Review on “Impacts of anomalies in Arctic sea ice outflow on sea ice in the Barents and Greenland Seas during the winter-to-summer seasons of 2020” by Zhang et al.

The Arctic sea ice reacts strongly to climate change. The common and well-recognized features are thinning of sea ice thickness and the shrinking of sea ice extent. On a local or regional scale, especially for those areas that are either associated with 1) the major sea ice formation, e.g., the northern coast of Greenland, the Lincoln Sea or 2) the major sea ice outflow transportation, e.g., the Fram Strait are of the great research interests. Because any perturbation or anomalies occurring in those areas are likely to have a major impact on the total sea ice budget and the downstream marine environment and the ocean circulation in general.

This paper focuses on sea ice outflow in two regional seas (the Barents Sea and the Greenland Sea). The topic belongs to the second category of great research interests (in my own opinion) stated in the first paragraph above. So, for this reason, I am very pleased to see such research has been carried out. The study subject fits well with the scope of TC. I recommend this work to be accepted by TC but with a precondition that the authors will carry out further improvement following my comments below because I do see there are still rooms to improve this work:  

Major comments

1 Title:

I would like to see a better title. The current title read: “Impacts of anomalies in Arctic sea ice outflow on sea ice in the Barents and Greenland Seas during the winter-to-summer seasons of 2020” When I looked at the following text, I felt the manuscript is mainly dealing with the factor that creates the Arctic sea ice outflow anomalies, as the author stated in the abstract (L9): “the impacts and feedback mechanisms on a seasonal scale of anomalies” So, I suggest authors speak out what “impacts”, e.g. atmospheric circulations. One possibility could be: The impact of atmospheric circulations on the anomalies of sea ice outflow and their feedback mechanisms in the Barents Sea and Greenland Sea

2 Abstract:

There are 322 words. I think it is too long, please compact it to e.g., 250 words. However, if TC accepts a long abstract, so be it, but please add some compact analyses/statements to echo the latest state-of-the-art findings. I am sure there are papers dealing with Arctic sea ice outflow

3 Introduction:

This part is largely ok, but as I have stated in the previous point, please consider echoing your work with UpToDate findings. I recommend authors check Sumata et al., 2022. This paper should be cited in your work. Some comparisons would be even better in the results/discussion section. The language can still be improved. This comment is valid for the entire manuscript.

4 Data and method:

I am quite impressed that such comprehensive data sets are used in this work, well done. I wish the authors could make further elaborate on data accuracy and comment/assess the data consistency, for example, the authors wrote “Here, we used the CryoSat- 122 2/SMOS SIT from December to April, and the PIOMAS SIT from May to June in 2011–2020 to estimate the anomaly in SIT during the study year of 2020”. Do I need to worry about the inconsistency of the data sets applied here?

Line: 180-182: “We note,,,”. So, this does suggest that in study deals with the impact of the atmosphere on sea ice, not the ocean at all. I suggest authors express this argument explicitly already in the beginning, e.g., introduction.

5 Results:

Figure 2 is very comprehensive and informative, yet in the main text, I see only once Figure 2 (the first column of Figure 2), please add more instruction on what text explains/analyses other columns of Figure 2.

Figure 3 is also very informative; I suggest you separate the last row of each panel to make it clearer and easier to distinguish from others. Furthermore, any patterns can be extracted from this figure?

The 3.3 section is very interesting. However, in order to prove the effectiveness of the reconstructed results of the ice floe backward drift trajectory, it would be interesting to compare the ice floe backward drift trajectory with forwarded observed buoy drift trajectories for example, under the scenarios of AO+, AO-, CAI+, CAI – and see whether the buoys’ drift trajectories are consistent with the reconstructed results. If not, any impact on your results and conclusions

Could you elaborate further on whether or not the abnormal AO and CAI would have impacts on sea ice thermodynamics, e.g. total ice mass balance before ice floes reached the Fram Strait? I would like to see more discussions.

6 Discussion:

Please strengthen the linkages of your work with the latest state of the art of research e.g., Sumata et al., (2022). In such extreme seasons, what are the possible impacts of Arctic sea ice outflow on the a) sea ice state and b) marine hydrographical and c) ecological conditions in the Barents Sea and the Greenland Sea?

7 Conclusions and recommendations:

I suggest you drop recommendations because you merely “recommended to further collect the in situ observation,,,, in the study region” which is not necessarily entitled as recommendations, unless you recommend some specific concrete parameters/variables or some specific instrumentation to be observed or to be used and further to be linked to each other.

Minor comments:

• Line 50: “plays a crucial role in shaping the icescape in this region”-- change the “shaping” to “proving the preconditions”
• Figure 1: There is no need to define geometrically regular study regions for the BGS. Its northern boundary can be consistent with the defined passageways, and the area bordering Greenland and other islands can be consistent with the shoreline.
• Line 107 “In addition, we used data from the NSIDC Sea Ice Index version 3 (Fetterer et al., 2017) to obtain monthly SIA changes in the Northern Hemisphere in 2020.” The purpose of using this data is unclear. In addition, data from the Arctic should be used instead of data from the Northern Hemisphere.
• Line 205 “regulating the sea ice outflow from the TPD region to the BGS” changes to “regulating the sea ice outflow from the Arctic Ocean to the BGS”
• Line 234 “resulting in relatively low SIAs of” change to related to..., Arctic sea ice outflow is only one of the factors affecting the reduction of Arctic sea ice.
• Line 266 “was insensitive to the changes in the TPD intensity or the CAI pattern” delete “the TPD intensity” because you did not directly quantify the strength of TPD.
• Line 309 “the monthly surface heat fluxes” (and also other text) changes to “the monthly atmospheric surface heat fluxes”
• Line 323 “sea ice during winter and early summer 2020.” change to “sea ice during spring and early summer 2020.”
• Line 370 “the absorption of incoming solar radiation” delete the “incoming”.
• Line 376 “are larger in the southern BGS (76°–80°N) than in the northern part (72°–76°N)”-- this should be a mistake.
• Line 395 “the complex interactions between SST, SIC and Chl-a” change to “the complex interactions between SST and SIC”.
• Line 401 “the year” changes to “the study year”.
• Line 403 “the abnormal Arctic sea ice flow” changes to “the abnormal Arctic sea ice outflow”.
• Section 4.3, 1) also consider the scenario with low Arctic sea ice outflow, 2) Does North Atlantic Oscillation have a significant regulatory effect on the marine environment of BGS?

15)Tables in Appendix: consider using simple expressions to indicate different significant levels, e.g., text in bold, or Italic.

It would be nice to apply professional language service for the entire manuscript

Reference:

Sumata, H., de Steur, L., Gerland, S. et al. Unprecedented decline of Arctic sea ice outflow in 2018. Nat Commun 13, 1747 (2022).  https://doi.org/10.1038/s41467-022-29470-7.

The manuscript  “Impacts of anomalies in Arctic sea ice outflow on sea ice in the Barents and Greenland Seas during the winter-to-summer seasons of 2020” by Zhang et al. studied the impacts of anomalous Arctic Oscillation and Central Arctic west-east air pressure gradient in 2020 on the Arctic sea ice outflow and feedback mechanisms in the Barents and Greenland Seas. It’s a good case study on atmosphere-sea ice-ocean-ecosystem interactions in the Arctic. Overall, the manuscript is well organized and presented in a logic way. I only have a few minor comments.

Minor comments:

Lines 19–20: I think “an extremely low Chlorophyll-a concentration observed over the BGS in April” is not supported by Figure 9. The anomaly of Chlorophyll-a concentration over the BGS in April 2020 is small shown by Figure 9a.

Line 47: Change “in the BGS” to “in the Barents Sea”.

Figure 2: Add ice drift velocity anomalies to the second and fourth columns.

Lines 251–253: “the anomalies of sea ice volume outflow in winter–spring 2020 were expected more obvious than the SIAF anomalies” may be not true. Because the ice thickness has larger negative trend than ice concentration during winter–spring over these regions, so it may lead to the volume outflow anomaly ranking is less obvious than the SIAF anomalies.

Lines 319–320: How to estimate SIT changes using the Eq.4 is not very clear. I think the surface heat flux anomalies shown in Figure 7 are the results for the whole BGS. Considering parts of BGS are ice-free, it is not a good idea to estimate SIT using the anomalies over the whole BGS. Using the surface heat flux anomalies only over the climatological ice-covered regions is more reasonable.

Line 343-345: How many years with extreme high or low are examined here?

Line 388-389: It’s good to add an appendix figure to support the conclusion of “The Chl-a over the southern Greenland Sea in April 2020 was smaller compared to the previous 5 years”.

Zhang et al. use preexisting ice, atmosphere and ocean datasets to examine sea ice inflows from the Arctic Ocean into the Fram Strait and the Barents Sea during the first half of 2020 - a period of a strong positive anomaly of the Arctic Oscillation index. They detail the atmospheric conditions in this period and analyse the satellite-derived sea ice area flux across gateways between the Arctic Ocean and the Greenland/Barents Seas on a background of the climatological mean over recent decades. They show sea ice extent and thickness, chlorophyll concentrations, and sea surface temperature in these two peripheral seas and discuss observed anomalies in the context of the observed sea ice fluxes.

The study is well-motivated, and the motivation is well laid out by the authors. The Atlantic sector of the Arctic Ocean is a focal point of system-wide ongoing environmental change, and studies of the interaction between ocean, ice and atmosphere in this region are of wide importance. Several studies have indicated that the northerly import of sea ice has a strong impact on regional atmospheric and oceanic conditions in these areas, yet large knowledge gaps remain in our understanding of the mechanisms that govern sea ice export to these peripheral seas and the impacts of variability in this export. It is exciting to see the combination of many different data products, and I commend the authors on the substantial effort that undoubtedly has gone into this work.

The quantitative analysis presented in the paper seems thorough and sound. Nicely made figures are helpful to the analysis, and the choice of methodology and data products seems appropriate for the most part. The methods section introduces the datasets and methodology in a clear way, and the introduction section gives a good overview of the background with appropriate references.

Unfortunately, the manuscript as a whole lacks for clarity in my view. I have read this manuscript several times, and I am still not sure I can summarise the main findings. There is a lot of interesting information in the paper, but I find it hard to follow the *reasoning* - it was often not clear to me what one should make of the specific results that were presented, and I often had a hard time understanding on what basis the authors arrived at their conclusions. The discussion section currently consists in large parts of new sets of results from an extended analysis, and does not in my view do much to aid the reader in the interpretation of the study. Throughout the paper, there is a general lack of separation between qualified speculation, knowledge based on existing studies, and conclusions substantiated in the analysis. 

There are many numerical quantities in this paper that are given importance (correlations, averages, anomalies etc). In my view, it is not always clear what these actually are - e.g., what is being averaged and over what domain, which variables are being correlated, how averages are computed, etc. It is important that such ambiguities are addressed in a final version of the manuscript (but I think this should be fairly straight forward to remedy).

I am hesitant to recommend the publication of this manuscript, and I would not recommend its publication in its current form. Yet, I believe that the authors have done some very interesting work here, and I think that this paper could be a valuable contribution to literature - but in my view, this would require a substantial effort in revising the paper.

I have included more specific comments below, but they should not be taken as an extensive list. My main recommendation to the authors would be to focus on making it much clearer to the reader what their key findings are and how they arrive at their interpretations. 

## MAJOR COMMENTS ##

There are many statements through the manuscript of 2020 being a year of very high sea ice exports through the gateways. The clearest example is perhaps L443, which states that area flux through the gateways was “extremely large compared to the 1988-2020 climatology”. I find that this is an inaccurate description of the data as described. E.g. in figure 3: 2020 looks like a year of fairly high exports, but does not seem to particularly stand out; for example, area exports were higher in all three gates preceding year of 2019. Elsewhere, it is stated that the JFM 2020 area flux through the FS gateway was 1.2 times the 1988-2020 mean (L219) - this does not strike me as extreme given the large interannual variability (and there is a corresponding ~20% negative SIAF anomaly through the FJL-NZ gate in 2020). Similarly for “extremely low” chlorophyll concentrations in April (L20) - is this really substantiated by the analysis? I think it is important to be quite careful with language here; strong statements need to be supported by a clear justification based on underlying data. (Note: I do not think this season being less “extreme” means that it is not worth studying! For example, an observation that an extreme AO year only had a moderate impact on sea ice outflow would in my view be a valuable contribution..).

The Results and Discussion sections in my view need reorganisation. There are many instances where some fairly strong claims which are not based on the present analysis are included in Results. Conversely, much of the Discussion section consists of the presentation of whole new sets of results and analysis that have not previously appeared in the manuscript. I strongly suggest reorganising the manuscript such that the Results section is reserved for the presentation of the outcome of the analysis (including that of SST and chlorophyll), and the Discussion section for the authors’ interpretation of the data/comparison with literature/qualified speculation etc. Given the complex topic and the many different datasets involved, the Discussion section of this paper is a great opportunity to carefully guide the reader through each key argument with reference to results from the different analyses.

• An example of discussion in the Results section is found in ~L288 - L293. Here, there are some fairly broad statements about stratification which are not actually substantiated in the data. Similarly with e.g., L233-L235 /  L239-241 / L299-L301 etc. These are important points but they require explanation and are not in my view obvious from the data alone.
• An example of results in the Discussion section is the correlation analysis in 4.1. This reads very much like new results to me. Same with the SST/chlorophyll analysis that follows.

If I have understood correctly, the atmospheric/sea ice area/chlorophyll analysis was done using averages across the area defined on L82 and shown as black polygons in Figure 1 BGS (if this is not the case, I suggest making it clearer which regions are used). The GS and BS boxes cover a vast area spanning quite different climatic environments and quite different ecosystems. I would guess that about half of this area is more or less never ice-covered in the present epoch; if chlorophyll in the southeastern part of the domain is influenced by sea ice inflows it may be in a rather indirect way. Likewise, if Figure 7 shows surface heat budget for the entire BGS area, I am not sure how meaningful they are. The authors need to justify this choice of a study area and discuss the implications of an analysis spanning widely different domains (or, if I have misunderstood, I would suggest that they clarify their methods..).

I would recommend going through the manuscript in general for clarity. In quite a few instances, the meaning of a sentence is ambiguous or, I suspect, not in line with the intended meaning. I’ve included some examples below under “technical/language”, but they do not constitute a complete list. 

I found the discussion of ecological conditions/phytoplankton to be severely lacking. First, it looks like the discussion is based on satellite chl-a concentrations from areas not directly influenced by sea ice (spring blooms in southern BS, eastern/southern GS). It was unclear to me whether the authors believe that conditions were favourable (L398) or unfavourable (L19) to biological production. Repeated statements about extremely low chlorophyll-a in April 2020 seem to refer to sea ice-covered areas where there are in fact no satellite measurements (Fig 9a). The statement that “phytoplankton seeding” and “residue of marine nutrients” were responsible for high primary production (L399-L400) is neither explained nor substantiated at all. I would suggest removing the chlorophyll discussion from the manuscript unless it is completely overhauled. 

I would say something along the same lines about the section on surface heat fluxes (L304 - L323). First, there is the issue of (apparently) using integrated fluxes over areas including huge ice-free areas of the north Atlantic to assess ice melting in the small northern/westernmost portions the domain (I would expect there to be large heat losses from the ocean to the atmosphere in the southern BS and over the West Spristbergen Current in winter, for example - how do they affect this estimate?). Second, the values that are given for estimated ice thickness change spans a huge range (1 to 41 cm). Lastly, I think it would be quite helpful to plot the actual delta h alongside the actual fluxes to clarify what we are actually looking at.

## MINOR COMMENTS ##

It might be nice to show time series of AO/CAI for context since these are pretty central to the paper. It’s not strictly necessary, but I would suggest adding this as a small figure, at least in the supplementary.

It is stated in the abstract (L14) and elsewhere that “the variability of.. total SIAF was dominated by changes in ice motion speed”, listing a high and significant correlation. In the text I could not find any details about how you actually calculated this, and given that this is a key point in the abstract I think the authors need to elaborate on actually how this quantity is computed, e.g.: 

• Is this a correlation between monthly values of a) SIAF across all three gates and b) mean speed across all three gates? Over what time period? 
• How are no-ice instances counted (for SIM and SIAF)? As zero, or are they not included?

Along similar lines, it should be clearly stated whether “sea ice thickness anomalies” include “thickness” from ice-free periods. (Fine either way, but “thinner ice” and “less frequently ice-covered waters” are different things, for example). 

I think it should be discussed whether comparing trajectories from one single day (e.g. 31 May 2020) with a climatological mean vector field (e.g. May 1988-2020 mean drift) may (or may not) influence the analysis. Is a trajectory along an average field different from a time average of several trajectories? Does the long-time average introduce a low speed/short trajectory bias? (I don’t really know - but I think it warrants at least a brief comment in the paper given that this is an important point in your study).

SIAF is basically the integrated product of SIM and SIC, so if it were not SIM that controlled SIAF variability it would presumably be SIC (or am I misunderstanding?). You should therefore probably include the corresponding correlation between SIAF and SICs somewhere in the paragraph at L236 in support of your statement that SIM controls SIAF. 

L45: I would revise the statement that sea ice outflow “contributes” to deep water formation in the north Atlantic. Please clarify what is meant here (presumably that high sea ice export -> high fw input -> increased stratification -> *inhibited* dw formation - or am I missing something?). I also can’t see that Lemke et al. 2000 is a great reference here - they show a large variability in sea ice export but don’t actually look at DW formation as far as I can tell? 

Figure 1 is excellent. The authors could consider adding mean SIC or similar (not necessary but might give some additional context to readers unfamiliar with the region).

The switch from CS2SMOS to PIOMAS seems to warrant a (quick) comparison of the two in the overlapping period. Should add at least a sentence about this somewhere around L123..

Which temperature does the oiSST give for fully ice-covered waters? What about for partial ice concentrations? Should be included in the presentation of this product somewhere after L124, and may be relevant for the interpretation of the SST data. 

L159: I believe that there is no SIM vector for SIC<15% in this product? I don’t believe that is a significant problem for your methods, but it is probably worth a mention here. 

L161: Was one trajectory computed from each grid point on each gateway?

L169: This seems to imply forward propagation, I would suggest reordering the equations to show backward propagation if that is what you do. Also, if delta t is negative as suggested in the preceding paragraph, I think this equation is in fact incorrect (wrong sign of second term)?

L223. If you attribute the 20% positive anomaly at FS to the AO anomaly, it seems strange to say that 20% negative anomaly at FJL-NZ means that flow across this gate was “not sensitive”. Could it not be a similar-magnitude response of opposite sign? Please elaborate.

L226: See above. This seems like a huge negative anomaly, and the idea that positive=response and negative=insensitivity needs an explanation at least. I also have a hard time seeing the 85% negative anomaly in Figure 3c - is this because the absolute values are small? 

L230: The phrasing here is a bit ambiguous - should make it clear what exactly these percentages are. (I assume that they are the fraction of SIAF through all three gates that went though FS - but it can read as the percentage of the *anomaly*). Same for L13.

L233: In my view, this is a strong statement that needs substantiation. What is your evidence that the low Arctic SIA was a result of increased outflow? Probably also a better fit in Discussion.

L237: Please explain exactly what this correlation is.

L251: “ the anomalies of sea ice volume outflow..” This is a bit confusing. You don’t have these numbers, right - is it that you would expect the volume export anomaly to be more pronounced than the area export anomaly? Rephrase for clarity.  

L267-L268: This seems true for the FS, but not for the two other gates?

Figure 4: I suggest using much thinner lines to show that these are trajectories and not a continuous scalar field - I think this figure is nice, but it took me quite a while to figure out what was going on.. Given that trajectories don’t run past January, I don’t think the colour by date actually adds that much, so I wouldn’t be worried if thinner lines show the colours less well. Should also change the labels in abc from e.g. “April 2020” to “30 April 2020” if you indeed only show back-trajectories from one single day in these.. (You should also explain why you chose to compare one day per month with monthly climatological means).

L298: It’s not clear to me from Figure 6a that there was a positive SIT anomaly in the BS - please explain.

L301: Similarly, it does *not* look to me like an overall positive anomaly in the GS in Figure 6d. And it seems strange not to address the east-west pattern in the GS here. And, as mentioned elsewhere, I think you also need to say something about whether the positive anomalies are a result of anomalously high sea ice *extent* or anomalously thick ice since this could impact your argument about increased import of thicker ice from the central Arctic (probably depends on how you compute these means).

L337-8: Is it surprising that you did *not* find a significant correlation between SIAF and AO in months other than February? Seems to warrant at least a mention.

L340: What about R?

L345-351. I found it difficult to see this in Figure A1ab, please explain how you arrive at this conclusion. To me, it looks like there is little difference except perhaps that the drift is stronger in the negative AO phase (A1b). It also looks like these are the trajectories of ice arriving at the gate in mid-summer - so is it actually representative of AO anomalies in winter? Please clarify. 

L365: Are these SST anomalies integrated over the whole area? Also, are the maps in Figure 8 really anomalies from the climatological mean for the month? To me they look like anomalies from the SST average across the entire season or something (+4C anomalies in the southern BS in July and -3C anomalies in the northernmost range in April both strike me as weird). I could certainly be wrong - but it might be a good idea to double check.

L389 onward: Again, it seems strange to me to do this sort of analysis across this very large area - you at least need to discuss the implications of 1) including large perennially ice-free areas, and 2) there being no satellite measurements of chlorophyll in ice.

I found the Conclusions section to be quite clarifying and well-written. I would encourage the authors to use a similar style in an updated Discussion section. 

## TECHNICAL/LANGUAGE ##

L13: “77.6%”: I gather from the conclusion that this is the fraction of the total flux through the three gateways that goes through the FS gate. That was not clear to me when reading the abstract alone.

L51: “More primary productivities” - please use more precise language.

L55: I would suggest reordering this sentence for clarity.

L63: “Relatively extreme” - please use more precise language.

L69: “Thereby.. ..the BGS.” - Meaning of this sentence is unclear.

L82: Studies -> Study?

L107: “abnormal” - “anomaly”?

L124: “Could be used as the best proxies for..” - please rephrase. “We use SST and chl as proxies for..”?

L137: Suggest replacing “ a significantly more advanced” with “an advanced” or similar (“more” relative to what?).

L143: “According to”  - “In accordance with” or similar?

L146: These directions aren’t really S/N? Maybe replace “southward” with “toward the BGS” or something.

L185: “with the values maintaining the top three..” - This phrasing is a bit unclear (here and elsewhere in the manuscript). 

L215-L217: Suggest moving this sentence to data/methods (and adding e.g. “relative to the 1988-2020 climatology” after “June 2020” in L218).

L238: Maybe “monthly SIM speed” to avoid confusion.

L256: “very analogous to” - “similar to”?

L261: “confluence of the Kara Sea and the CAO” - please explain.

L262: “exhibited a relatively high tortuous feature” - unclear-please explain.

L280: Sentence needs fixing; meaning of “since then” is unclear. 

L280-L288: I strongly suggest revisiting this entire section for clarity; I found it quite difficult to follow most of it.

L287: What is this correlation? Is it between the sum of the SIAF through all three passageways and SIA in the BGS? Also: “has been identified” should be rephrased for clarity (e.g. “we found a correlation..” if that is the meaning).

L304: “dominated” - “governed”?

L337: “most active” - rephrase for clarity.

L347: “spatial scope” - “spatial extent”?

L376: typo (southern/northern).

L396: “a single SST”  - rephrase for clarity.

I suggest looking over sentences starting with “thereby”, “furthermore” etc. and checking that these preserve the intended meaning. 

Figure A1: Please label clearly which plots correspond to positive/negative phases, or at least explain/refer to labels in the caption (“AO+sd” is not very intuitive on its own). I would also suggest adding the number of years that go into the mean (e.g. “n=3”) somewhere on each subpanel (not necessary but would aid interpretation).

Review of TC-2022-246“Impacts of anomalies in Arctic sea ice outflow on sea ice in the Barents and Greenland Seas during the winter-to-summer seasons of 2020” by Zhang et al.

The outflow of Arctic sea ice is important for the mass balance of Arctic sea ice as well as Arctic sea ice change in a warming climate. As the bearer of Arctic sea ice outflow to the Atlantic Ocean, the Barents and Greenland Seas have strong atmosphere-ocean interactions and are also sensitive to changes in sea ice outflow from the Arctic Ocean. In this study, Zhang et al. showed the changes of Arctic sea ice outflow under anomalous atmospheric circulation in the winter to summer seasons of 2020, and analyzed the influence of Arctic sea ice outflow on sea ice and ocean conditions in the Barents and Greenland Seas. This study presents some interesting results that contributes to the understanding of the modulation of atmospheric circulation on changes in Arctic sea ice outflow and the subsequent effects on sea ice conditions in the Barents and Greenland Seas. The manuscript is well organized and the data analysis is generally adequate. I believe some extra efforts are necessary to improve the reasoning and presentation of their idea in the paper. Detailed comments are given below.

General comments:

1. What is the reason to choose these two atmospheric circulation patterns for assessing the effects of large-scale atmospheric circulation on the changes in Arctic sea ice outflow? I would suggest that the authors also look at whether there is an abnormal North Atlantic Oscillation in 2020, and if so, the influence of the North Atlantic Oscillation on Arctic sea ice outflow should be discussed.
2. Section 3.3 focus on the comparison of the reconstructed sea ice backward trajectories in 2020 with the 1988-2020 climatology. The comparison assumes that the reconstructed sea ice backward trajectories are convincing. However the validation of the reconstructed backward trajectory method is not sufficient. I would suggest that the authors provide more assessments on the validity of the reconstructed trajectories using buoy observations.
3. Section 3.4 discussed the anomalies of the sea ice area and thickness in the Barents and Greenland Seas. The data analysis on sea ice area is relatively adequate. However the analysis of sea ice thickness anomalies is mostly qualitative. I would suggest that the authors provide more quantitative results on sea ice thickness anomalies and discuss them in details.
4. If the abnormal Arctic Oscillation and Arctic sea ice outflow do not occur in winter, but in other seasons, will the effect be different for the ice and marine environment conditions in the Barents and Greenland Seas? It would be better to add more discussion on this.
5. It is a last resort to use different sea ice thickness products (radar altimeter and PIOMAS model-based data) in different seasons. My concern is whether using different data produces inconsistent results. For example, during the freeze-up period, whether there is deviation or even contradiction between the qualitative conclusion and quantitative results using PIOMAS model-based data and radar altimeter?

Specially comments:

1. Line 72, change “during winter-spring 2020” to specific month, since you do not define the range of winter and spring months before that.
2. Line120-122, “We regridded the monthly SIT data on the 25-km EASE-Grid to maintain consistency with the CryoSat-2/SMOS SIT data.” These two datasets also have different temporal resolutions. How was this difference addressed in your study?
3. Line 161, change "restructured" to "reconstructed" to unify the expression and apply to the entire manuscript.
4. Line 271, “enhanced sea ice meridional motion”, remove the “meridional”, since you do not directly calculate the meridional sea ice motion speed.
5. Line 294, The text on the right side of Figure 5 is too busy and not intuitive enough. It is preferable to express the trend of sea ice area graphically.
6. Line 319, “The anomalies in cumulative surface heat fluxes from January to June 2020 can be related to a reduced decrease of 0.01–0.41 m in SIT, estimated using the Eq. 4” This is ambiguous. Does cumulative mean cumulative over time or across net surface heat fluxes?
7. Line 327, change the y-axis title in Figure 7 to “Surface heat fluxes anomaly”, because it is surface heat fluxes anomaly instead of surface heat fluxes in the figure caption.
8. Line 377-378, “In addition, we examined the statistical relationship between the April SIA and the monthly SST with a lag of 1–3 months in the BGS (Table A4).” Is this correlation for the detrended SIA and SST? This comment applies to the entire manuscript.
9. Line 388, “The Chl-a over the southern Greenland Sea in April 2020 was smaller compared to the previous 5 years.” Give the latitudinal range of the southern Greenland Sea.
10. Line 389, “A significant negative correlation between Chl-a and SIA in April over 1998–2020 was identified”. The geographical scope of this sentence is unclear. Is there a negative correlation between Chl-a and SIA in the BGS or only in the Greenland/Barents Sea?
11. Some illustrations need to be further revised, for example, the definition of the study area in Figure 1 is not so normative.