the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 15 Sep 2023
Probability estimation of March 1989-like geomagnetic storms and their relevance for the insurance industry
Abstract. This study employs Extreme Value Theory (EVT) to estimate the probability of geomagnetic storms of comparable magnitude to the March 1989 event and to assess the implications of such storms for the insurance industry. To calculate return periods for extreme events, historical Dst data from the World Data Centre for Geomagnetism are combined with the Generalized Extreme Value (GEV) distribution, maximum likelihood estimation, and the Peaks Over Threshold (POT) approach. The findings suggest that there is a 7.14 % to 8.33 % chance of a geomagnetic storm of equivalent severity occurring during the next 70 years (with a 95 % confidence interval). This study helps us understand the frequency and severity of extreme geomagnetic storms and helps the insurance industry make judgments about risk management.
- Preprint
(809 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2023-1800', Anonymous Referee #1, 24 Sep 2023
-
AC1: 'Reply on RC1', Deniz Güney Akkor, 25 Sep 2023
Dear Referee,
We would like to express our gratitude for your comprehensive review of our manuscript. Your expert input is invaluable, and we appreciate the time you've dedicated to evaluating our work. We've made it our priority to address all the concerns and suggestions you've raised.
We've taken steps to clarify the scope and limitations of our study, giving particular attention to the statistical methods and their applicability in the context of the data available. We also acknowledge your point on the lack of economic analysis; however, the focus of our current research is strictly on applying statistical data science methods to existing, concrete scientific data. In doing so, we believe our study meets the scientific requirements for such an application and the subsequent interpretation of the results.
We've endeavored to provide a more precise explanation for the return period related to the March 1989 event, emphasizing the EVT's value in answering questions about events that may not be present in the existing data records. We have taken note of your suggestion to be cautious in interpreting results that extend beyond the limitations of our dataset, and we intend to focus on this aspect in our future research.
Concerning the incorrect and incomplete references, we sincerely regret these oversights and assure you they will be rectified. We are also in agreement to update the tables and figure labels according to the journal's standards.
We have consulted pioneering studies in this field such as Rogers et al. (2021) and Chapman et al. (2020) prior to the conception of this work. However, we found that a constant index like Dst was a better fit for our modeling needs. This study is aligned with the existing literature and contributes to ongoing debates in the space weather science community, especially those related to events most likely to occur within our lifetimes.
Lastly, we are eagerly awaiting the updated RAE report by Cannon et al., as it promises to provide additional context and relevance for our work.
If our responses to your concerns are found to be satisfactory, we would like to proceed with the necessary revisions to improve the quality of the manuscript for re-submission. Thank you again for your valuable feedback, and we look forward to your response.
Best regards,
D. Güney Akkor
-
RC2: 'Reply on AC1', Anonymous Referee #1, 29 Sep 2023
This paper might be worth referencing as well:
Bergin, A., Chapman, S. C., Watkins, N. W., Moloney, N. R., & Gjerloev, J. W. (2023). Extreme event statistics in Dst, SYM-H, and SMR geomagnetic indices. Space Weather, 21, e2022SW003304. https://doi.org/10.1029/2022SW003304Citation: https://doi.org/10.5194/egusphere-2023-1800-RC2 -
AC2: 'Reply on RC2', Deniz Güney Akkor, 29 Sep 2023
Dear Referee,
We express our gratitude for dedicating your time to reviewing our manuscript and providing us with your valuable feedback. We acknowledge and value your recommendation to cite the aforementioned scholarly article.
Bergin, A., Chapman, S. C., Watkins, N. W., Moloney, N. R., and Gjerloev, J. W. (2023). This study focuses on the statistical analysis of extreme events in three geomagnetic indices, namely Dst, SYM-H, and SMR. The article titled "Space Weather, 21, e2022SW003304" can be accessed at the following DOI: https://doi.org/10.1029/2022SW003304.
After conducting a thorough examination of the aforementioned paper, we concur that it provides valuable insights that are pertinent to our research. The reference will be integrated into our revised manuscript, and its findings will be thoroughly examined and discussed within the framework of our research.
We express our gratitude for your meticulous focus on detail and constructive suggestions, which enhance the quality of our work.
We have updated our responses to your initial comments in the attached PDF (those highlighted in green are new additions based on your feedback, which we will incorporate into our draft and share a follow-up version with you if you find our responses reasonable, those highlighted in yellow are still open for comment, and those highlighted in red have not yet been made by us).
We look forward to receiving your feedback.
Best regards,
D. Güney Akkor
- AC3: 'Reply on RC2', Deniz Güney Akkor, 06 Oct 2023
-
AC4: 'Reply on RC2', Deniz Güney Akkor, 29 Oct 2023
Dear Refree 1,
Thank you for your insightful comments and suggestions, particularly regarding the utilization of the SMY and SMU indices. I would like to clarify that we have indeed followed a methodology similar to the one you recommended and have performed estimations employing these indices. Given the context of our study and its primary objectives, we've determined that granularity at the minute-level isn't imperative. Thus, our findings, without resorting to such granularity, have further reinforced the robustness and validity of our results.
From the perspective of insurance and risk management, we believe that our current approach in estimating the global effects of geomagnetic storms provides a sufficient foundation for risk modeling at this juncture. However, we do recognize the potential advantages of a more granular analysis using the SMY and SMU indices and are planning to incorporate them in our subsequent research endeavors to provide a more comprehensive view.
Your constructive feedback has been instrumental in shaping our future research direction. We sincerely appreciate your valuable contribution to this work and look forward to considering your insights in our forthcoming studies.
Kind regards,
D. Güney Akkor
Citation: https://doi.org/10.5194/egusphere-2023-1800-AC4
-
AC2: 'Reply on RC2', Deniz Güney Akkor, 29 Sep 2023
-
RC2: 'Reply on AC1', Anonymous Referee #1, 29 Sep 2023
-
AC1: 'Reply on RC1', Deniz Güney Akkor, 25 Sep 2023
-
RC3: 'Comment on egusphere-2023-1800', Anonymous Referee #2, 10 Jan 2024
The authors study the extremes of the Dst time series from 1957 - 2023
to estimate the risk of a future March 1989 event for the insurance
industry.- Section 1.2 suggests that the "relevance of these storms for the
insurance sector" will be determined, as well as emphasizing a "holistic
strategy that considers... power grid models, geological influences
and transmission system features." But I find no such discussion in
the paper.- In Section 2 I would expect at least some brief summary on the use
of Dst (as opposed to other observables) for measuring storm
strength, and e.g. why on long timescales one need not consider the
solar cycle, etc.- Section 2.3 suggests that knowledge of "potential effects of extreme
geomagnetic storms on numerous infrastructure and industry
sectors..." will be advanced. But, again, no such discussion
appears.- Section 3.1 (line 139) is slightly ambiguous as to who did the
pre-processing of the Dst data. Did the authors pre-process the data
(if so, how?), or they used the publicly available Dst data?- Section 3.1 (line 141): why are negative thresholds used? Haven't
the values been made positive as per the previous sentence?- Section 3.2 confuses somewhat the choice of EVT fitting
distribution. Section 2.4 announced it would be the Weibull
distribution, while the title of Section 3.2 suggests it is rather
the GEV distribution, usually used for block maxima, which includes
the Weibull family as a special case. The authors appear to settle
on a GEV distribution. But I find no mention of the block size, or
why a block maxima approach was favored over, say, a GPD fit to the
empirical exceedance distribution. Meanwhile, Figure 2 suggests a
GPD fit.- Section 4.1 (line 233) appears to refer to declustering on short
timescales. A more detailed discussion of the 24-hour gap is
required. How do the authors pick a representative Dst value for a
particular storm?- Figure 2 features exponential models(?) which are not discussed in the
text (the caption is also inadequate). Is this a GPD with shape
parameter zero (which should reduce to an exponential)?- What is there new to learn from Figure 3 which isn't already
included in Figure 1? Didn't we rather need to see the mean excess
plot to identify the onset of linear behavior previously mentioned?
Furthermore, isn't the mean excess (aka expected shortfall) a
central quantity for the insurance industry? Why is there no plot
of the GPD fit to the empirical tail distribution?- Section 4 (line 273). The authors appear to settle on 330nT as a
threshold. But subsequent analysis uses 300nT. The whole 150 vs
300nT discussion was unclear to me. Isn't the goal to use the lowest
threshold possible, while remaining within the "asymptotic" tail regime,
as evidenced by linear mean excess plot regions and stability
of shape parameters? One can discuss robustness with respect to
threshold choice by sweeping over threshold choices, rather than
singling out 150nT in particular.- Can't Figures 5 and 6 be combined? Would that not illustrate
that the 300nT threshold choice is responsible for the larger
confidence intervals, owing to the fewer exceedances remaining? As
mentioned by another referee, the extrapolation to 1,000 year return
periods is far-fetched, particularly for the insurance industry.- The paper is unfortunately full of vague and overpromising sentences
such as (line 357) "Our findings enable more effective application
of updated risk assessment methods for geomagnetic storms in the
context of the insurance industry." While the analysis presented
doesn't advance the state-of-the art in modelling space weather
extremes, one might hope for a detailed discussion of how
the analysis could be used in the insurance industry (e.g. to
estimate potential damage in dollars, say). But no such discussion
is presented.Citation: https://doi.org/10.5194/egusphere-2023-1800-RC3 -
AC5: 'Reply on RC3', Deniz Güney Akkor, 10 Jan 2024
Dear Referee 2,
We sincerely value your thorough and perceptive evaluation of our manuscript. We greatly appreciate your expertise and the thoroughness of your feedback. Your input is extremely valuable to us, and we are thankful for the time and effort you have dedicated to reviewing our work.
We have carefully examined all of your comments and suggestions and are fully dedicated to addressing them in a comprehensive manner. In our response, we will provide a comprehensive account of the measures we have implemented to improve our study, with a specific emphasis on augmenting its lucidity, pertinence, and scientific meticulousness.
We recognize the necessity for a more precise discourse concerning the utilization of Dst in contrast to other measurable quantities, and the significance of taking into account the solar cycle over extended periods of time. We acknowledge the significance of further exploring the potential impacts of severe geomagnetic storms on different infrastructure and industry sectors. We will guarantee that these discussions are expanded upon and clearly communicated.
The insights you provided on the pre-processing of the Dst data, the selection of thresholds, and the application of EVT distributions are extremely valuable. We will elucidate these methodological aspects, guaranteeing that our selections and their ramifications are evident and thoroughly elucidated.
We will meticulously address the ambiguity you identified in our visual representations and specific sections of our analysis. Our objective is to communicate our research results in a manner that is both scientifically rigorous and easily understandable to readers, including those in the insurance sector.
We acknowledge the significance of basing our research on pragmatic implementations, particularly in relation to the insurance sector. Based on your feedback, we will improve our manuscript by providing a more concentrated analysis of how our findings can be practically implemented, potentially incorporating instances of estimating potential financial harm.
We are in the process of crafting a comprehensive reply that will specifically tackle each of your concerns. Our aim is to make all the required modifications to meet the rigorous criteria of scientific investigation and publishing.
We appreciate your comprehensive evaluation and valuable suggestions. We anticipate improving our manuscript accordingly and resubmitting it for your evaluation.
Best regards,
D. Güney Akkor
Citation: https://doi.org/10.5194/egusphere-2023-1800-AC5 -
AC6: 'Reply on RC3', Deniz Güney Akkor, 22 Jan 2024
Dear Referee,
I hope this message finds you well. I am writing to express my gratitude for your insightful and constructive feedback on our manuscript titled "Probability estimation of March 1989-like geomagnetic storms and their relevance for the insurance industry" Your expert review has provided us with valuable perspectives that are instrumental in enhancing the quality of our work.We have carefully considered each of your comments and have taken steps to address the concerns and suggestions raised. Allow me to briefly summarize our responses and the corresponding revisions made to the manuscript:
-
Relevance to the Insurance Sector: We have expanded our discussion to emphasize the predictive aspect of geomagnetic storms, which forms the foundation of risk management in the insurance sector. This includes a more in-depth analysis of the impact on infrastructure and industry sectors, particularly focusing on the implications for insurance coverage components.
-
Methodology Clarifications: We've clarified the methodologies used in our study, specifically addressing the use of the Generalized Pareto Distribution (GPD) in our analysis. We have ensured that the revised manuscript accurately reflects our approach and rationale behind the choice of statistical models and thresholds.
-
Threshold Selection and Analysis: We've provided a more detailed explanation of our dual-threshold approach and the rationale behind our threshold selections. This includes a clearer presentation of our methodology in distinguishing between individual geomagnetic storm events and the selection of representative Dst values for each storm.
-
Model Fitting and Distribution Choice: We have revised Sections 3.2 and 2.4 to accurately reflect our exclusive use of the GPD, addressing the confusion around the GEV distribution and the Weibull Law. We've ensured that these sections are now consistent and provide a clear understanding of our analytical framework.
-
General Refinements: In response to your other valuable insights, we have made additional refinements throughout the manuscript to enhance its clarity, coherence, and overall academic rigor. This includes revising figures and captions for better representation and understanding.
We believe these revisions address your concerns and significantly improve the manuscript. We are grateful for the opportunity to refine our work based on your feedback, and we are hopeful that the revised manuscript will meet your expectations and the high standards of the journal.
Thank you once again for your time and valuable insights. We look forward to any further suggestions or comments you may have.
Best regards,
D. Güney Akkor -
-
AC5: 'Reply on RC3', Deniz Güney Akkor, 10 Jan 2024
Status: closed
-
RC1: 'Comment on egusphere-2023-1800', Anonymous Referee #1, 24 Sep 2023
-
AC1: 'Reply on RC1', Deniz Güney Akkor, 25 Sep 2023
Dear Referee,
We would like to express our gratitude for your comprehensive review of our manuscript. Your expert input is invaluable, and we appreciate the time you've dedicated to evaluating our work. We've made it our priority to address all the concerns and suggestions you've raised.
We've taken steps to clarify the scope and limitations of our study, giving particular attention to the statistical methods and their applicability in the context of the data available. We also acknowledge your point on the lack of economic analysis; however, the focus of our current research is strictly on applying statistical data science methods to existing, concrete scientific data. In doing so, we believe our study meets the scientific requirements for such an application and the subsequent interpretation of the results.
We've endeavored to provide a more precise explanation for the return period related to the March 1989 event, emphasizing the EVT's value in answering questions about events that may not be present in the existing data records. We have taken note of your suggestion to be cautious in interpreting results that extend beyond the limitations of our dataset, and we intend to focus on this aspect in our future research.
Concerning the incorrect and incomplete references, we sincerely regret these oversights and assure you they will be rectified. We are also in agreement to update the tables and figure labels according to the journal's standards.
We have consulted pioneering studies in this field such as Rogers et al. (2021) and Chapman et al. (2020) prior to the conception of this work. However, we found that a constant index like Dst was a better fit for our modeling needs. This study is aligned with the existing literature and contributes to ongoing debates in the space weather science community, especially those related to events most likely to occur within our lifetimes.
Lastly, we are eagerly awaiting the updated RAE report by Cannon et al., as it promises to provide additional context and relevance for our work.
If our responses to your concerns are found to be satisfactory, we would like to proceed with the necessary revisions to improve the quality of the manuscript for re-submission. Thank you again for your valuable feedback, and we look forward to your response.
Best regards,
D. Güney Akkor
-
RC2: 'Reply on AC1', Anonymous Referee #1, 29 Sep 2023
This paper might be worth referencing as well:
Bergin, A., Chapman, S. C., Watkins, N. W., Moloney, N. R., & Gjerloev, J. W. (2023). Extreme event statistics in Dst, SYM-H, and SMR geomagnetic indices. Space Weather, 21, e2022SW003304. https://doi.org/10.1029/2022SW003304Citation: https://doi.org/10.5194/egusphere-2023-1800-RC2 -
AC2: 'Reply on RC2', Deniz Güney Akkor, 29 Sep 2023
Dear Referee,
We express our gratitude for dedicating your time to reviewing our manuscript and providing us with your valuable feedback. We acknowledge and value your recommendation to cite the aforementioned scholarly article.
Bergin, A., Chapman, S. C., Watkins, N. W., Moloney, N. R., and Gjerloev, J. W. (2023). This study focuses on the statistical analysis of extreme events in three geomagnetic indices, namely Dst, SYM-H, and SMR. The article titled "Space Weather, 21, e2022SW003304" can be accessed at the following DOI: https://doi.org/10.1029/2022SW003304.
After conducting a thorough examination of the aforementioned paper, we concur that it provides valuable insights that are pertinent to our research. The reference will be integrated into our revised manuscript, and its findings will be thoroughly examined and discussed within the framework of our research.
We express our gratitude for your meticulous focus on detail and constructive suggestions, which enhance the quality of our work.
We have updated our responses to your initial comments in the attached PDF (those highlighted in green are new additions based on your feedback, which we will incorporate into our draft and share a follow-up version with you if you find our responses reasonable, those highlighted in yellow are still open for comment, and those highlighted in red have not yet been made by us).
We look forward to receiving your feedback.
Best regards,
D. Güney Akkor
- AC3: 'Reply on RC2', Deniz Güney Akkor, 06 Oct 2023
-
AC4: 'Reply on RC2', Deniz Güney Akkor, 29 Oct 2023
Dear Refree 1,
Thank you for your insightful comments and suggestions, particularly regarding the utilization of the SMY and SMU indices. I would like to clarify that we have indeed followed a methodology similar to the one you recommended and have performed estimations employing these indices. Given the context of our study and its primary objectives, we've determined that granularity at the minute-level isn't imperative. Thus, our findings, without resorting to such granularity, have further reinforced the robustness and validity of our results.
From the perspective of insurance and risk management, we believe that our current approach in estimating the global effects of geomagnetic storms provides a sufficient foundation for risk modeling at this juncture. However, we do recognize the potential advantages of a more granular analysis using the SMY and SMU indices and are planning to incorporate them in our subsequent research endeavors to provide a more comprehensive view.
Your constructive feedback has been instrumental in shaping our future research direction. We sincerely appreciate your valuable contribution to this work and look forward to considering your insights in our forthcoming studies.
Kind regards,
D. Güney Akkor
Citation: https://doi.org/10.5194/egusphere-2023-1800-AC4
-
AC2: 'Reply on RC2', Deniz Güney Akkor, 29 Sep 2023
-
RC2: 'Reply on AC1', Anonymous Referee #1, 29 Sep 2023
-
AC1: 'Reply on RC1', Deniz Güney Akkor, 25 Sep 2023
-
RC3: 'Comment on egusphere-2023-1800', Anonymous Referee #2, 10 Jan 2024
The authors study the extremes of the Dst time series from 1957 - 2023
to estimate the risk of a future March 1989 event for the insurance
industry.- Section 1.2 suggests that the "relevance of these storms for the
insurance sector" will be determined, as well as emphasizing a "holistic
strategy that considers... power grid models, geological influences
and transmission system features." But I find no such discussion in
the paper.- In Section 2 I would expect at least some brief summary on the use
of Dst (as opposed to other observables) for measuring storm
strength, and e.g. why on long timescales one need not consider the
solar cycle, etc.- Section 2.3 suggests that knowledge of "potential effects of extreme
geomagnetic storms on numerous infrastructure and industry
sectors..." will be advanced. But, again, no such discussion
appears.- Section 3.1 (line 139) is slightly ambiguous as to who did the
pre-processing of the Dst data. Did the authors pre-process the data
(if so, how?), or they used the publicly available Dst data?- Section 3.1 (line 141): why are negative thresholds used? Haven't
the values been made positive as per the previous sentence?- Section 3.2 confuses somewhat the choice of EVT fitting
distribution. Section 2.4 announced it would be the Weibull
distribution, while the title of Section 3.2 suggests it is rather
the GEV distribution, usually used for block maxima, which includes
the Weibull family as a special case. The authors appear to settle
on a GEV distribution. But I find no mention of the block size, or
why a block maxima approach was favored over, say, a GPD fit to the
empirical exceedance distribution. Meanwhile, Figure 2 suggests a
GPD fit.- Section 4.1 (line 233) appears to refer to declustering on short
timescales. A more detailed discussion of the 24-hour gap is
required. How do the authors pick a representative Dst value for a
particular storm?- Figure 2 features exponential models(?) which are not discussed in the
text (the caption is also inadequate). Is this a GPD with shape
parameter zero (which should reduce to an exponential)?- What is there new to learn from Figure 3 which isn't already
included in Figure 1? Didn't we rather need to see the mean excess
plot to identify the onset of linear behavior previously mentioned?
Furthermore, isn't the mean excess (aka expected shortfall) a
central quantity for the insurance industry? Why is there no plot
of the GPD fit to the empirical tail distribution?- Section 4 (line 273). The authors appear to settle on 330nT as a
threshold. But subsequent analysis uses 300nT. The whole 150 vs
300nT discussion was unclear to me. Isn't the goal to use the lowest
threshold possible, while remaining within the "asymptotic" tail regime,
as evidenced by linear mean excess plot regions and stability
of shape parameters? One can discuss robustness with respect to
threshold choice by sweeping over threshold choices, rather than
singling out 150nT in particular.- Can't Figures 5 and 6 be combined? Would that not illustrate
that the 300nT threshold choice is responsible for the larger
confidence intervals, owing to the fewer exceedances remaining? As
mentioned by another referee, the extrapolation to 1,000 year return
periods is far-fetched, particularly for the insurance industry.- The paper is unfortunately full of vague and overpromising sentences
such as (line 357) "Our findings enable more effective application
of updated risk assessment methods for geomagnetic storms in the
context of the insurance industry." While the analysis presented
doesn't advance the state-of-the art in modelling space weather
extremes, one might hope for a detailed discussion of how
the analysis could be used in the insurance industry (e.g. to
estimate potential damage in dollars, say). But no such discussion
is presented.Citation: https://doi.org/10.5194/egusphere-2023-1800-RC3 -
AC5: 'Reply on RC3', Deniz Güney Akkor, 10 Jan 2024
Dear Referee 2,
We sincerely value your thorough and perceptive evaluation of our manuscript. We greatly appreciate your expertise and the thoroughness of your feedback. Your input is extremely valuable to us, and we are thankful for the time and effort you have dedicated to reviewing our work.
We have carefully examined all of your comments and suggestions and are fully dedicated to addressing them in a comprehensive manner. In our response, we will provide a comprehensive account of the measures we have implemented to improve our study, with a specific emphasis on augmenting its lucidity, pertinence, and scientific meticulousness.
We recognize the necessity for a more precise discourse concerning the utilization of Dst in contrast to other measurable quantities, and the significance of taking into account the solar cycle over extended periods of time. We acknowledge the significance of further exploring the potential impacts of severe geomagnetic storms on different infrastructure and industry sectors. We will guarantee that these discussions are expanded upon and clearly communicated.
The insights you provided on the pre-processing of the Dst data, the selection of thresholds, and the application of EVT distributions are extremely valuable. We will elucidate these methodological aspects, guaranteeing that our selections and their ramifications are evident and thoroughly elucidated.
We will meticulously address the ambiguity you identified in our visual representations and specific sections of our analysis. Our objective is to communicate our research results in a manner that is both scientifically rigorous and easily understandable to readers, including those in the insurance sector.
We acknowledge the significance of basing our research on pragmatic implementations, particularly in relation to the insurance sector. Based on your feedback, we will improve our manuscript by providing a more concentrated analysis of how our findings can be practically implemented, potentially incorporating instances of estimating potential financial harm.
We are in the process of crafting a comprehensive reply that will specifically tackle each of your concerns. Our aim is to make all the required modifications to meet the rigorous criteria of scientific investigation and publishing.
We appreciate your comprehensive evaluation and valuable suggestions. We anticipate improving our manuscript accordingly and resubmitting it for your evaluation.
Best regards,
D. Güney Akkor
Citation: https://doi.org/10.5194/egusphere-2023-1800-AC5 -
AC6: 'Reply on RC3', Deniz Güney Akkor, 22 Jan 2024
Dear Referee,
I hope this message finds you well. I am writing to express my gratitude for your insightful and constructive feedback on our manuscript titled "Probability estimation of March 1989-like geomagnetic storms and their relevance for the insurance industry" Your expert review has provided us with valuable perspectives that are instrumental in enhancing the quality of our work.We have carefully considered each of your comments and have taken steps to address the concerns and suggestions raised. Allow me to briefly summarize our responses and the corresponding revisions made to the manuscript:
-
Relevance to the Insurance Sector: We have expanded our discussion to emphasize the predictive aspect of geomagnetic storms, which forms the foundation of risk management in the insurance sector. This includes a more in-depth analysis of the impact on infrastructure and industry sectors, particularly focusing on the implications for insurance coverage components.
-
Methodology Clarifications: We've clarified the methodologies used in our study, specifically addressing the use of the Generalized Pareto Distribution (GPD) in our analysis. We have ensured that the revised manuscript accurately reflects our approach and rationale behind the choice of statistical models and thresholds.
-
Threshold Selection and Analysis: We've provided a more detailed explanation of our dual-threshold approach and the rationale behind our threshold selections. This includes a clearer presentation of our methodology in distinguishing between individual geomagnetic storm events and the selection of representative Dst values for each storm.
-
Model Fitting and Distribution Choice: We have revised Sections 3.2 and 2.4 to accurately reflect our exclusive use of the GPD, addressing the confusion around the GEV distribution and the Weibull Law. We've ensured that these sections are now consistent and provide a clear understanding of our analytical framework.
-
General Refinements: In response to your other valuable insights, we have made additional refinements throughout the manuscript to enhance its clarity, coherence, and overall academic rigor. This includes revising figures and captions for better representation and understanding.
We believe these revisions address your concerns and significantly improve the manuscript. We are grateful for the opportunity to refine our work based on your feedback, and we are hopeful that the revised manuscript will meet your expectations and the high standards of the journal.
Thank you once again for your time and valuable insights. We look forward to any further suggestions or comments you may have.
Best regards,
D. Güney Akkor -
-
AC5: 'Reply on RC3', Deniz Güney Akkor, 10 Jan 2024
Data sets
Dst index World Data Center for Geomagnetism, Kyoto http://wdc.kugi.kyoto-u.ac.jp/
Model code and software
Pyextremes George Bocharov https://georgebv.github.io/pyextremes/
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 486 | 163 | 46 | 695 | 27 | 32 |
- HTML: 486
- PDF: 163
- XML: 46
- Total: 695
- BibTeX: 27
- EndNote: 32
Viewed (geographical distribution)
| Country | # | Views | % |
|---|---|---|---|
| United States of America | 1 | 277 | 41 |
| Germany | 2 | 87 | 13 |
| China | 3 | 39 | 5 |
| United Kingdom | 4 | 30 | 4 |
| Turkey | 5 | 29 | 4 |
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
- 277