ScintPi measurements of low-latitude ionospheric irregularity drifts using the spaced-receiver technique and SBAS signals

Gomez Socola, Josemaria; Rodrigues, Fabiano; Wrigth, Isaac; Paulino, Igo; Buriti, Ricardo

doi:https://doi.org/10.5194/egusphere-2024-2262

Preprints

https://doi.org/10.5194/egusphere-2024-2262

Preprints

25 Jul 2024

| 25 Jul 2024

ScintPi measurements of low-latitude ionospheric irregularity drifts using the spaced-receiver technique and SBAS signals

Josemaria Gomez Socola, Fabiano Rodrigues, Isaac Wrigth, Igo Paulino, and Ricardo Buriti

Abstract. Previous efforts have used pairs of closely-spaced specialized receivers to measure Global Navigation Satellite System (GNSS) signals and to estimate ionospheric irregularity drifts. The relatively high cost associated with commercial GNSS-based ionospheric receivers limited somewhat their deployment and the estimation of ionospheric drifts. The development of an alternative low-cost GNSS-based scintillation monitor (ScintPi) motivated us to investigate the possibility of using it to overcome this limitation. ScintPi monitors can observe signals from geostationary satellites, which can greatly simplify the estimation of the drifts. We present results of an experiment to evaluate the use of ScintPi 3.0 to estimate ionospheric irregularity drifts. The experiment consisted of two ScintPi 3.0 deployed in Campina Grande, Brazil (7.213° S, 35.907° W, dip latitude ~14° S). The monitors were spaced by 140 m in the magnetic east-west direction and targeted the estimation of the zonal drifts associated with scintillation-causing equatorial spread F (ESF) irregularities. Routine observations throughout an entire ESF season (September 2022 – April 2023) were made as part of the experiment. We focused on results of irregularity drifts derived from geostationary satellite signals. The results show that the local time variation in the estimated irregularity zonal drifts is in good agreement with previous measurements and with the expected behavior of the background zonal plasma drifts. Our results also reveal a seasonal trend in the irregularity zonal drifts. The trend follows the seasonal behavior of the zonal component of the thermospheric neutral winds as predicted by the Horizontal Wind Model (HMW14). This is explained by the fact that low latitude ionospheric F-region plasma drifts are controlled, in great part, by Pedersen conductivity weighted flux tube integrated zonal neutral winds. The results confirm that ScintPi has the potential to contribute to new, cost-effective measurements of ionospheric irregularity drifts in addition to scintillation and total electron content. Furthermore, the results indicate that these new ScintPi measurements can provide insight into the ionosphere-thermosphere coupling.

Received: 19 Jul 2024 – Discussion started: 25 Jul 2024

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Josemaria Gomez Socola, Fabiano Rodrigues, Isaac Wrigth, Igo Paulino, and Ricardo Buriti

Status: closed

RC1:
'Comment on egusphere-2024-2262', Anonymous Referee #1, 26 Aug 2024
The manuscript presents a new way to estimate the zonal drift velocity of scintillation-related equatorial plasma bubbles by utilizing measurements made by low-cost ground-based receivers. The method to derive the zonal drift is well explained, and the results are interesting and promising. The manuscript is well-organized and will be improved after addressing the following comments.
Line 105, could you use one sentence to summarize the relation between the spacing, irregularity drift, and sampling rate?

In Figure 3, do you use a threshold of |r| to select high correlation data points? If so, would this method be only applicable during strong scintillation conditions?

In Figures 4b and c, while it seems that zonal drifts and winds show similarity in the local time and daily variation, it might be better to have a figure to show the correlation between them.

The IPP calculation considers the altitude of the ionospheric F-region peak density at 450 km. How do different choices of this value impact the estimation of drift velocities? It might be better to have some discussion or background literature on this.

Minor comments:
Line 267-268, please check “It also shows that scintillation occurrence in most days”.

Line 274, please check “Eastward drifts around 20-80 m/s in the midnight and post-midnight sector”.

Line 180, is there any reason to use west receiver instead of east receiver?

Line 216, could you explain more for the two values in θ?
Citation: https://doi.org/10.5194/egusphere-2024-2262-RC1
- AC1: 'Reply on RC1', Josemaria Gomez Socola, 05 Dec 2024
  
  Dear Referee#1, below you will find the answers to your questions and comments.
  1. Line 105, could you use one sentence to summarize the relation between the spacing, irregularity drift, and sampling rate?
  As requested, we now provide a sentence in section 2.1 summarizing the relationship between spacing, irregularity drift, and sampling rate.
  2. In Figure 3, do you use a threshold of |r| to select high-correlation data points? If so, would this method be only applicable during strong scintillation conditions?
  Following Referee#1’s comment, we changed the text to clarify that the velocity estimation only uses scintillation patterns with |r| > 0.5. This threshold is now stated in the caption of Figure 3c.
  The method is not only applicable during strong scintillation conditions since selection is based on |r|. Even during moderate or weak scintillation, estimates of irregularity drifts can be made. This is how we obtain irregularity velocities near local midnight when L-Band scintillation is typically weak due to reduced background ionospheric densities. The reduction in scintillation severity late at night while |r| remains high can be seen in Figure 3.
  For completeness, Section 4.1 states that “high cross-correlation coefficients persisted even during weak scintillation activity (i.e., S4 in the 0.2 - 0.4 range), indicating that one can compute irregularity drifts even during periods of weak L-Band scintillation.”
  3. In Figures 4b and c, while it seems that zonal drifts and winds show similarity in the local time and daily variation, it might be better to have a figure to show the correlation between them.
  We appreciate the comment by the Referee#1. We initially considered to show individual curves and correlation values. We quickly realized that it would be misleading to show those since we would be comparing measurements against climatological models. The measurements have day-to-day variabilities and solar flux effects in them. HMW does not take into consideration solar flux. Additionally, the relationship between winds and ionospheric drifts is non-linear through the Pedersen conductivity weighted integral that takes into consideration the contribution of winds at different heights along magnetic field lines. Showing Figures 4(b) and 4(c) is the best compromise we found to those variabilities in local time and season.
  4. The IPP calculation considers the altitude of the ionospheric F-region peak density at 450 km. How do different choices of this value impact the estimation of drift velocities? It might be better to have some discussion or background literature on this.
  This highlights the benefits of using signals from geostationary satellites. Values of the F-region peak density in a realistic range (say, 300 to 600 km) have minimal impact on the estimation of drift velocities. Of relevance for our calculation are the magnetic field vector components, Bx, By, and Bz at the IPP location, which do not change much for typical F-region heights. This characteristic was leveraged to estimate the scattering height of irregularities by comparing velocities derived from geostationary and moving satellites, as demonstrated by Cerruti et al., 2006. Also, following referee#1’s suggestion, we added clarification about the impact of F-region height in the drift calculation in Section 3.1.
  Minor comments:
  1. Line 267-268, please check “It also shows that scintillation occurrence in most days”.
  Corrected. Thank you.
  2. Line 274, please check “Eastward drifts around 20-80 m/s in the midnight and post-midnight sector”.
  Corrected. Thank you.
  3. Line 180, is there any reason to use west receiver instead of east receiver?
  The west receiver seems to be less susceptible to multipath.
  4. Line 216, could you explain more for the two values in θ?
  The primary reason is that θ is the azimuth angle of the geostationary satellite with respect to magnetic north (definition earlier on in Section 3.1). The magnetic declination for our setup is -21.39 deg. and the geographic azimuth angle of the geostationary satellite is 84deg,. Therefore, θ = 84 + 21.39 = 105.39 deg. To avoid confusion, we now only provide the final θ value, 105.39 deg.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2262-AC1
RC2:
'Comment on egusphere-2024-2262', Anonymous Referee #2, 29 Oct 2024
The manuscript presents a technique for estimating the drift velocity of equatorial plasma bubbles using measurements made by low-cost ground-based receivers. The distinguishing feature of this work is the ability of these monitors to capture signals transmitted by geostationary satellites. The main advantage of using geostationary signals is that the satellite's velocity is not considered in the calculation of ionospheric plasma bubble drift. The methodology is well explained, and the results are interesting. The manuscript is well-structured and will be enhanced after incorporating the following comments.

In equation (6), a value of 0.899 was obtained. I suggest including the values used to reach this result, perhaps in a table.

Was a threshold value applied to define a high correlation |r|? Figure 3c suggests that this threshold might be 0.5. If so, this should be explicitly stated in the manuscript.

The altitude of the ionospheric F-region peak density is mentioned as 450 km several times in the manuscript. Why were 450 km chosen? If there is a specific reason, please provide justification in the text.
Citation: https://doi.org/10.5194/egusphere-2024-2262-RC2
- AC2:
  'Reply on RC2', Josemaria Gomez Socola, 05 Dec 2024
  Dear Referee#2, below you will find the answers to your questions and comments.
  1. In equation (6), a value of 0.899 was obtained. I suggest including the values used to reach this result, perhaps in a table.
  We agree that showing these values clarifies our calculations. Following your suggestion we now specify these values in Equation 6 and remind the reader that the variables are described in Section 2.2.
  2. Was a threshold value applied to define a high correlation |r|? Figure 3c suggests that this threshold might be 0.5. If so, this should be explicitly stated in the manuscript
  In the caption of Figure 3, we have made more explicit the threshold of |r| > 0.5. Additionally, in Section 4.1, we also add an explanation that these correlation values can be obtained even during weak scintillation conditions.
  3. The altitude of the ionospheric F-region peak density is mentioned as 450 km several times in the manuscript. Why were 450 km chosen? If there is a specific reason, please provide justification in the text
  While 350 km is often used, there are studies showing that the optimal F-region height used for IPP calculation can vary between 400 km and 500 km for low and mid-latitude regions (Nava et al., 2007). Also, there are studies of hmF2 that show pre-midnight values around 450 km (e.g., Abdu et al., 2006) for low-latitude stations. The IPP was computed using a mean ionospheric altitude of 450 km which is thought to be adequate for the location and solar flux conditions of the observations. More importantly, different choices of ionospheric altitude (say, 300 to 600 km) do not affect the drift estimates and results. Following the referee#2’s comments, these are now better explained in Section 3.1.
  References:
  Abdu, M. A., Batista, I. S., Reinisch, B. W., Sobral, J. H. A., & Carrasco, A. J. (2006). Equatorial F region evening vertical drift, and peak height, during southern winter months: A comparison of observational data with the IRI descriptions. Advances in Space Research, 37(5), 1007-1017.
  
  Nava B, Radicella SM, Leitinger R, Coïsson P (2007) Use of total electron content data to analyze ionosphere electron density gradients. Adv Space Res 39(8):1292–1297. doi:10.1016/j.asr.2007.01.041
  
  Citation: https://doi.org/10.5194/egusphere-2024-2262-AC2
RC3:
'Comment on egusphere-2024-2262', Anonymous Referee #3, 12 Nov 2024

The manuscript presents the measurements of the zonal drift of the ionospheric irregularities using two low-cost GNSS-based scintillation monitor(ScintPi) installed 140 m apart in the magnetic east-west direction. The measurements were performed from September 2022 to April 2023 at Campina Grande Brazil using signal from a Geostationary satellite. Even though the data resolution is low 20 HZ and with a low C/N signal the irregularity zonal drifts determined by these ScintPi receivers were in good agreement with previous works that used 50 Hz data resolution and using very expensive GNSS receivers. The methodology used is well explained and the authors even found a seasonal trend in the irregularity zonal drifts that was well correlated with the behavior of the zonal component of the thermospheric neutral winds as predicted by the Horizontal 340 Wind Model (HMW14). Based on the above comment I recommend the manuscript to be accepted as it is.

Citation: https://doi.org/10.5194/egusphere-2024-2262-RC3
- AC3: 'Reply on RC3', Josemaria Gomez Socola, 05 Dec 2024
  
  Dear Referee#3,
  
  Thank you for your comments and positive feedback on our manuscript. We are glad that you found our methodology well-explained. Your recognition of the observed seasonal trends and their correlation with the Horizontal Wind Model (HWM14) is particularly encouraging. We are also glad to see a good performance of the low-cost ScintPi receivers. We sincerely appreciate your recommendation for acceptance.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2262-AC3

Status: closed

RC1:
'Comment on egusphere-2024-2262', Anonymous Referee #1, 26 Aug 2024
The manuscript presents a new way to estimate the zonal drift velocity of scintillation-related equatorial plasma bubbles by utilizing measurements made by low-cost ground-based receivers. The method to derive the zonal drift is well explained, and the results are interesting and promising. The manuscript is well-organized and will be improved after addressing the following comments.
Line 105, could you use one sentence to summarize the relation between the spacing, irregularity drift, and sampling rate?

In Figure 3, do you use a threshold of |r| to select high correlation data points? If so, would this method be only applicable during strong scintillation conditions?

In Figures 4b and c, while it seems that zonal drifts and winds show similarity in the local time and daily variation, it might be better to have a figure to show the correlation between them.

The IPP calculation considers the altitude of the ionospheric F-region peak density at 450 km. How do different choices of this value impact the estimation of drift velocities? It might be better to have some discussion or background literature on this.

Minor comments:
Line 267-268, please check “It also shows that scintillation occurrence in most days”.

Line 274, please check “Eastward drifts around 20-80 m/s in the midnight and post-midnight sector”.

Line 180, is there any reason to use west receiver instead of east receiver?

Line 216, could you explain more for the two values in θ?
Citation: https://doi.org/10.5194/egusphere-2024-2262-RC1
- AC1: 'Reply on RC1', Josemaria Gomez Socola, 05 Dec 2024
  
  Dear Referee#1, below you will find the answers to your questions and comments.
  1. Line 105, could you use one sentence to summarize the relation between the spacing, irregularity drift, and sampling rate?
  As requested, we now provide a sentence in section 2.1 summarizing the relationship between spacing, irregularity drift, and sampling rate.
  2. In Figure 3, do you use a threshold of |r| to select high-correlation data points? If so, would this method be only applicable during strong scintillation conditions?
  Following Referee#1’s comment, we changed the text to clarify that the velocity estimation only uses scintillation patterns with |r| > 0.5. This threshold is now stated in the caption of Figure 3c.
  The method is not only applicable during strong scintillation conditions since selection is based on |r|. Even during moderate or weak scintillation, estimates of irregularity drifts can be made. This is how we obtain irregularity velocities near local midnight when L-Band scintillation is typically weak due to reduced background ionospheric densities. The reduction in scintillation severity late at night while |r| remains high can be seen in Figure 3.
  For completeness, Section 4.1 states that “high cross-correlation coefficients persisted even during weak scintillation activity (i.e., S4 in the 0.2 - 0.4 range), indicating that one can compute irregularity drifts even during periods of weak L-Band scintillation.”
  3. In Figures 4b and c, while it seems that zonal drifts and winds show similarity in the local time and daily variation, it might be better to have a figure to show the correlation between them.
  We appreciate the comment by the Referee#1. We initially considered to show individual curves and correlation values. We quickly realized that it would be misleading to show those since we would be comparing measurements against climatological models. The measurements have day-to-day variabilities and solar flux effects in them. HMW does not take into consideration solar flux. Additionally, the relationship between winds and ionospheric drifts is non-linear through the Pedersen conductivity weighted integral that takes into consideration the contribution of winds at different heights along magnetic field lines. Showing Figures 4(b) and 4(c) is the best compromise we found to those variabilities in local time and season.
  4. The IPP calculation considers the altitude of the ionospheric F-region peak density at 450 km. How do different choices of this value impact the estimation of drift velocities? It might be better to have some discussion or background literature on this.
  This highlights the benefits of using signals from geostationary satellites. Values of the F-region peak density in a realistic range (say, 300 to 600 km) have minimal impact on the estimation of drift velocities. Of relevance for our calculation are the magnetic field vector components, Bx, By, and Bz at the IPP location, which do not change much for typical F-region heights. This characteristic was leveraged to estimate the scattering height of irregularities by comparing velocities derived from geostationary and moving satellites, as demonstrated by Cerruti et al., 2006. Also, following referee#1’s suggestion, we added clarification about the impact of F-region height in the drift calculation in Section 3.1.
  Minor comments:
  1. Line 267-268, please check “It also shows that scintillation occurrence in most days”.
  Corrected. Thank you.
  2. Line 274, please check “Eastward drifts around 20-80 m/s in the midnight and post-midnight sector”.
  Corrected. Thank you.
  3. Line 180, is there any reason to use west receiver instead of east receiver?
  The west receiver seems to be less susceptible to multipath.
  4. Line 216, could you explain more for the two values in θ?
  The primary reason is that θ is the azimuth angle of the geostationary satellite with respect to magnetic north (definition earlier on in Section 3.1). The magnetic declination for our setup is -21.39 deg. and the geographic azimuth angle of the geostationary satellite is 84deg,. Therefore, θ = 84 + 21.39 = 105.39 deg. To avoid confusion, we now only provide the final θ value, 105.39 deg.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2262-AC1
RC2:
'Comment on egusphere-2024-2262', Anonymous Referee #2, 29 Oct 2024
The manuscript presents a technique for estimating the drift velocity of equatorial plasma bubbles using measurements made by low-cost ground-based receivers. The distinguishing feature of this work is the ability of these monitors to capture signals transmitted by geostationary satellites. The main advantage of using geostationary signals is that the satellite's velocity is not considered in the calculation of ionospheric plasma bubble drift. The methodology is well explained, and the results are interesting. The manuscript is well-structured and will be enhanced after incorporating the following comments.

In equation (6), a value of 0.899 was obtained. I suggest including the values used to reach this result, perhaps in a table.

Was a threshold value applied to define a high correlation |r|? Figure 3c suggests that this threshold might be 0.5. If so, this should be explicitly stated in the manuscript.

The altitude of the ionospheric F-region peak density is mentioned as 450 km several times in the manuscript. Why were 450 km chosen? If there is a specific reason, please provide justification in the text.
Citation: https://doi.org/10.5194/egusphere-2024-2262-RC2
- AC2:
  'Reply on RC2', Josemaria Gomez Socola, 05 Dec 2024
  Dear Referee#2, below you will find the answers to your questions and comments.
  1. In equation (6), a value of 0.899 was obtained. I suggest including the values used to reach this result, perhaps in a table.
  We agree that showing these values clarifies our calculations. Following your suggestion we now specify these values in Equation 6 and remind the reader that the variables are described in Section 2.2.
  2. Was a threshold value applied to define a high correlation |r|? Figure 3c suggests that this threshold might be 0.5. If so, this should be explicitly stated in the manuscript
  In the caption of Figure 3, we have made more explicit the threshold of |r| > 0.5. Additionally, in Section 4.1, we also add an explanation that these correlation values can be obtained even during weak scintillation conditions.
  3. The altitude of the ionospheric F-region peak density is mentioned as 450 km several times in the manuscript. Why were 450 km chosen? If there is a specific reason, please provide justification in the text
  While 350 km is often used, there are studies showing that the optimal F-region height used for IPP calculation can vary between 400 km and 500 km for low and mid-latitude regions (Nava et al., 2007). Also, there are studies of hmF2 that show pre-midnight values around 450 km (e.g., Abdu et al., 2006) for low-latitude stations. The IPP was computed using a mean ionospheric altitude of 450 km which is thought to be adequate for the location and solar flux conditions of the observations. More importantly, different choices of ionospheric altitude (say, 300 to 600 km) do not affect the drift estimates and results. Following the referee#2’s comments, these are now better explained in Section 3.1.
  References:
  Abdu, M. A., Batista, I. S., Reinisch, B. W., Sobral, J. H. A., & Carrasco, A. J. (2006). Equatorial F region evening vertical drift, and peak height, during southern winter months: A comparison of observational data with the IRI descriptions. Advances in Space Research, 37(5), 1007-1017.
  
  Nava B, Radicella SM, Leitinger R, Coïsson P (2007) Use of total electron content data to analyze ionosphere electron density gradients. Adv Space Res 39(8):1292–1297. doi:10.1016/j.asr.2007.01.041
  
  Citation: https://doi.org/10.5194/egusphere-2024-2262-AC2
RC3:
'Comment on egusphere-2024-2262', Anonymous Referee #3, 12 Nov 2024

The manuscript presents the measurements of the zonal drift of the ionospheric irregularities using two low-cost GNSS-based scintillation monitor(ScintPi) installed 140 m apart in the magnetic east-west direction. The measurements were performed from September 2022 to April 2023 at Campina Grande Brazil using signal from a Geostationary satellite. Even though the data resolution is low 20 HZ and with a low C/N signal the irregularity zonal drifts determined by these ScintPi receivers were in good agreement with previous works that used 50 Hz data resolution and using very expensive GNSS receivers. The methodology used is well explained and the authors even found a seasonal trend in the irregularity zonal drifts that was well correlated with the behavior of the zonal component of the thermospheric neutral winds as predicted by the Horizontal 340 Wind Model (HMW14). Based on the above comment I recommend the manuscript to be accepted as it is.

Citation: https://doi.org/10.5194/egusphere-2024-2262-RC3
- AC3: 'Reply on RC3', Josemaria Gomez Socola, 05 Dec 2024
  
  Dear Referee#3,
  
  Thank you for your comments and positive feedback on our manuscript. We are glad that you found our methodology well-explained. Your recognition of the observed seasonal trends and their correlation with the Horizontal Wind Model (HWM14) is particularly encouraging. We are also glad to see a good performance of the low-cost ScintPi receivers. We sincerely appreciate your recommendation for acceptance.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2262-AC3

Josemaria Gomez Socola, Fabiano Rodrigues, Isaac Wrigth, Igo Paulino, and Ricardo Buriti

Viewed

Total article views: 359 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
249	88	22	359	10	10

HTML: 249
PDF: 88
XML: 22
Total: 359
BibTeX: 10
EndNote: 10

Views and downloads (calculated since 25 Jul 2024)

Month	HTML	PDF	XML	Total
Jul 2024	59	21	3	83
Aug 2024	59	22	12	93
Sep 2024	31	7	2	40
Oct 2024	31	7	1	39
Nov 2024	27	4	1	32
Dec 2024	42	27	3	72

Cumulative views and downloads (calculated since 25 Jul 2024)

Month	HTML	PDF	XML	Total
Jul 2024	59	21	3	83
Aug 2024	59	22	12	93
Sep 2024	31	7	2	40
Oct 2024	31	7	1	39
Nov 2024	27	4	1	32
Dec 2024	42	27	3	72

Viewed (geographical distribution)

Total article views: 331 (including HTML, PDF, and XML) Thereof 331 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 13 Dec 2024

Short summary

New low-cost, off-the-shelf GNSS receivers enable the estimation of zonal ionospheric irregularity drifts using the "scintillation spaced receiver" technique, which was previously tested only with commercial GNSS receivers. Despite their low C/No resolution (1 dB-Hz), we demonstrate that the raw data obtained from the receivers can be used to estimate irregularity velocities. Furthermore, our observations are consistent with the behavior of an empirical model of the thermospheric winds.


Total:	0
HTML:	0
PDF:	0
XML:	0