Global ocean and sea ice variability simulated in eddy-permitting climate models

Morioka, Yushi; Maisonnave, Eric; Masson, Sébastien; Rousset, Clement; Kornblueh, Luis; Giorgetta, Marco; Nonaka, Masami; Behera, Swadhin K.

doi:https://doi.org/10.5194/egusphere-2025-2258

Preprints

https://doi.org/10.5194/egusphere-2025-2258

Preprints

28 May 2025

| 28 May 2025

Global ocean and sea ice variability simulated in eddy-permitting climate models

Yushi Morioka, Eric Maisonnave, Sébastien Masson, Clement Rousset, Luis Kornblueh, Marco Giorgetta, Masami Nonaka, and Swadhin K. Behera

Abstract. Ocean mesoscale eddies, which have a horizontal scale with an order of 100 km, play a prominent role in global ocean heat transport that regulates the Earth's climate. Most of climate models, however, cannot fully resolve the ocean mesoscale eddies because of the constraint of computational resources. To mitigate this shortcoming, we newly develop an eddy-permitting climate model, SINTEX-F3, which has the ocean resolution with the order of 25 km. Compared to other eddy-permitting climate models available from the CMIP6 HighResMIP, the SINTEX-F3 represents a cold bias in the mid-high latitudes and weaker El Niño-Southern Oscillation (ENSO). Despite the weaker ENSO, the SINTEX-F3 realistically reproduces other tropical climate phenomena such as the Indian Ocean Dipole and Atlantic Niño/Niña, indicating that these modes are less dependent on ENSO in the model. In the subtropical-midlatitudes, the SINTEX-F3 well captures mesoscale sea surface temperature and surface heat flux variability, particularly in the western and eastern boundary current regions. Furthermore, the SINTEX-F3 simulates the mean state and variability of sea ice in the Antarctic Sea more accurately than in the Arctic Sea, likely due to improvements in sea ice model physics and the increased ocean model resolution. While further efforts are needed to address the cold bias and the weaker ENSO representation, the SINTEX-F3 shows significant potential for simulating and predicting global ocean and sea ice variability at an eddy-permitting scale.

Received: 14 May 2025 – Discussion started: 28 May 2025

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 6121 KB)

Supplement (24508 KB)

Download & links

Preprint (6121 KB)
Metadata XML
Supplement (24508 KB)
BibTeX
EndNote

Yushi Morioka, Eric Maisonnave, Sébastien Masson, Clement Rousset, Luis Kornblueh, Marco Giorgetta, Masami Nonaka, and Swadhin K. Behera

Status: final response (author comments only)

CEC1:
'Comment on egusphere-2025-2258 - No compliance with the policy of the journal', Juan Antonio Añel, 21 Jun 2025

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
in your "Code Availability" statement you fail to identify a repository for the code used to produce your work, which includes the SINTEX-F2 and SINTEX-F3 models. I am sorry to have to be so outspoken, but this is completely unacceptable, forbidden by our policy, and your manuscript should have never been accepted for Discussions because of it. Our policy clearly states that all the code and data necessary to produce a manuscript must be published openly and freely to anyone before submission of a manuscript.
Also, for the data, the links to sites that you provide, which a main portals in some cases, not specific repositories for the data that you use for your work, are not acceptable.
Therefore, we are granting you a short time to solve this situation. You have to reply to this comment in a prompt manner with the information for the repositories containing all the models, code and data that you use to produce and replicate your manuscript. The reply must include the link and permanent identifier (e.g. DOI). Also, any future version of your manuscript must include the modified section with the new information.
In the meantime, I advise the Topical Editor to stop the peer-review process for your manuscript, as we can not waste the time of reviewers to review something that is not acceptable for publication in our journal in its current form.
Please, note that if you do not fix these problems as requested, we will have to reject your manuscript for publication in our journal.
Juan A. Añel
Geosci. Model Dev. Executive Editor

Citation: https://doi.org/10.5194/egusphere-2025-2258-CEC1
- AC1:
  'Reply on CEC1', Yushi Morioka, 22 Jun 2025
  
  Dear Chief Editor,
  Thank you so much for kindly raising the technical issues on the limited availability of the codes and data used in our study.
  Following the editor's comments and the journal guidelines, we have provided the model source codes, configuration files, documentation files, and run scripts for both SINTEX-F2 and SINTEX-F3 models at the first author's account of the GitHub website, considering a huge number of files and file sizes. The model codes are now publicly available and accessible with the GitHub account (sign-up for free).
  Also, we have provided the specific URL for the observation data and reanalysis product which are used as reference for the calculation of the model biases. However, we find it difficult to specify the URL for each of CMIP6 models because of the integrated nature of global model repository thereby staying as it was.
  We will send you the revised file including the updated Code and Data availability sections, separately in a reply to the journal's editorial email.
  Best regards,
  Yushi Morioka
  
  Citation: https://doi.org/10.5194/egusphere-2025-2258-AC1
  - CEC2: 'Reply on AC1', Juan Antonio Añel, 22 Jun 2025
    
    Dear authors,
    Again, I do not think you have read our policy and paid the due attention and given full consideration to my previous comment. In your reply you state that you have shared the information, but the reply does not contain it. You must publish here, in reply to this comment, all the information which I have requested previously and is required according to the policy of the journal. Also, you must modify the "Code and data availability" section in any future version of your manuscript including it.
    Moreover, you mention that you are sharing the information via a GitHub site. If you had paid attention to our policy, you would see that it clearly states that GitHub sites are not acceptable repositories.
    Therefore, again, please, read the policy of the journal, and share here the information requested on valid repositories according to our policy, containing all the code and data necessary to replicate your work.
    Juan A. Añel
    Geosci. Model Dev. Executive Editor
    
    Citation: https://doi.org/10.5194/egusphere-2025-2258-CEC2
    
    AC2: 'Reply on CEC2', Yushi Morioka, 23 Jun 2025
    
    Dear Chief Editor,
    Thank you again for raising the remaining issues on the code and data availability in our paper.
    Following the editor's comments and the journal policy here (https://www.geoscientific-model-development.net/policies/code_and_data_policy.html), we have made the repositories publicly available and archived with the DOI (see below) for the SINTEX-F2 and SINTEX-F3 models used in this paper. Also, we have clarified the specific URLs (see below) where we downloaded the observation data, the reanalysis product, and the CMIP6 model output.
    We will replace the code and data availability statements with the following ones when we submit the revised manuscript.
    If we need further modification on the statements, please let us know specifically where we should make changes in the statements.
    Code Availability
    The exact version of the SINTEX-F2 model used to produce the results in this paper is archived on the Zenodo’s GitHub repository under the DOI of https://doi.org/10.5281/zenodo.15717852 (Morioka 2025), whereas that of the SINTEX-F3 model is archived on the repository under the DOI of https://doi.org/10.5281/zenodo.15717900 (Morioka 2025).
    Data Availability
    The observed SST data is available from the NOAA’s OISSTv2 website (https://downloads.psl.noaa.gov/Datasets/noaa.oisst.v2/new/, Reynolds et al. 2007). The SIC data is also downloaded from the NSIDC website (https://cmr.earthdata.nasa.gov/virtual-directory/collections/C3177837840-NSIDC_CPRD, DiGirolamo et al., 2022). The subsurface ocean temperature and salinity data can also be obtained from the UK Met Office’s EN4 website (https://www.metoffice.gov.uk/hadobs/en4/download-en4-2-2.html, Good et al. 2013). The atmospheric variables can be downloaded from the ECMWF and Copernicus Climate Data Store (https://cds.climate.copernicus.eu/datasets/reanalysis-era5-single-levels-monthly-means?tab=download, Hersbach et al., 2020). The HighResMIP model output can be downloaded from any of the CMIP6 ESGF nodes (https://esgf-ui.ceda.ac.uk/cog/search/cmip6-ceda/, Haarsma et al., 2016).
    
    Best regards,
    Yushi Morioka
    
    Citation: https://doi.org/10.5194/egusphere-2025-2258-AC2
    
    CEC3: 'Reply on AC2', Juan Antonio Añel, 23 Jun 2025
    
    Dear authors,
    Many thanks for addressing this issue. Only one additional note, there is not such thing as "Zenodo’s GitHub repository", but it is a "Zenodo repository".
    Juan A. Añel
    Geosci. Model Dev. Executive Editor
    
    Citation: https://doi.org/10.5194/egusphere-2025-2258-CEC3
    
    AC3: 'Reply on CEC3', Yushi Morioka, 23 Jun 2025
    
    Dear Chief Editor,
    Thank you so much for the kind confirmation of our statements on the code and data availability.
    We will modify the words referring to the repository accordingly, when we submit the revised manuscript.
    Best regards,
    Yushi Morioka
    
    Citation: https://doi.org/10.5194/egusphere-2025-2258-AC3
RC1:
'Comment on egusphere-2025-2258', Anonymous Referee #1, 02 Jul 2025
Review of “Global ocean and sea ice variability simulated in eddy-permitting climate models”, by Yushi Morioka, Eric Maisonnave, Sébastien Masson, Clement Rousset, Luis Kornblueh, Marco Giorgetta, Masami Nonaka, Swadhin K. Behera, submitted to Geoscientific Model Development.
Oceanic mesoscale eddies account for 70% of kinetic energy in the global ocean and play a significant role in the transport of mass, heat, and nutrients. The accurate representation of oceanic mesoscale eddies in numerical models is thus essential. By developing an oceanic eddy-permitting (~25 km) coupled climate model SINTEX-F3 and comparing with the coarse-resolution counterparts, this study assesses the model performance at various resolution in simulating global ocean and sea ice variability as well as the climate variability over different regions including the tropical ocean, western/eastern boundary current regions, and polar regions. I recommend major revision prior to the publication of this manuscript and my concerns are listed below.
Major comments
In this study, the authors attributed most improvement in the western/eastern boundary current regions and sea ice variability to increased oceanic resolution. However, the findings are primarily descriptive and lack sufficient mechanism analysis or plausible mechanism conjecture on how the increased oceanic resolution, and thus, the better presented mesoscale eddies contribute to these improvements through various oceanic dynamic processes or air-sea interaction processes. In addition, models with finer oceanic resolution are typically accompanied by higher atmospheric resolution, and the simple comparison across LR and HR ensembles is not sufficient to highlight the relative importance of refining oceanic and atmospheric resolution to benefit the model simulation. The resolution hierarchy including solely increasing oceanic and atmospheric resolution for ECMWF-IFS and HadGEM-GC31 may provide a testable approach to address this issue.

The authors categorized 13 CMIP6 HighResMIP models into 3 low-resolution (CMIP6-LR) models and 10 high-resolution (CMIP6-HR) models for comparison with the SINTEX-F2 and SINTEX-F3 models. In fact, following Hewitt et al. (2020), the simulations can be classified into three regimes according to the resolution of their ocean components under the guidance of Rossby deformation radius: eddy-free (≥50 km), eddy-present (~25 km), and eddy-rich (~10 km). They suggested that whether the ocean mesoscale is explicitly represented in eddy-rich regime or parameterised in eddy-free/present regimes affects not only the mean state of the ocean but also the climate variability and the future climate response. Given such difference between eddy-rich and eddy-present models, is it reasonable to simply classify them together into CMIP6-HR? In addition, I noticed a considerable intermodal dispersion indicated by the large standard deviation among the CMIP6-HR ensemble. Is it partly attributable to disparities in different resolution regimes? Since the metric calculated from CMIP6-LR is included in the one standard deviation interval of CMIP6-HR, taking Figure 2b as reference, is the difference between CMIP6-LR and HR considered to be statistically significant?

In this study, the authors defined a series of climate indices within the same region and made comparisons of the simulation performance across multi-models at various resolutions. Could this introduce additional errors into the mode evaluation? Taking Figure 9 as an example, it seems that the defined region is not focused on where the salient center of ocean heat release is located. In addition, it has been widely recognized that the simulated western boundary currents might present meridional displacement bias compared to observation. Should it be more suitable to adjust the reanalysis regions according to the model behavior?

This study demonstrated limited improvement of simulation for the tropical climate variability such as ENSO in most of the eddy-permitting models, and found a remarkably reduced warm bias over the tropical ocean in SINTEX-F3 model compared to the coarse-resolution SINTEX-F2. This is different from the finding of Liu et al. (2022), which highlights that eddy-present models improve ENSO patterns because of the realistic mean state and associated SST-net heat flux feedback and underscores the reduced model cold biases of the equatorial SST mean state in higher-resolution models, which is attributed to increased oceanic resolution, and thus, better resolved eddy-driven heat transport. What factors contribute to these divergent conclusions?

Reference:
Hewitt, H. T., and Coauthors, 2020: Resolving and parameterizing the ocean mesoscale in Earth system models. Curr. Climate Change Rep., 6, 137–152, https://doi.org/10.1007/s40641-020-00164-w.
Liu, B., and Coauthors, 2022: Will increasing climate model resolution be beneficial for ENSO simulation? Geophys. Res. Lett., 49, e2021GL096932, https://doi.org/10.1029/2021GL096932.

Minor comments
In the introduction section, the authors mentioned the role of ocean mesoscale eddies as well as their impacts on the overlying atmosphere in the climate system. In fact, recent studies (Gan et al. 2023; Yu et al. 2024) reveal that such mesoscale and frontal-scale ocean–atmosphere coupling could significantly increase the formation of subtropical mode water, which is likely exerting far-reaching impact on the regional and global climate. It should be explicitly addressed in the introduction.

Reference:
Yu, J., B. Gan, H. Yang, Z. Chen, L. Xu, and L. Wu, 2024: Mesoscale Ocean–Atmosphere Coupling Effects on the North Pacific Subtropical Mode Water. J. Phys. Oceanogr., 54, 1467–1488, https://doi.org/10.1175/JPO-D-23-0148.1.
Gan, B., and Coauthors, 2023: North Atlantic subtropical mode water formation controlled by Gulf Stream fronts. Natl. Sci. Rev., 10, nwad133, https://doi.org/10.1093/nsr/nwad133.
Table 1: Spatial resolution of “Ocean-Sea Ice” should be specified for the SINTEX-F2 and SINTEX-F3 coupled models, which represents one of the most significant differences between the two versions of the SINTEX-F model.

Line 216: Replace "Is" with "is".

Line 222: Replace "^5th" with "5th".

Line 245: You mentioned “the atmospheric component of the SINTEX-F2 is based on the ECHAM6”. This disagrees with the model information listed in Table 1. Which is correct? Please check it carefully.

Section 2.3: As for the CMIP6 HighResMIP models, which variant labels did you use? Please clarify. Looking closer to Table 2, it is better to express oceanic resolution in kilometer units to keep correspondence with the atmospheric component as well as the information provided on the official website. In addition, it is wrong that the HadGEM3-GC31-HH model is categorized into LR.

Lines 256–258 and Lines 273–275: A linear trend of the variables in SINTEX-F model was removed to avoid the influence from model drift. Given that the 30-50 years spin-up period in CMIP6 may be insufficient especially for HR models, was identical data processing applied to the CMIP6 model outputs?

The shaded regions denoting one standard deviation of the model spreads for LR and HR ensemble overlap with each other in most of the pictures in the manuscript. Maybe try changing the color or increasing the transparency to make it clear.

It is mentioned many times in the figure caption that because of the limited number of the CMIP6-LR models, we did not apply statistical test to the regression values for the CMIP6-LR models. How do you apply the statistical test for multi-model ensemble? Should you conduct the regression analysis before calculating the multi-model ensemble mean, or directly conduct regression in the multimodel-averaged time series? Why do the model numbers limit the statistical test? Please be specific.

Remove line numbers from blank lines.

The Results section should provide a concise synthesis rather than exhaustive phenomenological descriptions without interpretative conclusions.

The manuscript needs to be further polished and have more professional phrasing. I recommend having the text checked for English style.
Citation: https://doi.org/10.5194/egusphere-2025-2258-RC1
RC2: 'Comment on egusphere-2025-2258', Anonymous Referee #2, 04 Aug 2025

The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2025/egusphere-2025-2258/egusphere-2025-2258-RC2-supplement.pdf

Citation: https://doi.org/10.5194/egusphere-2025-2258-RC2

Yushi Morioka, Eric Maisonnave, Sébastien Masson, Clement Rousset, Luis Kornblueh, Marco Giorgetta, Masami Nonaka, and Swadhin K. Behera

Supplement

https://doi.org/10.5194/egusphere-2025-2258-supplement

Yushi Morioka, Eric Maisonnave, Sébastien Masson, Clement Rousset, Luis Kornblueh, Marco Giorgetta, Masami Nonaka, and Swadhin K. Behera

Viewed

Total article views: 969 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	Supplement	BibTeX	EndNote
859	82	28	969	60	14	28

HTML: 859
PDF: 82
XML: 28
Total: 969
Supplement: 60
BibTeX: 14
EndNote: 28

Views and downloads (calculated since 28 May 2025)

Month	HTML	PDF	XML	Total
May 2025	54	5	2	61
Jun 2025	112	30	13	155
Jul 2025	60	24	10	94
Aug 2025	373	10	2	385
Sep 2025	260	13	1	274

Cumulative views and downloads (calculated since 28 May 2025)

Month	HTML	PDF	XML	Total
May 2025	54	5	2	61
Jun 2025	112	30	13	155
Jul 2025	60	24	10	94
Aug 2025	373	10	2	385
Sep 2025	260	13	1	274

Viewed (geographical distribution)

Total article views: 942 (including HTML, PDF, and XML) Thereof 942 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 11 Sep 2025

Short summary

Ocean mesoscale eddies, which have a horizontal scale with an order of 100 km, play a prominent role in global ocean heat transport that regulates Earth climate. Here we newly develop an eddy-permitting climate model to demonstrate that the increased ocean model resolution improves representation of air-sea interaction in the western and eastern boundary current regions, while the improved sea ice model physics benefit realistic simulation of sea ice variability.


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