Design and Evaluation of an Efficient High-Precision Ocean Surface Wave Model with a Multiscale Grid System (MSG_Wav1.0)

Li, Jiangyu; Zhang, Shaoqing; Liu, Qingxiang; Yu, Xiaolin; Zhang, Zhiwei

doi:https://doi.org/10.5194/gmd-2023-24

Preprints

https://doi.org/10.5194/gmd-2023-24

Preprints

Submitted as: development and technical paper

13 Mar 2023

Submitted as: development and technical paper |

| 13 Mar 2023

Status: a revised version of this preprint was accepted for the journal GMD and is expected to appear here in due course.

Design and Evaluation of an Efficient High-Precision Ocean Surface Wave Model with a Multiscale Grid System (MSG_Wav1.0)

Jiangyu Li, Shaoqing Zhang, Qingxiang Liu, Xiaolin Yu, and Zhiwei Zhang

Abstract. Ocean surface waves induced by wind forcing and topographic effects are a crucial physical process at the air-sea interface, which significantly affect typhoon development, ocean mixing, etc. Higher-resolution wave modeling can simulate more accurate wave states, but requires huge computational resources, making it difficult for Earth system models to include ocean waves as a fast-response physical process. Given that high-resolution Earth system models are in demand, efficient high-precision wave simulation is necessary and urgent. Based on the wave dispersion relation, we design a new wave modeling framework using a multiscale grid system. It has the fewest number of fine grids and reasonable grid spacing in deep water areas. We compare the performance of wave simulation using different spatial propagation schemes, reveal the different reasons for wave simulation differences in the westerly zone and the active tropical cyclone region, and quantify the matching of spatial resolutions between wave models and wind forcing. A series of numerical experiments show that this new modeling framework can more precisely simulate wave states in shallow water areas without losing accuracy in the deep ocean while costing a small fraction of traditional simulations with uniform fine-gridding space. With affordable computational expenses, the new ocean surface wave modeling can be implemented into high-resolution Earth system models, which may significantly improve the simulation of the atmospheric planetary boundary layer and upper-ocean mixing.

Received: 09 Feb 2023 – Discussion started: 13 Mar 2023

Download & links

Jiangyu Li et al.

Status: closed

CEC1:
'Comment on gmd-2023-24', Juan Antonio Añel, 06 Apr 2023

Dear authors,
Unfortunately, after checking your manuscript, it has come to our attention that it does not comply with our "Code and Data Policy".

https://www.geoscientific-model-development.net/policies/code_and_data_policy.html
Neither of the code or data repositories that you mention in the "Code and data availability" section of your manuscript (GitHub, NOAA or Copernicus servers, and Aquaveo) are suitable repositories for scientific publishing. You must store all the code and data relevant to your manuscript in one acceptable according to our policy. Currently, only the Zenodo repository for a few files is in compliance. Moreover, the link to NOAA and Copernicus servers could not be enough specific to locate the exact files necessary to replicate your work.
I highlight that the situation with your submission is currently irregular, as your manuscript should have never been accepted for Discussions with such a problem. Therefore, we will have to reject your manuscript unless you address these issues in a prompt manner.
In this way, you must reply to this comment as soon as possible with the DOIs and links for the new repositories containing the code and data.
Also, I would like to note that in the only repository that you have shared correctly, many files are in .mat format. .mat is a proprietary binary data container format. In this way, these files can only be accessed with a given version of proprietary software, which precludes the ability to access the data. I would suggest you store these files in an open format, accessible without the need for specific software (ASCII, NetCDF, etc.)
I emphasize that failing to comply with this request to store the code and data, we will reject your manuscript.
Juan A. Añel
Geosci. Model Dev. Executive Editor

Citation: https://doi.org/10.5194/gmd-2023-24-CEC1
- AC1:
  'Reply on CEC1', Jiangyu Li, 10 Apr 2023
  
  Thank you very much for your comments, we will strictly follow the code and data policy of GMD. According to your comments and the policy, we have made the following changes.
  (1) We have stored all the codes and data related to this paper in the Zenodo: https:/doi.org/10.5281/zenodo.7811681 (last access: April 10, 2023). They include a) the wind forcing data, the configure, boundary, topography, and control files required for the wave model run, and b) the observation data, the model output data, and its post-processing codes. Based on these codes and data, readers can repeat our work completely.
  (2) We have converted all the files with .mat format to NetCDF or ASCII format.
  Thanks again.
  
  Citation: https://doi.org/10.5194/gmd-2023-24-AC1
  - CEC2: 'Reply on AC1', Juan Antonio Añel, 13 Apr 2023
    
    Dear authors,
    Unfortunately, I have to bring to your attention (and to the Topical Editor) that despite depositing some data files in Zenodo, your submission continues not to comply with our policy. Therefore, unless you fix it (please, in a prompt manner), we will have to proceed to reject your submission for publication.
    We continue missing the repository with the code of WaveWatch III and the Surface-water Modeling System software (SMS). Please, upload them to one of the acceptable repositories and reply to this comment with the information to access them (link and DOI).
    Juan A. Añel
    Geosci. Model Dev. Executive Editor
    
    Citation: https://doi.org/10.5194/gmd-2023-24-CEC2
    
    AC2: 'Reply on CEC2', Jiangyu Li, 14 Apr 2023
    
    Thank you very much for your comments.
    We have uploaded everything related to this paper to Zenodo: https:/doi.org/10.5281/zenodo.7827541 (last access: April 14, 2023). They include:
    a) the source code of wave model WW3 (wwatch3.v5.16.tar.gz),
    b) the Surface-water Modeling System software (SMS) software (software_SMS.zip),
    c) the wind forcing, boundary, topography, configuration, and control files required for the wave model run (wind_file.rar, boundary_file_*.rar, topography_file.rar, configuration_file.rar, and control_file.rar), and
    d) the observation data, the model output data, and its post-processing codes (output_data_and_code.rar).
    We hope these can comply with the code and data policy of GMD. Thanks again.
    
    Citation: https://doi.org/10.5194/gmd-2023-24-AC2
    
    CEC3: 'Reply on AC2', Juan Antonio Añel, 15 Apr 2023
    
    Dear authors,
    Many thanks for your reply and for adding part of the requested material to the new repository. It is someway disappointing that for the SMS software, you can only share a binary .exe file. In such a way, it is not possible to verify if your results, in the end, are correct, as it is necessary to trust that this proprietary software works correctly. It is not possible to check it.
    However, you have probably done your best to try to comply with our requests, and the main problem is that when you designed your research, you chose the SMS software to carry it on. This hardly has a solution now. However, if you, the topical editor or the reviewers know of a software or library that you can use to replace the use of SMS in your work, I encourage you to replace SMS to comply with the principle of scientific reproducibility and submit a version of the manuscript with results using FLOSS.
    Beyond this issue, I consider that you now comply with our policy regarding code, and we can continue with the review process for your manuscript.
    Juan A. Añel
    Geosci. Model Dev. Executive Editor
    
    Citation: https://doi.org/10.5194/gmd-2023-24-CEC3
    
    AC3: 'Reply on CEC3', Jiangyu Li, 19 Apr 2023
    
    Thanks for your comments and suggestions. We couldn’t agree more with your opinions. If the topic editor or the reviewers know a more powerful software replacing the SMS tool, we will follow their request and suggestions reproducing the process of generating unstructured grid files.
    
    Thanks a lot!
    
    Citation: https://doi.org/10.5194/gmd-2023-24-AC3
RC1:
'Comment on gmd-2023-24', Anonymous Referee #1, 06 Jun 2023
This paper illustrates a significant work in high-resolution wave model development. The multiscale griding framework shows it can mitigate the hight computing cost while keeping a good performance in coastal regions. The reasoning is clear supported by evidence. I believe this technique will make a significant contribution to HRM development. I only have a few minor comments/suggestions. See them below.
Line 103: ‘acceptable’ is a subjective word. Suggest removing it or changing to an objective word.

Line 207-208: what is 'a reasonable range'? Better to say it explicitly.

Lines 259-261: is it better to move this to discussion section?

Line 295: again, ‘acceptable’ is a subjective word.

Paragraph with Line 300: any better to move it to discussion section?

Paragraph withLine 325: is it true that WS_multi3 and WS_multi3(new) have the same grid number? Resolution in Area1 is higher in WS_multi3(new), meaning a (slightly?) larger number of grid than in WS_multi3.

Lines 514-517: there are no analysis on era-i and CFSR2 forced wave simulations in this paper. Remove this or put as a discussion point?

Line ~440: WS_multi4 has the finest resolution only in SCS, which is a small proportion in the coastal regions of the whole domain. If the gridding strategy in WS_multi4 is applied to the whole domain, the computing demand could increase significantly. It deserves a clarification for this point, otherwise it may cause a misleading.

Future work around line 445: the 1^st and 2^nd points are not distinctive as some thoughts are mixed up. Suggest rephrasing these two paragraphs.

Future work around line 452: it’s worth mentioning in a coupling context, whether the feedback of waves to the atmosphere and ocean is sensitive to wave multiscale resolution. For example, one may assume that a coarser resolution wave model produces less wave effect on atm and ocean. Does this mean inhomogeneity in wave feedbacks in this multiscale framework?

Summary and discussions: Are the conclusions and recommendations in this paper applicable to other wave models as well e.g. SWAN and WAM? It’s worth having a thought.
Citation: https://doi.org/10.5194/gmd-2023-24-RC1
- AC5:
  'Reply on RC1', Jiangyu Li, 22 Sep 2023
  This paper illustrates a significant work in high-resolution wave model development. The multiscale gridding framework shows it can mitigate the high computing cost while keeping a good performance in coastal regions. The reasoning is clear supported by evidence. I believe this technique will make a significant contribution to HRM development. I only have a few minor comments/suggestions. See them below.
  RE: Thank you very much for your endorsement and excellent suggestions, which give us more thinking, especially in the applicability of wave modeling with a multiscale grid in other wave models and Earth system models. In this revision, we (1) have added more discussions about the applicability of the wave modeling framework with a multiscale system; (2) have moved several pieces of content to a more suitable place and further optimized them; (3) have revised inappropriate language description, such as: using objective words instead of subjective words; giving clearer descriptions for vague expressions.
  Line 103: ‘acceptable’ is a subjective word. Suggest removing it or changing to an objective word.
  
  RE: Thanks for your good suggestion. We have changed this inaccurate description. Please see lines 114-115.
  Line 207-208: what is 'a reasonable range'? Better to say it explicitly.
  
  RE: Thanks for your kind reminder. We have given the specific range used in this manuscript. Please see line 228.
  Lines 259-261: is it better to move this to discussion section?
  
  RE: Thanks for your good suggestion. We have moved this part to Section 5 and further optimized it. Please see lines 541-545.
  Line 295: again, ‘acceptable’ is a subjective word.
  
  RE: Thanks for your good suggestion. We have changed this improper expression. Please see lines 326-328.
  Paragraph with Line 300: any better to move it to discussion section?
  
  RE: Thanks for your good advice. We have moved this paragraph to Section 5 and further optimized it. Please see lines 546-550.
  Paragraph with Line 325: is it true that WS_multi3 and WS_multi3(new) have the same grid number? Resolution in Area1 is higher in WS_multi3(new), meaning a (slightly?) larger number of grid than in WS_multi3.
  
  RE: Thanks for your kind reminder. We have changed this inaccurate description. Please see lines 380-382.
  Lines 414-417: there are no analysis on era-i and CFSR2 forced wave simulations in this paper. Remove this or put as a discussion point?
  
  RE: Thanks for your kind reminder. We have moved it to Section 5 as a discussion point. Please see lines 561-564.
  Line ~440: WS_multi4 has the finest resolution only in SCS, which is a small proportion in the coastal regions of the whole domain. If the gridding strategy in WS_multi4 is applied to the whole domain, the computing demand could increase significantly. It deserves a clarification for this point, otherwise it may cause a misleading.
  
  RE: Thanks for your excellent suggestion. We have added a more detailed description to make it clearer. Please see lines 520-529.
  Future work around line 445: the 1^stand 2^nd points are not distinctive as some thoughts are mixed up. Suggest rephrasing these two paragraphs.
  
  RE: Thanks for your helpful suggestion. We have rewritten these two paragraphs to make their meanings independent. Please see lines 541-550.
  Future work around line 452: it’s worth mentioning in a coupling context, whether the feedback of waves to the atmosphere and ocean is sensitive to wave multiscale resolution. For example, one may assume that a coarser resolution wave model produces less wave effect on atm and ocean. Does this mean inhomogeneity in wave feedbacks in this multiscale framework?
  
  RE: Thanks for your thoughtful advice. We have refined this discussion about the application of this wave modeling framework to Earth system models in Section 5. Please see lines 572-578.
  Summary and discussions: Are the conclusions and recommendations in this paper applicable to other wave models as well e.g. SWAN and WAM? It’s worth having a thought.
  
  RE: Thanks for your excellent suggestion. We have added a discussion about the application of this multiscale grid to other wave models in Section 5. Please see lines 567-571.
  
  Citation: https://doi.org/10.5194/gmd-2023-24-AC5
CC1:
'Comment on gmd-2023-24', Chu-En Hsu, 26 Jul 2023

Respected colleagues,
Thanks for your time and consideration. While the authors have done an excellent job of presenting a new wave modeling framework, I would like to highlight Hsu et al.'s (2023; https://doi.org/10.3390/jmse11061152) relevant analyses and discussions on wave model performance using a different modeling system (COAWST) during various tropical cyclones. The evolution of wave energy spectra and model performance for wave energy distribution during hurricanes with varying storm characteristics were documented. Scenarios in which model performance may be less than ideal were identified and explored.
Our study focused on three historical hurricanes affecting the U.S. East Coast: Matthew (2016), Dorian (2019), and Isaias (2020). It'll be fascinating to see if these findings can be expanded on and debated further utilizing the new modeling framework.
Best wishes,

Chu-En Hsu

Citation: https://doi.org/10.5194/gmd-2023-24-CC1
- AC6: 'Reply on CC1', Jiangyu Li, 22 Sep 2023
  
  Thank you very much for your interest in our work and your constructive suggestions. We learn a lot from your excellent study about the influences of three hurricanes on ocean surface waves.
  As you suggested, we initially wanted to study the atmosphere-ocean-wave interaction (especially the driving and feedback between typhoons and ocean waves) using a coupled model. Different versions of ultra-high resolution atmosphere-ocean coupled models based on the COAWST model have been developed in Asia-Pacific areas (https:/doi.org/10.1016/j.scib.2020.07.022). However, we couldn’t afford very expensive computational resources when a high-resolution wave model component is incorporated into this coupled system, so we develop an efficient high-precision wave modeling framework to solve this problem.
  In the next work, we will achieve an online atmosphere-ocean-wave information exchange using this efficient wave modeling and continue to study the interaction of typhoons and ocean waves. We are very willing to test the performance of your findings using this framework. If possible, we hope to have a deep communication and cooperation when doing this study.
  Thanks a lot for your comments!
  
  Citation: https://doi.org/10.5194/gmd-2023-24-AC6
RC2:
'Comment on gmd-2023-24', Anonymous Referee #2, 07 Sep 2023
The study of wave numerical simulation is of great value to both practical application and scientific research. However, development of a high-precision wave model with highly-efficient computation is very difficult. This paper studies the effect of model resolution on wave simulation in shallow and deep water areas and then designs a reasonable unstructured multiscale grid system to balance simulation accuracy and computational efficiency. The novelty and application value are high. I recommend publication after minor revision. See below for specific points.
Line 47: about the sentence “It is inconvenient for high-precision operational wave forecasting …”, operational wave forecasting systems pay attention to both simulation accuracy and calculation efficiency.

Line 55: about the sentence “… we design a new … based on a multiscale grid system”, you can simply describe this “system” to give readers a general impression.

Lines 92-93: which way do you use in this paper? Being approximated with obstruction grids? Being parameterized with a source term? Or something else?

Lines 103-105: please give a more detailed explanation about the dramatic increase in computing time.

Lines 109-111: about the sentence “In the future, … the atmosphere-ocean coupled models”, I feel that it is a half sentence, the meaning of the expression is incomplete.

Lines 153-154: about the sentence “As we expected … gradually decrease”, the increase of model resolution not only improves the representativeness of topographic features but also improves the response to local wind.

Lines 211-212: about Figs 6c and 6d, is it only a small area drawn for clarity? In Tab. 2, some of the nine stations are located outside this area.

Line 247: about the sentence “… shallow and deep water areas …”, you can note their definition from section 2.3 to avoid confusion with the classical definition.

Line 255: how to identify and deal with poor meshes? Does SMS have this function?

In section 4.1, about the description of “first-order accuracy” and “second-order accuracy”, the word “precision” is more commonly used than “accuracy”.

Lines 288-290: about the sentence “It should be noted that…especially if the wave model uses the CRD-FCT scheme”, is it because the CRD-FCT scheme has a two-order precision?

Line 291: does the reference also use four schemes or only use the CRD-N scheme?

Lines 300-302: about the sentence “Although the implicit N scheme … domain decomposition”, there is an ambiguity here. Does the new algorithm help save calculation time or improve simulation accuracy?

Lines 315-316: I don’t know the intent of the sentence “Chen et al. (2018) …”.

Lines 321-323: about the sentence “As shown in Fig.7, … remains the same as before”, leaving everything else the same and only dealing with the filled area?

Lines 328-330: about the sentence “While there are still … with the explicit scheme”, the spatial resolution of wind forcing and wave models is inconsistent, which also is an intuitive reason when the wind speed is very high and the wind direction changes quickly.

Line 337: you have better to delete “from wind sea and swell both” because you don’t mention them in the following.

Lines 378-379: is this a mistake? “no observed wave data in ERA5 dataset”?

Lines 384-389: I don’t understand these sentences “Furthermore, … in coastal areas”, it seems to have a gap with the previous meaning.

In section 5: Recommend authors to add some discussions on how to extend this grid to global domain.
Citation: https://doi.org/10.5194/gmd-2023-24-RC2
- AC4:
  'Reply on RC2', Jiangyu Li, 22 Sep 2023
  The study of wave numerical simulation is of great value to both practical application and scientific research. However, development of a high-precision wave model with highly-efficient computation is very difficult. This paper studies the effect of model resolution on wave simulation in shallow and deep water areas and then designs a reasonable unstructured multiscale grid system to balance simulation accuracy and computational efficiency. The novelty and application value are high. I recommend publication after minor revision. See below for specific points.
  RE: Thanks for your constructive comments, which have helped us greatly to evaluate the performance of wave modeling with a multiscale grid system more clearly. In this revision, we have modified the manuscript in strict accordance with your comments and suggestions, mainly including four aspects: (1) adding a clearer description of the design and operation of the control and reference experiments; (2) adding more wind information in some figures; (3) giving more reasonable description in Section Evaluation of wave simulations; (4) giving more detailed expression about some unclear or inaccurate sentences and method choices.
  Line 47: about the sentence “It is inconvenient for high-precision operational wave forecasting …”, operational wave forecasting systems pay attention to both simulation accuracy and calculation efficiency.
  
  RE: Thanks for your good comment. We have rewritten this sentence to get a better expression. Please see lines 47-50.
  Line 55: about the sentence “… we design a new … based on a multiscale grid system”, you can simply describe this “system” to give readers a general impression.
  
  RE: Thanks for your good advice. We have added the corresponding description to give the reader a general impression. Please see line 62.
  Lines 92-93: which way do you use in this paper? Being approximated with obstruction grids? Being parameterized with a source term? Or something else?
  
  RE: Thanks for your kind reminder. We have mentioned the way we used in this manuscript. Please see line 105.
  Lines 103-105: please give a more detailed explanation about the dramatic increase in computing time.
  
  RE: Thanks for your excellent suggestion. We have added a more detailed explanation of this situation. Please see lines 116-121.
  Lines 109-111: about the sentence “In the future, … the atmosphere-ocean coupled models”, I feel that it is a half sentence, the meaning of the expression is incomplete.
  
  RE: Thanks for your good advice. We have added the corresponding content to make the meaning clearer. Please see line 128.
  Lines 153-154: about the sentence “As we expected … gradually decrease”, the increase of model resolution not only improves the representativeness of topographic features but also improves the response to local wind.
  
  RE: Thanks for your great comment. We have added the corresponding content to get a more proper description. Please see lines 173-174.
  Lines 211-212: about Figs 6c and 6d, is it only a small area drawn for clarity? In Tab. 2, some of the nine stations are located outside this area.
  
  RE: Thanks for your kind reminder. We have mentioned this point to give readers a clearer understanding. Please see lines 232-233.
  Line 247: about the sentence “… shallow and deep water areas …”, you can note their definition from section 2.3 to avoid confusion with the classical definition.
  
  RE: Thanks for your kind advice. We have added the definition used in this manuscript to avoid confusion with the classical definition. Please see lines 268-269.
  Line 255: how to identify and deal with poor meshes? Does SMS have this function?
  
  RE: Thanks for your helpful comment. We have given a more detailed description of how to identify and deal with poor meshes using the SMS tool, and the identification criteria used in this manuscript. Please see lines 276-280.
  In section 4.1, about the description of “first-order accuracy” and “second-order accuracy”, the word “precision” is more commonly used than “accuracy”.
  
  RE: Thanks for your kind reminder. We have replaced this inaccurate word. Please see lines 299 and 309.
  Lines 288-290: about the sentence “It should be noted that…especially if the wave model uses the CRD-FCT scheme”, is it because the CRD-FCT scheme has a two-order precision?
  
  RE: Thanks for your thoughtful comment. We have added the corresponding description. Please see lines 320-321.
  Line 291: does the reference also use four schemes or only use the CRD-N scheme?
  
  RE: Thanks for your kind reminder. We have added a clearer description of the reference design to make the comparison between the reference and control experiment easier to follow. Please see lines 323-325.
  Lines 300-302: about the sentence “Although the implicit N scheme … domain decomposition”, there is an ambiguity here. Does the new algorithm help save calculation time or improve simulation accuracy?
  
  RE: Thanks for your thoughtful comment. We have rewritten the corresponding description to make the meaning clearer and moved it to Section 5. Please see lines 546-550.
  Lines 315-316: I don’t know the intent of the sentence “Chen et al. (2018) …”.
  
  RE: Thanks for your helpful comment. We have rewritten the study about the influence of a smoothing wind on ocean surface waves from Chen et al. (2018). Please see lines 355-357.
  Lines 321-323: about the sentence “As shown in Fig.7, … remains the same as before”, leaving everything else the same and only dealing with the filled area?
  
  RE: Thanks for your kind reminder. We have added a detailed description of generating a new multiscale grid. Please see line 362.
  Lines 328-330: about the sentence “While there are still … with the explicit scheme”, the spatial resolution of wind forcing and wave models is inconsistent, which also is an intuitive reason when the wind speed is very high and the wind direction changes quickly.
  
  RE: Thanks for your excellent suggestion. We have rewritten the reasons for visible simulation differences in boreal winter in the North Pacific Ocean. Please see lines 369-378.
  Line 337: you have better to delete “from wind sea and swell both” because you don’t mention them in the following.
  
  RE: Thanks for your good advice. We have deleted these words. Please see line 385.
  Lines 378-379: is this a mistake? “no observed wave data in ERA5 dataset”?
  
  RE: Thanks for your kind reminder. We have changed this inaccurate description. Please see lines 438-440.
  Lines 384-389: I don’t understand these sentences “Furthermore, … in coastal areas”, it seems to have a gap with the previous meaning.
  
  RE: Thanks for your helpful comment. We have rewritten this part and moved some of it to a more suitable place. Please see lines 389-391 and 446-452.
  In section 5: Recommend authors to add some discussions on how to extend this grid to global domain.
  
  RE: Thanks for your thoughtful advice. We have added some discussions about how to optimize multiscale grids (including the extension of a regional grid to a global domain) in Section 5. Please see lines 541-545.
  
  Citation: https://doi.org/10.5194/gmd-2023-24-AC4

Status: closed

CEC1:
'Comment on gmd-2023-24', Juan Antonio Añel, 06 Apr 2023

Dear authors,
Unfortunately, after checking your manuscript, it has come to our attention that it does not comply with our "Code and Data Policy".

https://www.geoscientific-model-development.net/policies/code_and_data_policy.html
Neither of the code or data repositories that you mention in the "Code and data availability" section of your manuscript (GitHub, NOAA or Copernicus servers, and Aquaveo) are suitable repositories for scientific publishing. You must store all the code and data relevant to your manuscript in one acceptable according to our policy. Currently, only the Zenodo repository for a few files is in compliance. Moreover, the link to NOAA and Copernicus servers could not be enough specific to locate the exact files necessary to replicate your work.
I highlight that the situation with your submission is currently irregular, as your manuscript should have never been accepted for Discussions with such a problem. Therefore, we will have to reject your manuscript unless you address these issues in a prompt manner.
In this way, you must reply to this comment as soon as possible with the DOIs and links for the new repositories containing the code and data.
Also, I would like to note that in the only repository that you have shared correctly, many files are in .mat format. .mat is a proprietary binary data container format. In this way, these files can only be accessed with a given version of proprietary software, which precludes the ability to access the data. I would suggest you store these files in an open format, accessible without the need for specific software (ASCII, NetCDF, etc.)
I emphasize that failing to comply with this request to store the code and data, we will reject your manuscript.
Juan A. Añel
Geosci. Model Dev. Executive Editor

Citation: https://doi.org/10.5194/gmd-2023-24-CEC1
- AC1:
  'Reply on CEC1', Jiangyu Li, 10 Apr 2023
  
  Thank you very much for your comments, we will strictly follow the code and data policy of GMD. According to your comments and the policy, we have made the following changes.
  (1) We have stored all the codes and data related to this paper in the Zenodo: https:/doi.org/10.5281/zenodo.7811681 (last access: April 10, 2023). They include a) the wind forcing data, the configure, boundary, topography, and control files required for the wave model run, and b) the observation data, the model output data, and its post-processing codes. Based on these codes and data, readers can repeat our work completely.
  (2) We have converted all the files with .mat format to NetCDF or ASCII format.
  Thanks again.
  
  Citation: https://doi.org/10.5194/gmd-2023-24-AC1
  - CEC2: 'Reply on AC1', Juan Antonio Añel, 13 Apr 2023
    
    Dear authors,
    Unfortunately, I have to bring to your attention (and to the Topical Editor) that despite depositing some data files in Zenodo, your submission continues not to comply with our policy. Therefore, unless you fix it (please, in a prompt manner), we will have to proceed to reject your submission for publication.
    We continue missing the repository with the code of WaveWatch III and the Surface-water Modeling System software (SMS). Please, upload them to one of the acceptable repositories and reply to this comment with the information to access them (link and DOI).
    Juan A. Añel
    Geosci. Model Dev. Executive Editor
    
    Citation: https://doi.org/10.5194/gmd-2023-24-CEC2
    
    AC2: 'Reply on CEC2', Jiangyu Li, 14 Apr 2023
    
    Thank you very much for your comments.
    We have uploaded everything related to this paper to Zenodo: https:/doi.org/10.5281/zenodo.7827541 (last access: April 14, 2023). They include:
    a) the source code of wave model WW3 (wwatch3.v5.16.tar.gz),
    b) the Surface-water Modeling System software (SMS) software (software_SMS.zip),
    c) the wind forcing, boundary, topography, configuration, and control files required for the wave model run (wind_file.rar, boundary_file_*.rar, topography_file.rar, configuration_file.rar, and control_file.rar), and
    d) the observation data, the model output data, and its post-processing codes (output_data_and_code.rar).
    We hope these can comply with the code and data policy of GMD. Thanks again.
    
    Citation: https://doi.org/10.5194/gmd-2023-24-AC2
    
    CEC3: 'Reply on AC2', Juan Antonio Añel, 15 Apr 2023
    
    Dear authors,
    Many thanks for your reply and for adding part of the requested material to the new repository. It is someway disappointing that for the SMS software, you can only share a binary .exe file. In such a way, it is not possible to verify if your results, in the end, are correct, as it is necessary to trust that this proprietary software works correctly. It is not possible to check it.
    However, you have probably done your best to try to comply with our requests, and the main problem is that when you designed your research, you chose the SMS software to carry it on. This hardly has a solution now. However, if you, the topical editor or the reviewers know of a software or library that you can use to replace the use of SMS in your work, I encourage you to replace SMS to comply with the principle of scientific reproducibility and submit a version of the manuscript with results using FLOSS.
    Beyond this issue, I consider that you now comply with our policy regarding code, and we can continue with the review process for your manuscript.
    Juan A. Añel
    Geosci. Model Dev. Executive Editor
    
    Citation: https://doi.org/10.5194/gmd-2023-24-CEC3
    
    AC3: 'Reply on CEC3', Jiangyu Li, 19 Apr 2023
    
    Thanks for your comments and suggestions. We couldn’t agree more with your opinions. If the topic editor or the reviewers know a more powerful software replacing the SMS tool, we will follow their request and suggestions reproducing the process of generating unstructured grid files.
    
    Thanks a lot!
    
    Citation: https://doi.org/10.5194/gmd-2023-24-AC3
RC1:
'Comment on gmd-2023-24', Anonymous Referee #1, 06 Jun 2023
This paper illustrates a significant work in high-resolution wave model development. The multiscale griding framework shows it can mitigate the hight computing cost while keeping a good performance in coastal regions. The reasoning is clear supported by evidence. I believe this technique will make a significant contribution to HRM development. I only have a few minor comments/suggestions. See them below.
Line 103: ‘acceptable’ is a subjective word. Suggest removing it or changing to an objective word.

Line 207-208: what is 'a reasonable range'? Better to say it explicitly.

Lines 259-261: is it better to move this to discussion section?

Line 295: again, ‘acceptable’ is a subjective word.

Paragraph with Line 300: any better to move it to discussion section?

Paragraph withLine 325: is it true that WS_multi3 and WS_multi3(new) have the same grid number? Resolution in Area1 is higher in WS_multi3(new), meaning a (slightly?) larger number of grid than in WS_multi3.

Lines 514-517: there are no analysis on era-i and CFSR2 forced wave simulations in this paper. Remove this or put as a discussion point?

Line ~440: WS_multi4 has the finest resolution only in SCS, which is a small proportion in the coastal regions of the whole domain. If the gridding strategy in WS_multi4 is applied to the whole domain, the computing demand could increase significantly. It deserves a clarification for this point, otherwise it may cause a misleading.

Future work around line 445: the 1^st and 2^nd points are not distinctive as some thoughts are mixed up. Suggest rephrasing these two paragraphs.

Future work around line 452: it’s worth mentioning in a coupling context, whether the feedback of waves to the atmosphere and ocean is sensitive to wave multiscale resolution. For example, one may assume that a coarser resolution wave model produces less wave effect on atm and ocean. Does this mean inhomogeneity in wave feedbacks in this multiscale framework?

Summary and discussions: Are the conclusions and recommendations in this paper applicable to other wave models as well e.g. SWAN and WAM? It’s worth having a thought.
Citation: https://doi.org/10.5194/gmd-2023-24-RC1
- AC5:
  'Reply on RC1', Jiangyu Li, 22 Sep 2023
  This paper illustrates a significant work in high-resolution wave model development. The multiscale gridding framework shows it can mitigate the high computing cost while keeping a good performance in coastal regions. The reasoning is clear supported by evidence. I believe this technique will make a significant contribution to HRM development. I only have a few minor comments/suggestions. See them below.
  RE: Thank you very much for your endorsement and excellent suggestions, which give us more thinking, especially in the applicability of wave modeling with a multiscale grid in other wave models and Earth system models. In this revision, we (1) have added more discussions about the applicability of the wave modeling framework with a multiscale system; (2) have moved several pieces of content to a more suitable place and further optimized them; (3) have revised inappropriate language description, such as: using objective words instead of subjective words; giving clearer descriptions for vague expressions.
  Line 103: ‘acceptable’ is a subjective word. Suggest removing it or changing to an objective word.
  
  RE: Thanks for your good suggestion. We have changed this inaccurate description. Please see lines 114-115.
  Line 207-208: what is 'a reasonable range'? Better to say it explicitly.
  
  RE: Thanks for your kind reminder. We have given the specific range used in this manuscript. Please see line 228.
  Lines 259-261: is it better to move this to discussion section?
  
  RE: Thanks for your good suggestion. We have moved this part to Section 5 and further optimized it. Please see lines 541-545.
  Line 295: again, ‘acceptable’ is a subjective word.
  
  RE: Thanks for your good suggestion. We have changed this improper expression. Please see lines 326-328.
  Paragraph with Line 300: any better to move it to discussion section?
  
  RE: Thanks for your good advice. We have moved this paragraph to Section 5 and further optimized it. Please see lines 546-550.
  Paragraph with Line 325: is it true that WS_multi3 and WS_multi3(new) have the same grid number? Resolution in Area1 is higher in WS_multi3(new), meaning a (slightly?) larger number of grid than in WS_multi3.
  
  RE: Thanks for your kind reminder. We have changed this inaccurate description. Please see lines 380-382.
  Lines 414-417: there are no analysis on era-i and CFSR2 forced wave simulations in this paper. Remove this or put as a discussion point?
  
  RE: Thanks for your kind reminder. We have moved it to Section 5 as a discussion point. Please see lines 561-564.
  Line ~440: WS_multi4 has the finest resolution only in SCS, which is a small proportion in the coastal regions of the whole domain. If the gridding strategy in WS_multi4 is applied to the whole domain, the computing demand could increase significantly. It deserves a clarification for this point, otherwise it may cause a misleading.
  
  RE: Thanks for your excellent suggestion. We have added a more detailed description to make it clearer. Please see lines 520-529.
  Future work around line 445: the 1^stand 2^nd points are not distinctive as some thoughts are mixed up. Suggest rephrasing these two paragraphs.
  
  RE: Thanks for your helpful suggestion. We have rewritten these two paragraphs to make their meanings independent. Please see lines 541-550.
  Future work around line 452: it’s worth mentioning in a coupling context, whether the feedback of waves to the atmosphere and ocean is sensitive to wave multiscale resolution. For example, one may assume that a coarser resolution wave model produces less wave effect on atm and ocean. Does this mean inhomogeneity in wave feedbacks in this multiscale framework?
  
  RE: Thanks for your thoughtful advice. We have refined this discussion about the application of this wave modeling framework to Earth system models in Section 5. Please see lines 572-578.
  Summary and discussions: Are the conclusions and recommendations in this paper applicable to other wave models as well e.g. SWAN and WAM? It’s worth having a thought.
  
  RE: Thanks for your excellent suggestion. We have added a discussion about the application of this multiscale grid to other wave models in Section 5. Please see lines 567-571.
  
  Citation: https://doi.org/10.5194/gmd-2023-24-AC5
CC1:
'Comment on gmd-2023-24', Chu-En Hsu, 26 Jul 2023

Respected colleagues,
Thanks for your time and consideration. While the authors have done an excellent job of presenting a new wave modeling framework, I would like to highlight Hsu et al.'s (2023; https://doi.org/10.3390/jmse11061152) relevant analyses and discussions on wave model performance using a different modeling system (COAWST) during various tropical cyclones. The evolution of wave energy spectra and model performance for wave energy distribution during hurricanes with varying storm characteristics were documented. Scenarios in which model performance may be less than ideal were identified and explored.
Our study focused on three historical hurricanes affecting the U.S. East Coast: Matthew (2016), Dorian (2019), and Isaias (2020). It'll be fascinating to see if these findings can be expanded on and debated further utilizing the new modeling framework.
Best wishes,

Chu-En Hsu

Citation: https://doi.org/10.5194/gmd-2023-24-CC1
- AC6: 'Reply on CC1', Jiangyu Li, 22 Sep 2023
  
  Thank you very much for your interest in our work and your constructive suggestions. We learn a lot from your excellent study about the influences of three hurricanes on ocean surface waves.
  As you suggested, we initially wanted to study the atmosphere-ocean-wave interaction (especially the driving and feedback between typhoons and ocean waves) using a coupled model. Different versions of ultra-high resolution atmosphere-ocean coupled models based on the COAWST model have been developed in Asia-Pacific areas (https:/doi.org/10.1016/j.scib.2020.07.022). However, we couldn’t afford very expensive computational resources when a high-resolution wave model component is incorporated into this coupled system, so we develop an efficient high-precision wave modeling framework to solve this problem.
  In the next work, we will achieve an online atmosphere-ocean-wave information exchange using this efficient wave modeling and continue to study the interaction of typhoons and ocean waves. We are very willing to test the performance of your findings using this framework. If possible, we hope to have a deep communication and cooperation when doing this study.
  Thanks a lot for your comments!
  
  Citation: https://doi.org/10.5194/gmd-2023-24-AC6
RC2:
'Comment on gmd-2023-24', Anonymous Referee #2, 07 Sep 2023
The study of wave numerical simulation is of great value to both practical application and scientific research. However, development of a high-precision wave model with highly-efficient computation is very difficult. This paper studies the effect of model resolution on wave simulation in shallow and deep water areas and then designs a reasonable unstructured multiscale grid system to balance simulation accuracy and computational efficiency. The novelty and application value are high. I recommend publication after minor revision. See below for specific points.
Line 47: about the sentence “It is inconvenient for high-precision operational wave forecasting …”, operational wave forecasting systems pay attention to both simulation accuracy and calculation efficiency.

Line 55: about the sentence “… we design a new … based on a multiscale grid system”, you can simply describe this “system” to give readers a general impression.

Lines 92-93: which way do you use in this paper? Being approximated with obstruction grids? Being parameterized with a source term? Or something else?

Lines 103-105: please give a more detailed explanation about the dramatic increase in computing time.

Lines 109-111: about the sentence “In the future, … the atmosphere-ocean coupled models”, I feel that it is a half sentence, the meaning of the expression is incomplete.

Lines 153-154: about the sentence “As we expected … gradually decrease”, the increase of model resolution not only improves the representativeness of topographic features but also improves the response to local wind.

Lines 211-212: about Figs 6c and 6d, is it only a small area drawn for clarity? In Tab. 2, some of the nine stations are located outside this area.

Line 247: about the sentence “… shallow and deep water areas …”, you can note their definition from section 2.3 to avoid confusion with the classical definition.

Line 255: how to identify and deal with poor meshes? Does SMS have this function?

In section 4.1, about the description of “first-order accuracy” and “second-order accuracy”, the word “precision” is more commonly used than “accuracy”.

Lines 288-290: about the sentence “It should be noted that…especially if the wave model uses the CRD-FCT scheme”, is it because the CRD-FCT scheme has a two-order precision?

Line 291: does the reference also use four schemes or only use the CRD-N scheme?

Lines 300-302: about the sentence “Although the implicit N scheme … domain decomposition”, there is an ambiguity here. Does the new algorithm help save calculation time or improve simulation accuracy?

Lines 315-316: I don’t know the intent of the sentence “Chen et al. (2018) …”.

Lines 321-323: about the sentence “As shown in Fig.7, … remains the same as before”, leaving everything else the same and only dealing with the filled area?

Lines 328-330: about the sentence “While there are still … with the explicit scheme”, the spatial resolution of wind forcing and wave models is inconsistent, which also is an intuitive reason when the wind speed is very high and the wind direction changes quickly.

Line 337: you have better to delete “from wind sea and swell both” because you don’t mention them in the following.

Lines 378-379: is this a mistake? “no observed wave data in ERA5 dataset”?

Lines 384-389: I don’t understand these sentences “Furthermore, … in coastal areas”, it seems to have a gap with the previous meaning.

In section 5: Recommend authors to add some discussions on how to extend this grid to global domain.
Citation: https://doi.org/10.5194/gmd-2023-24-RC2
- AC4:
  'Reply on RC2', Jiangyu Li, 22 Sep 2023
  The study of wave numerical simulation is of great value to both practical application and scientific research. However, development of a high-precision wave model with highly-efficient computation is very difficult. This paper studies the effect of model resolution on wave simulation in shallow and deep water areas and then designs a reasonable unstructured multiscale grid system to balance simulation accuracy and computational efficiency. The novelty and application value are high. I recommend publication after minor revision. See below for specific points.
  RE: Thanks for your constructive comments, which have helped us greatly to evaluate the performance of wave modeling with a multiscale grid system more clearly. In this revision, we have modified the manuscript in strict accordance with your comments and suggestions, mainly including four aspects: (1) adding a clearer description of the design and operation of the control and reference experiments; (2) adding more wind information in some figures; (3) giving more reasonable description in Section Evaluation of wave simulations; (4) giving more detailed expression about some unclear or inaccurate sentences and method choices.
  Line 47: about the sentence “It is inconvenient for high-precision operational wave forecasting …”, operational wave forecasting systems pay attention to both simulation accuracy and calculation efficiency.
  
  RE: Thanks for your good comment. We have rewritten this sentence to get a better expression. Please see lines 47-50.
  Line 55: about the sentence “… we design a new … based on a multiscale grid system”, you can simply describe this “system” to give readers a general impression.
  
  RE: Thanks for your good advice. We have added the corresponding description to give the reader a general impression. Please see line 62.
  Lines 92-93: which way do you use in this paper? Being approximated with obstruction grids? Being parameterized with a source term? Or something else?
  
  RE: Thanks for your kind reminder. We have mentioned the way we used in this manuscript. Please see line 105.
  Lines 103-105: please give a more detailed explanation about the dramatic increase in computing time.
  
  RE: Thanks for your excellent suggestion. We have added a more detailed explanation of this situation. Please see lines 116-121.
  Lines 109-111: about the sentence “In the future, … the atmosphere-ocean coupled models”, I feel that it is a half sentence, the meaning of the expression is incomplete.
  
  RE: Thanks for your good advice. We have added the corresponding content to make the meaning clearer. Please see line 128.
  Lines 153-154: about the sentence “As we expected … gradually decrease”, the increase of model resolution not only improves the representativeness of topographic features but also improves the response to local wind.
  
  RE: Thanks for your great comment. We have added the corresponding content to get a more proper description. Please see lines 173-174.
  Lines 211-212: about Figs 6c and 6d, is it only a small area drawn for clarity? In Tab. 2, some of the nine stations are located outside this area.
  
  RE: Thanks for your kind reminder. We have mentioned this point to give readers a clearer understanding. Please see lines 232-233.
  Line 247: about the sentence “… shallow and deep water areas …”, you can note their definition from section 2.3 to avoid confusion with the classical definition.
  
  RE: Thanks for your kind advice. We have added the definition used in this manuscript to avoid confusion with the classical definition. Please see lines 268-269.
  Line 255: how to identify and deal with poor meshes? Does SMS have this function?
  
  RE: Thanks for your helpful comment. We have given a more detailed description of how to identify and deal with poor meshes using the SMS tool, and the identification criteria used in this manuscript. Please see lines 276-280.
  In section 4.1, about the description of “first-order accuracy” and “second-order accuracy”, the word “precision” is more commonly used than “accuracy”.
  
  RE: Thanks for your kind reminder. We have replaced this inaccurate word. Please see lines 299 and 309.
  Lines 288-290: about the sentence “It should be noted that…especially if the wave model uses the CRD-FCT scheme”, is it because the CRD-FCT scheme has a two-order precision?
  
  RE: Thanks for your thoughtful comment. We have added the corresponding description. Please see lines 320-321.
  Line 291: does the reference also use four schemes or only use the CRD-N scheme?
  
  RE: Thanks for your kind reminder. We have added a clearer description of the reference design to make the comparison between the reference and control experiment easier to follow. Please see lines 323-325.
  Lines 300-302: about the sentence “Although the implicit N scheme … domain decomposition”, there is an ambiguity here. Does the new algorithm help save calculation time or improve simulation accuracy?
  
  RE: Thanks for your thoughtful comment. We have rewritten the corresponding description to make the meaning clearer and moved it to Section 5. Please see lines 546-550.
  Lines 315-316: I don’t know the intent of the sentence “Chen et al. (2018) …”.
  
  RE: Thanks for your helpful comment. We have rewritten the study about the influence of a smoothing wind on ocean surface waves from Chen et al. (2018). Please see lines 355-357.
  Lines 321-323: about the sentence “As shown in Fig.7, … remains the same as before”, leaving everything else the same and only dealing with the filled area?
  
  RE: Thanks for your kind reminder. We have added a detailed description of generating a new multiscale grid. Please see line 362.
  Lines 328-330: about the sentence “While there are still … with the explicit scheme”, the spatial resolution of wind forcing and wave models is inconsistent, which also is an intuitive reason when the wind speed is very high and the wind direction changes quickly.
  
  RE: Thanks for your excellent suggestion. We have rewritten the reasons for visible simulation differences in boreal winter in the North Pacific Ocean. Please see lines 369-378.
  Line 337: you have better to delete “from wind sea and swell both” because you don’t mention them in the following.
  
  RE: Thanks for your good advice. We have deleted these words. Please see line 385.
  Lines 378-379: is this a mistake? “no observed wave data in ERA5 dataset”?
  
  RE: Thanks for your kind reminder. We have changed this inaccurate description. Please see lines 438-440.
  Lines 384-389: I don’t understand these sentences “Furthermore, … in coastal areas”, it seems to have a gap with the previous meaning.
  
  RE: Thanks for your helpful comment. We have rewritten this part and moved some of it to a more suitable place. Please see lines 389-391 and 446-452.
  In section 5: Recommend authors to add some discussions on how to extend this grid to global domain.
  
  RE: Thanks for your thoughtful advice. We have added some discussions about how to optimize multiscale grids (including the extension of a regional grid to a global domain) in Section 5. Please see lines 541-545.
  
  Citation: https://doi.org/10.5194/gmd-2023-24-AC4

Jiangyu Li et al.

Viewed

Total article views: 658 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
478	151	29	658	4	3

HTML: 478
PDF: 151
XML: 29
Total: 658
BibTeX: 4
EndNote: 3

Views and downloads (calculated since 13 Mar 2023)

Month	HTML	PDF	XML	Total
Mar 2023	145	27	4	176
Apr 2023	139	27	12	178
May 2023	26	7	1	34
Jun 2023	39	23	2	64
Jul 2023	39	22	2	63
Aug 2023	24	21	0	45
Sep 2023	66	24	8	98

Cumulative views and downloads (calculated since 13 Mar 2023)

Month	HTML	PDF	XML	Total
Mar 2023	145	27	4	176
Apr 2023	139	27	12	178
May 2023	26	7	1	34
Jun 2023	39	23	2	64
Jul 2023	39	22	2	63
Aug 2023	24	21	0	45
Sep 2023	66	24	8	98

Viewed (geographical distribution)

Total article views: 627 (including HTML, PDF, and XML) Thereof 627 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 28 Sep 2023

Short summary

Ocean surface waves play an important role at the air-sea interface but are rarely activated in high-resolution Earth system simulations due to their expensive computational costs. To alleviate this situation, this paper designs a new wave modeling framework with a multiscale grid system. Evaluations of a series of numerical experiments show that it has good feasibility and applicability in WW3, and can achieve the goals of efficient and high-precision wave simulation.


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