Influence of different types of ionospheric disturbances on GPS signals at polar latitudes

Belakhovsky, Vladimir B.; Jin, Yaqi; Miloch, Wojciech J.

doi:https://doi.org/10.5194/angeo-39-687-2021

Articles | Volume 39, issue 4

https://doi.org/10.5194/angeo-39-687-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/angeo-39-687-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 39, issue 4

Regular paper

|

20 Jul 2021

Regular paper |

| 20 Jul 2021

Influence of different types of ionospheric disturbances on GPS signals at polar latitudes

Vladimir B. Belakhovsky, Yaqi Jin, and Wojciech J. Miloch

Download

Final revised paper (published on 20 Jul 2021)
Preprint (discussion started on 21 Jan 2021)

Interactive discussion

Status: closed

RC1:
'Comment on angeo-2020-93', Anonymous Referee #1, 11 Mar 2021

Review of "Influence of different types of ionospheric disturbances on GPS signals at polar latitudes"

The paper presents short case studies of ionospheric conditions and GPS disturbances during 5 different types of events;

dayside/cusp particle precipitation, substorm particle precipitation, daytime PCP, nighttime PCP and precipitation due to interplanetary shock arrival.

Its main conclusion is that all these event types lead to phase scintillation, but that the strongest phase scintillation is correlated with particle precipitation during substorms.

(For more detailed conclusions, see Conclusion section in the paper)

It is mentioned in the paper that 150 events have been considered, but there is only data from 5 events used in the paper.

This raises doubt about whether the conclusion in the paper is generally valid, or only valid for this set of events.

If 150 events have been analyzed, I would expect there to be at least some summary of the results included in the paper.

It should include at least some key statistics derived from the analysis of the 150 events.

It would give much more support to the conclusions if they are based on statistics from 150 events than if they are just based

on one case study for each type of event.

(This does not mean that the case studies should be removed. They are very nice as examples of the different event types.)

There are numerous language issues throughout the paper.

While the errors are not critical for the understanding of the paper, it would be significantly improved by careful proof-reading.

Lines 84-91:

You define the phase scintillation index twice.

Fortunately, the definitions are the same.

But you may want to delete one of the sentences.

Line 125: The reference [Belakhovsky et al., 2019] does not match the reference list.

The Belakhovsky et al. papers listed in the reference list are from the years 2016, 2017 and 2020.

Probably, your "2019" reference should be "2020"?

Please check and correct.

Line 147:

"Possibly it’s due to the field of view of EISCAT radar not coincides with the field of view of GPS receivers."

It is clear that the radar and the GNSS observations cannot fully overlap.

But, is it possible to provide any more assessment of the degree of overlap?

(approximately how much of the GNSS observations are typically in the region of the ionosphere that is

observed by EISCAT?)

(for the case studies, have you checked which of the satellites (if any) actually passed through the

region of the ionosphere that is observed by EISCAT?)

A fully detailed description is not necessary, but please consider if there is any useful information you could include.

Lines 187-188:

"The PCP is also identified in the aurora intensity variations as forms propagating from the polar to low

latitudes in 630.0 nm (red line) emission (Figure 7) at 19.00-23.00 UT according to LYR all-sky camera observations."

The keogram seems to show multiple patches, not just one.

Are you using the abbreviation "PCP" to refer to a multitude of patches, instead of just one patch?

(And also reinforcing the misunderstanding by using the singular "is" instead of the plural "are")

I would prefer it if you used "PCPs" when referring to multiple patches, and "PCP" when referring to a single patch.

So that sentence would start with: "The PCPs are "

Please check the usage throughout the paper, and adjust where necessary.

Lines 191-192:

"At latitudes of SKN (TRO) stations the PCP manifests itself as a long lasting substorm (more than 4

hours duration) with the amplitude 200-250 nT."

A patch is not the same as a substorm.

You need to rewrite this sentence.

For example:

"At latitudes of the SKN (TRO) stations the PCPs are observed during a long lasting substorm (more than 4

hours duration) with the amplitude 200-250 nT."

Line 195 (Daytime polar cap patches):

For a CIR event, I am interested to know the approximate solar wind properties.

In particular, what was the maximum value of the solar wind speed.

It is not required to include a plot of the solar wind data, just state the value in the text.

Lines 204-205:

"cusp precipitation has stronger

influence on GPS phase scintillation when it combined with the PCP. Our analyses also confirm this finding."

You stated that "PCPes were registered in time interval 06.00-12.00 UT".

You have not stated the presence of precipitation at any time interval for this event.

Looking at the plot, there appears to be enhanced density at 100 to 200 km from 13:00 to 15:00 UT, indicating

particle precipitation at that time. But, this does not correspond to the time period for which you state

the occurrence of PCPs, and also not with the time period of strong scintillations.

Please explain how the presented data confirms the finding.

To reach your conclusion, did you make assumptions regarding the particle precipitation also for times

when EISCAT did not directly observe precipitation?

If so, please state the assumptions clearly in the text.

Line 217:

"module of the interplanetary magnetic field"

Do you mean "magnitude of the interplanetary magnetic field"?

Line 219:

Please spell out the abbreviation "SSC" in full at its first occurrence.

Lines 243-244:

"Possibly low values of amplitude scintillations at high latitudes are caused by the low elevation angles of

GPS satellites at these regions."

Please explain how observing at low elevation decreases the amount/magnitude of amplitude scintillation.

Citation: https://doi.org/10.5194/angeo-2020-93-RC1
- AC1: 'Reply on RC1', Vladimir Belakhovsky, 06 Apr 2021
  
  We thank to the reviewer for the careful reading of our paper and helpful comments and remarks. Below are our answers to the remarks.
  
  Reviewer:
  «It is mentioned in the paper that 150 events have been considered, but there is only data from 5 events used in the paper. This raises doubt about whether the conclusion in the paper is generally valid, or only valid for this set of events. If 150 events have been analyzed, I would expect there to be at least some summary of the results included in the paper. It should include at least some key statistics derived from the analysis of the 150 events…»
  
  Answer:
  To describe statistically our research we have added 4 tables and section 3.5 to the paper. These tables include 121 events: 33 cusp/daytime precipitations, 36 substorms, 14 SSC events and 38 PCPs. It is less than 150 events which were mentioned in previous variant of the paper. But the tables already are quite large. We have selected mainly interesting events. Of course we have considered much more event for the 2010-2017 years. But the paper cannot include all of these events. The data from the tables confirms our case studies.
  We have decided to remove Figure 8 because now it is too much Figures and Tables. At the same time the dayside PCPs were also observed during another considered event 22 January 2012. The characteristics of the 7 November 2013 event are presented now in the Table 4. We have changed numeration of the Figures.
  We also have combined Figures 5 and Figure 6 on same plot to better compare the level of GPS scintillations on Svalbard and in Scandinavia. Now it is Figure 7.
  
  Reviewer:
  «There are numerous language issues throughout the paper. While the errors are not critical for the understanding of the paper, it would be significantly improved by careful proof-reading».
  Answer:
  We have tried to remove the language issues.
  
  Reviewer:
  «Lines 84-91: You define the phase scintillation index twice. Fortunately, the definitions are the same. But you may want to delete one of the sentences».
  Answer: We have corrected. Now we present only one definition of the phase scintillation index in the paper (Line 92)
  
  Reviewer:
  «Line 125: The reference [Belakhovsky et al., 2019] does not match the reference list…»
  
  Answer: We have corrected. Lines 130.
  
  Reviewer:
  «Line 147: "Possibly it’s due to the field of view of EISCAT radar not coincides with the field of view of GPS receivers." It is clear that the radar and the GNSS observations cannot fully overlap.
  
  But, is it possible to provide any more assessment of the degree of overlap?...
  …for the case studies, have you checked which of the satellites (if any) actually passed through the region of the ionosphere that is observed by EISCAT?».
  
  Answer: To show the degree of overlap of GPS satellites and EISCAT 42m radar we have plotted on Figure 4 the ionosphere projections of the GPS satellites, location of EISCAT radar and location of GPS receiver at NYA. Of course for the large scale disturbances (300-500 km and more) which was considered in the paper this overlap is not significant because several ionosphere projection points of GPS satellites (2-4) are always located near field of view of the EISCAT radar. So practically we have continuous measurements of the ionosphere parameters determined by the NYA GPS receiver.
  
  Reviewer:
  «Lines 187-188:
  
  "The PCP is also identified in the aurora intensity variations as forms propagating from the polar to low latitudes in 630.0 nm (red line) emission (Figure 7) at 19.00-23.00 UT according to LYR all-sky camera observations."
  The keogram seems to show multiple patches, not just one.
  
  Are you using the abbreviation "PCP" to refer to a multitude of patches, instead of just one patch?... ».
  Answer: It is corrected. Now we use abbreviation "PCPs" everywhere in the paper.
  
  Reviewer:
  «Lines 191-192: "At latitudes of SKN (TRO) stations the PCP manifests itself as a long lasting substorm (more than 4 hours duration) with the amplitude 200-250 nT."
  A patch is not the same as a substorm…»
  Answer: Yes, we agree. We rewrote this sentence.
  Lines 217:
  «During appearance of PCPs near the Svalbard at latitudes of the SKN (TRO) stations a long lasting substorm (more than 4 hours duration) with the amplitude 200-250 nT is observed».
  Line 319:
  «Comparison of the EISCAT observations on Svalbard and in Tromso shows that during PCPs appearance on Svalbard a typical substorm at lower latitudes (Tromso) was observed. The level of the phase scintillations are quite comparable at high (Tromso) and polar (Svalbard) latitudes».
  
  Reviewer:
  «Line 195 (Daytime polar cap patches): For a CIR event, I am interested to know the approximate solar wind properties. In particular, what was the maximum value of the solar wind speed. It is not required to include a plot of the solar wind data, just state the value in the text».
  Answer: The solar wind speed for this event was about 390 km/s according to the OMNI database. But we have decided to remove this event as mentioned above due to large numbers of Figures and Tables. The characteristics of the 7 November 2013 event are presented now in the Table 4.
  
  Reviewer:
  «Lines 204-205: "cusp precipitation has stronger influence on GPS phase scintillation when it combined with the PCP. Our analyses also confirm this finding." You stated that "PCPes were registered in time interval 06.00-12.00 UT".
  
  …
  You have not stated the presence of precipitation at any time interval for this event. To reach your conclusion, did you make assumptions regarding the particle precipitation also for times when EISCAT did not directly observe precipitation? If so, please state the assumptions clearly in the text».
  
  Answer: Yes, you are right. For the event 7 November 2013 there is no particle precipitation during PCPs appearance. We remove Figure 8 due to large number of Figures and Tables.
  We rewrite this sentence (Line 261):
  « [Jin et al., 2017] investigated the GPS scintillations around cusp region and found that cusp precipitation has stronger influence on GPS phase scintillation when it combined with the PCPs. Our analyses show that daytime PCPs can produce stronger GPS phase scintillations than cusp/dayside precipitations».
  
  Reviewer:
  «Line 217: "module of the interplanetary magnetic field". Do you mean "magnitude of the interplanetary magnetic field"?».
  Answer: Yes, we mean magnitude of the interplanetary magnetic field. It is corrected. Line 196.
  
  Reviewer:
  «Line 219: Please spell out the abbreviation "SSC" in full at its first occurrence.
  Answer: We have indicated (Line 192).
  
  Reviewer:
  «Lines 243-244: "Possibly low values of amplitude scintillations at high latitudes are caused by the low elevation angles of GPS satellites at these regions". Please explain how observing at low elevation decreases the amount/magnitude of amplitude scintillation».
  Answer: This is only hypothesis. The plasma irregularities producing high-latitudes scintillations mainly formed along the geomagnetic field. At polar latitudes (near Svalbard) the geomagnetic field is close to vertical. So radiowave beam of GPS satellite penetrate through the ionosphere not along geomagnetic field. If we will have satellite with higher inclination angle the amplitude scintillation possibly can be detected. But this hypothesis needs to be tested. Line 290.
  
  Citation: https://doi.org/10.5194/angeo-2020-93-AC1
- AC2: 'Reply on RC1', Vladimir Belakhovsky, 06 Apr 2021
  
  We thank to the reviewer for the careful reading of our paper and helpful comments and remarks. Below are our answers to the remarks.
  
  Reviewer:
  «It is mentioned in the paper that 150 events have been considered, but there is only data from 5 events used in the paper. This raises doubt about whether the conclusion in the paper is generally valid, or only valid for this set of events. If 150 events have been analyzed, I would expect there to be at least some summary of the results included in the paper. It should include at least some key statistics derived from the analysis of the 150 events…»
  
  Answer:
  To describe statistically our research we have added 4 tables and section 3.5 to the paper. These tables include 121 events: 33 cusp/daytime precipitations, 36 substorms, 14 SSC events and 38 PCPs. It is less than 150 events which were mentioned in previous variant of the paper. But the tables already are quite large. We have selected mainly interesting events. Of course we have considered much more event for the 2010-2017 years. But the paper cannot include all of these events. The data from the tables confirms our case studies.
  We have decided to remove Figure 8 because now it is too much Figures and Tables. At the same time the dayside PCPs were also observed during another considered event 22 January 2012. The characteristics of the 7 November 2013 event are presented now in the Table 4. We have changed numeration of the Figures.
  We also have combined Figures 5 and Figure 6 on same plot to better compare the level of GPS scintillations on Svalbard and in Scandinavia. Now it is Figure 7.
  
  Reviewer:
  «There are numerous language issues throughout the paper. While the errors are not critical for the understanding of the paper, it would be significantly improved by careful proof-reading».
  Answer:
  We have tried to remove the language issues.
  
  Reviewer:
  «Lines 84-91: You define the phase scintillation index twice. Fortunately, the definitions are the same. But you may want to delete one of the sentences».
  Answer: We have corrected. Now we present only one definition of the phase scintillation index in the paper (Line 92)
  
  Reviewer:
  «Line 125: The reference [Belakhovsky et al., 2019] does not match the reference list…»
  
  Answer: We have corrected. Lines 130.
  
  Reviewer:
  «Line 147: "Possibly it’s due to the field of view of EISCAT radar not coincides with the field of view of GPS receivers." It is clear that the radar and the GNSS observations cannot fully overlap.
  
  But, is it possible to provide any more assessment of the degree of overlap?...
  …for the case studies, have you checked which of the satellites (if any) actually passed through the region of the ionosphere that is observed by EISCAT?».
  
  Answer: To show the degree of overlap of GPS satellites and EISCAT 42m radar we have plotted on Figure 4 the ionosphere projections of the GPS satellites, location of EISCAT radar and location of GPS receiver at NYA. Of course for the large scale disturbances (300-500 km and more) which was considered in the paper this overlap is not significant because several ionosphere projection points of GPS satellites (2-4) are always located near field of view of the EISCAT radar. So practically we have continuous measurements of the ionosphere parameters determined by the NYA GPS receiver.
  
  Reviewer:
  «Lines 187-188:
  
  "The PCP is also identified in the aurora intensity variations as forms propagating from the polar to low latitudes in 630.0 nm (red line) emission (Figure 7) at 19.00-23.00 UT according to LYR all-sky camera observations."
  The keogram seems to show multiple patches, not just one.
  
  Are you using the abbreviation "PCP" to refer to a multitude of patches, instead of just one patch?... ».
  Answer: It is corrected. Now we use abbreviation "PCPs" everywhere in the paper.
  
  Reviewer:
  «Lines 191-192: "At latitudes of SKN (TRO) stations the PCP manifests itself as a long lasting substorm (more than 4 hours duration) with the amplitude 200-250 nT."
  A patch is not the same as a substorm…»
  Answer: Yes, we agree. We rewrote this sentence.
  Lines 217:
  «During appearance of PCPs near the Svalbard at latitudes of the SKN (TRO) stations a long lasting substorm (more than 4 hours duration) with the amplitude 200-250 nT is observed».
  Line 319:
  «Comparison of the EISCAT observations on Svalbard and in Tromso shows that during PCPs appearance on Svalbard a typical substorm at lower latitudes (Tromso) was observed. The level of the phase scintillations are quite comparable at high (Tromso) and polar (Svalbard) latitudes».
  
  Reviewer:
  «Line 195 (Daytime polar cap patches): For a CIR event, I am interested to know the approximate solar wind properties. In particular, what was the maximum value of the solar wind speed. It is not required to include a plot of the solar wind data, just state the value in the text».
  Answer: The solar wind speed for this event was about 390 km/s according to the OMNI database. But we have decided to remove this event as mentioned above due to large numbers of Figures and Tables. The characteristics of the 7 November 2013 event are presented now in the Table 4.
  
  Reviewer:
  «Lines 204-205: "cusp precipitation has stronger influence on GPS phase scintillation when it combined with the PCP. Our analyses also confirm this finding." You stated that "PCPes were registered in time interval 06.00-12.00 UT".
  
  …
  You have not stated the presence of precipitation at any time interval for this event. To reach your conclusion, did you make assumptions regarding the particle precipitation also for times when EISCAT did not directly observe precipitation? If so, please state the assumptions clearly in the text».
  
  Answer: Yes, you are right. For the event 7 November 2013 there is no particle precipitation during PCPs appearance. We remove Figure 8 due to large number of Figures and Tables.
  We rewrite this sentence (Line 261):
  « [Jin et al., 2017] investigated the GPS scintillations around cusp region and found that cusp precipitation has stronger influence on GPS phase scintillation when it combined with the PCPs. Our analyses show that daytime PCPs can produce stronger GPS phase scintillations than cusp/dayside precipitations».
  
  Reviewer:
  «Line 217: "module of the interplanetary magnetic field". Do you mean "magnitude of the interplanetary magnetic field"?».
  Answer: Yes, we mean magnitude of the interplanetary magnetic field. It is corrected. Line 196.
  
  Reviewer:
  «Line 219: Please spell out the abbreviation "SSC" in full at its first occurrence.
  Answer: We have indicated (Line 192).
  
  Reviewer:
  «Lines 243-244: "Possibly low values of amplitude scintillations at high latitudes are caused by the low elevation angles of GPS satellites at these regions". Please explain how observing at low elevation decreases the amount/magnitude of amplitude scintillation».
  Answer: This is only hypothesis. The plasma irregularities producing high-latitudes scintillations mainly formed along the geomagnetic field. At polar latitudes (near Svalbard) the geomagnetic field is close to vertical. So radiowave beam of GPS satellite penetrate through the ionosphere not along geomagnetic field. If we will have satellite with higher inclination angle the amplitude scintillation possibly can be detected. But this hypothesis needs to be tested. Line 290.
  
  Citation: https://doi.org/10.5194/angeo-2020-93-AC2
RC2:
'Comment on angeo-2020-93', Anonymous Referee #2, 07 Apr 2021
Dear Authors,

your work is surely valuable because it deals with a quite high number of events detected at high latitudes, from auroral latitudes to polar cap. I had the same comment posted by the other referee inviting you to report the statistics on the entire events collection and I read you modified the manuscript accordingly.

My main concern is about the meaning of the phase scintillation index.

Electron density gradients result in refractive index variations, which give rise to refraction and diffraction processes distorting the original wave front of the received signals. Scintillations are the effects due to the diffractive (stochastic) process, while the bulk of the refraction is considered deterministic (Mushini et al., 2012 and references therein). The threat to GNSS signals is due to the stochastic contribution, identified by the S4 enhancement. What you are observing in your case studies are likely phase fluctuations (misinterpreted as phase scintillations), not (or weakly) accompanied by S4 (light) enhancements. This misinterpretation derives from the phase detrending adopted to compute _φ that is often not appropriate to account for the plasma velocity at high latitudes ( Spogli et al., 2021; Ghobadi et al., 2020 and references therein). Indeed, a choice of the cutoff frequency at 0.1 Hz is often not appropriate to remove the plasma convection velocity at high latitudes. So, what you are observing with _φ maxima are phase fluctuations due to the presence of large scale irregularities (above the Fresnel radius) and cannot be considered scintillations. De Franceschi et al. (2009) have demonstrated the difficulty in finding the optimum cutoff frequency for statistical studies and they proposed to look for actual scintillations investigating the simultaneous occurrence of S4 and _φ increase.

I am aware of the confusion in the terms used in a number of papers (also among the ones you cite) but I think that after the recent debate (Rino et al., 2019; McCaffrey and Jayachandran, 2019; Ghobadi et al., 2020; Spogli et al., 2021) it is important not to call phase scintillation what is phase fluctuation. It is important also to stress that the scintillations are the real threats to GNSS signal propagation.

This implies a substantial revision of your paper following two possible (alternative) options: keep the selected events considering them as phase fluctuation occurrences, or select other events in which S4 levels (considering the elevation and the projection to the vertical, see comment below) greater than 0.25 occur. For the statistical description you could use the method adopted by De Franceschi et al. (2019).

Another option, much more demanding, is to recalculate the _φ finding the optimal cutoff frequency case by case. This could be also carried on in a future work.

About the way you analyse the GNSS data:

Are you applying an elevation mask to minimize the multipath?

Did you projected the scintillation indices to the vertical to minimize the geometrical effects as suggested by Spogli et al. (2013)?

Minor comments:

Line 45: “…in the magnetosphere tail and then released into the auroral ionosphere”

Lines 51-52: “At polar latitudes polar cap patches can produce severe ionosphere disturbances.”

Lines 195-196: Reword the sentence “It was also analyzed the influence of GPS signals scintillations the polar cap patches propagating on the dayside”.

Lines 243-244: “Possibly low values of amplitude scintillations at high latitudes are caused by the low elevation angles of GPS satellites at these regions.”

The issue is the opposite! When the elevation is low the S4 could be higher because of the contributions from longer path from the transmitter to the receiver.

References

De Franceschi et al., “The ionospheric irregularities climatology over Svalbard from solar cycle 23”. Sci. Rep., vol. 9, no. 1, pp. 1-14, 2019.

Ghobadi, H.; Spogli, L.; Alfonsi, L.; Cesaroni, C.; Cicone, A.; Linty, N.; Romano, V.; Cafaro, M. Disentangling ionospheric refraction and diffraction effects in GNSS raw phase through fast iterative filtering technique. GPS Solutions 2020, 24, 1-13.

M. McCaffrey & P. T. Jayachandran, “Determination of the Refractive Contribution to GPS Phase “Scintillation”. J Geophys Res-Space, vol. 124, no. 2, pp. 1454-1469, 2019.

C. Mushini et al., “Improved amplitude-and phase-scintillation indices derived from wavelet detrended high-latitude GPS data”, GPS solut., vol. 16, no. 3, pp. 363-373, 2012.

L. Rino et al., “Stochastic TEC structure characterization.” J Geophys Res-Space, vol. 124, no. 12, pp. 10571-10579, 2019.

Spogli, L.; Alfonsi, L.; Cilliers, P.J.; Correia, E.; De Franceschi, G.; Mitchell, C.N.; Romano, V.; Kinrade, J.; Cabrera, M.A. GPS scintillations and total electron content climatology in the southern low, middle and high latitude regions. 2013, , 0220

Spogli et al., "Adaptive Phase Detrending for GNSS Scintillation Detection: A Case Study Over Antarctica," in IEEE Geoscience and Remote Sensing Letters, doi: 10.1109/LGRS.2021.3067727.
Citation: https://doi.org/10.5194/angeo-2020-93-RC2
- AC3: 'Reply on RC2', Vladimir Belakhovsky, 22 Apr 2021
  
  We thank to the referee 2 for interest to our paper, careful reading and helpful comments and remarks. Below are our answers to the remarks.
  Edition board asks us not to upload corrected manuscript with this answer. We will upload the corrected manuscript to the special link as I understand.
  
  Reviewer:
  «My main concern is about the meaning of the phase scintillation index…»
  Answer:
  We agree with the referee that the problem of the GNSS signal detrending for the high-latitude disturbances is actual, the fixed cut-off frequency is not always able to remove the refractive (deterministic) effects at high latitudes. At the same time the choice of the optimal cut-off frequency is still open problem.
  There are a lot of papers and results since [Fremouw et al., 1978] where the standard cut-off frequency (0.1 Hz) was used. So the classic definition of the phase and amplitude scintillation indexes already is established in literature.
  The best way in this situation in our opinion is to introduce new scintillations indexes. Therefore, there will be clarity through the researchers. For example [Mushini, 2012] introduced improved phase-scintillation index, σ_chain, [Forte, 2005] introduced Sφ index.
  We kept our events and we used the standard cut-off frequency (0.1 Hz) for the detrending. We mentioned in the paper about problem of the GPS signal detrending, we wrote that the term “phase scintillation index” used in our study means the phase fluctuations due to the presence of large-scale irregularities.
  
  - Fremouw E.J., Leadabrand R.L., Livingston R.C., Cousins M.D., Rino C.L., Fair B.C., Long R.A. Early results from the DNA wideband satellite experiment—complex-signal scintillation // Radio Sci. 13(1):167–187. doi:10.1029/RS013i001p0016, 1978.
  - Mushini S.C., Jayachandran P.T., Langley R.B., MacDougall J.W., Pokhotelov D. Improved amplitude and phase-scintillation indices derived from wavelet detrended high-latitude GPS data // GPS Solutions, vol. 16, no. 3, pp. 363-373, https://doi.org/10.1007/s10291-011-0238-4, 2012.
  - Forte B. Optimum detrending of raw GPS data for scintillation measurements at auroral latitudes // Journal of Atmospheric and Solar-Terrestrial Physics. 67. 1100–1109. 2005.
  
  This text we pasted to the Introduction:
  «In order to calculate scintillation indices a long-term trend caused by the satellite motion in relation to the receiver and ionosphere changes needs to be removed. A standard cutoff frequency (0.1 Hz) is commonly used for signal detrending since [Fremouw et al., 1978]. This cutoff frequency is adequate for the equatorial and midlatitude ionosphere. But the high-latitude ionoshere is characterized by the high and variable ionospheric drift velocity (~100 m/s–1 500 m/s). The value of the cutoff frequency affects on the phase scintillation index. So it often leads to the strong phase scintillations without amplitude scintillation. Some researchers to solve this problem introduce new scintillations indexes: [Mushini et al., 2012] introduced improved phase-scintillation index (σ_chain), [Forte, 2005] introduced Sφ index. The fast iterative filtering signal decomposition technique was used to find optimal cutoff frequency [Ghobadi et al., 2020; Spogli, 2021].
  In our paper we use the standard detrending (0.1 Hz) in order to be able to compare our results with previous results obtained with using GPS receiver of Oslo University on Svalbard [Clausen et al., 2016; Jin et al., 2015, 2016, 2017, 2018]. So term “phase scintillation index” used in our study means the phase fluctuations due to the presence of large-scale irregularities (above the Fresnel radius)».
  
  Reviewer:
  «Are you applying an elevation mask to minimize the multipath?
  Did you projected the scintillation indices to the vertical to minimize the geometrical effects as suggested by Spogli et al. (2013)»
  Answer:
  We did not apply an elevation mask to minimize the multipath. We did not project the scintillation indices to the vertical to minimize the geometrical effects.
  We use the data of the same GPS receiver on Svalbard that was used in other papers published in rating journals. So we apply the same data processing in order to be able to compare our results with results in other papers.
  
  - Clausen L. B. N., Moen J. I., Hosokawa K., Holmes J. M.: GPS scintillations in the high latitudes during periods of dayside and nightside reconnection, J. Geophys. Res., 121, 3293–3309, https://doi.org/10.1002/2015JA022199, 2016.
  - Jin Y., Moen J. I., and Miloch W. J.: On the collocation of the cusp aurora and the GPS phase scintillation: A statistical study, J. Geophys. Res., 120, 9176–9191, https://doi.org/10.1002/2015JA021449, 2015.
  - Jin Y., Moen J. I., Miloch W. J., Clausen L. B. N., and Oksavik K.: Statistical study of the GNSS phase scintillation associated with two types of auroral blobs, J. Geophys. Res., 121, 4679–4697, https://doi.org/10.1002/2016JA022613, 2016.
  - Jin Yaqi, Zhou Xiaoyan, Moen Jøran I., Hairston Marc. The auroral ionosphere TEC response to an interplanetary shock // Geophysical Research Letters, Vol. 43, Issue 5, pp. 1810-1818. https://doi.org/10.1002/2016GL067766. 2016.
  - Jin Y., Moen J. I., Oksavik K., Spicher A., Clausen L. B.N., Miloch W. J. GPS scintillations associated with cusp dynamics and polar cap patches: J. Space Weather Space Clim., 7, A23. https://doi.org/10.1051/swsc/2017022, 2017.
  - Jin Y., Oksavik K. GPS scintillations and losses of signal lock at high latitudes during the 2015 St. Patrick’s Day storm, J. Geophys. Res., 123, 7943–7957, https://doi.org/10.1029/2018JA025933, 2018.
  - Jin Y., Moen J.I., Spicher A., Oksavik K., Miloch W. J., Clausen L. B. N., et al. Simultaneous rocket andscintillation observations of plasmairregularities associated with a reversed ﬂow event in the cusp ionosphere, Journal of Geophysical Research, 124, 7098–7111. https://doi .org/10.1029/2019JA026942. 2019.
  - Chernyshov A.A., Miloch W.J., Jin Y., Zakharov V.I. Relationship between TEC jumps and auroral substorm in the high-latitude ionosphere, Scientific Reports. 10:6363. https://doi.org/10.1038/s41598-020-63422-9. 2020.
  
  Reviewer:
  «Minor comments: …»
  Answer: It is corrected.
  
  Citation: https://doi.org/10.5194/angeo-2020-93-AC3
- AC4: 'Reply on RC2', Vladimir Belakhovsky, 23 Apr 2021
  
  Reviewer:
  «Lines 243-244: “Possibly low values of amplitude scintillations at high latitudes are caused by the low elevation angles of GPS satellites at these regions.”
  The issue is the opposite! When the elevation is low the S4 could be higher because of the contributions from longer path from the transmitter to the receiver».
  Answer:
  This is hypothesis. The plasma irregularities producing high-latitudes scintillations mainly formed along the geomagnetic field. At polar latitudes (near Svalbard) the geomagnetic field is close to vertical. So radiowave beam of GPS satellite penetrate through the ionosphere not along geomagnetic field. If we will have satellite with higher inclination angle the amplitude scintillation possibly can be detected. But this hypothesis needs to be tested.
  
  Citation: https://doi.org/10.5194/angeo-2020-93-AC4

Peer review completion

AR: Author's response | RR: Referee report | ED: Editor decision | EF: Editorial file upload

ED: Reconsider after major revisions (further review by editor and referees) (24 Apr 2021) by Ana G. Elias

AR by Vladimir Belakhovsky on behalf of the Authors (30 Apr 2021)

EF by Svenja Lange (03 May 2021) Manuscript

EF by Svenja Lange (03 May 2021) Author's response

ED: Referee Nomination & Report Request started (04 May 2021) by Ana G. Elias

RR by Anonymous Referee #1 (01 Jun 2021)

RR by Anonymous Referee #2 (03 Jun 2021)

ED: Publish subject to revisions (further review by editor and referees) (04 Jun 2021) by Ana G. Elias

AR by Vladimir Belakhovsky on behalf of the Authors (06 Jun 2021) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (07 Jun 2021) by Ana G. Elias

RR by Anonymous Referee #1 (08 Jun 2021)

ED: Publish as is (10 Jun 2021) by Ana G. Elias

AR by Vladimir Belakhovsky on behalf of the Authors (10 Jun 2021)

Short summary

The high-latitude ionosphere has a very dynamic structure and causes a negative influence via radio wave propagation on, for example, global positioning system (GPS) signals and satellite communication. In this study, we have determined which type of high-latitude ionosphere disturbance has a stronger influence on GPS signals using a GPS receiver on Svalbard. It is shown that a substorm ionosphere disturbance leads to stronger GPS signal fluctuations than other types of ionosphere disturbances.