the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 03 Feb 2025
Analysis of Long-Term Changes in Extreme Waves in the Northwest Pacific Over the Past 60 Years
Abstract. This study analyses wave height trends in the Northwest Pacific over the past 60 years and estimates design wave heights across various return periods to assess the resilience of marine and coastal structures to extreme wave events. Design wave height is a critical parameter for evaluating structural stability and safety, especially during typhoon season (May to October), when strong winds and rapid movements often trigger extreme waves, significantly impacting offshore structures, coastlines, and ports. To avoid underestimating risks during typhoon season, this study simulated wave heights from 1961 to 2020 using historical wind field data from the EC-Earth3 climate model and the WAVEWATCH III wave model. The 95th percentile was chosen as the threshold for extreme wave events, and the Generalized Pareto Distribution (GPD) model was applied for fitting. Finally, the bootstrap resampling method was used to quantify uncertainties in return periods to ensure reliable assessments of design wave heights. The analysis shows a slight increase in design wave heights with longer return periods (10 to 200 years) near Taiwan, with significantly higher wave heights observed in the southern and eastern regions, indicating a need to enhance disaster resilience in marine infrastructure designs for these areas.
- Preprint
(2262 KB) - Metadata XML
- BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2024-3954', Anonymous Referee #1, 30 Mar 2025
This manuscript deals with storm surge characteristics around the northern coast of Kyushu Island in Japan. The topic is interesting, and it should be published. However, the manuscript needs to be improved before publication.
Major points
- Section 2 is too short. It should be combined with the introduction.
- The description of the two assumptions in section 3.2 is incorrect. It should be updated correctly.
- The locations of the time series in Figures 1-4 are not easy to understand for non-East Asian readers. It should be indicated.
- The validation of historical runs needs to be improved. The author can compare the historical runs comparing with the altimeters.
- The author estimated a low probability of extreme wave heights, but the results do not match those of previous studies. For example, the extreme wave heights over 30-50-year return periods show particular typhoon tracks due to low probability. However, Figure 6 does not show such characteristics of extreme waves but has strange spotters-like noises. The author needs to clarify these characteristics.
Minor points
- The notions of horizontal and vertical winds in section 3 should be reworded as longitudinal and latitudinal winds.
Citation: https://doi.org/10.5194/egusphere-2024-3954-RC1 -
AC1: 'Reply on RC1', Yang-Ming Fan, 23 Jun 2025
Dear Reviewer,
Thank you for your valuable and constructive comments, which have contributed to improving the quality of the manuscript. All comments have been carefully reviewed and fully responded to. Detailed responses to all comments are provided below.
Reviewer’s comment 1:
Section 2 is too short. It should be combined with the introduction.
Response:
Section 2 has been substantially revised by adding a new paragraph at the end to explicitly define the research scope. This new content emphasizes that the application of a deep learning-based downscaling approach enables the generation of high-resolution wave data from coarse-resolution global climate model outputs, which is essential for advancing the analysis of wave variability and impacts in Taiwan’s coastal regions. Additionally, the Introduction (Section 1) has been revised to more clearly present the originality and scientific contribution of this study.
Reviewer’s comment 2:
The description of the two assumptions in section 3.2 is incorrect. It should be updated correctly.
Response:
The model assumptions described in Section 3.2 have been revised for accuracy. The current text now correctly reflects the fundamental assumptions of the WAVEWATCH III wave model.
Reviewer’s comment 3:
The locations of the time series in Figures 1-4 are not easy to understand for non-East Asian readers. It should be indicated.
Response:
A schematic map indicating the locations has been added as a new figure, with a corresponding explanation provided in the revised manuscript. The relevant sections have been revised accordingly, and details will be provided in the revised manuscript.
Reviewer’s comment 4:
The validation of historical runs needs to be improved. The author can compare the historical runs comparing with the altimeters.
Response:
Validation of the historical simulations has been improved by comparison with offshore data buoy observations. A new subsection, “Validation with In-situ Observations” has been included in the revised manuscript. In this subsection, the reliability of the high-resolution wave simulations generated by the CRNN model is evaluated against in-situ buoy observations from coastal waters around Taiwan. The results demonstrate that both the timing and magnitude of major wave events are generally well represented, and statistical metrics such as RMSE and SI further indicate good agreement between the simulations and observations.
We acknowledge that satellite altimeter data are commonly used for model validation on a global scale. However, because the primary objective of this study is to investigate wave variability in nearshore regions, validation was focused on offshore buoys located only a few kilometers from the coast. These buoys provide high-frequency, site-specific measurements that are more representative of nearshore wave conditions than satellite altimeter data, which are generally limited in spatial and temporal resolution near the coast. Therefore, model validation in this study is primarily based on in-situ buoy observations.
Reviewer’s comment 5:
The author estimated a low probability of extreme wave heights, but the results do not match those of previous studies. For example, the extreme wave heights over 30-50-year return periods show particular typhoon tracks due to low probability. However, Figure 6 does not show such characteristics of extreme waves but has strange spotters-like noises. The author needs to clarify these characteristics.
Response:
The spatial distribution of extreme wave heights for 30–50-year return periods in this study does not always correspond to specific typhoon tracks as expected, but instead shows some localized spotter-like features in certain regions. In this analysis, the maximum significant wave height from each typhoon event at each grid point was used as the independent extreme sample for estimating design wave heights. While this approach is widely used in extreme value analysis, it can result in the spatial distribution of extreme values being dominated by a small number of intense typhoon events, particularly in regions where historical typhoon tracks are sparse or highly variable. Consequently, the extrapolated return period estimates may display discontinuities or localized spotter-like patterns, reflecting the influence of individual extreme events rather than continuous storm tracks.
Further examination of the historical wave height simulations indicates that the spatial distribution is continuous and physically reasonable, without the presence of such spotter-like features, demonstrating that the model can effectively reproduce major historical events. The occurrence of spotter-like features is therefore mainly associated with the statistical extrapolation process, rather than deficiencies in the wave model itself. This methodological aspect and its implications for interpreting return period maps have been clarified and discussed in the revised manuscript.
Reviewer’s comment 6:
The notions of horizontal and vertical winds in section 3 should be reworded as longitudinal and latitudinal winds.
Response:
The original text in Section 3 has been fully revised to highlight the rationale for using wave data generated by the WAVEWATCH III model driven by wind fields from the EC-Earth3 model. The revised text no longer refers to "horizontal" or "vertical" winds.All relevant sections have been updated accordingly, and details will be provided in the revised manuscript. It is hoped that the responses and proposed revisions provided can sufficiently address the concerns raised. Further feedback would be greatly appreciated.
Best regards,
Yang-Ming FanCitation: https://doi.org/10.5194/egusphere-2024-3954-AC1
-
RC2: 'Comment on egusphere-2024-3954', Anonymous Referee #2, 18 May 2025
The originality of the study and its clear contribution to the literature are not presented in the article. The data used to achieve the objective of the study is not appropriate. How accurate is the calculation of the design waves of a region using climate change data? It is a question mark. Why such an approach is needed when there is historical climate data? Why climate change is not examined if climate change data is to be used? The introduction and literature presented in the study are far from the main objective. The coarse resolution climate data is converted to high resolution using CRNN, but no evidence is presented regarding the accuracy of the data produced. No detailed information about the method applied is given. It is doubtful that the data inserted into the GPD are independent storms.
The objective of the study and the information given in the literature do not match. Why is it that extreme waves in the past 60 years are analyzed but future scenarios are mentioned?
Why is the analysis of extreme waves in the past period done with climate change data but not with past climate data?
Similar expressions appear in different sections. Repetitive expressions disrupt the flow.
Do you need table 1?
Why monthly average data was calculated? This part is out of the main scope.
Do the data in Figs. 1-4 represent an average of the data of the grid points of a region or a location in the respective regions? The boundaries of the regions should be shown on a map.
What was used as output for developing the model in CRNN?
Were all values above the 95th percentile considered as extreme events? How were independent events sorted?
Citation: https://doi.org/10.5194/egusphere-2024-3954-RC2 -
AC2: 'Reply on RC2', Yang-Ming Fan, 23 Jun 2025
Dear Reviewer,
Thank you for the detailed and critical comments, which have been invaluable in identifying areas for improvement in the manuscript. For clarity and consistency, related comments have been grouped and addressed together where appropriate. All comments have been carefully reviewed and fully responded to. Detailed responses to all comments are provided below.
Reviewer’s comment 1:
The originality of the study and its clear contribution to the literature are not presented in the article.
Response:
The manuscript has been revised to more clearly present the originality and scientific contribution of this study. The background, motivation, and the novelty of applying a deep learning-based downscaling approach for regional wave analysis and design wave estimation have been explicitly highlighted. These revisions address the reviewer’s concern by clarifying both the unique aspects and the scientific value of the study in the context of existing literature.
Reviewer’s comment 2:
The data used to achieve the objective of the study is not appropriate. How accurate is the calculation of the design waves of a region using climate change data? It is a question mark. Why such an approach is needed when there is historical climate data? Why climate change is not examined if climate change data is to be used?
The coarse resolution climate data is converted to high resolution using CRNN, but no evidence is presented regarding the accuracy of the data produced. No detailed information about the method applied is given.
The objective of the study and the information given in the literature do not match. Why is it that extreme waves in the past 60 years are analyzed but future scenarios are mentioned?
Why is the analysis of extreme waves in the past period done with climate change data but not with past climate data?
Response:
Observational and reanalysis data are widely recognized as the most reliable sources for examining past wave climate. However, to provide a consistent modeling basis for both historical analysis and future projections—which cannot be achieved using only observational or reanalysis datasets—a climate model-based approach was adopted in this study. This allows for the application of a unified framework to investigate long-term wave trends and ensures internal consistency between past and future analyses. Although the current manuscript focuses on historical wave simulations for the period 1961–2020, future wave projections have also been developed and will be analyzed in a separate study.
Further, to address concerns regarding the accuracy of design wave calculations using climate model data, a new subsection, “Validation with In-situ Observations” has been included. In this subsection, the reliability of the high-resolution wave simulations generated by the CRNN model is evaluated against in-situ buoy observations from coastal waters around Taiwan. The results demonstrate that both the timing and magnitude of major wave events are generally well represented. Statistical metrics such as RMSE and SI further indicate good agreement in the overall trends between the simulations and observations.
In addition, the revised manuscript has included additional details about the CRNN method applied, including the model architecture, data processing, training procedures, and hyperparameter settings.
Reviewer’s comment 3:
The introduction and literature presented in the study are far from the main objective.
Response:
The Introduction section has been thoroughly revised to better align with the main objective of the study. The revised text now clearly presents the scientific motivation, summarizes relevant literature with a focus on the limitations of existing wave climate projections for Taiwan, and explicitly highlights the knowledge gap addressed by this work. The main objectives and contributions of the present study are now clearly articulated at the end of the Introduction.
Reviewer’s comment 4:
It is doubtful that the data inserted into the GPD are independent storms.
Were all values above the 95th percentile considered as extreme events? How were independent events sorted?
Response:
The procedure for selecting independent extreme events for GPD analysis has been clarified in the revised manuscript. Specifically, the maximum significant wave height from each typhoon event was extracted to ensure statistical independence among samples. Subsequently, only those event maxima exceeding the 95th percentile of all typhoon maxima during 1961–2020 were included in the GPD analysis.
Reviewer’s comment 5:
Similar expressions appear in different sections. Repetitive expressions disrupt the flow.
Response:
The revised manuscript has been carefully reviewed to eliminate unnecessary repetition across different sections.
Reviewer’s comment 6:
Do you need table 1?
Response:
Table 1 has been removed from the revised manuscript because the present study focuses on historical wave conditions, while Table 1 described future scenario settings that are not analyzed in this study.
Reviewer’s comment 7:
Why monthly average data was calculated? This part is out of the main scope.
Response:
The calculation of monthly mean wave height data provides important information on long-term variability and general trends in regional wave conditions. This analysis offers essential background context, enabling readers to interpret the main study results within the broader framework of regional wave climate changes over time.
Reviewer’s comment 8:
Do the data in Figs. 1–4 represent an average of the data of the grid points of a region or a location in the respective regions? The boundaries of the regions should be shown on a map.
Response:
The data presented in Figures 1–4 represent spatial averages of all grid points within each defined region. In addition, a schematic map illustrating the boundaries of each region has been added to the revised manuscript.
Reviewer’s comment 9:
What was used as output for developing the model in CRNN?
Response:
The output used for developing the CRNN model was the high-resolution significant wave height field.
All relevant sections have been updated accordingly, and details will be provided in the revised manuscript. It is hoped that the responses and proposed revisions provided can sufficiently address the concerns raised. Further feedback would be greatly appreciated.Best regards,
Yang-Ming FanCitation: https://doi.org/10.5194/egusphere-2024-3954-AC2
-
AC2: 'Reply on RC2', Yang-Ming Fan, 23 Jun 2025
Status: closed
-
RC1: 'Comment on egusphere-2024-3954', Anonymous Referee #1, 30 Mar 2025
This manuscript deals with storm surge characteristics around the northern coast of Kyushu Island in Japan. The topic is interesting, and it should be published. However, the manuscript needs to be improved before publication.
Major points
- Section 2 is too short. It should be combined with the introduction.
- The description of the two assumptions in section 3.2 is incorrect. It should be updated correctly.
- The locations of the time series in Figures 1-4 are not easy to understand for non-East Asian readers. It should be indicated.
- The validation of historical runs needs to be improved. The author can compare the historical runs comparing with the altimeters.
- The author estimated a low probability of extreme wave heights, but the results do not match those of previous studies. For example, the extreme wave heights over 30-50-year return periods show particular typhoon tracks due to low probability. However, Figure 6 does not show such characteristics of extreme waves but has strange spotters-like noises. The author needs to clarify these characteristics.
Minor points
- The notions of horizontal and vertical winds in section 3 should be reworded as longitudinal and latitudinal winds.
Citation: https://doi.org/10.5194/egusphere-2024-3954-RC1 -
AC1: 'Reply on RC1', Yang-Ming Fan, 23 Jun 2025
Dear Reviewer,
Thank you for your valuable and constructive comments, which have contributed to improving the quality of the manuscript. All comments have been carefully reviewed and fully responded to. Detailed responses to all comments are provided below.
Reviewer’s comment 1:
Section 2 is too short. It should be combined with the introduction.
Response:
Section 2 has been substantially revised by adding a new paragraph at the end to explicitly define the research scope. This new content emphasizes that the application of a deep learning-based downscaling approach enables the generation of high-resolution wave data from coarse-resolution global climate model outputs, which is essential for advancing the analysis of wave variability and impacts in Taiwan’s coastal regions. Additionally, the Introduction (Section 1) has been revised to more clearly present the originality and scientific contribution of this study.
Reviewer’s comment 2:
The description of the two assumptions in section 3.2 is incorrect. It should be updated correctly.
Response:
The model assumptions described in Section 3.2 have been revised for accuracy. The current text now correctly reflects the fundamental assumptions of the WAVEWATCH III wave model.
Reviewer’s comment 3:
The locations of the time series in Figures 1-4 are not easy to understand for non-East Asian readers. It should be indicated.
Response:
A schematic map indicating the locations has been added as a new figure, with a corresponding explanation provided in the revised manuscript. The relevant sections have been revised accordingly, and details will be provided in the revised manuscript.
Reviewer’s comment 4:
The validation of historical runs needs to be improved. The author can compare the historical runs comparing with the altimeters.
Response:
Validation of the historical simulations has been improved by comparison with offshore data buoy observations. A new subsection, “Validation with In-situ Observations” has been included in the revised manuscript. In this subsection, the reliability of the high-resolution wave simulations generated by the CRNN model is evaluated against in-situ buoy observations from coastal waters around Taiwan. The results demonstrate that both the timing and magnitude of major wave events are generally well represented, and statistical metrics such as RMSE and SI further indicate good agreement between the simulations and observations.
We acknowledge that satellite altimeter data are commonly used for model validation on a global scale. However, because the primary objective of this study is to investigate wave variability in nearshore regions, validation was focused on offshore buoys located only a few kilometers from the coast. These buoys provide high-frequency, site-specific measurements that are more representative of nearshore wave conditions than satellite altimeter data, which are generally limited in spatial and temporal resolution near the coast. Therefore, model validation in this study is primarily based on in-situ buoy observations.
Reviewer’s comment 5:
The author estimated a low probability of extreme wave heights, but the results do not match those of previous studies. For example, the extreme wave heights over 30-50-year return periods show particular typhoon tracks due to low probability. However, Figure 6 does not show such characteristics of extreme waves but has strange spotters-like noises. The author needs to clarify these characteristics.
Response:
The spatial distribution of extreme wave heights for 30–50-year return periods in this study does not always correspond to specific typhoon tracks as expected, but instead shows some localized spotter-like features in certain regions. In this analysis, the maximum significant wave height from each typhoon event at each grid point was used as the independent extreme sample for estimating design wave heights. While this approach is widely used in extreme value analysis, it can result in the spatial distribution of extreme values being dominated by a small number of intense typhoon events, particularly in regions where historical typhoon tracks are sparse or highly variable. Consequently, the extrapolated return period estimates may display discontinuities or localized spotter-like patterns, reflecting the influence of individual extreme events rather than continuous storm tracks.
Further examination of the historical wave height simulations indicates that the spatial distribution is continuous and physically reasonable, without the presence of such spotter-like features, demonstrating that the model can effectively reproduce major historical events. The occurrence of spotter-like features is therefore mainly associated with the statistical extrapolation process, rather than deficiencies in the wave model itself. This methodological aspect and its implications for interpreting return period maps have been clarified and discussed in the revised manuscript.
Reviewer’s comment 6:
The notions of horizontal and vertical winds in section 3 should be reworded as longitudinal and latitudinal winds.
Response:
The original text in Section 3 has been fully revised to highlight the rationale for using wave data generated by the WAVEWATCH III model driven by wind fields from the EC-Earth3 model. The revised text no longer refers to "horizontal" or "vertical" winds.All relevant sections have been updated accordingly, and details will be provided in the revised manuscript. It is hoped that the responses and proposed revisions provided can sufficiently address the concerns raised. Further feedback would be greatly appreciated.
Best regards,
Yang-Ming FanCitation: https://doi.org/10.5194/egusphere-2024-3954-AC1
-
RC2: 'Comment on egusphere-2024-3954', Anonymous Referee #2, 18 May 2025
The originality of the study and its clear contribution to the literature are not presented in the article. The data used to achieve the objective of the study is not appropriate. How accurate is the calculation of the design waves of a region using climate change data? It is a question mark. Why such an approach is needed when there is historical climate data? Why climate change is not examined if climate change data is to be used? The introduction and literature presented in the study are far from the main objective. The coarse resolution climate data is converted to high resolution using CRNN, but no evidence is presented regarding the accuracy of the data produced. No detailed information about the method applied is given. It is doubtful that the data inserted into the GPD are independent storms.
The objective of the study and the information given in the literature do not match. Why is it that extreme waves in the past 60 years are analyzed but future scenarios are mentioned?
Why is the analysis of extreme waves in the past period done with climate change data but not with past climate data?
Similar expressions appear in different sections. Repetitive expressions disrupt the flow.
Do you need table 1?
Why monthly average data was calculated? This part is out of the main scope.
Do the data in Figs. 1-4 represent an average of the data of the grid points of a region or a location in the respective regions? The boundaries of the regions should be shown on a map.
What was used as output for developing the model in CRNN?
Were all values above the 95th percentile considered as extreme events? How were independent events sorted?
Citation: https://doi.org/10.5194/egusphere-2024-3954-RC2 -
AC2: 'Reply on RC2', Yang-Ming Fan, 23 Jun 2025
Dear Reviewer,
Thank you for the detailed and critical comments, which have been invaluable in identifying areas for improvement in the manuscript. For clarity and consistency, related comments have been grouped and addressed together where appropriate. All comments have been carefully reviewed and fully responded to. Detailed responses to all comments are provided below.
Reviewer’s comment 1:
The originality of the study and its clear contribution to the literature are not presented in the article.
Response:
The manuscript has been revised to more clearly present the originality and scientific contribution of this study. The background, motivation, and the novelty of applying a deep learning-based downscaling approach for regional wave analysis and design wave estimation have been explicitly highlighted. These revisions address the reviewer’s concern by clarifying both the unique aspects and the scientific value of the study in the context of existing literature.
Reviewer’s comment 2:
The data used to achieve the objective of the study is not appropriate. How accurate is the calculation of the design waves of a region using climate change data? It is a question mark. Why such an approach is needed when there is historical climate data? Why climate change is not examined if climate change data is to be used?
The coarse resolution climate data is converted to high resolution using CRNN, but no evidence is presented regarding the accuracy of the data produced. No detailed information about the method applied is given.
The objective of the study and the information given in the literature do not match. Why is it that extreme waves in the past 60 years are analyzed but future scenarios are mentioned?
Why is the analysis of extreme waves in the past period done with climate change data but not with past climate data?
Response:
Observational and reanalysis data are widely recognized as the most reliable sources for examining past wave climate. However, to provide a consistent modeling basis for both historical analysis and future projections—which cannot be achieved using only observational or reanalysis datasets—a climate model-based approach was adopted in this study. This allows for the application of a unified framework to investigate long-term wave trends and ensures internal consistency between past and future analyses. Although the current manuscript focuses on historical wave simulations for the period 1961–2020, future wave projections have also been developed and will be analyzed in a separate study.
Further, to address concerns regarding the accuracy of design wave calculations using climate model data, a new subsection, “Validation with In-situ Observations” has been included. In this subsection, the reliability of the high-resolution wave simulations generated by the CRNN model is evaluated against in-situ buoy observations from coastal waters around Taiwan. The results demonstrate that both the timing and magnitude of major wave events are generally well represented. Statistical metrics such as RMSE and SI further indicate good agreement in the overall trends between the simulations and observations.
In addition, the revised manuscript has included additional details about the CRNN method applied, including the model architecture, data processing, training procedures, and hyperparameter settings.
Reviewer’s comment 3:
The introduction and literature presented in the study are far from the main objective.
Response:
The Introduction section has been thoroughly revised to better align with the main objective of the study. The revised text now clearly presents the scientific motivation, summarizes relevant literature with a focus on the limitations of existing wave climate projections for Taiwan, and explicitly highlights the knowledge gap addressed by this work. The main objectives and contributions of the present study are now clearly articulated at the end of the Introduction.
Reviewer’s comment 4:
It is doubtful that the data inserted into the GPD are independent storms.
Were all values above the 95th percentile considered as extreme events? How were independent events sorted?
Response:
The procedure for selecting independent extreme events for GPD analysis has been clarified in the revised manuscript. Specifically, the maximum significant wave height from each typhoon event was extracted to ensure statistical independence among samples. Subsequently, only those event maxima exceeding the 95th percentile of all typhoon maxima during 1961–2020 were included in the GPD analysis.
Reviewer’s comment 5:
Similar expressions appear in different sections. Repetitive expressions disrupt the flow.
Response:
The revised manuscript has been carefully reviewed to eliminate unnecessary repetition across different sections.
Reviewer’s comment 6:
Do you need table 1?
Response:
Table 1 has been removed from the revised manuscript because the present study focuses on historical wave conditions, while Table 1 described future scenario settings that are not analyzed in this study.
Reviewer’s comment 7:
Why monthly average data was calculated? This part is out of the main scope.
Response:
The calculation of monthly mean wave height data provides important information on long-term variability and general trends in regional wave conditions. This analysis offers essential background context, enabling readers to interpret the main study results within the broader framework of regional wave climate changes over time.
Reviewer’s comment 8:
Do the data in Figs. 1–4 represent an average of the data of the grid points of a region or a location in the respective regions? The boundaries of the regions should be shown on a map.
Response:
The data presented in Figures 1–4 represent spatial averages of all grid points within each defined region. In addition, a schematic map illustrating the boundaries of each region has been added to the revised manuscript.
Reviewer’s comment 9:
What was used as output for developing the model in CRNN?
Response:
The output used for developing the CRNN model was the high-resolution significant wave height field.
All relevant sections have been updated accordingly, and details will be provided in the revised manuscript. It is hoped that the responses and proposed revisions provided can sufficiently address the concerns raised. Further feedback would be greatly appreciated.Best regards,
Yang-Ming FanCitation: https://doi.org/10.5194/egusphere-2024-3954-AC2
-
AC2: 'Reply on RC2', Yang-Ming Fan, 23 Jun 2025
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 221 | 76 | 16 | 313 | 12 | 29 |
- HTML: 221
- PDF: 76
- XML: 16
- Total: 313
- BibTeX: 12
- EndNote: 29
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1