = 59.5 • (orbit 16234) and 0.9 A/m at MLat = 55.8 • (orbit 16235).Hence, the latitudinal peaks of the WE vary during nighttime in phase with the intensification of SYM/H (storm development) before the main phase commences at higher latitudes, while shifting to the equator during the maximum of the main phase.The peak intensities changes both in phase and in antiphase 350 with the SYM/H intensity.During the recovery phase, the peak density of the WE current is smaller than 0.2 A/m, while the eastward currents within the polar cap are too small to be recorded.4.2 Observations related to ASYM/H variations and to high-latitude currentsIn the dayside sector during the existence of the EE (orbits 16229-16236), the peak current intensities and the peak latitude positions vary synchronous with the ASYM/H changes, except of one orbit 355 (16235) during the main phase.During this orbit, the ASYM/H index abruptly intensifies to 145 nT with a pertaining small density of the EE with ∼0.29 A/m and a shift of MLat by 1.8 • .For the WE, the change in latitude and density of the peak currents is in phase with the ASYM/H variations during the storm, with the exception of orbit 16235.In the nighttime sector, the intensity of the peaks and their latitude (except orbit 16235) change in 360 phase with the ASYM/H variations.Summarizing the results of Hall current observations by the CHAMP satellite during the magnetic disturbance period of 29-30 May 2003 in the daytime and nighttime sectors (12-16 MLT and 00-04 MLT, respectively) we conclude: -Intense >1 A/m eastward and westward electrojets can occur at latitudes of the auroral zone 365 during substorm periods, which precede the magnetic storm, and during the beginning of its main phase.During the maximum of the main phase, the density of the Hall currents as well Ann.Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.as the substorms diminish in antiphase to an increase of the SYM/H index.-A fast decay of the EE and WE occurs during the recovery phase at auroral latitudes both during daytime and nighttime hours.The westward or the eastward currents can be influenced 370 during this storm phase by the existence of a PE at 73 • <MLat<80 • in the region of the dayside cusp.The direction of the current in the PE is determined by the IMF B y component: for B y > 0 the current is eastward, for B y < 0 westward.The change of the current direction within the PE can occur several times during the storm development, but always in accordance with the 375 change of the IMF B y orientation.-The Hall currents in the auroral ionosphere, both the EE and the WE, vary usually in phase with the SYM/H and ASYM/H variations (but sometimes also in antiphase).There are time intervals, where any correlation between the geomagnetic activity indices and the Hall current parameters is missing.There is a closer connection of the current density and the MLat 380 variations with ASYM/H than with SYM/H.-In the daytime sector (14-16 MLT) during a period of intense substorms, the EE is located in a latitude range 56 • <MLat<64 • , while the WE is at 64 • <MLat<70 • .During the main phase of the storm, the EE shifts to 58 • <MLat<62 • , while the WE is situated at 64 • <MLat<73 • , and during the recovery phase, finally, the WE is observed at latitudes of 73 • <MLat<76 • .385 Therefore, the EE stays at about the same latitudes during the both the intense substorms and the main phase of the storm, attaining extreme equatorward values of MLat∼ 56 • .An analogue situation exists with regard to the change of position for the WE in various storm phases, but during daytime hours the WE is located about 6 • closer to the pole.-In the nighttime sector (02-04 MLT), there exists practically only the WE, which is located 390 during substorms at 61 • <MLat<64 • , and during the main storm phase at 56 • <MLat<60 • .Therefore, extremal positions of the WE and EE can reach latitudes below 60 • .This occurs in the daytime sector for the EE, while in the nighttime for the WE.The detailed description of the Hall current dynamics during five further magnetic summer storm intervals is transferred to the Appendix.3955 Discussion In the section 4 and in the appendices A1-A5 we have investigated several geomagnetic storm periods, based on magnetometer measurements onboard the CHAMP satellite.The Hall currents in the high-latitude upper ionosphere of the Northern Hemisphere were analysed for various MLT sectors 14 Ann.Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.with regard to their position in geomagnetic latitude, their density and direction.The empirical de-400 scription concerned the appearance of the EE, the WE, and the PE during various storm phases and was carried out primarily qualitatively.

Table 2 .
Fig. 3a-f shows the magnetic latitude (left side panels) and the Hall current density I (right side panels) obtained by CHAMP satellite crossings over the polar electrojets during 6 geomagnetic 415

Fig. 3a differentiatesFig. 3 .
Fig.3adifferentiates the current measurements with regard to the azimuthal IMF component (By), i.e., between those, obtained during By>0 and those during By<0 conditions.It is clearly seen that the direction of the current within the PE is determined by the IMF By sign.For intervals with positive IMF By>0, we observe for all cases an eastward directed Hall current; for negative IMF

455
along the magnetic field lines, the electric field E z can exist only in the region of open field lines, rooted at the polar caps, and will be short-circuited along closed field lines.Thus, the boundary between closed and open field lines (OCB) will be the line of zero potential for the field E z , and U =0 at this boundary in the ionosphere.Feldstein et al. (1975b) estimated the effectiveness of the electric field penetration from the solar 460 wind to the cusp.During summer season (July-August 1969) the integral Hall current within the PE Ann.Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.wasestimated to ∼220 kA.The integral conductivity in the ionosphere during summer around noon is ∼7 mhos, i.e., the potential drop in the cusp is about U ∼30 kV.The potential difference between the northern and southern boundaries of the magnetotail amount to U m ∼600 kV.The following assumptions were made for the estimation: solar wind speed V x ∼400 km/s, IMF B y =6 nT, and 465 the magnetospheric size along the z-axis ∼40 R E .If the voltage drop is the same in the northern and southern hemispheres, then the efficiency of the electric field penetration from the solar wind to the high-latitude ionosphere during summer season is ∼10%.

470
During orbit 16238 of 30 May 2003 we observe a maximum Hall current density at 06:15 UT.

495
According to Fig.3d, there is a high correlation between the ASYM/H index and the PE currentI (r ∼ 0.74) and an absence of correlation with the MLat position of the PE (r ∼ −0.29).The PE current density has therefore a direct relation to the intensity of the ASYM/H current system: with Ann.Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.increasing longitudinal asymmetry increases the PE current density, but the latitudinal position of the PE does not depend on ASYM/H. 500

Figures 4 -
Figures 4-7 consider the MLat positions (left columns) and current densitites I (right columns) during the moments of extreme values of current density in dependence on the SYM/H, ASYM/H, AL, and IndN indices.As in Fig. 3a, data points of electrojets with an eastward direction are indicated by red colour and those with westward direction by blue colour. 530

Fig. 4
Fig.4shows only those cases of AE appearance in the daytime sector with changing SYM/H

Fig. 4 .
Fig. 4. Daytime sector (09-14 MLT): Dependence of the magnetic latitude MLat (degrees) position of the peak(left column) and of the density (I in A/m, right column) of Hall current in the WE (blue) and EE (red) on the geomagnetic activity indices SYM/H (a), ASYM/H (b), AL (c), and the solar wind coupling function IndN (d).For the cases of correlations with r > 0.46, the correlation coefficients (r) and the dispersion (σ) according to a linear regression are shown as labels.

6056. 1 --
Fig. 3a-f shows the results of the correlation analysis of the PE characteristics and the IMF By component as well as various activity indices.The values of the correlation coefficients r and the 630

-
Ann. Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.There is a correlation between the current density I in the PE and the density of the partial ring current in the magnetosphere (PRC, index ASYM/H), but practically no correlation of 650 this index with MLat of the PE.

655
Auroral electrojets are located at auroral latitudes (MLat<72 • during daytime hours, and MLat<68 • during nighttime) exist during all MLT.The amount of electrojet current in a certain latitude range, the structure of the currents in them, the interconnection with concrete magnetospheric domains, depends on the level of disturbance, which is controlled by UT as well as local time (MLT) at the observational points.Therefore we present the conclusions from the observations for each of the 660 four time sectors: daytime, evening, nighttime, and morning hours.Daytime sector (09-14 MLT): -The MLat positions of the auroral electrojets, both WE and EE, correlate with the activity indices ASYM/H, AL, and IndN.The auroral electrojets shift toward lower latitudes with increasing activity.For ASYM/H ∼220 nT the EE shifts to MLAT ∼57 • .

670-
Significant values of the correlation coefficients r with activity indices exist both for MLat and for the Hall current density I.

675--
The EE and WE shift equatorward with increasing activity.This shift occurs with respect to all activity indices inspected here.The EE is located equatorward of MLat ∼60 • during magnetic storm periods; the farthest shift attains MLat ∼53 • .Ann. Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.The equatorward shift for increasing activity is accompanied by increasing current densities from I < 0.2A/m for ASYM/H∼40 nT to ∼1.7 A/m in the EE and ∼1.3 A/m in the WE for 680 ASYM/H∼380 nT.The current density of the EE increases hence stronger than that of the WE (by about 30%).

690-----
The MLat(WE) position correlates only with the SYM/H index, shifting equatorward from 62 • to 58 • for a SYM/H increase from -40 nT to -170 nT.The lowest possible MLat is ∼58 • .The equatorward shift of the WE is accpmpanied by an increase of the current density from I < 0.2A/m to ∼1.5 A/m.Morning sector (02-09 MLT): 695 The characteristics of the auroral electrojets is almost identical for midnight and morning hours.-The MLat position at that MLT is controlled for the most part by the SYM/H activity index, i.e., by the density of the ring current.In the morning sector, there exists almost exclusively the WE only.The current density in the WE correlates with the ASYM/H and the AL indices with maximum 700 values of r = 0.69 with respect to ASYM/H.The current density increases from 0.5 A/m to 2.1 A/m for intensifications of ASYM/H from 40 nT to 200 nT.
Fig. A2.1 shows the variations of the SYM/H and ASYM/H indices for the magnetic storm of June 18th, 2003.The storm phases are represented by the orbit numbers 16532 and 16533 for the creation phase, 16534-16536 for the main phase, and 16537-16541 for the recovery phase.Extreme values 845

Fig. A4. 1 .
Fig. A4.1.One-minute values of the ASYM/H, SYM/H, and AL indices and of the By and Bz components of the for the storm of 15 May 2005 (analysis interval 00:00-19:00 UT, orbits 27423-27432).The time of each orbit and its orbit number are indicated as in Fig. A1.1.

Fig. A5. 1 .
Fig. A5.1.One-minute values of the ASYM/H, SYM/H, and AL indices and of the By and Bz components of the IMF for the storm of 18 Aug 2003 (analysis interval 00:00-23:00 UT, orbits 17480-17494).The time of each orbit and its orbit number are indicated as in Fig. A1.1.

Table 1 .
Overview of CHAMP satellite orbits used for this study.

Table 3 .
The dependent (X) and the independent variable (Y ), their correlation coefficients (r), the coefficients A and B of the regression equations X = A + B * Y , and their dispersions σ, listed for four different MLT 2. the Hall current direction in the AE does not depend uniquely from the orientation of the IMF By component.
Indeed, the By component appeared to be at a steady neagative value during the orbits 16537-16541.As a rule, the ionospheric currents in the nighttime sector are westward directed in the MLat range of 57.8 • -63.0 • with I∼0.5 A/m.Only during two orbits in the creation and main phases, the current • with I∼0.43 A/m, and a WE at MLat ∼72 • with I∼0.42 A/m.Both electrojets are retained during the main storm phase with 850 an EE of I∼0.8 A/m at MLat ∼62 • and a WE of I∼0.5 A/m at MLat ∼67 • .The WE only persists during the recovery phase with I∼0.3 A/m at MLat ∼78 • (orbits 16537 and 16538) This high-latitude westward current near MLat ∼77 • with I∼0.4 A/m does not vanish till the end of the recovery phase.Such a high-latitude position of a westward current near noontime MLT gives reason to suggest that this is a polar electrojet (PE).This assumption would apply, if the IMF By component is negative.855 -The quite strong geomagnetic storm (according to the SYM/H<-150 nT index value during 865 the main phase) is accompanied by substorms with AL up to -1500 nT and with the lowermost index value for the asymmetry of the field ASYM/H<-100 nT.The peculiarities of this storm Ann.Geophys.Discuss., https://doi.org/10.5194/angeo-2018-31Manuscript under review for journal Ann.Geophys.Discussion started: 25 April 2018 c Author(s) 2018.CC BY 4.0 License.