| 17 Oct 2025
Magnetotail response to corotating interaction region driven geomagnetic storms: Cluster observations
Abstract. In this work we have selected 40 corotating/stream interaction (CIR) driven geomagnetic storms that occurred between 2001 and 2016, and statistically studied their impacts on the magnetotail. The wavelet transform was applied to the interplanetary magnetic field (IMF) Bz component, magnetotail Bx component, and the auroral electrojet (AE) index during geomagnetic storms. The cross-wavelet technique was applied to determine the periods of higher correlation between the IMF Bz × Tail Bx, IMF Bz × AE index and Tail Bx × AE index. More than 80 % of the most energetic periods in the IMF Bz and magnetotail Bx are found to be shorter than 4 hours, independently of the storm phase. The AE index presented the range between 2–4 hours as the most common with energetic periods for both storm phases. In the recovery phase, periodicities in the AE index are more spread (≤ 12 hours) than for the main phase (≤ 8 hours) probably due to the presence of high-intensity long-duration continuous AE activities (HILDCAAs). From the cross-wavelet analysis (IMF Bz × magnetotail Bx, IMF Bz × AE index and magnetotail Bx × AE index), periods ≤ 4 hours are found to be dominant in both storm phases, which coincide with the cyclic substorm periods. The power spectral analysis showed that the IMF Bz and magnetotail Bx time series follow the Kolmogorov (-5/3) power law. Additionally, the mean values of the spectral indices for the magnetotail Bx and AE index are higher during the recovery phase than the main phase. This suggests that turbulence is more pronounced during the recovery phase of geomagnetic storms driven by CIRs.
This paper studies the response of the magnetotail and the auroral electrojet to geomagnetic storms driven by corotating interaction regions (CIR) or stream interaction regions. The wavelet and cross-wavelet analyses and Fourier transform of the IMF Bz, the magnetotail Bx, and the AE index show the characteristic periods of fluctuations of these parameters, the correlations between them, and the spectral index for the storm main and recovery phases, including the characteristic substorm cycles and the large spectral index during the storm recovery phase or possibly high-intensity long-duration continuous AE activities (HILDCAAs). The results are interesting and will potentially deepen our understanding of the coupling of the solar wind, the magnetosphere, and the ionosphere during storms. I think that the manuscript is relatively well organized and written. I, however, have a few concerns listed below to be considered before I can recommend accepting this paper for publication. In particular, I am concerned about the new points of this study, the magnetotail Bx, and the cone of influence for the wavelet analysis.
I think that the consistency with previous studies are well described, but it seems to me that it is necessary to describe or emphasize what are the new points or the differences, compared with previous studies.
Lines 91-96: In general, the observed magnetotail |Bx| increases during the substorm growth phase, associated with energy accumulation and plasma sheet thinning, while |Bx| decreases during the substorm expansion phase, associated with energy release and plasma sheet thickening. On the other hand, |Bx| has a spatial variation, that is, the polar-orbiting Cluster spacecraft observe the relatively small |Bx| in the plasma sheet and the large |Bx| in the lobe. This spatial variation is observed not only by the spacecraft motion but also by flapping motion of the magnetotail. To remove the spatial (Z-directional) variations in the magnetotail |Bx|, I suggest that instead of Bx, the lobe magnetic field be used, which can be estimated from the total pressure (the sum of the plasma (ion) and magnetic pressures), assuming the total pressure balance in the Z direction in the magnetotail: B_lobe^2/2 mu0 = NkT + B^2/2 mu0.
Lines 249-251 (for all wavelet spectrograms): As the authors wrote, the wavelet power outside the cone of influence should be discarded due to edge effects. The discarded wavelet powers occur in the interval of interest (the interval of the plot) in the present analysis. Hence, to make the meaningful interval longer, particularly for long-period variations, I suggest that the interval of wavelet analysis be extended behind and ahead, from ~8 hours or half a day before the beginning of the plot to ~8 hours or half a day after the end of the plot (in the case of Figure 4, from 7 or 11 UT on August 31 to 16:40 or 20:40 UT on September 1 for the wavelet analyses), and then the additional intervals before and after the interval of the plot be cut (for Figure 4, show only the data from 19 UT on August 31 to 08:40 UT on September 1). Thus, this method will make the wavelet powers shown in the figures meaningful (inside the cone of influence).
Minor comments:
Lines 101-103: What criterion (or minimum Dst or Sym-H) did this study adopt to define the geomagnetic storms? It should be written in the text.
Figures 1 and 2: Please draw the line of IMF Bz=0 in the top-left panel to identify the southward Bz easily.
Lines 183-184: Please write the substorm signatures (AE increase and tail |Bx| decrease) specifically in the previous paragraph or somewhere.
Line 246: The variation is not quasiperiodic but quasicontinuous?
Line 265, Figure 5: The period ranges (or the maximum period range) for the histograms are different for different panels. For easier comparison between the three parameters and the two storm phases, it is better to make the format of the plot the same for all panels by showing all the period ranges, 0-2, 2-4,..., and 10-12 hours, even if the percentage is 0%.
Line 299 (Figure 6): What are the definitions of the four categories of local, intermittent, quasi-continuous, and continuous? They should be written in the text.
Figure 11: Do the upper panels show the results of event 7 of the storm main phase? And do the bottom panels show the results of event 1 of the storm main phase? These events should be mentioned in the text and in the figure caption. Confusingly, the authors discuss the mean values in the text, but Figure 11 seems to show the results of the specific events.
Other minor corrections:
Throughout the text: Units should be written in roman, not in italic (for example, km s-1 at line 30).
Line 15: both storm phases --> both storm main and recovery phases
Line 28: Add Schwenn (2006) to the references list.
Line 33: Add Sanchez-Garcia (2023) to the references list.
Line 42: Add Echer et al. (2013) to the references list.
Lines 83 and 86: I think "between" is better than "among".
Line 90: http --> https
Line 99: http --> https
Line 101: Leymory --> Iyemori
Lines 151 and 152: Write the units of Dst: -100 nT and -250 nT.
Line 255: Add Souza et al. (2018) to the references list.
Line 274: Add "For the magnetotail Bx," before "Although the range between 2 and 4 hours...."
Line 280: Delete "in the tail".
Line 317: Delete the comma after "both".
Line 387: MP and RP should be spelled out.
Line 466: Delete the comma after "Similarly".
Line 489: B= --> Bx
Line 525: Add "https://" before "wdc.kugi...".
Line 526: Add "https://" before "omniweb.gsfc...", and delete the space between "....gov/" and "form/...".
Line 554: Cite Chian et al. (2024) in the text, or delete it from the references list.
Line 618: Lyemori --> Iyemori
Line 646-647: The information of Runov et al. (2005) seems incorrect. Please correct it.
Figure 3, the middle panel: The title of this panel seems to be wrong. "IMF Bx" should be "Tail Bx" or something like that.