the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Jun 2025
A suite of coupled ocean-sea ice simulations examining the effect of changes in sea-ice thickness distribution on ice-ocean interaction in the Arctic Ocean
Abstract. A major shift in Arctic sea ice occurred in 2007, transitioning from thicker, deformed ice to thinner, more uniform ice with reduced surface roughness. This abrupt change likely altered the dynamic and thermodynamic interactions between sea ice and ocean, with potential implications for nutrient and biogeochemical cycles in both sea ice and the upper ocean. In this study, we present a suite of regional coupled ocean-sea ice simulations designed to assess the potential impact of the regime shift on sea ice-ocean interactions, with a regional focus on the Atlantic sector of the Arctic Ocean. The different sea ice regimes are represented by changes in ice thickness distribution described by ice thickness classes in the sea ice model, and the effects of the different regimes are simulated through variations in the drag coefficient diagnosed from the ice thickness distribution. We emulate different sea ice regimes by prescribing sea ice properties at the model's lateral boundaries. We describe the experiment setups and the use of observational data that supports a comparison between pre and post regime shift sea ice conditions. Key differences in the simulated physical environment are highlighted, with a focus on sea ice-ocean interactions and upper ocean stratification. The simulation framework and the physical analyses presented here serve as a basis for ocean biogeochemical modelling studies that aim at understanding ocean ecosystem responses to changing Arctic sea ice.
- Preprint
(2809 KB) - Metadata XML
-
Supplement
(1589 KB) - BibTeX
- EndNote
Status: final response (author comments only)
- RC1: 'Comment on egusphere-2025-3022', Anna Nikolopoulos, 22 Aug 2025
-
RC2: 'Comment on egusphere-2025-3022', Samuel Brenner, 02 Sep 2025
The manuscript A suite of coupled ocean-sea ice simulations examining the effect of changes in sea-ice thickness distribution on ice-ocean interaction in the Arctic Ocean by Sumata et al. presents an interesting view of changes to ice-ocean coupling before and after 2007, when a notable regime shift in ice thickness and deformation characteristics was observed. Using a set of models initialized with either pre- or post-2007 ice thickness distributions (ITDs) at the boundaries to isolate the role of changes to the ITD, along with variable atmosphere-ice and ice-ocean drag coefficients, the authors show that the post-2007 shift to a thinner, less deformed ice cover also results in a weaker ice-ocean dynamic coupling in the Atlantic sector of the Arctic, with accompanying impacts on ocean surface fluxes and upper-ocean stratification.
The study is of generally high quality and interesting, the arguments are well laid out, and the writing and presentation are mostly clear. However, by focusing on changes to ice-ocean drag, I think that the study does not provide a complete picture of the net effects of the shift in ice cover. Importantly, as mentioned in my general comments (below), there is a lack of discussion of the interrelated changes to sea ice internal stresses, which were also likely impacted by the modelled change in ITD, and which have an important and well-documented role in ice-ocean dynamic coupling. That said, I think that this study will likely warrant publication after some changes to address the concerns listed below.
General comments
- How do rheological terms in the ice momentum equation fit into this story? The manuscript focuses on the role of drag coefficient variations in modifying ice-ocean coupling. However, the change in mean ice thickness in PRE and POST runs will also impact internal ice stress (usually modelled as having a strength that is a function of ice thickness), as shown here by the change in ice strength between PRE and POST runs in Fig. S2. As the internal stresses can act as a sink of momentum, they modify ice drift speeds and change the net transfer of momentum from the wind into the ice (e.g., Martin et al., 2014; Gimbert et al., 2012; Brenner et al., 2023; Muilwijk et al., 2024, and others). Differences between the runs POST, POST.lvl, and POST.rdg may help elucidate the effects of these forces.
- The idea of an increased seasonal cycle of sea ice in the POST experiments appears a few times throughout the study (e.g., L320-321), and the impacts on heat and freshwater fluxes are shown in Fig. 5. However, Figs. 3 and 4, which show differences in some fields between PRE and POST experiments, primarily show only winter time periods (except for Fig. 4c). Since the increased ice melting in the POST run is an important underlying cause of some of the other details of the manuscript, it would be helpful to show it more explicitly.
- For both of the above comments, I wonder if some direct inferences can be made to the results of previous studies that have looked at the role of form drag compared to constant drag coefficient (e.g., Tsamados et al., 2014; Martin et al., 2016; Castellani et al., 2018; Sterlin et al., 2023). While a comparison between variable versus constant drag doesn’t map exactly to the results here with two differing regimes of variable drag, some discussion of these past results (and particularly, changes in the ocean response in cases they were investigated) may still be informative. Does the post-2007 regime look more like a constant drag case?
Specific comments- L25-26: It would be nice if the abstract contained a very brief description of what the key differences are, rather than just saying that they will be highlighted.
- L26-28: The biogeochemical modelling studies mentioned here are not a part of this manuscript, nor are the results discussed in the context of biogeochemical or ecosystem studies, so I would not include this sentence in the abstract.
- Fig 1b-c: While I appreciate schematic panels like this, these versions do not necessarily add much to the figure (perhaps partly because it is not apparent that there are any changes aside from the ITD, when you could schematically show changes in velocities too).
- L105: Wrong sub-panel reference: should be Fig. 2b
- L154: Do you use a total or relative keel depth? Do you assume uniformly sized keels?
- Table 1: Lmax=300 m is a common choice, but is it appropriate? (e.g., Sterlin et al., 2023 show their results are sensitive to this choice).
- L186/191-192: Is one year sufficient for the ITD to grow to a roughly equilibrated state?
- Figure 3: It seems that the authors could better maximize the information density and impact of the figures by choosing map domain extents that match the model domain. Then there would be less wasted white space in each panel. This is especially pertinent given the small size of the panels necessary to fit all of them in the figure. (Comment also applicable to Figure 4)
- Figure 3: the only simulated field shown for the PRE run is the mean ice thickness, while all other panels show only differences between PRE and POST. But some of the fields themselves would be instructive to see. Particularly, the drag coefficient values, which are a major focus of this study. Figure 3f shows POST-PRE difference values as high as 12×10-3—a value higher than the previous reported maximum values of the coefficient in that region (in Tsamados 2014, Fig. 5f/6f). With so high a value in the differences, it would be beneficial to know what the baseline drag coefficient values are (and how they vary though the year)
- L229-234: The authors should comment here (and elsewhere as appropriate) on how the interplay of changes in the atmosphere-ice and ice-ocean drag coefficients play out. The statements in the paragraph are logical, but do not tell the complete story. For example, it’s stated that a lower ice-ocean drag leads to faster wind-driven ice motion; however, this overlooks the fact that a lower atmosphere-ice drag (which would be similarly expected due to a reduction in sails) should act in the opposite sense, causing a slowdown in the wind-driven ice motion. The increase in ice speed is the net result of these effects. Does it mean that atmospheric ridge form drag is less impacted than oceanic drag? This is briefly hinted at in the conclusions (L322-323), but a more complete explanation is needed.
- L236-237: Awkard phrasing in “ocean currents exhibit a logarithmic decline downward”
- Fig 3 and 4: Can you show an ice extent contour on the maps? In Fig 4b–c, it almost appears that the mean summer ice extent is effectively the same as in winter (based on the extent of the coloured regions), but I’m sure that’s not correct.
- L261/265: What is the form of the temporal filter (moving average? Butterworth?...)
- L267-269: How deep do the differences in MKE extend?
- L271-274: The separation into MKE and EKE, and subsequent description, indicates an attribution of short-timescale fluctuations (<30 days) to mesoscale eddy processes. Are mesoscales well resolved in a model with a horizontal resolution of 4km? Would a response to storm events be included in the EKE?
- L298: I think more explanation is warranted regarding the interpretation of POST.lvl versus POST.rdg experiments. If I am understanding correctly, POST.rdg has a higher fraction of ridged ice (L196-197), thus it should have a stronger ice-ocean drag and stronger dynamical coupling than POST.lvl. That seems consistent with the results shown and the explanations given, but it could be stated more explicitly to aid readers.
- Figure 6c: How can there be a positive-definite salinity difference in POST.lvl – POST.rdg if there was no change in freshwater flux (Fig. 6b and L299)?
- Figure S1. The “jet” colourmap in panels b and c is not perceptually uniform and can create issues with interpretation; it should be avoided. Additionally, if the purpose of the figure is to demonstrate eddy activity in the ocean, perhaps ocean vorticity would be a better metric to plot than velocity. (And since velocity is a vector quantity, perhaps direction “quiver” arrows could be included)References:
Brenner, S., Thomson, J., Rainville, L., Crews, L., & Lee, C. M. (2023). Wind-driven motions of the ocean surface mixed layer in the Western Arctic. Journal of Physical Oceanography, 53 (7), 1787–1804. doi: 10.1175/JPO-D-22-0112.1
Castellani, G., Losch, M., Ungermann, M., & Gerdes, R. (2018). Sea-ice drag as a function of deformation and ice cover: Effects on simulated sea ice and ocean circulation in the Arctic. Ocean Model., 128 , 48–66. doi: 10.1016/j.ocemod.2018.06.002
Gimbert, F., Jourdain, N. C., Marsan, D., Weiss, J., & Barnier, B. (2012). Recent mechanical weakening of the Arctic sea ice cover as revealed from larger inertial oscillations. Journal of Geophysical Research: Oceans, 117 (C11). doi: 10.1029/2011JC007633
Martin, T., Steele, M., & Zhang, J. (2014). Seasonality and long-term trend of Arctic Ocean surface stress in a model. J. Geophys. Res. Oceans, 119 (3), 1723–1738. doi: 10.1002/2013JC009425
Martin, T., Tsamados, M., Schroeder, D., & Feltham, D. L. (2016). The impact of variable sea ice roughness on changes in Arctic Ocean surface stress: A model study. J. Geophys. Res. Oceans, 121 (3), 1931–1952. doi: 10.1002/2015JC011186
Muilwijk, M., Hattermann, T., Martin, T., & Granskog, M. A. (2024). Future sea ice weakening amplifies wind-driven trends in surface stress and Arctic Ocean spin-up. Nat Commun, 15 (1), 6889. doi: 10.1038/s41467-024-50874-0
Tsamados, M., Feltham, D. L., Schroeder, D., Flocco, D., Farrell, S. L., Kurtz, N., . . . Bacon, S. (2014). Impact of variable atmospheric and oceanic form drag on simulations of Arctic sea ice. J. Phys. Oceanogr., 44 (5), 1329–1353. doi: 10.1175/JPO-D-13-0215.1
Citation: https://doi.org/10.5194/egusphere-2025-3022-RC2
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 976 | 45 | 14 | 1,035 | 24 | 19 | 33 |
- HTML: 976
- PDF: 45
- XML: 14
- Total: 1,035
- Supplement: 24
- BibTeX: 19
- EndNote: 33
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments:
I found this to be a very concise, insightful, and well-written paper. It was a pleasure to read, both in terms of content and structure. As a reader, I was guided through the background and motivation of the study in a transparent way. While questions arose at times, they were often immediately followed by explanations, indicating a well thought-through manuscript.
The Arctic Ocean sea ice (and ocean) characteristics change as we speak, and it is vital to make progress on deciphering the implications for the entire air-sea ice-ocean system, across all white/blue/green science disciplines where ocean stratification is a key parameter/indicator in all of them. Modelling efforts are central in that aspect, providing us both with the global and long-term scales for climate aspects and the 'topical experimental boxes' needed to explore the complex system in systematic ways (as in this study).
With the drastic 2007 regime shift recently detected and explored by Sumata and others, it is crucial to follow up with studies as the current one, on the meaning of such a shift. The current effort for improving our understanding was focused on the effect of smoother and thinner ice on upper (< 50 m depth) ice-ocean exchanges.
The methodology/approach is clever and effective, with both the main runs (PRE/POST) and the sensitivity runs increasing the range for the POST conditions. The methodology builds upon established modeling tools and parameter values (eg. for the drag coefficients), perhaps not granting 'excellent' scores for pure novelty but nevertheless leading to robustness and also comparability towards related studies based on the same setup.
The simulations indicate that the seasonal cycle of the ice itself (melt/formation), freshwater and salinity is amplified due to the mechanically weaker ice in the POST regime. The behaviour and drift of the altered sea ice is also simulated to change (I found the TPD velocity profiles intriguing!), with implications on mixing properties (decreases) and stratification (increases) of the upper ocean with the thinner and less deformed ice.
Specific comments/reflections:
1. On the use of the 2012-2015 simulation period for both the PRE and POST runs (ie. pre-2007-ice conditions superposed on post-2012 ocean background): I understand one has to choose for consistency and for limiting variations for your background 'items', but could you elaborate on potential implications for the results, as I imagine that the background conditions may have been different before 2007, from 2012-2015?
Also, you consider the simulation period as short (L191: 'preserving the similarity of the background ocean stratification of the twin experiments'). I find this confusing, since the upper ocean characterisics surely are variable enough for potential changes to arise within a span of 3-4 years? Can you clarify that?
2. You present estimates for the 'TPD' box outlined in black, and I understand and agree with the motivation behind examining the effect in this important 'funnel area' for Arctic Sea ice. Did you ever consider other placements of your boxes? It would be enlightening to see if/how the shown effects apply to other areas as well, within the model region.
3. In your conlusions I would find it useful with some more words on the 'hands-on' usability for your results, particularly with respect to the BGC work. Could your results be implemented directly in the BGC modelling hands-on, beyond contributing to improved understanding and explanatory value also for that context? Gaps and challenges to still overcome in this context?
Technical (more hands-on) corrections:
Further comments and suggestions for minor edits/clarifications on text and figures are incorporated into the attached PDF document as I find it more time efficient to do this during the read-throughs. I hope this works as format of such feedback (instead of pasting in more text here).