Effects of sea surface and air temperatures on interannual variations and trends of Arctic sea ice concentration in summer and autumn&nbsp;

Chen, Di; Sun, Qizhen; Vihma, Timo

doi:https://doi.org/10.5194/egusphere-2024-2359

Preprints

https://doi.org/10.5194/egusphere-2024-2359

Preprints

12 Aug 2024

| 12 Aug 2024

Effects of sea surface and air temperatures on interannual variations and trends of Arctic sea ice concentration in summer and autumn

Di Chen, Qizhen Sun, and Timo Vihma

Editorial note: the affiliation of the first author could not be verified and has been changed to 'independent researcher'.

Abstract. We analyzed the Arctic atmosphere – sea ice – ocean relationships to reveal their intrinsic connections and the roles of the sea surface temperature (SST) and surface air temperature (SAT) on the interannual variations and trends of Arctic sea ice concentration (SIC) in July to October during 1951 to 2021. Both SST and SIC have significant impacts on Arctic SIC. SST affects both interannual variations and decadal trends of SIC, whereas SAT has more significant effects on interannual variations of SIC. In addition, SAT affects SIC trends with a seven-month lead time, which is due to the much stronger warming trend in winter than summer. Statistically SST explained 53 % and SAT 35 % of the detrended interannual variance of SIC. SIC trends will continue to decline in the future as SAT and SST continue to rise.

Received: 26 Jul 2024 – Discussion started: 12 Aug 2024

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 preprint. The responsibility to include appropriate place names lies with the authors.

Download & links

Preprint (PDF, 3783 KB)

Please read the editorial note first before accessing the preprint.
Preprint (3783 KB)

Download & links

Please read the editorial note first before accessing the preprint.

Di Chen, Qizhen Sun, and Timo Vihma

Editorial note: the affiliation of the first author could not be verified and has been changed to 'independent researcher'.

Status: closed

RC1:
'Comment on egusphere-2024-2359', Anonymous Referee #1, 08 Sep 2024

I gave Fair to 1-4, but recommended rejection, because there is a conceptual issue in the title, overall work arrangement and conclusion.
Considering that sea surface temperature (SST) of the whole Arctic directly depends on Arctic ice-covered area - whether partial open-ice area is used (meaning that the more Arctic covered by ice the lower temperature as a smaller area will form a larger relative fraction of near-ice zone and therefore will be colder, and vice versa) or if the ice-covered Arctic is used as well (meaning that it will naturally give lower SST), any statement that sea surface temperature is the reason for changes in Arctic ice extent can be misleading. I realize that, yes, warmer water facilitates ice melt, however, Pearson correlation analysis is not the way to analyze the actual impact and separate it from feedback.
There is nothing wrong in studying correlations, but there need to be some basic understanding of the processes projected on that study, and here I am not convinced if sea-ice extent changes made SST different or the other way around or both. So, I see no gain for the understanding of this complex coupling mechanism.
Sorry ...

Citation: https://doi.org/10.5194/egusphere-2024-2359-RC1
- CC1:
  'Reply on RC1', Qizhen Sun, 09 Sep 2024
  
  Your comment contains the following statement “any statement that sea surface temperature is the reason for changes in Arctic ice extent can be misleading”.Please provide the basis for this viewpoint.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2359-CC1
  - RC2: 'Reply on CC1', Anonymous Referee #1, 09 Sep 2024
    
    ok, I will explain it in very easy terms, imagine ground partially covered by snow, you measure surface temperature of the whole area (or only of the snow-bare part, either way supports my point). Then the Sun comes out and melts snow, which retreats. The same Sun also heats the ground not even close to the snow age. You measure snow area and surface temperature again. Will they be correlated? I guess they will be correlated. Is it now correct to conclude that the warmer ground melted the snow? Or maybe, it is the retreat of the snow allowed more ground to be heated, raising its temperature. This is what I meant in the two statements: (1) the correlation of the two is expected, and (2) it is not obvious from simple correlation approach that warmer water (over the whole ocean, including the locations remote from ice) was the reason for ice melt. Yes, rapid warming by the ice edge accelerates ice melting, but this needs to be studied differently than suggested. Otherwise. refer to my analogy with snow on the ground ...
    Thanks!
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC2
    
    CC2: 'Reply on RC2', Di Chen, 09 Sep 2024
    
    This paper only focus on the the impact of sea surface and air temperature on Arctic sea ice. It doesn't deal with snow. The title is pretty clear.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-CC2
    
    RC4: 'Reply on CC2', Anonymous Referee #1, 09 Sep 2024
    
    It seems like, you do not understand what I am trying to explain - I explained my point to you using snow as analogy just to make the idea more transparent. If you replace snow with ice and ground with sea, the point will be the same. Since you did not understand my point that the correlation of the "geometrically" connected processes (less ice means higher SST without any look in processes, just because of areas change), I thought it might be easier to use the snow analogy.
    If you want an example of correlation meaning impact or effect, this would be winter convection in the ocean - ocean loses heat to atmosphere and convects, but the depth of its convection does not have a significant return effect on air. Do you see the difference?
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC4
    
    CC4: 'Reply on RC4', Qizhen Sun, 09 Sep 2024
    
    These explanations of yours are completely irrelevant to this investigation.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-CC4
    
    RC6: 'Reply on CC4', Anonymous Referee #1, 09 Sep 2024
    
    Another disrespectful statement from the author, without even trying to understand what the reviewer meant or proving the reviewer wrong.
    I strongly disagree, and repeat that the way the problem was approached gives a kind of outcome that was expected simply because the high correlation is a partly if not mainly a result of ice-covered water being colder than ice-free. I am just impressed by the amount of effort it takes to explain something obvious, and the author should be thankful that I ignore their disrespectful conduct.
    Thinking of the problem as a whole, one could use the whole basin SST (which includes sea ice), or just over the ice-free part. The first approach gives expected result as the mean SST itself is function of ice extent. The second approach has a major caveat - every time the region is different than in the other times, so the comparison is not of exact locations, and therefore the mean SST filed contributes there differently.
    The same problem could be approached differently, but that would be a different study, a different paper.
    The high correlation of 0.9 fully supports my point - it is so high because it is ice-to-icefree ratio is included in it.
    My apologies to the disrespectful authors that I could not explain it even simpler than that. Hopefully, the other reviewer will understand my points and do better job explaining it to the author.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC6
    
    CC3: 'Reply on RC2', Di Chen, 09 Sep 2024
    
    BTW, your example of snow and sea ice is completely different from your previous statement that “any statement that sea surface temperature is the reason for changes in Arctic ice extent can be misleading”. “ which are two completely different issues. You are completely misleading.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-CC3
    
    RC5: 'Reply on CC3', Anonymous Referee #1, 09 Sep 2024
    
    Read carefully what I just said about the snow example - it is just another case of correlation not meaning a direct impact or a direct effect. And I do not take an offense when you call me "misleading". You did not get my point, which I thought was obvious for any oceanographer, and so I went into trouble explaining my point to you as best as I could. You are trying to offend me instead ;).
    Once again, do a simple math with changing ice area and see who it impacts SST.
    It is a shame young generation do not have respect to those who are trying to help them.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC5
  - RC3: 'Reply on CC1', Anonymous Referee #1, 09 Sep 2024
    
    "The temporal coefficients of the Arctic SAT and SIC fields (Figure 4c) show an obvious inverse phase relationship (r = -0.90). This analysis shows that the SST has a greater impact on the SIC than the SAT."
    A high correlation of two coupled processes does not mean that one has a great impact - liquid water by all means is warmer than sea ice, and your conclusion is just a result of simple shift in the sea ice cover.
    Furthermore, SST and AT are correlated, and therefore, and statement about the relative role of AT over SST can be misleading as weel. If the two processes are correlated, it is not that easy to partition their impact on the third process. To explain it simpler - AT may not be a factor at all (although it is, but let's assume opposite), but it may be correlated with sea ice only because SST is correlated with the latter.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC3
    
    CC5: 'Reply on RC3', Timo Vihma, 09 Sep 2024
    
    Thank you for your valuable and insightful comments on the manuscript!
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-CC5
    
    RC8: 'Reply on CC5', Anonymous Referee #1, 09 Sep 2024
    
    Thank you, Timo,
    Hope it all makes sense!
    I have done lots of correlations to be able to tell where we can state that some effects are real and direct, and when the variables are just "forced" to be correlated. SST is a tricky thing - it is both result and factor at the same time.
    The case of this review was an exception of my rule of no anonymity when reviewing (the reason was that I did not go into a detailed review spotting an issue at start). I can share my name and email, if further explanations, clarifications and suggestions are needed.
    Thanks,
    
    IY
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC8
RC7:
'Comment on egusphere-2024-2359', Anonymous Referee #2, 09 Sep 2024
Few questions need to be addressed and answered by the author before making further decisions:
Why July-October were selected?

The SIC data used in this study was from Hadley Center. Does the pixels with value ＜15% need to be eliminated just like the SIC data retrieved from satellite remote sensing data.

When calculating SIC in different sea regions of the Arctic Ocean, there should be a geographical boundary, how to define it?

In figure 1, how was the multiyear ice edge (green line) identified? What the other colored lines represent?

Why only v-wind component was analyzed?

The results of statistical analysis can only show that there is a relationship between SST and SIC, but can not show that SST causes the change of SIC. The mechanism should be analyzed or studied. In addition, SST and SAT affect each other, how do you think about their relationship when doing dominance analysis.

Line 165-170: Those sentences describe the results of SVD statistics, which is OK. However, that are not enough to support the conclusion of the last sentence “This analysis shows that the SST has a greater impact on the SIC than the SAT”. Your study focuses on July to October, which is the polar day time in Arctic. Large amounts of solar shortwave radiation reach the earth surface. Open water regions absorb much more solar energy and get warmer because of lacking of sea ice coverage. So these regions get higher SST. Does the higher SST lead to low SIC? The higher SST may lead to the SIC decline in marginal sea ice zone, but what’s the influence extent of SST on SIC far from marginal sea ice zone? In figure 4a, another question comes out. Are there effective SST values under sea ice?

All figures in the manuscript have low resolution.

As stated in the line306-307 of the manuscript, more work is needed to better distinguish between statistical and causal relationships. So far, authors have done some statistical analysis. However, these are not sufficient to prove that SST and SAT have a significant effect on SIC.
Citation: https://doi.org/10.5194/egusphere-2024-2359-RC7
- RC9:
  'Reply on RC7', Anonymous Referee #1, 09 Sep 2024
  
  "The results of statistical analysis can only show that there is a relationship between SST and SIC, but can not show that SST causes the change of SIC. The mechanism should be analyzed or studied (1). In addition, SST and SAT affect each other, how do you think about their relationship when doing dominance analysis (2)."
  These are exactly the two points I made (thanks!!!!), which the authors claimed to be irrelevant, and my viewpoint misleading, but I red-flagged as critical for the whole story.
  In my further explanations I pointed that any ocean-wide or ice-free-wide SST value is directly affected by sea ice. SST and SAT are indeed dependent, and if someone goes to process study, SST over the whole Arctic may not be a direct precursor of the ice extent change as the actual interaction happened in a certain zone.
  Apologies for writing to you, but expressed my hope last night that the other reviewer may be more successful in translating my points to the authors.
  Thanks,
  
  IY
  
  Citation: https://doi.org/10.5194/egusphere-2024-2359-RC9
  - RC10: 'Reply on RC9', Anonymous Referee #2, 11 Sep 2024
    
    You are welcome. Here is my viewpoint. The temperature is just a representation of energy. The fundamental cause of temperature rise or fall is that there is an exchange of energy,no matter for ocean or atmosphere. The same is true for melting or freezing of sea ice. So I think authors should investigate and analyze the relationships of ocean-sea ice-atmosphere from the perspective of energy exchange.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC10
    
    RC11: 'Reply on RC10', Anonymous Referee #1, 11 Sep 2024
    
    Thank you, once again!
    Absolutely agree with your viewpoint - an actual process of energy exchange (specific for each case) needs to be studied to explain a change. As a matter of fact, I do exactly heat budget study to explain ocean heat content change and convection in the Labrador Sea. E.g., Figures 4 and 5 in "Intensification and shutdown of deep convection in the Labrador Sea were caused by changes in atmospheric and freshwater dynamics | Communications Earth & Environment (nature.com)"
    However, I did not go into detail of how I would perform a sea-ice budget study in my brief review (shortest I have ever written), because the authors took a totally different approach. It was tempting to share some ideas, of course, but they replied quite aggressively (no apologies have followed whatsoever) ... Comparing three results (SST, SAT and SIE) of the same global cause (heat exchange in the coupled ocean-atmosphere-ice system), is a completely different task, totally unrelated to what we just saw. The present approach, as said, is misleading, as it claims SST and SAT to be the primary causes of SIE change, while they are just members of the same strongly coupled bundle...
    Sincerely,
    
    IY
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC11
    
    RC12: 'Reply on RC10', Anonymous Referee #1, 11 Sep 2024
    
    Thank you, once again!
    Absolutely agree with your viewpoint - an actual process of energy exchange (specific for each case) needs to be studied to explain a change. As a matter of fact, I do exactly heat budget study to explain ocean heat content change and convection in the Labrador Sea. E.g., Figures 4 and 5 in "Intensification and shutdown of deep convection in the Labrador Sea were caused by changes in atmospheric and freshwater dynamics | Communications Earth & Environment (nature.com)"
    However, I did not go into detail of how I would perform a sea-ice budget study in my brief review (shortest I have ever written), because the authors took a totally different approach. It was tempting to share some ideas, of course, but they replied quite aggressively (no apologies have followed whatsoever) ... Comparing three results (SST, SAT and SIE) of the same global cause (heat exchange in the coupled ocean-atmosphere-ice system), is a completely different task, totally unrelated to what we just saw. The present approach, as said, is misleading, as it claims SST and SAT to be the primary causes of SIE change, while they are just members of the same strongly coupled bundle...
    Sincerely,
    
    IY
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC12

Status: closed

RC1:
'Comment on egusphere-2024-2359', Anonymous Referee #1, 08 Sep 2024

I gave Fair to 1-4, but recommended rejection, because there is a conceptual issue in the title, overall work arrangement and conclusion.
Considering that sea surface temperature (SST) of the whole Arctic directly depends on Arctic ice-covered area - whether partial open-ice area is used (meaning that the more Arctic covered by ice the lower temperature as a smaller area will form a larger relative fraction of near-ice zone and therefore will be colder, and vice versa) or if the ice-covered Arctic is used as well (meaning that it will naturally give lower SST), any statement that sea surface temperature is the reason for changes in Arctic ice extent can be misleading. I realize that, yes, warmer water facilitates ice melt, however, Pearson correlation analysis is not the way to analyze the actual impact and separate it from feedback.
There is nothing wrong in studying correlations, but there need to be some basic understanding of the processes projected on that study, and here I am not convinced if sea-ice extent changes made SST different or the other way around or both. So, I see no gain for the understanding of this complex coupling mechanism.
Sorry ...

Citation: https://doi.org/10.5194/egusphere-2024-2359-RC1
- CC1:
  'Reply on RC1', Qizhen Sun, 09 Sep 2024
  
  Your comment contains the following statement “any statement that sea surface temperature is the reason for changes in Arctic ice extent can be misleading”.Please provide the basis for this viewpoint.
  
  Citation: https://doi.org/10.5194/egusphere-2024-2359-CC1
  - RC2: 'Reply on CC1', Anonymous Referee #1, 09 Sep 2024
    
    ok, I will explain it in very easy terms, imagine ground partially covered by snow, you measure surface temperature of the whole area (or only of the snow-bare part, either way supports my point). Then the Sun comes out and melts snow, which retreats. The same Sun also heats the ground not even close to the snow age. You measure snow area and surface temperature again. Will they be correlated? I guess they will be correlated. Is it now correct to conclude that the warmer ground melted the snow? Or maybe, it is the retreat of the snow allowed more ground to be heated, raising its temperature. This is what I meant in the two statements: (1) the correlation of the two is expected, and (2) it is not obvious from simple correlation approach that warmer water (over the whole ocean, including the locations remote from ice) was the reason for ice melt. Yes, rapid warming by the ice edge accelerates ice melting, but this needs to be studied differently than suggested. Otherwise. refer to my analogy with snow on the ground ...
    Thanks!
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC2
    
    CC2: 'Reply on RC2', Di Chen, 09 Sep 2024
    
    This paper only focus on the the impact of sea surface and air temperature on Arctic sea ice. It doesn't deal with snow. The title is pretty clear.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-CC2
    
    RC4: 'Reply on CC2', Anonymous Referee #1, 09 Sep 2024
    
    It seems like, you do not understand what I am trying to explain - I explained my point to you using snow as analogy just to make the idea more transparent. If you replace snow with ice and ground with sea, the point will be the same. Since you did not understand my point that the correlation of the "geometrically" connected processes (less ice means higher SST without any look in processes, just because of areas change), I thought it might be easier to use the snow analogy.
    If you want an example of correlation meaning impact or effect, this would be winter convection in the ocean - ocean loses heat to atmosphere and convects, but the depth of its convection does not have a significant return effect on air. Do you see the difference?
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC4
    
    CC4: 'Reply on RC4', Qizhen Sun, 09 Sep 2024
    
    These explanations of yours are completely irrelevant to this investigation.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-CC4
    
    RC6: 'Reply on CC4', Anonymous Referee #1, 09 Sep 2024
    
    Another disrespectful statement from the author, without even trying to understand what the reviewer meant or proving the reviewer wrong.
    I strongly disagree, and repeat that the way the problem was approached gives a kind of outcome that was expected simply because the high correlation is a partly if not mainly a result of ice-covered water being colder than ice-free. I am just impressed by the amount of effort it takes to explain something obvious, and the author should be thankful that I ignore their disrespectful conduct.
    Thinking of the problem as a whole, one could use the whole basin SST (which includes sea ice), or just over the ice-free part. The first approach gives expected result as the mean SST itself is function of ice extent. The second approach has a major caveat - every time the region is different than in the other times, so the comparison is not of exact locations, and therefore the mean SST filed contributes there differently.
    The same problem could be approached differently, but that would be a different study, a different paper.
    The high correlation of 0.9 fully supports my point - it is so high because it is ice-to-icefree ratio is included in it.
    My apologies to the disrespectful authors that I could not explain it even simpler than that. Hopefully, the other reviewer will understand my points and do better job explaining it to the author.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC6
    
    CC3: 'Reply on RC2', Di Chen, 09 Sep 2024
    
    BTW, your example of snow and sea ice is completely different from your previous statement that “any statement that sea surface temperature is the reason for changes in Arctic ice extent can be misleading”. “ which are two completely different issues. You are completely misleading.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-CC3
    
    RC5: 'Reply on CC3', Anonymous Referee #1, 09 Sep 2024
    
    Read carefully what I just said about the snow example - it is just another case of correlation not meaning a direct impact or a direct effect. And I do not take an offense when you call me "misleading". You did not get my point, which I thought was obvious for any oceanographer, and so I went into trouble explaining my point to you as best as I could. You are trying to offend me instead ;).
    Once again, do a simple math with changing ice area and see who it impacts SST.
    It is a shame young generation do not have respect to those who are trying to help them.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC5
  - RC3: 'Reply on CC1', Anonymous Referee #1, 09 Sep 2024
    
    "The temporal coefficients of the Arctic SAT and SIC fields (Figure 4c) show an obvious inverse phase relationship (r = -0.90). This analysis shows that the SST has a greater impact on the SIC than the SAT."
    A high correlation of two coupled processes does not mean that one has a great impact - liquid water by all means is warmer than sea ice, and your conclusion is just a result of simple shift in the sea ice cover.
    Furthermore, SST and AT are correlated, and therefore, and statement about the relative role of AT over SST can be misleading as weel. If the two processes are correlated, it is not that easy to partition their impact on the third process. To explain it simpler - AT may not be a factor at all (although it is, but let's assume opposite), but it may be correlated with sea ice only because SST is correlated with the latter.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC3
    
    CC5: 'Reply on RC3', Timo Vihma, 09 Sep 2024
    
    Thank you for your valuable and insightful comments on the manuscript!
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-CC5
    
    RC8: 'Reply on CC5', Anonymous Referee #1, 09 Sep 2024
    
    Thank you, Timo,
    Hope it all makes sense!
    I have done lots of correlations to be able to tell where we can state that some effects are real and direct, and when the variables are just "forced" to be correlated. SST is a tricky thing - it is both result and factor at the same time.
    The case of this review was an exception of my rule of no anonymity when reviewing (the reason was that I did not go into a detailed review spotting an issue at start). I can share my name and email, if further explanations, clarifications and suggestions are needed.
    Thanks,
    
    IY
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC8
RC7:
'Comment on egusphere-2024-2359', Anonymous Referee #2, 09 Sep 2024
Few questions need to be addressed and answered by the author before making further decisions:
Why July-October were selected?

The SIC data used in this study was from Hadley Center. Does the pixels with value ＜15% need to be eliminated just like the SIC data retrieved from satellite remote sensing data.

When calculating SIC in different sea regions of the Arctic Ocean, there should be a geographical boundary, how to define it?

In figure 1, how was the multiyear ice edge (green line) identified? What the other colored lines represent?

Why only v-wind component was analyzed?

The results of statistical analysis can only show that there is a relationship between SST and SIC, but can not show that SST causes the change of SIC. The mechanism should be analyzed or studied. In addition, SST and SAT affect each other, how do you think about their relationship when doing dominance analysis.

Line 165-170: Those sentences describe the results of SVD statistics, which is OK. However, that are not enough to support the conclusion of the last sentence “This analysis shows that the SST has a greater impact on the SIC than the SAT”. Your study focuses on July to October, which is the polar day time in Arctic. Large amounts of solar shortwave radiation reach the earth surface. Open water regions absorb much more solar energy and get warmer because of lacking of sea ice coverage. So these regions get higher SST. Does the higher SST lead to low SIC? The higher SST may lead to the SIC decline in marginal sea ice zone, but what’s the influence extent of SST on SIC far from marginal sea ice zone? In figure 4a, another question comes out. Are there effective SST values under sea ice?

All figures in the manuscript have low resolution.

As stated in the line306-307 of the manuscript, more work is needed to better distinguish between statistical and causal relationships. So far, authors have done some statistical analysis. However, these are not sufficient to prove that SST and SAT have a significant effect on SIC.
Citation: https://doi.org/10.5194/egusphere-2024-2359-RC7
- RC9:
  'Reply on RC7', Anonymous Referee #1, 09 Sep 2024
  
  "The results of statistical analysis can only show that there is a relationship between SST and SIC, but can not show that SST causes the change of SIC. The mechanism should be analyzed or studied (1). In addition, SST and SAT affect each other, how do you think about their relationship when doing dominance analysis (2)."
  These are exactly the two points I made (thanks!!!!), which the authors claimed to be irrelevant, and my viewpoint misleading, but I red-flagged as critical for the whole story.
  In my further explanations I pointed that any ocean-wide or ice-free-wide SST value is directly affected by sea ice. SST and SAT are indeed dependent, and if someone goes to process study, SST over the whole Arctic may not be a direct precursor of the ice extent change as the actual interaction happened in a certain zone.
  Apologies for writing to you, but expressed my hope last night that the other reviewer may be more successful in translating my points to the authors.
  Thanks,
  
  IY
  
  Citation: https://doi.org/10.5194/egusphere-2024-2359-RC9
  - RC10: 'Reply on RC9', Anonymous Referee #2, 11 Sep 2024
    
    You are welcome. Here is my viewpoint. The temperature is just a representation of energy. The fundamental cause of temperature rise or fall is that there is an exchange of energy,no matter for ocean or atmosphere. The same is true for melting or freezing of sea ice. So I think authors should investigate and analyze the relationships of ocean-sea ice-atmosphere from the perspective of energy exchange.
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC10
    
    RC11: 'Reply on RC10', Anonymous Referee #1, 11 Sep 2024
    
    Thank you, once again!
    Absolutely agree with your viewpoint - an actual process of energy exchange (specific for each case) needs to be studied to explain a change. As a matter of fact, I do exactly heat budget study to explain ocean heat content change and convection in the Labrador Sea. E.g., Figures 4 and 5 in "Intensification and shutdown of deep convection in the Labrador Sea were caused by changes in atmospheric and freshwater dynamics | Communications Earth & Environment (nature.com)"
    However, I did not go into detail of how I would perform a sea-ice budget study in my brief review (shortest I have ever written), because the authors took a totally different approach. It was tempting to share some ideas, of course, but they replied quite aggressively (no apologies have followed whatsoever) ... Comparing three results (SST, SAT and SIE) of the same global cause (heat exchange in the coupled ocean-atmosphere-ice system), is a completely different task, totally unrelated to what we just saw. The present approach, as said, is misleading, as it claims SST and SAT to be the primary causes of SIE change, while they are just members of the same strongly coupled bundle...
    Sincerely,
    
    IY
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC11
    
    RC12: 'Reply on RC10', Anonymous Referee #1, 11 Sep 2024
    
    Thank you, once again!
    Absolutely agree with your viewpoint - an actual process of energy exchange (specific for each case) needs to be studied to explain a change. As a matter of fact, I do exactly heat budget study to explain ocean heat content change and convection in the Labrador Sea. E.g., Figures 4 and 5 in "Intensification and shutdown of deep convection in the Labrador Sea were caused by changes in atmospheric and freshwater dynamics | Communications Earth & Environment (nature.com)"
    However, I did not go into detail of how I would perform a sea-ice budget study in my brief review (shortest I have ever written), because the authors took a totally different approach. It was tempting to share some ideas, of course, but they replied quite aggressively (no apologies have followed whatsoever) ... Comparing three results (SST, SAT and SIE) of the same global cause (heat exchange in the coupled ocean-atmosphere-ice system), is a completely different task, totally unrelated to what we just saw. The present approach, as said, is misleading, as it claims SST and SAT to be the primary causes of SIE change, while they are just members of the same strongly coupled bundle...
    Sincerely,
    
    IY
    
    Citation: https://doi.org/10.5194/egusphere-2024-2359-RC12

Di Chen, Qizhen Sun, and Timo Vihma

Editorial note: the affiliation of the first author could not be verified and has been changed to 'independent researcher'.

Di Chen, Qizhen Sun, and Timo Vihma

Editorial note: the affiliation of the first author could not be verified and has been changed to 'independent researcher'.

Viewed

Total article views: 499 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
326	76	97	499	3	4

HTML: 326
PDF: 76
XML: 97
Total: 499
BibTeX: 3
EndNote: 4

Views and downloads (calculated since 12 Aug 2024)

Month	HTML	PDF	XML	Total
Aug 2024	114	29	7	150
Sep 2024	181	40	89	310
Oct 2024	29	7	1	37
Nov 2024	2	0	2

Cumulative views and downloads (calculated since 12 Aug 2024)

Month	HTML	PDF	XML	Total
Aug 2024	114	29	7	150
Sep 2024	181	40	89	310
Oct 2024	29	7	1	37
Nov 2024	2	0	2

Viewed (geographical distribution)

Total article views: 499 (including HTML, PDF, and XML) Thereof 499 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 07 Nov 2024

Download

Please read the editorial note first before accessing the preprint.

Preprint (3783 KB)
Metadata XML

Short summary

We investigates the variations and trends in Arctic sea ice during summer and autumn, focusing on the impacts of sea surface temperature (SST) and surface air temperature (SAT). Both SST and SAT significantly influence Arctic sea ice concentration. SST affects both interannual variations and decadal trends, while SAT primarily influences interannual variations. Additionally, SAT's impact on sea ice concentration leads by seven months, due to a stronger warming trend in winter than in summer.


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