Volume 47, Issue 4 e2019GL086676
Research Letter
Free Access

Explicit IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0002 Dependence in Geomagnetic Activity: Modulation of Precipitating Electrons

L. Holappa

Corresponding Author

L. Holappa

ReSoLVE Centre of Excellence, Space Physics and Astronomy Research Unit, University of Oulu, Oulu, Finland

Correspondence to: L. Holappa,

[email protected]

Search for more papers by this author
T. Asikainen

T. Asikainen

ReSoLVE Centre of Excellence, Space Physics and Astronomy Research Unit, University of Oulu, Oulu, Finland

Search for more papers by this author
K. Mursula

K. Mursula

ReSoLVE Centre of Excellence, Space Physics and Astronomy Research Unit, University of Oulu, Oulu, Finland

Search for more papers by this author
First published: 06 February 2020
Citations: 18

Abstract

The most important driver of geomagnetic activity is the north–south ( urn:x-wiley:grl:media:grl60214:grl60214-math-0003) component of the interplanetary magnetic field (IMF), which dominates the solar wind-magnetosphere coupling and all solar wind coupling functions. While the east–west ( urn:x-wiley:grl:media:grl60214:grl60214-math-0004) IMF component is also included in most coupling functions, its effect is always assumed to be symmetric with respect of its sign. However, recent studies have shown that, for a fixed value of any coupling function, geomagnetic activity is stronger for urn:x-wiley:grl:media:grl60214:grl60214-math-0005 than for urn:x-wiley:grl:media:grl60214:grl60214-math-0006 in Northern Hemisphere winter. In Southern Hemisphere winter, the dependence on the urn:x-wiley:grl:media:grl60214:grl60214-math-0007 sign is reversed. In this paper, we use measurements of National Oceanic and Atmospheric Administration Polar-Orbiting Operational Environmental Satellites to show that the flux of magnetospheric electrons precipitating into the atmosphere also exhibits an explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0008 dependence. This urn:x-wiley:grl:media:grl60214:grl60214-math-0009 dependence is strong in the midnight and dawn sectors where it causes a related urn:x-wiley:grl:media:grl60214:grl60214-math-0010 effect in the westward electrojet and geomagnetic activity by modulating ionospheric conductivity.

Key Points

  • Fluxes of energetic electrons precipitating into atmosphere are strongly dependent on the sign and amplitude of IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0011
  • This IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0012 effect on electron fluxes is strong around midnight and at dawn but weak at dusk
  • This IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0013 effect is expected to modulate ionospheric conductivity and to affect a similar effect in geomagnetic activity

1 Introduction

Short-term variability of the Earth's magnetic field, also known as geomagnetic activity, is one of the most significant manifestations of space weather. Geomagnetic activity is caused by various electric currents in the near-Earth space, maintained by the interaction between solar wind and the magnetosphere. The most significant factor contributing to geomagnetic activity is the north–south ( urn:x-wiley:grl:media:grl60214:grl60214-math-0014) component of the interplanetary magnetic field (IMF), measured in the Geocentric Solar Magnetospheric coordinate system. While the east–west ( urn:x-wiley:grl:media:grl60214:grl60214-math-0015) IMF component is also known to contribute to the reconnection rate (Laitinen et al., 2007; Sonnerup, 1974), its effect is symmetric with respect to its sign, that is, if other solar wind parameters are held constant, the reconnection rate does not change if the sign of urn:x-wiley:grl:media:grl60214:grl60214-math-0016 is reversed. This symmetric effect of IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0017 is included in most solar wind-magnetosphere coupling functions, such as the Newell universal coupling function (Newell et al., 2007)
urn:x-wiley:grl:media:grl60214:grl60214-math-0018(1)
where urn:x-wiley:grl:media:grl60214:grl60214-math-0019 is solar wind speed, urn:x-wiley:grl:media:grl60214:grl60214-math-0020, and urn:x-wiley:grl:media:grl60214:grl60214-math-0021 is the so-called IMF clock angle.

Recent observations of the geomagnetic field by polar-orbiting satellites (Friis-Christensen et al., 2017; Smith et al., 2017) found that the westward auroral electrojet in both hemispheres is more intense in Northern Hemisphere (NH) winter for urn:x-wiley:grl:media:grl60214:grl60214-math-0022 than for urn:x-wiley:grl:media:grl60214:grl60214-math-0023. For Southern Hemisphere (SH) winter, the effect of urn:x-wiley:grl:media:grl60214:grl60214-math-0024 is opposite: the westward electrojet is more intense for urn:x-wiley:grl:media:grl60214:grl60214-math-0025 than for urn:x-wiley:grl:media:grl60214:grl60214-math-0026. Holappa and Mursula (2018) quantified this new, explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0027 effect in detail and showed that during NH winter solstice, for a given value of urn:x-wiley:grl:media:grl60214:grl60214-math-0028, the value of the urn:x-wiley:grl:media:grl60214:grl60214-math-0029 index (measuring the NH westward electrojet) is 40–50% greater for urn:x-wiley:grl:media:grl60214:grl60214-math-0030 than for urn:x-wiley:grl:media:grl60214:grl60214-math-0031. During NH summer, the urn:x-wiley:grl:media:grl60214:grl60214-math-0032 index is about 10% stronger for urn:x-wiley:grl:media:grl60214:grl60214-math-0033 than for urn:x-wiley:grl:media:grl60214:grl60214-math-0034. Holappa and Mursula (2018) also studied the urn:x-wiley:grl:media:grl60214:grl60214-math-0035 dependence of the SH westward electrojet using the urn:x-wiley:grl:media:grl60214:grl60214-math-0036 index at Syowa station in Antarctica and found a strong urn:x-wiley:grl:media:grl60214:grl60214-math-0037 effect (higher urn:x-wiley:grl:media:grl60214:grl60214-math-0038 index for urn:x-wiley:grl:media:grl60214:grl60214-math-0039) in SH winter and a weak urn:x-wiley:grl:media:grl60214:grl60214-math-0040 effect in SH summer. Thus the urn:x-wiley:grl:media:grl60214:grl60214-math-0041 effect in geomagnetic activity has the same sign globally, but it is clearly stronger in local winter than in local summer. Holappa and Mursula (2018) also showed that the urn:x-wiley:grl:media:grl60214:grl60214-math-0042 effect maximizes when the Earth's dipole axis points towards midnight (5 UT in NH). This combined seasonal/UT dependence suggests that the urn:x-wiley:grl:media:grl60214:grl60214-math-0043 effect in ionospheric currents may work most efficiently under low ionospheric conductivity.

This explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0044 effect is currently not included in any space weather prediction models. However, Holappa et al. (2019) showed that the urn:x-wiley:grl:media:grl60214:grl60214-math-0045 effect strongly affects geomagnetic activity even at subauroral latitudes during geomagnetic storms driven by coronal mass ejections. Thus, inclusion of the urn:x-wiley:grl:media:grl60214:grl60214-math-0046 effect to models predicting space weather is important for the future predictions of space weather hazards, such as geomagnetically induced currents.

The physical mechanism of the urn:x-wiley:grl:media:grl60214:grl60214-math-0047 effect in geomagnetic activity is not yet known. Friis-Christensen et al. (2017) suggested that IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0048 modulates the strength of the substorm current wedge, which probably makes the largest contribution to the urn:x-wiley:grl:media:grl60214:grl60214-math-0049 index. While this mechanism is consistent with observations (Friis-Christensen et al., 2017; Holappa & Mursula, 2018; Holappa et al., 2019), it is still unclear how IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0050 is connected to the substorm current wedge. In principle, in order to be able to modulate ionospheric currents, IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0051 must modulate either the ionospheric electric fields or ionospheric conductivity or both. Radar studies (Pettigrew et al., 2010; Thomas & Shepherd, 2018) have shown that the cross-polar cap potential exhibits a qualitatively similar urn:x-wiley:grl:media:grl60214:grl60214-math-0052 dependence as high-latitude geomagnetic activity, indicating a urn:x-wiley:grl:media:grl60214:grl60214-math-0053 dependence in ionospheric electric fields that may contribute to ionospheric currents. A recent study gives evidence that there is an explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0054 dependence also in polar cap area, indicating a possible explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0055 effect in the dayside reconnection rate (Reistad et al., 2020). This is consistent with the above radar observations since the modulation of the reconnection rate would modulate magnetospheric convection and electric fields mapped into the ionosphere. The detailed mechanism of the urn:x-wiley:grl:media:grl60214:grl60214-math-0056 effect in the dayside reconnection is not well understood. However, several studies have shown that the geometry of the reconnection X-line varies with (seasonally changing) dipole tilt angle and IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0057 (Liu et al., 2012; Park et al., 2006; Zhu et al., 2015). The X-line may also be split into the two hemispheres if the IMF vector is dominated by the urn:x-wiley:grl:media:grl60214:grl60214-math-0058 component (Connor et al., 2015; Trattner et al., 2012). More research is needed to identify which of these produce the urn:x-wiley:grl:media:grl60214:grl60214-math-0059 dependence. However, it is important to note that the urn:x-wiley:grl:media:grl60214:grl60214-math-0060 dependence of the dayside reconnection rate alone does not explain why the urn:x-wiley:grl:media:grl60214:grl60214-math-0061 effect in geomagnetic activity is strong only in local winter (Holappa & Mursula, 2018; Smith et al., 2017).

In addition to electric fields, ionospheric currents (such as the substorm current wedge) also depend on ionospheric conductivity, which is controlled by solar illumination and the precipitation of magnetospheric particles into the ionosphere. The goal of this paper is to study whether there is a urn:x-wiley:grl:media:grl60214:grl60214-math-0062 effect in the flux of magnetospheric particles precipitating into high-latitude ionosphere. Understanding the relation between IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0063 and particle precipitation is an important step toward a better understanding of the urn:x-wiley:grl:media:grl60214:grl60214-math-0064 effect, as particle precipitation is an important source of ionospheric conductivity (especially in the nightside ionosphere), thus strongly contributing to the westward electrojet.

2 Energetic Electron Measurements

In this paper, we utilize the long database of magnetospheric energetic electron measurements by the Medium Energy Proton and Electron Detector instruments on board National Oceanic and Atmospheric Administration (NOAA) Polar-Orbiting Operational Environmental Satellites (NOAA15–NOAA19). These fluxes have been calibrated and corrected for instrument efficiency and cross contamination by energetic protons to form a homogenous long-term data set of energetic electron fluxes (Asikainen & Mursula, 2013). In this paper, we analyze the flux of electrons in the lowest energy channel (>30 keV) precipitating into the atmosphere (0° detector). The polar orbits of different NOAA satellites are on different magnetic local time (MLT) planes and vary slowly with time (Asikainen & Ruopsa, 2019), which allows us to study the urn:x-wiley:grl:media:grl60214:grl60214-math-0065 effect in different MLT sectors. In this paper, we analyze the electron flux data in three different (8 hr wide) MLT bins: 20-04 MLT (night), 04-12 MLT (dawn), and 12-20 MLT (dusk), in order to study possible MLT dependence of the urn:x-wiley:grl:media:grl60214:grl60214-math-0066 effect. The motivation for using such a wide MLT bin width is to ensure sufficient statistics in all bins.

3 IMF By-Effect in Electron Fluxes

In order to test whether there is an explicit IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0067-effect in the flux of precipitating electrons, we use a similar methodology as Holappa and Mursula (2018). Figures 1a and 1b show the average fluxes ( urn:x-wiley:grl:media:grl60214:grl60214-math-0068) of precipitating electrons for different values of IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0069 and urn:x-wiley:grl:media:grl60214:grl60214-math-0070 around NH winter solstice (December 21 urn:x-wiley:grl:media:grl60214:grl60214-math-0071 days) separately for NH and SH, respectively. (Figure 1 shows fluxes in a logarithmic scale. The original fluxes are given in the units 1/[cm urn:x-wiley:grl:media:grl60214:grl60214-math-0072 sr s].) The electron fluxes in Figure 1 are averaged over 20-04 MLT and urn:x-wiley:grl:media:grl60214:grl60214-math-0073(50°…70°) corrected geomagnetic latitude using all available data from NOAA15–NOAA19 satellites between 1998 and 2017. The coupling function urn:x-wiley:grl:media:grl60214:grl60214-math-0074 is normalized by its long-term average urn:x-wiley:grl:media:grl60214:grl60214-math-0075 calculated over all data in 1998–2017. Figures 1a and 1b clearly show that for a given value of urn:x-wiley:grl:media:grl60214:grl60214-math-0076, the flux of precipitating electrons in NH winter is greater for urn:x-wiley:grl:media:grl60214:grl60214-math-0077 than for urn:x-wiley:grl:media:grl60214:grl60214-math-0078 in both hemispheres. This explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0079 dependence is very similar in both hemispheres. This indicates that the urn:x-wiley:grl:media:grl60214:grl60214-math-0080 effect in >30-keV electrons is not related to polar rain, that is, precipitation of electrons of few hundred eV on the northern polar cap for urn:x-wiley:grl:media:grl60214:grl60214-math-0081 (away sector) and for the southern polar cap for urn:x-wiley:grl:media:grl60214:grl60214-math-0082 (toward sector) (Fairfield & Scudder, 1985). Note that the overall electron flux in SH is somewhat greater than in NH (different color scales in Figures 1a and 1b), probably due to the South Atlantic anomaly and other hemispheric asymmetries in the Earth's magnetic field (Meredith et al., 2011; Vampola & Gorney, 1983).

Details are in the caption following the image
Average logarithmic flux ( urn:x-wiley:grl:media:grl60214:grl60214-math-0083) of precipitating electrons for different values of the solar wind coupling function urn:x-wiley:grl:media:grl60214:grl60214-math-0084 and IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0085 in NH winter solstice urn:x-wiley:grl:media:grl60214:grl60214-math-0086 days in (a) Northern Hemisphere ( urn:x-wiley:grl:media:grl60214:grl60214-math-0087urn:x-wiley:grl:media:grl60214:grl60214-math-0088° CGM latitude) and (b) Southern Hemisphere ( urn:x-wiley:grl:media:grl60214:grl60214-math-008950°  to urn:x-wiley:grl:media:grl60214:grl60214-math-009070° CGM latitude). The values of urn:x-wiley:grl:media:grl60214:grl60214-math-0091 are normalized by the average of the coupling function ( urn:x-wiley:grl:media:grl60214:grl60214-math-0092) in 2003–2017. For a given value of urn:x-wiley:grl:media:grl60214:grl60214-math-0093, the electron flux is higher for urn:x-wiley:grl:media:grl60214:grl60214-math-0094 than for urn:x-wiley:grl:media:grl60214:grl60214-math-0095 in both hemispheres, that is, electron fluxes show an explicit IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0096 dependence. Panels (c) and (d) depict the normalized column averages ( urn:x-wiley:grl:media:grl60214:grl60214-math-0097) of the flux matrices in panels (a) and (b), quantifying the explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0098 dependence for NH and SH, respectively. Vertical bars denote the urn:x-wiley:grl:media:grl60214:grl60214-math-0099 standard errors.

We further quantify the strength of the explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0100 dependence in Figures 1c and 1d that show weighed column averages ( urn:x-wiley:grl:media:grl60214:grl60214-math-0101) of the flux matrices in Figures 1a and 1b together with their urn:x-wiley:grl:media:grl60214:grl60214-math-0102 errors. Before calculating the column means, each row is first normalized (divided) by its mean. This ensures that fluxes corresponding to different values of the solar wind coupling function are weighed equally. Figures 1c and 1d show that urn:x-wiley:grl:media:grl60214:grl60214-math-0103 increases rather linearly with IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0104 in both hemispheres. The ratio urn:x-wiley:grl:media:grl60214:grl60214-math-0105 is about 1.05 for NH (1.06 for SH). This is quite significant in linear scale since, for example, for a flux of urn:x-wiley:grl:media:grl60214:grl60214-math-0106, this corresponds to about 40% relative effect in NH (50% in SH). The relative effect grows with the level of electron precipitation: for the two uppermost rows in the flux matrix with the greatest urn:x-wiley:grl:media:grl60214:grl60214-math-0107 (and electron fluxes), the relative effect is about 60% in NH (50% in SH). This is consistent with the study of Shue et al. (2001) who found that the relative impact of IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0108 on auroral brightness increases with increasing southward urn:x-wiley:grl:media:grl60214:grl60214-math-0109. Interestingly, the urn:x-wiley:grl:media:grl60214:grl60214-math-0110effect in the electron fluxes is roughly equally strong in the winter and summer hemispheres (the difference being statistically insignificant). This winter–summer symmetry of the urn:x-wiley:grl:media:grl60214:grl60214-math-0111 effect is not seen in geomagnetic activity (Holappa & Mursula, 2018; Smith et al., 2017).

Figures 2a and 2b are similar to Figures 1a and 1b but show the electron fluxes for NH summer (June 21  urn:x-wiley:grl:media:grl60214:grl60214-math-0112 30 days). Figures 2a and 2b show that in SH winter (NH summer), the flux of precipitating electrons is greater for urn:x-wiley:grl:media:grl60214:grl60214-math-0113 than for urn:x-wiley:grl:media:grl60214:grl60214-math-0114 in both hemispheres. Thus the dependence on the sign of urn:x-wiley:grl:media:grl60214:grl60214-math-0115 is opposite for NH winter and SH winter. Again, the overall electron flux is greater in SH than in NH, but the relative strength of the urn:x-wiley:grl:media:grl60214:grl60214-math-0116 effect is roughly the same in both hemispheres (see Figures 2c and 2d). The ratio urn:x-wiley:grl:media:grl60214:grl60214-math-0117 is about 1.04 in NH (1.05 in SH). For a flux urn:x-wiley:grl:media:grl60214:grl60214-math-0118, this corresponds to about 30% effect in linear scale in NH (40% in SH). These ratios are quite close to those obtained from Figures 1c and 1d, indicating that the explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0119 effect is roughly equally strong in magnitude during both solstices.

Details are in the caption following the image
Same as Figure 1 but for SH winter (NH summer) solstice  urn:x-wiley:grl:media:grl60214:grl60214-math-0120 days.

Figures 3 and 4 show the electron fluxes in NH winter for dawn/prenoon (04-12 MLT) and dusk/afternoon (12-20 MLT) sectors, respectively. The electron fluxes in the dawn sector are slightly higher than in the midnight sector (Figures 1 and 2) and much higher than in the dusk sector (note the different color scales). This is due to the fact that energetic electrons injected into inner magnetosphere are partly precipitated into the atmosphere mostly in the dawn sector before drifting further on to dusk (Lam et al., 2010). The urn:x-wiley:grl:media:grl60214:grl60214-math-0121 effect in the dawn sector in NH winter (Figure 3) has the same magnitude as in the midnight sector (Figure 1) with urn:x-wiley:grl:media:grl60214:grl60214-math-0122 of about 1.05 in NH and 1.06 in SH (40% and 50% in linear scale for urn:x-wiley:grl:media:grl60214:grl60214-math-0123, respectively). The urn:x-wiley:grl:media:grl60214:grl60214-math-0124 effect at dawn (Figure 3) is also remarkably symmetric between the winter and summer hemispheres. Even though the dusk electron fluxes are low, they do show a weak urn:x-wiley:grl:media:grl60214:grl60214-math-0125 dependence in Figure 4 in both hemispheres. However, urn:x-wiley:grl:media:grl60214:grl60214-math-0126 does not increase linearly with urn:x-wiley:grl:media:grl60214:grl60214-math-0127, and the ratio urn:x-wiley:grl:media:grl60214:grl60214-math-0128 is about 1.02 for NH (1.03 for SH). For a typical dusk electron flux ( urn:x-wiley:grl:media:grl60214:grl60214-math-0129), this corresponds to about 10% effect in NH (20% in SH) in linear scale. Thus the explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0130 dependence in the dusk sector is significantly weaker than in the two other sectors.

Details are in the caption following the image
Same as Figure 1 but for dawn (04-12 MLT).
Details are in the caption following the image
Same as Figure 1 but for dusk (12-20 MLT).

4 Discussion and Conclusions

In this paper, we have shown that for a given fixed value of Newell solar wind-coupling function, the IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0131 component affects the flux of energetic (>30 keV) electrons precipitating into the atmosphere in the latitude range of urn:x-wiley:grl:media:grl60214:grl60214-math-0132(50–70°) corrected geomagnetic latitude. This explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0133 effect exhibits a systematic seasonal dependence: in NH winter, the flux of precipitating electrons is greater for urn:x-wiley:grl:media:grl60214:grl60214-math-0134 than for urn:x-wiley:grl:media:grl60214:grl60214-math-0135 in both hemispheres. In SH winter, the dependence on the urn:x-wiley:grl:media:grl60214:grl60214-math-0136 sign is reversed. We found that the magnitude of the explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0137 effect is roughly equal in both hemispheres and solstices.

Previous studies (Holappa & Mursula, 2018; Smith et al., 2017) have shown that the explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0138 effect in geomagnetic activity is much stronger in the winter hemisphere than in the summer hemisphere. Furthermore, Holappa and Mursula (2018) showed that the urn:x-wiley:grl:media:grl60214:grl60214-math-0139 effect in NH geomagnetic activity maximizes at 5 UT, when the Earth's magnetic dipole axis points towards midnight and the northern auroral region is maximally in darkness. These results imply that the urn:x-wiley:grl:media:grl60214:grl60214-math-0140 effect in geomagnetic activity works most efficiently under low ionospheric conductivity.

In contrast to geomagnetic activity, we have found that the urn:x-wiley:grl:media:grl60214:grl60214-math-0141 dependence of precipitating electrons is roughly equally strong in the winter and summer hemispheres. Thus the urn:x-wiley:grl:media:grl60214:grl60214-math-0142 effect to ionospheric conductivity, through the ionization by electron precipitation, is equally strong in the winter and summer hemispheres. However, in local winter, under low photoionization, particle precipitation, and thereby IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0143 control, a relatively large fraction of the total (photoionization + precipitation) conductivity. We suggest that this explains why the urn:x-wiley:grl:media:grl60214:grl60214-math-0144 effect in geomagnetic activity is relatively strong in local winter than summer.

In this paper, we also showed that while the urn:x-wiley:grl:media:grl60214:grl60214-math-0145 effect in the flux of precipitating electrons is strong in the midnight (20-04 MLT) and dawn (04-12 MLT) sectors, it is weak in the dusk sector (12-20 MLT). Moreover, we found that the overall level of electron precipitation is much lower in the dusk sector than in the two other sectors. This implies that IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0146 modulates ionospheric conductivity significantly in midnight and dawn sectors but much less in the dusk sector. This is in excellent agreement with the earlier finding that the urn:x-wiley:grl:media:grl60214:grl60214-math-0147 effect modulates the westward electrojet (located in the midnight and dawn sectors) but not the eastward electrojet located in the dusk sector (Holappa & Mursula, 2018). This is also supported by the fact that the westward electrojet is relatively strongly dependent on conductivity controlled by precipitating electrons (Østgaard et al., 2002) while the conductivity for the eastward electrojet is mainly controlled by solar illumination (Finch et al., 2008).

The physical mechanism of the urn:x-wiley:grl:media:grl60214:grl60214-math-0148 effect is still poorly understood. We note that a recent study (Reistad et al., 2020) gives evidence that the polar cap size in both hemispheres also shows an explicit urn:x-wiley:grl:media:grl60214:grl60214-math-0149 dependence, possibly indicating a stronger reconnection at the dayside magnetopause for urn:x-wiley:grl:media:grl60214:grl60214-math-0150 ( urn:x-wiley:grl:media:grl60214:grl60214-math-0151) in NH (SH) winter. Reistad et al. (2020) also found that the urn:x-wiley:grl:media:grl60214:grl60214-math-0152 effect in polar cap area is equally strong in both hemispheres, which is also expected if IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0153 indeed modulates the dayside reconnection rate and, thus, the global energy input into the magnetosphere. This paper is consistent with this mechanism as we found equally strong urn:x-wiley:grl:media:grl60214:grl60214-math-0154 effects in the precipitating electron fluxes in both hemispheres.

This paper gives strong evidence that ionospheric currents and related high-latitude geomagnetic activity are strongly modulated by IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0155-dependent fluxes of precipitating electrons. However, more research is needed for a better understanding of the chain of mechanisms connecting IMF urn:x-wiley:grl:media:grl60214:grl60214-math-0156 to particle precipitation and other magnetospheric and ionospheric phenomena.

Acknowledgments

We acknowledge the financial support by the Academy of Finland to the ReSoLVE Centre of Excellence (project 272157), to the postdoctoral researcher project of LH (322459), and to the PROSPECT project (321440). The solar wind data were downloaded from the OMNI2 database (http://omniweb.gsfc.nasa.gov/). All the original POES/MEPED energetic particle data used here are archived in the NOAA/NGDC dataserver (http://www.ngdc.noaa.gov/stp/satellite/poes/index.html).