the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 12 May 2025
Apparent vertical ionospheric drift: A Comparative Assessment of Digisonde and Ionogram-Based Methods
Abstract. Reliable estimation of vertical plasma drift in the ionosphere is crucial for interpreting ionospheric dynamics and enhancing the accuracy of space weather models. This study provides a comparative assessment of direct Digisonde Drift Measurements (DDM) and indirect ionogram-based methods using parameters such as hmF2, h′F2, h′(3.5 MHz), and h′(0.8foF2). Two high cadence measurement campaigns were conducted at the mid-latitude observatory in Pruhonice, Czech Republic, during different phases of the solar cycle. The analysis focuses on evaluating measurement consistency, temporal coherence, and the influence of sampling step and averaging strategy on drift estimation. While DDM yields stable and robust results even at one-minute resolution, ionogram-derived methods are strongly affected by measurement uncertainty and ambiguity in virtual height interpretation – particularly at short time scales. However, at night, all methods converge when a 15-minute time interval is consistently applied both as the computation step and for subsequent smoothing. Under these conditions, coherent wave-like features in the vertical drift are reliably captured. The study outlines the strengths and limitations of each technique and provides recommendations for optimizing temporal resolution in ionospheric drift measurements, supporting improved methodology for future observational campaigns and model validation.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1054 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1054 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-1811', Anonymous Referee #1, 04 Jun 2025
-
AC1: 'Reply on RC1', Daniel Kouba, 19 Jun 2025
We would like to sincerely thank the reviewer for their thorough and thoughtful assessment of our manuscript. We appreciate the critical insights and constructive suggestions, which helped us identify several areas for improvement. We have carefully considered each comment and revised the manuscript accordingly. Below, we provide a detailed response to all points raised.
Reviewer Comment: The scientific value and methodological rigor of the manuscript are questionable in its current form. The study relies on only two days of observational data to evaluate several methods for estimating vertical plasma drift, comparing them to a Doppler-based method that appears relatively more reliable. For any objective reader their results, shown, for example, in Figs. 5 and 6, is proof that indirect methods cannot be used for any serious science...
Author Response: We acknowledge the reviewer’s concern regarding the limited data set. We agree that a comprehensive validation of different methods for estimating vertical plasma drift would ideally require a much larger set of observations, covering various seasons, levels of geomagnetic activity, and different geographic locations.
However, the objective of our study was not to provide a complete characterization of each method under all possible conditions. Instead, we deliberately chose a controlled, quiet period at a mid-latitude station and performed specially configured high-cadence measurements—including both ionograms and drift observations. This setting allowed us to explore the intrinsic behavior of the methods under idealized conditions. Our reasoning is that if significant inconsistencies arise even under quiet conditions, the discrepancies will likely be amplified during geomagnetically disturbed periods.
Furthermore, this study required manual post-processing of ionograms to extract height parameters for the different indirect methods, which is labor-intensive due to the high temporal resolution. While our data set is limited in scope, the added value lies in its temporal granularity and the ability to perform direct side-by-side comparisons on a minute-by-minute basis.
We do not claim to generalize the results to all geophysical conditions but aim to highlight potential methodological issues and limitations. We believe this is a necessary step toward a more extensive evaluation of vertical drift estimation methods.
We also agree with the reviewer that Doppler-based DDM is, when properly processed, the most reliable and consistent method available. However, many studies continue to use indirect methods—particularly when drift measurements are unavailable (e.g., for historical ionogram data). We therefore consider it meaningful to analyze differences among methods and identify conditions under which they produce comparable or diverging results.
Finally, in response to the reviewer’s valid point about the lack of stronger critical commentary in the original manuscript, we have revised the discussion section to include a more explicit evaluation of the limitations of each method. We now provide specific recommendations on their applicability depending on time of day, expected variability, and data availability.
Reviewer Comment: Second, the DDM method is based on least-squares fits and therefore not just offers more physically grounded and statistically supported results but also provides per-measurement error estimates... etc.
Author Response: We thank the reviewer for this helpful suggestion. In the revised version of the manuscript, we now include error estimates for each DDM measurement. These uncertainties are based on the least-squares fit used to determine the drift velocity vector and are visualized in Figures 5 and 6.
We also added a brief discussion explaining that the high-cadence mode used during our campaigns resulted in abnormally short measurement durations and, consequently, a lower number of reflected points in each SKYmap. This limitation naturally increases the uncertainty of the DDM results in our data set compared to standard measurements. However, even under these constrained conditions, the DDM-derived vertical drift estimates remain stable and coherent—highlighting the robustness of this technique. This point has also been clarified in the revised discussion.
Reviewer Comment: L. 69–70: Digisondes can detect specific reflection points and measure Doppler shifts...
It’s worth mentioning here that Digisonde requires dedicated, non-ionogram, fixed-frequency mode...
Author Response: We appreciate this observation. The text has been revised to explicitly state that Digisonde drift measurements are performed in a dedicated mode using fixed-frequency sounding, which does not provide full ionogram data during those intervals.
Reviewer Comment: L. 101: Such a data set allows accurate estimation of the drift velocity vector...
Scientific objectivity requires to mention that this technique was first developed by Wright and Pitteway (1994)...
Author Response: We thank the reviewer for pointing this out. The revised manuscript now includes a citation of the earlier work by Wright and Pitteway (1994), acknowledging their contribution to the development of Doppler-based drift estimation techniques.
Reviewer Comment: L. 203: Standard ionogram measurement takes typically several minutes.
The authors probably mean typical ionogram cadence.
Author Response: Correct. This sentence has been rephrased in the revised version to clarify that we refer to the typical cadence of ionogram measurements, not the duration of an individual sweep.
Reviewer Comment: Section 4.2: When the authors speak of smoothing, is it the time series of a parameter itself smoothed before calculating the time derivative...?
Author Response: Thank you for this clarifying question. In our analysis, the vertical drift values were calculated first using the difference in characteristic height parameters (e.g., ΔhmF2/Δt) between two adjacent ionograms. The resulting time series of drift values was then smoothed using a moving average. We have clarified this sequence in the revised manuscript to avoid any confusion. We also briefly mention that we tested both approaches (smoothing height parameters first vs. smoothing the derived drift values) and found no significant differences for the purposes of our analysis.
Citation: https://doi.org/10.5194/egusphere-2025-1811-AC1
-
AC1: 'Reply on RC1', Daniel Kouba, 19 Jun 2025
-
RC2: 'Comment on egusphere-2025-1811', Anonymous Referee #2, 01 Jul 2025
This is a simple but useful paper as it provides recommendations to consider when extracting
vertical drifts from virtual height measurements. Since not all Digisondes provide drift files
from which to extract velocities the arguments noted in this paper could be useful for any
researcher attempting to exploit drift velocities in their studies. In principle I agree with most
of the previous comments posted by the first anonymous referee (limited scientific value e.t.c).
However this paper has a definite practical value as I already pointed out. I also agree that
the authors should enrich their paper with more extended datasets not necessarily
under limited high cadence campaigns which was the primary focus of their study. Nowadays most Digisondes
adopt a 5 min resolution in their ionogram and drift measurement schedule and thus they could indeed apply their
methodology on monthly intervals across different seasons to examine their arguments more extensively.
I also encourage them to investigate the aspect of uncertainty in vertical drifts as expressed by the DDA
technique (as also pointed out by the first referee). Although the authors were clear that the uncertainty is much higher
in horizontal drift velocity measurements based on the actual principle of the DDA approach I think that since DDA generating
vertical drifts provides error bars they should introduce a discussion in their paper on "if and how" this uncertainty can be interpreted and possibly utilized
in studies exploiting vertical drifts. This will add more value in their paper.
Citation: https://doi.org/10.5194/egusphere-2025-1811-RC2 -
AC2: 'Reply on RC2', Daniel Kouba, 13 Jul 2025
We thank the reviewer for the constructive feedback and for recognizing the practical value of our study. In response to the specific suggestions:
1.Extended datasets:
We fully agree that applying the comparative methodology over longer periods and under standard observational regimes (e.g., 5-minute schedules) would significantly enhance the robustness of our conclusions. To address this, we have added a new paragraph in the Discussion section (Section 5) explicitly outlining how the methodology could be scaled to longer-term datasets. We also emphasize that many Digisondes now operate at a 5-minute cadence, making our proposed comparison techniques broadly applicable.2.Uncertainty in vertical drifts (DDA error bars):
Based on the reviewer’s recommendation (and in agreement with Reviewer #1), we have now included a dedicated subsection discussing the standard deviation values reported in the DDA-derived vertical drift velocities. In the revised Figure 2, we visualize the evolution of these uncertainties across different smoothing levels. The discussion addresses:- How the magnitude of error bars is often inversely related to the number of detected echoes during a given sounding.
- The practical implications of this uncertainty when interpreting short-term fluctuations versus long-term trends.
Furthermore, we explicitly comment on the differences in uncertainty levels between this short-campaign setup (with reduced sounding duration) and standard, longer Digisonde soundings. This comparison helps contextualize the magnitude of the reported standard deviations and suggests that uncertainty values in routine operations are likely lower and more stable.
We believe this addition improves the interpretability and applicability of our results and brings attention to an underexplored but important aspect of vertical drift analyses.
Citation: https://doi.org/10.5194/egusphere-2025-1811-AC2
-
AC2: 'Reply on RC2', Daniel Kouba, 13 Jul 2025
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2025-1811', Anonymous Referee #1, 04 Jun 2025
The scientific value and methodological rigor of the manuscript are questionable in its current form. The study relies on only two days of observational data to evaluate several methods for estimating vertical plasma drift, comparing them to a Doppler-based method that appears relatively more reliable. For any objective reader their results, shown, for example, in Figs. 5 and 6, is proof that indirect methods cannot be used for any serious science: they produce mostly false or highly distorted signatures of wave activity, both day and night. This conclusion would be even more obvious if the authors used more decisive comparison technique (see below). The authors currently avoid strong critical commentary on the performance of the tested methods, as well as on the Digisonde technique itself, which risks legitimizing approaches that may be unsuitable for serious scientific application.
A more rigorous comparison framework is necessary. First, regarding data volume: it is not sufficient to base conclusions on two isolated daily campaigns when working with a high-frequency sounding system capable of continuous autonomous operation. Such a limited dataset renders the study vulnerable to event-specific anomalies and does not support general conclusions. Second, the DDM method is based on least-squares fits and therefore not just offers more physically grounded and statistically supported results but also provides per-measurement error estimates. This is a valuable benchmark that should be used more proactively in the comparison. The analysis could be strengthened if the authors include error bars for the DDM-derived values shown as black dots in Figs. 5 and 6. These error bars (appropriately adjusted when smoothing is applied) would allow readers to better assess the reliability of DDM itself and to identify deviations in the other methods that fall outside acceptable uncertainty bounds.
In conclusion, while the manuscript may serve a purpose in highlighting deficiencies in certain vertical drift estimation methods, it must undergo significant methodological revision and adopt a more candid interpretation of its results before being considered for publication. Specifically, the scope of the data must be expanded, and the analysis must incorporate uncertainty quantification and a more critical evaluation of the tested methods.
A few specific technical and editorial comments follow below.
L. 69-70: Digisondes can detect specific reflection points and measure Doppler shifts at those points, enabling full vector drift estimations.
It’s worth mentioning here that Digisonde requires dedicated, non-ionogram, fixed-frequency mode of operation for their drift measurements, thus causing loss of all information the ionogram mode can provide.
L.101: Such a data set allows accurate estimation of the drift velocity vector (Reinisch et al., 1998).
The scientific objectivity requires to mention here that this technique was developed, implemented and published four years earlier in [Wright and Pitteway, 1994, https://doi.org/10.1016/0021-9169(94)90157-0].
L. 203: Standard ionogram measurement takes typically several minutes.
The authors probably mean typical ionogram cadence.
Section 4.2: When the authors speak of smoothing, is it the time series of a parameter itself (hmF2, h′F2, h′(3.5 MHz), or h′(0.8foF2)) smoothed before calculating the time derivative, or is it the time derivative calculated based on two adjacent values of those parameters smoothed? Have the authors tried both approaches? Is there a difference? Just for clarity, not that I expect critical improvement from any of these approaches.
Citation: https://doi.org/10.5194/egusphere-2025-1811-RC1 -
AC1: 'Reply on RC1', Daniel Kouba, 19 Jun 2025
We would like to sincerely thank the reviewer for their thorough and thoughtful assessment of our manuscript. We appreciate the critical insights and constructive suggestions, which helped us identify several areas for improvement. We have carefully considered each comment and revised the manuscript accordingly. Below, we provide a detailed response to all points raised.
Reviewer Comment: The scientific value and methodological rigor of the manuscript are questionable in its current form. The study relies on only two days of observational data to evaluate several methods for estimating vertical plasma drift, comparing them to a Doppler-based method that appears relatively more reliable. For any objective reader their results, shown, for example, in Figs. 5 and 6, is proof that indirect methods cannot be used for any serious science...
Author Response: We acknowledge the reviewer’s concern regarding the limited data set. We agree that a comprehensive validation of different methods for estimating vertical plasma drift would ideally require a much larger set of observations, covering various seasons, levels of geomagnetic activity, and different geographic locations.
However, the objective of our study was not to provide a complete characterization of each method under all possible conditions. Instead, we deliberately chose a controlled, quiet period at a mid-latitude station and performed specially configured high-cadence measurements—including both ionograms and drift observations. This setting allowed us to explore the intrinsic behavior of the methods under idealized conditions. Our reasoning is that if significant inconsistencies arise even under quiet conditions, the discrepancies will likely be amplified during geomagnetically disturbed periods.
Furthermore, this study required manual post-processing of ionograms to extract height parameters for the different indirect methods, which is labor-intensive due to the high temporal resolution. While our data set is limited in scope, the added value lies in its temporal granularity and the ability to perform direct side-by-side comparisons on a minute-by-minute basis.
We do not claim to generalize the results to all geophysical conditions but aim to highlight potential methodological issues and limitations. We believe this is a necessary step toward a more extensive evaluation of vertical drift estimation methods.
We also agree with the reviewer that Doppler-based DDM is, when properly processed, the most reliable and consistent method available. However, many studies continue to use indirect methods—particularly when drift measurements are unavailable (e.g., for historical ionogram data). We therefore consider it meaningful to analyze differences among methods and identify conditions under which they produce comparable or diverging results.
Finally, in response to the reviewer’s valid point about the lack of stronger critical commentary in the original manuscript, we have revised the discussion section to include a more explicit evaluation of the limitations of each method. We now provide specific recommendations on their applicability depending on time of day, expected variability, and data availability.
Reviewer Comment: Second, the DDM method is based on least-squares fits and therefore not just offers more physically grounded and statistically supported results but also provides per-measurement error estimates... etc.
Author Response: We thank the reviewer for this helpful suggestion. In the revised version of the manuscript, we now include error estimates for each DDM measurement. These uncertainties are based on the least-squares fit used to determine the drift velocity vector and are visualized in Figures 5 and 6.
We also added a brief discussion explaining that the high-cadence mode used during our campaigns resulted in abnormally short measurement durations and, consequently, a lower number of reflected points in each SKYmap. This limitation naturally increases the uncertainty of the DDM results in our data set compared to standard measurements. However, even under these constrained conditions, the DDM-derived vertical drift estimates remain stable and coherent—highlighting the robustness of this technique. This point has also been clarified in the revised discussion.
Reviewer Comment: L. 69–70: Digisondes can detect specific reflection points and measure Doppler shifts...
It’s worth mentioning here that Digisonde requires dedicated, non-ionogram, fixed-frequency mode...
Author Response: We appreciate this observation. The text has been revised to explicitly state that Digisonde drift measurements are performed in a dedicated mode using fixed-frequency sounding, which does not provide full ionogram data during those intervals.
Reviewer Comment: L. 101: Such a data set allows accurate estimation of the drift velocity vector...
Scientific objectivity requires to mention that this technique was first developed by Wright and Pitteway (1994)...
Author Response: We thank the reviewer for pointing this out. The revised manuscript now includes a citation of the earlier work by Wright and Pitteway (1994), acknowledging their contribution to the development of Doppler-based drift estimation techniques.
Reviewer Comment: L. 203: Standard ionogram measurement takes typically several minutes.
The authors probably mean typical ionogram cadence.
Author Response: Correct. This sentence has been rephrased in the revised version to clarify that we refer to the typical cadence of ionogram measurements, not the duration of an individual sweep.
Reviewer Comment: Section 4.2: When the authors speak of smoothing, is it the time series of a parameter itself smoothed before calculating the time derivative...?
Author Response: Thank you for this clarifying question. In our analysis, the vertical drift values were calculated first using the difference in characteristic height parameters (e.g., ΔhmF2/Δt) between two adjacent ionograms. The resulting time series of drift values was then smoothed using a moving average. We have clarified this sequence in the revised manuscript to avoid any confusion. We also briefly mention that we tested both approaches (smoothing height parameters first vs. smoothing the derived drift values) and found no significant differences for the purposes of our analysis.
Citation: https://doi.org/10.5194/egusphere-2025-1811-AC1
-
AC1: 'Reply on RC1', Daniel Kouba, 19 Jun 2025
-
RC2: 'Comment on egusphere-2025-1811', Anonymous Referee #2, 01 Jul 2025
This is a simple but useful paper as it provides recommendations to consider when extracting
vertical drifts from virtual height measurements. Since not all Digisondes provide drift files
from which to extract velocities the arguments noted in this paper could be useful for any
researcher attempting to exploit drift velocities in their studies. In principle I agree with most
of the previous comments posted by the first anonymous referee (limited scientific value e.t.c).
However this paper has a definite practical value as I already pointed out. I also agree that
the authors should enrich their paper with more extended datasets not necessarily
under limited high cadence campaigns which was the primary focus of their study. Nowadays most Digisondes
adopt a 5 min resolution in their ionogram and drift measurement schedule and thus they could indeed apply their
methodology on monthly intervals across different seasons to examine their arguments more extensively.
I also encourage them to investigate the aspect of uncertainty in vertical drifts as expressed by the DDA
technique (as also pointed out by the first referee). Although the authors were clear that the uncertainty is much higher
in horizontal drift velocity measurements based on the actual principle of the DDA approach I think that since DDA generating
vertical drifts provides error bars they should introduce a discussion in their paper on "if and how" this uncertainty can be interpreted and possibly utilized
in studies exploiting vertical drifts. This will add more value in their paper.
Citation: https://doi.org/10.5194/egusphere-2025-1811-RC2 -
AC2: 'Reply on RC2', Daniel Kouba, 13 Jul 2025
We thank the reviewer for the constructive feedback and for recognizing the practical value of our study. In response to the specific suggestions:
1.Extended datasets:
We fully agree that applying the comparative methodology over longer periods and under standard observational regimes (e.g., 5-minute schedules) would significantly enhance the robustness of our conclusions. To address this, we have added a new paragraph in the Discussion section (Section 5) explicitly outlining how the methodology could be scaled to longer-term datasets. We also emphasize that many Digisondes now operate at a 5-minute cadence, making our proposed comparison techniques broadly applicable.2.Uncertainty in vertical drifts (DDA error bars):
Based on the reviewer’s recommendation (and in agreement with Reviewer #1), we have now included a dedicated subsection discussing the standard deviation values reported in the DDA-derived vertical drift velocities. In the revised Figure 2, we visualize the evolution of these uncertainties across different smoothing levels. The discussion addresses:- How the magnitude of error bars is often inversely related to the number of detected echoes during a given sounding.
- The practical implications of this uncertainty when interpreting short-term fluctuations versus long-term trends.
Furthermore, we explicitly comment on the differences in uncertainty levels between this short-campaign setup (with reduced sounding duration) and standard, longer Digisonde soundings. This comparison helps contextualize the magnitude of the reported standard deviations and suggests that uncertainty values in routine operations are likely lower and more stable.
We believe this addition improves the interpretability and applicability of our results and brings attention to an underexplored but important aspect of vertical drift analyses.
Citation: https://doi.org/10.5194/egusphere-2025-1811-AC2
-
AC2: 'Reply on RC2', Daniel Kouba, 13 Jul 2025
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 682 | 51 | 28 | 761 | 23 | 43 |
- HTML: 682
- PDF: 51
- XML: 28
- Total: 761
- BibTeX: 23
- EndNote: 43
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Daniel Kouba
Zbyšek Mošna
Petra Koucká Knížová
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1054 KB) - Metadata XML
The scientific value and methodological rigor of the manuscript are questionable in its current form. The study relies on only two days of observational data to evaluate several methods for estimating vertical plasma drift, comparing them to a Doppler-based method that appears relatively more reliable. For any objective reader their results, shown, for example, in Figs. 5 and 6, is proof that indirect methods cannot be used for any serious science: they produce mostly false or highly distorted signatures of wave activity, both day and night. This conclusion would be even more obvious if the authors used more decisive comparison technique (see below). The authors currently avoid strong critical commentary on the performance of the tested methods, as well as on the Digisonde technique itself, which risks legitimizing approaches that may be unsuitable for serious scientific application.
A more rigorous comparison framework is necessary. First, regarding data volume: it is not sufficient to base conclusions on two isolated daily campaigns when working with a high-frequency sounding system capable of continuous autonomous operation. Such a limited dataset renders the study vulnerable to event-specific anomalies and does not support general conclusions. Second, the DDM method is based on least-squares fits and therefore not just offers more physically grounded and statistically supported results but also provides per-measurement error estimates. This is a valuable benchmark that should be used more proactively in the comparison. The analysis could be strengthened if the authors include error bars for the DDM-derived values shown as black dots in Figs. 5 and 6. These error bars (appropriately adjusted when smoothing is applied) would allow readers to better assess the reliability of DDM itself and to identify deviations in the other methods that fall outside acceptable uncertainty bounds.
In conclusion, while the manuscript may serve a purpose in highlighting deficiencies in certain vertical drift estimation methods, it must undergo significant methodological revision and adopt a more candid interpretation of its results before being considered for publication. Specifically, the scope of the data must be expanded, and the analysis must incorporate uncertainty quantification and a more critical evaluation of the tested methods.
A few specific technical and editorial comments follow below.
L. 69-70: Digisondes can detect specific reflection points and measure Doppler shifts at those points, enabling full vector drift estimations.
It’s worth mentioning here that Digisonde requires dedicated, non-ionogram, fixed-frequency mode of operation for their drift measurements, thus causing loss of all information the ionogram mode can provide.
L.101: Such a data set allows accurate estimation of the drift velocity vector (Reinisch et al., 1998).
The scientific objectivity requires to mention here that this technique was developed, implemented and published four years earlier in [Wright and Pitteway, 1994, https://doi.org/10.1016/0021-9169(94)90157-0].
L. 203: Standard ionogram measurement takes typically several minutes.
The authors probably mean typical ionogram cadence.
Section 4.2: When the authors speak of smoothing, is it the time series of a parameter itself (hmF2, h′F2, h′(3.5 MHz), or h′(0.8foF2)) smoothed before calculating the time derivative, or is it the time derivative calculated based on two adjacent values of those parameters smoothed? Have the authors tried both approaches? Is there a difference? Just for clarity, not that I expect critical improvement from any of these approaches.