Ionospheric Perturbations Due to Large Thunderstorms and the Resulting Mechanical and Acoustic Signatures

Round 1

Reviewer 1 Report

Comments on Ogunsua et al., “Ionospheric Perturbations due to large thunderstorms and the resulting Mechanical and Acoustic Signatures”

   The paper claims to study the influence of some thunderstorm events associated lightning activities on ionospheric perturbations. The wavelet analysis and Discrete Fourier Transform (DFT) were used to study the frequency modes and periodicities of the TEC deviation. However, there are some major issues in the present manuscript.

1)     In the section of TEC data extraction, vertical TEC data are mostly estimated by equation (2). It is also used to define the ionospheric pierce point of the vertical TEC and the corresponding disturbed area. As a result, I think that the method used in this study is not reasonable to use equation (2b) to estimate the vertical TEC. Please see the new published paper (Chowdhury et al., 2023), which is similar to this study.

2)     In terms of Ionospheric Perturbations induced by large thunderstorms, authors mentioned gravity waves lead to the ionospheric perturbations in the lower ionosphere. In fact, the perturbation in the lower ionosphere might not lead to the perturbation of the TEC data. In the mechanism, gravity waves induced by thunderstorms can propagate to the upper ionosphere (F region). Then, the disturbed plasma in the F region can cause the perturbation of the TEC data, because the electron density in the F region has a significant role in the TEC data. Please refer to Chowdhury et al. (2023). In the lower ionosphere, Davis and Johnson (2005) found that the occurrence of the Es layer might have a positive relationship with the thunderstorms.

3)     Related work is badly structured and difficult to follow as well, and the analysis method and observation results are hard to understand. For example, does Fig.3a show the hourly flash count for the entire area or for a single station? What do the different colored lines in Fig.3b represent? Which is the control group of TEC deviation value of each station? What is the corresponding time in Fig.3c and 3d. Such unclear statements make it difficult for readers to understand and give people a sense of incongruity.

4)     Lines 227-228, “The result from these analysis reveals the acoustic and gravity waves dynamics in the ionosphere during the thunderstorm events.” This view is not rigorous. Is the TEC disturbance period within the range of gravity wave period necessarily related to gravity wave? This needs to be verified in conjunction with other data.

Refs:

Chowdhury, S., Kundu, S., Ghosh, S., Sasmal, S., Brundell, J., & Chakrabarti, S. K. (2023). Statistical study of global lightning activity and thunderstorm-induced gravity waves in the ionosphere using WWLLN and GNSS-TEC. Journal of Geophysical Research: Space Physics, 128, e2022JA030516.

Davis, C. J., & Johnson, C. G. (2005). Lightning-induced intensification of the ionospheric sporadic E layer. Nature, 435(7043), 799– 801. https://doi.org/10.1038/nature03638。

    The methodology used in this paper might not be right. As a result, this manuscript in this form is not recommended for publication in remote sensing based on the major issues mentioned above. I suggest that it might be rejected in the current version.

Author Response

Reviewer 1

Comments on Ogunsua et al., “Ionospheric Perturbations due to large thunderstorms and the resulting Mechanical and Acoustic Signatures”

   The paper claims to study the influence of some thunderstorm events associated lightning activities on ionospheric perturbations. The wavelet analysis and Discrete Fourier Transform (DFT) were used to study the frequency modes and periodicities of the TEC deviation. However, there are some major issues in the present manuscript.

 1)    In the section of TEC data extraction, vertical TEC data are mostly estimated by equation (2). It is also used to define the ionospheric pierce point of the vertical TEC and the corresponding disturbed area. As a result, I think that the method used in this study is not reasonable to use equation (2b) to estimate the vertical TEC. Please see the new published paper (Chowdhury et al., 2023), which is similar to this study.

Thank you for the suggestion. The equation has been corrected as applied in the work.

Initially we decided to simplify the equation for the benefit of the general audience. However, we have corrected the content regarding the equation 2.

2)     In terms of Ionospheric Perturbations induced by large thunderstorms, authors mentioned gravity waves lead to the ionospheric perturbations in the lower ionosphere. In fact, the perturbation in the lower ionosphere might not lead to the perturbation of the TEC data. In the mechanism, gravity waves induced by thunderstorms can propagate to the upper ionosphere (F region). Then, the disturbed plasma in the F region can cause the perturbation of the TEC data, because the electron density in the F region has a significant role in the TEC data. Please refer to Chowdhury et al. (2023). In the lower ionosphere, Davis and Johnson (2005) found that the occurrence of the Es layer might have a positive relationship with the thunderstorms.

We agree that the comment on thunderstorms and “lower” atmosphere is not an accurate way to describe the thunderstorm impact on the ionosphere, however, there is no statement anywhere in the text referring to the effect of thunderstorm associated gravity wave  on “lower” atmosphere.

However, if at all it is written that way , the statement is meant to be …. perturbations in the “bottom-side” ionosphere ( rather than lower ionosphere). The bottom-side ionosphere is known to include the lower ionosphere from the D-region up to some part of F region. The sporadic E (Es) layer has been found to propagate the disturbance and plasma bubbles into the F-region.

A similar comment in the manuscript, which referred to the lower region of the ionosphere has been rephrased appropriately.

3)     Related work is badly structured and difficult to follow as well, and the analysis method and observation results are hard to understand. For example, does Fig.3a show the hourly flash count for the entire area or for a single station? What do the different colored lines in Fig.3b represent? Which is the control group of TEC deviation value of each station? What is the corresponding time in Fig.3c and 3d. Such unclear statements make it difficult for readers to understand and give people a sense of incongruity.

Thank you for the suggestion. Efforts have been made to properly clarify the target and the control station. The approach to this is to select a target location using WWLLN and compare the other stations relatively to the main location of the event, based on differences in proximity of the  selected stations to the thunderstorm events.

4)     Lines 227-228, “The result from these analysis reveals the acoustic and gravity waves dynamics in the ionosphere during the thunderstorm events.” This view is not rigorous. Is the TEC disturbance period within the range of gravity wave period necessarily related to gravity wave? This needs to be verified in conjunction with other data.

We appreciate your keen observation. The verification was carried out in section below with data analysis and two methods of frequency domain analysis were carried, which are DFT and wavelet analysis. However, we agree that the isolated Statement at this point of the paper might be a little bit confusing for some readers. With this consideration we have rephrased the statement and framed it in such a way that it directs the readers to the analysis carried out for the frequencies and the periodic properties of the signals.

Additional analysis has also been made to show the type of variations and the magnitude of variations for periods without thunderstorms withing the year (please see the supplementary).

Refs:

Chowdhury, S., Kundu, S., Ghosh, S., Sasmal, S., Brundell, J., & Chakrabarti, S. K. (2023). Statistical study of global lightning activity and thunderstorm-induced gravity waves in the ionosphere using WWLLN and GNSS-TEC. Journal of Geophysical Research: Space Physics, 128, e2022JA030516.

Davis, C. J., & Johnson, C. G. (2005). Lightning-induced intensification of the ionospheric sporadic E layer. Nature, 435(7043), 799– 801. https://doi.org/10.1038/nature03638。

    The methodology used in this paper might not be right. As a result, this manuscript in this form is not recommended for publication in remote sensing based on the major issues mentioned above. I suggest that it might be rejected in the current version.

The authors will like to add that the methodology used in this work have been verified as standard scientific approach with application of appropriate statistics and numerical method to show our scientific inquiry. Some aspects of the techniques used can be seen in Lay et al 2013 and some additional methods can be found in other scientific papers with similar numerical and statistical approach.

Reviewer 2 Report

This paper presents some ‘picked-cases’ analysis of the ionospheric perturbations caused by large thunderstorms using TEC data. The results show that TEC deviations reach up to 2.2 TECUs with dominant frequency modes in the gravity wave range and acoustic range. Additionally, they found that there was a 20-60-minute time delay between sprite events and the highest peak of acoustic-gravity wave perturbations. A possible mechanism for this phenomenon was proposed and discussed.

Overall, this work provides an interesting study that may enrich our understanding of the coupling between ionosphere and thunderstorms. The structure of the paper is organized in a reasonable way. However, the data and analysis presented now do not support the results that the variations of TEC are definitely caused by thunderstorms. The authors should address some comments as follows before the reviewer considers accepting the work.

Major concerns:

Ionosphere is very complex and varies from day to day. There are many factors that influence it, especially the effects of space weather, such as magnetic storms and solar activity. The influence of thunderstorms from the troposphere is only a very small part. You cannot contribute the perturbations of TEC to the thunderstorm as there still has large variations without thunderstorms! It is crucial to exclude the influence of other factors in the perturbation. Hence, the authors should provide more solid evidence and data analysis to support that the variations of TEC are caused by a thunderstorm.  It is suggested that you should provide the changes in TEC on the surrounding days to compare with the thunderstorm day.

Specific comments:

1.       There are so many grammar errors in the paper.  The authors should check it very carefully.

2.       In Line 18. The word “significant” is not very accurate. The ionosphere is dominated by space weather rather than atmospheric, Please replace it with a more accurate term, for example, “notable” or “intriguing”.

3.       In Figure 3.Adjust the order of the subplots in Figures 3 and 4 by making subplot a and subplot b vertically aligned.

In Figure 3 and Figure 4. The comparison between the flash distribution in subplot c and the peak TEC distribution in subplot d lacks clarity. To improve the visualization of the relationship between flash and peak TEC, please add terrain contour lines to subplot c, similar to subplot d.

4.       To ensure accuracy and consistency with standard notation, it is recommended that the misspelled unit abbreviation "TECu" in the colorbars of Figures 3, 4, and 5 be corrected to "TECU". Please make the necessary revisions accordingly.

5.       In Section 3.2, Line 316. This explanation lacks sufficient support and needs to be reworded with more evidence to strengthen the argument. It could be beneficial to provide more context and organization to make this point more compelling.

6.       In Section 4.2, Line 479. The sentence suggests that the effect of TLEs on TEC is negligible based on certain observations. However, it lacks sufficient support and explanation to make a convincing argument. More context is needed regarding TLEs and TEC, the observations made, and the methods used to measure the effect of TLEs on TEC to evaluate the validity of the statement.

Author Response

Reviewer 2

Major concerns:

Ionosphere is very complex and varies from day to day. There are many factors that influence it, especially the effects of space weather, such as magnetic storms and solar activity. The influence of thunderstorms from the troposphere is only a very small part. You cannot contribute the perturbations of TEC to the thunderstorm as there still has large variations without thunderstorms! It is crucial to exclude the influence of other factors in the perturbation. Hence, the authors should provide more solid evidence and data analysis to support that the variations of TEC are caused by a thunderstorm.  It is suggested that you should provide the changes in TEC on the surrounding days to compare with the thunderstorm day.

Thank you for these suggestions. Further analysis has been made considering days without thunderstorms in the year of selected events (that is the days in January 2013 and January 2016). This can be found in the supplementary material.

Specific comments:

1. There are so many grammar errors in the paper.  The authors should check it very carefully.

Thank you for the suggestion. A thorough check have been carried out and appropriate corrections have been made.

2. In Line 18. The word “significant” is not very accurate. The ionosphere is dominated by space weather rather than atmospheric, please replace it with a more accurate term, for example, “notable” or “intriguing”.

Done: We have replaced the word “significant” with the word “notable” in the text.

Even though we believe that these words are synonyms the word “Significant” has been replaced with “notable”. The word significant was initially used in context of TEC deviations due to the fact that the values are higher that most values obtain for internal forcing mechanisms like thunderstorm. However, we agree that this context may not be very clear at the first introduction of the concept.

Thank you for the suggestion.

3. In Figure 3 .Adjust the order of the subplots in Figures 3 and 4 by making subplot a and subplot b vertically aligned.

It has been re-aligned. Thank you for the suggestion

In Figure 3 and Figure 4. The comparison between the flash distribution in subplot c and the peak TEC distribution in subplot d lacks clarity. To improve the visualization of the relationship between flash and peak TEC, please add terrain contour lines to subplot c, similar to subplot d.

Thank you for the suggestions. The plots have been redrawn to show more clarity, as specified by you.

4. To ensure accuracy and consistency with standard notation, it is recommended that the misspelled unit abbreviation "TECu" in the colorbars of Figures 3, 4, and 5 be corrected to "TECU". Please make the necessary revisions accordingly.

It has been corrected. Thank you for the suggestion

5. In Section 3.2, Line 316. This explanation lacks sufficient support and needs to be reworded with more evidence to strengthen the argument. It could be beneficial to provide more context and organization to make this point more compelling.

We have corrected this portion of the paper based on your suggestion  and relevant reference backup references have been added to the statement. Thank you.

6. In Section 4.2, Line 479.The sentence suggests that the effect of TLEs on TEC is negligible based on certain observations. However, it lacks sufficient support and explanation to make a convincing argument. More context is needed regarding TLEs and TEC, the observations made, and the methods used to measure the effect of TLEs on TEC to evaluate the validity of the statement.

Thank you for the suggestion. We have added supporting explanations with literature (lines 523 – 529)

The required energy for photoionization in the ionosphere is usually up to 40 – 100 eV, whereas the photon energy from sprites usually within the range of 1.5 - 3.5 eV (Takashi et al), by which we can speculate that the photon energy from sprites can barely contribute to a bulk increase the ionospheric electron density. In addition to this, we strongly believe that the contribution of TLEs to the ionosphere would be mostly based on its impact on the ionospheric current and its resultant magnetic field considering that the ionosphere is part of the global electric circuit.

Reviewer 3 Report

In this paper, the authors presented the ionospheric perturbations due to thunderstorm by extracting the total electron Content (TEC) from GNSS measurements at the selected thunderstorm locations using polynomial filtering method. The thunderstorm in the selected area were mapped using the data from World Wide Lightning Location Network (WWLLN) to identify the higher concentration area of the thunderstorm event. Further analyses were conducted using wavelet analysis and Discrete Fourier Transform (DFT) to study the frequency modes and periodicities of the TEC deviation. The TEC deviations and frequency ranges in the scales of gravity wave range and acoustic range were discriminated from the analysis. The authors also proposed the possible phenomena by discussing the perturbations extracted from TEC deviations.

 From my opinion, the manuscript is well-written and original but minor revision is required to improve it further for the readers of the journal. The authors are directed to incorporate the following points in the revised version.

1.     The figure quality could have been improved for publishable form.

2.     It is not clear how did the authors got access to the WWLLN database and extracted the region of interest.

3.     In table1, the authors may indicate which stations are IGS and which are under non-IGS stations.

4.     Did the authors use their own program to extract TEC from GNSS observables or used any existing program for this.

5.     Line 189, the subsection title should start with Thunderstorm effect… instead of Storm effect….

6.     A conclusion section is expected after the Discussion, mentioning the important remarks from the analysis and future aspects.

Author Response

Reviewer 3

 From my opinion, the manuscript is well-written and original but minor revision is required to improve it further for the readers of the journal. The authors are directed to incorporate the following points in the revised version.

1. The figure quality could have been improved for publishable form.

Done. Improved have been made to improve upon some of the figures where necessary. Thank you for the suggestion.

2. It is not clear how did the authors got access to the WWLLN database and extracted the region of interest.

Our laboratory has a collaboration with the WWLLN company.

3. In table1, the authors may indicate which stations are IGS and which are under non-IGS stations.

All the station are from IGS. We have now indicated this in the table title.

4. Did the authors use their own program to extract TEC from GNSS observables or used any existing program for this.

We used Existing program and our own program. Thank you.

5. Line 189, the subsection title should start with Thunderstorm effect… instead of Storm effect….

The correction has been made. Thank you for the suggestion.

6. A conclusion section is expected after the Discussion, mentioning the important remarks from the analysis and future aspects.

A conclusion section has been added to this work. Thank you for the suggestion.

The authors will like to thank the reviewer for the suggestions. The suggestions have helped to improve the quality of the manuscript.

Round 2

Reviewer 1 Report

The authors have addressed some remarks on the previous manuscript. I think this manuscript can almost be published as it is now

Author Response

Thanks for your answer.

Reviewer 2 Report

The authors addressed most of my concerns from the first review. The supplementary information provides more evidence to support that the perturbations of TEC in the main text are likely caused by thunderstorms. Many sections with grammar errors and sentence structure have been improved. Further, the scientific discussion is consistent and appropriate. Therefore, I believe it is ready for publication after the comment below is taken into consideration.

In the first-round review, the reviewer recommends the authors supplement the surrounding days to compare with TECs variations of thunderstorm days in June 2016 and July 2013. Hence, the authors should supplement the TEC deviation without thunderstorm days in June 2016 and July 2013, not in January month as the TEC may be different greatly.

Author Response

Comment:

In the first-round review, the reviewer recommends the authors supplement the surrounding days to compare with TEC variations of thunderstorm days in June 2016 and July 2013. Hence, the authors should supplement the TEC deviation without thunderstorm days in June 2016 and July 2013, not in January month as the TEC may be different greatly.

Response:

Thank you for the suggestion. Additional analyses have been made for days around the thunderstorm cases were chosen.

Considering that the months of June, July and August are continually characterized by thunderstorms and lightning flashes. The days without thunderstorms have not been noticed after careful analysis for June 2016 and July 2013. In addition, further evaluation was made to consider the days around the thunderstorm and to compare the computed daily TEC deviation with respect to variations in the daily amount lightning flashes. These additional analyses are available with supplementary material. Also, please see the text (lines 321 to 329) for further explanations.