the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 27 Jul 2023
Atmospheric teleconnections between the Arctic and the Baltic Sea region as simulated by CESM1-LE
Abstract. This paper examines teleconnections between the Arctic and the Baltic Sea region and is based on two cases of CESM-LE climate model simulations': the stationary case with pre-industrial radiative forcing and the climate change case with measured and RCP8.5 radiative forcing.
Stationary control simulation 1800-year long time-series were used for stationary teleconnection and 40-member ensemble from the period 1920–2100 for teleconnections during ongoing climate change. We analyzed seasonal temperature at a 2-meter level, sea-level pressure, sea ice concentration, precipitations, geopotential height and 10-meter level wind speed. The Arctic was divided into seven areas.
The Baltic Sea region climate has strong teleconnections with the Arctic climate; the strongest connections are with Svalbard and Greenland region. There is high seasonality in the teleconnections, with the strongest correlations in winter and the lowest correlations in summer, when the local factors are stronger. The majority of teleconnections in winter and spring can be explained by climate indexes NAO and AO. During ongoing climate change, the teleconnection patterns did not show remarkable developments by the end of the 21st century. Minor pattern changes are between the Baltic Sea region temperature and the sea ice concentration.
To estimate different Arctic regions' collective statistical connections with the Baltic Sea region, we calculated the correlation between the parameter and its Ridge regression estimation. Seasonal coefficient of determination, R2, were highest for winter: for temperature R2 = 0.64, for surface pressure R2 = 0.44 and for precipitation R2 = 0.35. When doing the same for the seasons' previous month values in the Arctic, the relations are considerably weaker with the highest R2 = 0.09 for temperature in the spring. Hence, the forecasting capacity of Arctic climate data for the Baltic Sea region is weak.
Although there are statistically significant teleconnections between the Arctic and Baltic Sea region, the Arctic impacts are regional and mostly connected with climate indexes. There are no simple cause-and-effect pathways. By the end of the 21st century, the Arctic ice concentration has significantly decreased. Still, the general teleconnections pattern between the Arctic and the Baltic Sea region will not change considerably by the end of the 21st century.
-
Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
-
Preprint
(1713 KB)
-
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1713 KB) - Metadata XML
- BibTeX
- EndNote
- Final revised paper
Journal article(s) based on this preprint
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1637', Anonymous Referee #1, 10 Sep 2023
The manuscript “Atmospheric teleconnections between the Arctic and the Baltic Sea region as simulated by CESM1-LE” by Jakobson and Jakobson examines North Atlantic sea ice teleconnections to the Baltic Sea through the large-scale atmospheric pathways described by the Arctic Oscillation, North Atlantic Oscillation, and Barents Oscillation. The authors examine stationary (pre-industrial) versus modern climate forcing (RCP8.5 scenario) in their model simulations in an attempt to understanding natural versus anthropogenic forcing on these Arctic-Baltic physical connections.
Key findings are that Svalbard and Greenland Sea regions two-meter air temperature and surface pressure exhibit the strongest correlative relationships with the Baltic region climate, namely in winter due to stronger forcing by the modes of variability (NAO and AO) relative to the summer season. Under continued climate change from greenhouse gas emissions, the authors note that the end-of-century projections suggest these Arctic-Baltic relationships will remain rather consistent through time despite continued Arctic warming and sea ice loss.
The topic of Arctic change linkages with northern European climate is one that continues to receive much attention. That said, the authors could do a better job of reviewing this work to date, and bringing attention to what value is added by their new analysis and modeling approach. The methods, especially the ridge regression approach, could be more clearly defined and the results could be much more clearly stated. Moreover, the paper is difficult to follow due to numerous grammatical errors and redundancies in uncommon acronyms. The paper may benefit from English editing services. Several remarks along these lines are made by manuscript line number (L) in the comments that follow.
Major comments:
L105: Are the correlative results sensitive to changes in the Baltic and Arctic marginal seas domains? Did you test for this? Why were these geographic areas selected? More details are needed to provide some context for results. Citing previous literature that has used these or similar domains may help in this regard.
L155: While the correlative approaches are defined in Section 2, the Ridge Regression approach is not clearly described. What does it entail and why is it used? Readers will generally be familiar with correlation techniques, but less so with this specific regression method. It should be described in detail, including with justification for why it was selected in lieu of a simple parametric test (i.e., linear regression) as are used in the correlation analyses.
Minor comments and corrections:
L8: Suggest removing “measured and”
L15: What is meant by “local factors”? Please be more specific.
L16: Suggest revise to “NAO and AO climate indices”
L48: By “permanent” do you mean “seasonal” or “ephemeral” snow cover? Please clarify.
L53: Suggest substituting “knowing” for “studying” or similar word choice
L75: reanalyzes reanalyses
L76: Suggest substituting “search” for “examine” or similar word choice
L83: Suggest substituting “completes” for “concludes” or similar word choice
L98: Once these climate variables are defined, they do not need to redefined (e.g., L143, L192-193, etc) through the paper. Also, suggest using typical acronyms such as SLP for sea-level pressure and SIC for sea-ice concentration as they will make it easier for readers to follow results.
L102: Are the NAO, BO, and AO definitions adopted or adapted from previous studies? If so, the studies should be cited. If not, then some explanation should be used for modifying data and domains used to define these indices.
L107: Change “above” to “north of”
L126: Change “supposedly not” to “less”
L137-138: Please reword this sentence as the second half of it is confusing.
L161: Add “and” before “seasonally”
L178: “then 0.82 = 64%” – it is very difficult to follow what is meant here. Are you referencing squared correlations initially then their explained variance? Please clarify.
L199: “positive correlation” involving what? More specifics are needed to make results easier to follow.
L209: “earlier average month is confusing as worded” – please revise.
L224: What is meant by “self-consistent”
L225: r=0.046 is a pretty weak threshold for a physical relationship given that a random relationship could arise ~5% of the time. Why mention this threshold?
L228-231: Should some sort of teleconnection hypothesis be revisited in the introduction then touched upon here? This seems like a strange place to comment that the present study does not confirm the long-proposed linkage between sea ice around Greenland and European climate.
L252: This sentence is confusing, please reword.
L275: Change to “will significantly decrease”
L278: T2m and SLP from what Arctic region are best connected with Baltic climate?
Citation: https://doi.org/10.5194/egusphere-2023-1637-RC1 - AC1: 'Reply on RC1', Erko Jakobson, 13 Oct 2023
-
RC2: 'Comment on egusphere-2023-1637', Anonymous Referee #2, 12 Sep 2023
This manuscript investigates teleconnections between the Arctic and Baltic sea in both pre-industrial and climate change model simulations from the CESM-LE models. The authors find that the strongest correlations with the Baltic Sea are from Svalbard and Greenland particularly in boreal winter and that most of these correlations can be explained by NAO and AO. The authors also find that with continued climate change, these teleconnections do remain consistent to the end of the century.
I found this paper quite difficult to follow as I was unsure of many of the methods and therefore the results made little sense. I felt that most things were not clearly explained or explained using non-scientific language and the results were not very robust thus making it hard to understand the relevance and novelty of this study. I have listed both the major concerns I have with this manuscript and more detailed/minor comments below that.
Major comments
I have a few issues with the methods used:
- What is the ridge regression? There is no description as to what this actually does and so is difficult to understand what this means in the results and what especially when its described to be able to explain cause of winter variability in Baltic Sea.
- For the Arctic area that you assessed, this reaches to as far as 50N. Most studies start the Arctic from 60 or 67.5N. I think that this is especially important here because you are doing correlations of the Arctic from 50-90N to the Baltic Sea which is from 50-65N so essentially you correlate the Baltic sea area with itself when you are doing a whole Arctic correlation and so will therefore get high correlations in this region. I think it would be best to stick to an Arctic area that is commonly used in other studies.
Due to the issues with the methods used I also have a few concerns regarding the results presented:
- The description of the results were very confusing. Line 162 you say that the correlations with Svalbard T2M and ICEC are weaker, is this in the table because this is showing the correlation between TA T2M, PSL and PREC and, the IA T2M, PSL and ICEC so I am not sure how you are showing the correlation between Svalbard T2M and ICEC.
- Figure 3 is never referenced in the manuscript and although I assume that it refers to section 3.2 it but this is not made obvious. Within this section, from assessing figure 3, if this is the correct figure, I would say that there is only a slight weakening in the winter North Pole ICEC and TA T2m correlation.
- Why did you corelate seasons to previous months particularly when in the lines 220-222 you state that a monthly correlation will have a weaker influence on the next seasonal average, why not then do solely seasonal lagged correlations or monthly lagged correlations?
- Finally you mention that you analyse Z500 and U10 and yet this is not shown anywhere in the manuscript. I advise either add figures to show this or remove the reference to analysing this completely.
Minor/detailed comments
Line 17: Unsure of what you mean about how teleconnection patterns did not show remarkable developments with ongoing climate change, particularly the phrase ‘remarkable developments’
Line 36: There have been many studies that have discussed Arctic amplification and the causes behind it that should be referenced here.
Line 39: What is meant by; ‘AA is expected to be related to further changes that affect mid-latitudes’
Line 54: ‘Possibility to glance in to the future’ this is not accurate (or scientific enough) as to what scientists do when assessing the impact of Arctic on mid-latitudes. We assess the impact of Arctic on mid-latitudes to improve accuracy for forecasting or climate projections, we cannot glance in to the future but give a scenario based prediction of what may occur.
Lines 54- 62: I do not understand the relevance of this section. I suggest either removing it or being explicit about why the reader needs to know about the pace of climate change.
Line 66: This sounds like you have multiple models but with the CESM-LE you have one model that has multiple members.
Line 101: Can you give references to the classification of the AO, NAO and in particular the Barents Oscillation as it is less well known that the other two.
Line 104: Can you be more explicit here in your methods as to how the ‘correlations with and without the effect of teleconnection indices’ were analysed
Line 109: ‘were clearly weaker than with the remaining regions’ feels like there is a word(s) missing here.
Line 123: What do you mean by teleconnection transformations?
Line 126: ‘weaker correlations are supposedly not important’ – yes stronger correlations are more important
Line 137: This line is very confusing, please edit.
Line 147: Is row 1 or 2 the one that is not controlled by NAO and what is the sign of the strong teleconnection in the Atlantic to the east of Iceland, in my view the strongest correlations are off the coast of North America.
Table 2: Why did you separate East and West Greenland
Lines 181-185: I am not sure which figure you are referencing here, can you add this.
Line 202: The correlation becomes negligible off the coast of Siberia rather than fades away and Barents sea is not off the coast of Siberia but Russia. Also a positive correlation is found in the East Siberian Sea in 2080-2100 showing that not all correlation fades away.
Line 217-218: What rate are you talking about here and what is meant by incoming parameters?
Line 218-219: Revise this line starting ‘It turned out that quite common…’ as I am not sure what is meant in particular what is quite common?
Line 224: What is meant by CESM-LE self-consistent database?
Line 225: These correlations are very weak, plus this value differs from the one stated in the methods, are these values supposed to be different?
Line 230: Are there no more up to date papers than Hildebrandsson which is 109 years old?
Line 236: What differences in the model parameters and different periods from CESM-LE complicated a comparison with the Jacokson et al. 2017 paper? A lot of studies compare reanalysis and historical CESM simulations.
Line 237: The comparable T2m – comparable to what?
Line 253: What is natural SAT variability?
Line 260. This sentence is confusing, please reword.
Line 262: I am not sure what you mean here, can you reword.
Line 273: what local factors are you talking about?
Citation: https://doi.org/10.5194/egusphere-2023-1637-RC2 - AC2: 'Reply on RC2', Erko Jakobson, 13 Oct 2023
-
RC3: 'Comment on egusphere-2023-1637', Anonymous Referee #3, 14 Sep 2023
This submission follows up some earlier work by these authors which made use of reanalysis data (namely NCEP and ERA-I) for the period 1979-2015 to explore teleconnections between the Arctic and the Baltic. This present work attacks a similar issue using model data. The submission has the potential to make a significant contribution to the literature, but it is not quite there yet. Before I would be able to recommend acceptance, there are a number of issues which need to be addressed.
Lines 29-30: Also here to cite the recent analysis of
Mika Rantanen, Alexey Yu Karpechko, Antti Lipponen, Kalle Nordling, Otto Hyvärinen, Kimmo Ruosteenoja, Timo Vihma and Ari Laaksonen, 2022: The Arctic has warmed nearly four times faster than the globe since 1979. Communications Earth & Environment, 3, 168, doi: 10.1038/s43247-022-00498-3.
Line 30: What ‘average’ is referred to here?
Line 38: ‘Screens’ should be ‘Screen’
Line 42: Question raised here is to which specific region of the Arctic is of importance. Add to this reference to Wenqin Zhuo & co-authors, 2023: The key atmospheric drivers linking regional Arctic amplification with East Asian cold extremes. Atmospheric Research, 283, 106557, doi: 10.1016/j.atmosres.2022.106557.
Lines 44-45: Valuable here to cite the more recent analyses of
Overland, J. E., Ballinger, T. J., Cohen, J., Francis, J. A., Hanna, E., Jaiser, R., Kim, B.-M., Kim, S.-J., Ukita, J., Vihma, T., Wang, M. and Zhang, X. 2021. 'How do intermittency and simultaneous processes obfuscate the Arctic influence on midlatitude winter extreme weather events?', Env. Res. Lett. 16, 043002, doi: 10.1088/1748-9326/abdb5d,
Luo, D., X. Chen, J. Overland, et al., 2019: Weakened potential vorticity barrier linked to recent winter Arctic sea ice loss and midlatitude cold extremes. J. Climate, 32, 4235-4261, doi: 10.1175/JCLI-D-18-0449.1, and
Rudeva, & coauthors, 2021: Midlatitude winter extreme temperature events and connections with anomalies in the Arctic and tropics. J. Climate, 34, 3733-3749, doi: 10.1175/JCLI-D-20-0371.1.
Lines 79-80, …: The paper makes frequent allusions to ‘forecasting’, but it is not always clear what timescale is meant. Please make these parts more specific. One could argue that anything longer than 2-3 weeks is really an ‘outlook’.
Lines 121-129: The value of very long simulations as used here is that it is ‘easier’ to achieve statistical significance. However, such long integrations may not be able to reveal PHYSICAL significance. The threshold correlations here are very small and explain less than 1% of the variance. Strongly suggest the authors add some remarks to point out these issues.
Also, the time series considered here will possess considerable autocorrelation. This, in turn, with reduce the ‘effective’ degrees of freedom with which the tests for significance are conducted. Was allowance made for this effect (see, e.g., Bretherton C S et al 1999 The effective number of spatial degrees of freedom of a time-varying field J. Climate 12 1990-2009).
Lines 132-134: Not all readers will be familiar with Ridge regression. Helpful here to reference the recent (and accessible) monograph of Saleh, A. K. Md. Ehsanes; Arashi, Mohammad; Kibria, B. M. Golam (2019). Theory of Ridge Regression Estimation with Applications. New York: John Wiley & Sons. ISBN 978-1-118-64461-4.
Lines 146-150: The demonstrated link here to the NAO is interesting. A little more physics-based discussion is required on this. Make reference here to the investigation of
Luo, D., Y. Xiao, …, 2016: Impact of Ural Blocking on winter Warm Arctic–Cold Eurasian anomalies. Part II: The link to the North Atlantic Oscillation. J. Climate, 29, 3949-3971, doi: 10.1175/JCLI-D-15-0612.1.
That paper also implies a role of the Ural Mountains, and the atmospheric blocks situated over them. The Urals are only a short distance ‘downstream’ of the Baltic and would be expected to influence this local region. The paper would greatly benefit from some thoughts on this aspect, and of changes and variability in blocking. Beneficial in this to cite analysis of
Luo, Dehai & coauthors, 2017: Increased quasi-stationarity and persistence of winter Ural Blocking and Eurasian extreme cold events in response to Arctic warming. Part II: A theoretical explanation. J. Climate, 30, 3569–3587, doi: 10.1175/JCLI-D-16-0262.1.
Lines 181-188: The ‘local’ correlations are similar to what one would expect. It is not clear to me that they are ‘necessary to better understand the teleconnections …’ Please clarify.
Lines 193-194: See my earlier point on sample size and ‘statistically reliable results’. May wish to reword this.
Lines 195-206: I have trouble interpreting these results, as it is not made clear what the sea ice is doing in these simulations. The integrations to 2100 are performed with RCP8.5 forcing and the reader is entitled to some (limited) information as to how the Arctic ice is changing over the period. As the changes will almost certainly be large the correlations in the last epochs of this century will essentially refer to physical associations which are different from those in the earlier part of the 21st century. The interpretation here needs much more thought.
Of relevance to these considerations is the recent analysis of
Yeon-Hee Kim, Seung-Ki Min, Nathan P. Gillett, Dirk Notz and Elizaveta Malinina, 2023: Observationally-constrained projections of an ice-free Arctic even under a low emission scenario. Nature Communications, 14, 3139, doi: 10.1038/s41467-023-38511-8.
Line 216: Insert ‘of the variance’ after ‘10%’.
Line 228-233: An interesting point is made here in connection with the potential influence from the Greenland Sea. It would be worth mentioning in the text that Zhuo, Yao et al., 2023: The key atmospheric drivers linking regional Arctic amplification with East Asian cold extremes. Atmospheric Research, 283, 106557 found sea ice cover this Greenland region to be one of the most influential of all Arctic regions on teleconnections to, and conditions in, the Baltic.
Lines 243-245: What are the physics behind this correlation. Is it ‘on shore’ flow, Ekman effect, …?
Also here may be informative to reference the recent work of
Vavrus, S. J., and R. Alkama, 2022: Future trends of arctic surface wind speeds and their relationship with sea ice in CMIP5 climate model simulations. Climate Dyn., 59, 1833-1848, doi: 10.1007/s00382-021-06071-6.
Lines 263-265: Please to note that the year of Lantao Sun’s paper is ‘2018’, not ‘2016’. The reason(s) for this disagreement must be canvassed. Sun t al. used the GFDL model (with RCP8.5). Related to my earlier point part of this difference may be due to how the sea ice evolves over the century. A more critical examination is required here.
Line 411-416: These two reference are the same, except the dates are different. I strongly suspect that the authors meant the first of these to be
Lantao Sun, Judith Perlwitz and Martin Hoerling, 2016: What caused the recent "Warm Arctic, Cold Continents" trend pattern in winter temperatures? Geophysical Research Letters, 43, 5345-5352, doi: 10.1002/2016gl069024
But please check carefully.
Citation: https://doi.org/10.5194/egusphere-2023-1637-RC3 - AC3: 'Reply on RC3', Erko Jakobson, 13 Oct 2023
Interactive discussion
Status: closed
-
RC1: 'Comment on egusphere-2023-1637', Anonymous Referee #1, 10 Sep 2023
The manuscript “Atmospheric teleconnections between the Arctic and the Baltic Sea region as simulated by CESM1-LE” by Jakobson and Jakobson examines North Atlantic sea ice teleconnections to the Baltic Sea through the large-scale atmospheric pathways described by the Arctic Oscillation, North Atlantic Oscillation, and Barents Oscillation. The authors examine stationary (pre-industrial) versus modern climate forcing (RCP8.5 scenario) in their model simulations in an attempt to understanding natural versus anthropogenic forcing on these Arctic-Baltic physical connections.
Key findings are that Svalbard and Greenland Sea regions two-meter air temperature and surface pressure exhibit the strongest correlative relationships with the Baltic region climate, namely in winter due to stronger forcing by the modes of variability (NAO and AO) relative to the summer season. Under continued climate change from greenhouse gas emissions, the authors note that the end-of-century projections suggest these Arctic-Baltic relationships will remain rather consistent through time despite continued Arctic warming and sea ice loss.
The topic of Arctic change linkages with northern European climate is one that continues to receive much attention. That said, the authors could do a better job of reviewing this work to date, and bringing attention to what value is added by their new analysis and modeling approach. The methods, especially the ridge regression approach, could be more clearly defined and the results could be much more clearly stated. Moreover, the paper is difficult to follow due to numerous grammatical errors and redundancies in uncommon acronyms. The paper may benefit from English editing services. Several remarks along these lines are made by manuscript line number (L) in the comments that follow.
Major comments:
L105: Are the correlative results sensitive to changes in the Baltic and Arctic marginal seas domains? Did you test for this? Why were these geographic areas selected? More details are needed to provide some context for results. Citing previous literature that has used these or similar domains may help in this regard.
L155: While the correlative approaches are defined in Section 2, the Ridge Regression approach is not clearly described. What does it entail and why is it used? Readers will generally be familiar with correlation techniques, but less so with this specific regression method. It should be described in detail, including with justification for why it was selected in lieu of a simple parametric test (i.e., linear regression) as are used in the correlation analyses.
Minor comments and corrections:
L8: Suggest removing “measured and”
L15: What is meant by “local factors”? Please be more specific.
L16: Suggest revise to “NAO and AO climate indices”
L48: By “permanent” do you mean “seasonal” or “ephemeral” snow cover? Please clarify.
L53: Suggest substituting “knowing” for “studying” or similar word choice
L75: reanalyzes reanalyses
L76: Suggest substituting “search” for “examine” or similar word choice
L83: Suggest substituting “completes” for “concludes” or similar word choice
L98: Once these climate variables are defined, they do not need to redefined (e.g., L143, L192-193, etc) through the paper. Also, suggest using typical acronyms such as SLP for sea-level pressure and SIC for sea-ice concentration as they will make it easier for readers to follow results.
L102: Are the NAO, BO, and AO definitions adopted or adapted from previous studies? If so, the studies should be cited. If not, then some explanation should be used for modifying data and domains used to define these indices.
L107: Change “above” to “north of”
L126: Change “supposedly not” to “less”
L137-138: Please reword this sentence as the second half of it is confusing.
L161: Add “and” before “seasonally”
L178: “then 0.82 = 64%” – it is very difficult to follow what is meant here. Are you referencing squared correlations initially then their explained variance? Please clarify.
L199: “positive correlation” involving what? More specifics are needed to make results easier to follow.
L209: “earlier average month is confusing as worded” – please revise.
L224: What is meant by “self-consistent”
L225: r=0.046 is a pretty weak threshold for a physical relationship given that a random relationship could arise ~5% of the time. Why mention this threshold?
L228-231: Should some sort of teleconnection hypothesis be revisited in the introduction then touched upon here? This seems like a strange place to comment that the present study does not confirm the long-proposed linkage between sea ice around Greenland and European climate.
L252: This sentence is confusing, please reword.
L275: Change to “will significantly decrease”
L278: T2m and SLP from what Arctic region are best connected with Baltic climate?
Citation: https://doi.org/10.5194/egusphere-2023-1637-RC1 - AC1: 'Reply on RC1', Erko Jakobson, 13 Oct 2023
-
RC2: 'Comment on egusphere-2023-1637', Anonymous Referee #2, 12 Sep 2023
This manuscript investigates teleconnections between the Arctic and Baltic sea in both pre-industrial and climate change model simulations from the CESM-LE models. The authors find that the strongest correlations with the Baltic Sea are from Svalbard and Greenland particularly in boreal winter and that most of these correlations can be explained by NAO and AO. The authors also find that with continued climate change, these teleconnections do remain consistent to the end of the century.
I found this paper quite difficult to follow as I was unsure of many of the methods and therefore the results made little sense. I felt that most things were not clearly explained or explained using non-scientific language and the results were not very robust thus making it hard to understand the relevance and novelty of this study. I have listed both the major concerns I have with this manuscript and more detailed/minor comments below that.
Major comments
I have a few issues with the methods used:
- What is the ridge regression? There is no description as to what this actually does and so is difficult to understand what this means in the results and what especially when its described to be able to explain cause of winter variability in Baltic Sea.
- For the Arctic area that you assessed, this reaches to as far as 50N. Most studies start the Arctic from 60 or 67.5N. I think that this is especially important here because you are doing correlations of the Arctic from 50-90N to the Baltic Sea which is from 50-65N so essentially you correlate the Baltic sea area with itself when you are doing a whole Arctic correlation and so will therefore get high correlations in this region. I think it would be best to stick to an Arctic area that is commonly used in other studies.
Due to the issues with the methods used I also have a few concerns regarding the results presented:
- The description of the results were very confusing. Line 162 you say that the correlations with Svalbard T2M and ICEC are weaker, is this in the table because this is showing the correlation between TA T2M, PSL and PREC and, the IA T2M, PSL and ICEC so I am not sure how you are showing the correlation between Svalbard T2M and ICEC.
- Figure 3 is never referenced in the manuscript and although I assume that it refers to section 3.2 it but this is not made obvious. Within this section, from assessing figure 3, if this is the correct figure, I would say that there is only a slight weakening in the winter North Pole ICEC and TA T2m correlation.
- Why did you corelate seasons to previous months particularly when in the lines 220-222 you state that a monthly correlation will have a weaker influence on the next seasonal average, why not then do solely seasonal lagged correlations or monthly lagged correlations?
- Finally you mention that you analyse Z500 and U10 and yet this is not shown anywhere in the manuscript. I advise either add figures to show this or remove the reference to analysing this completely.
Minor/detailed comments
Line 17: Unsure of what you mean about how teleconnection patterns did not show remarkable developments with ongoing climate change, particularly the phrase ‘remarkable developments’
Line 36: There have been many studies that have discussed Arctic amplification and the causes behind it that should be referenced here.
Line 39: What is meant by; ‘AA is expected to be related to further changes that affect mid-latitudes’
Line 54: ‘Possibility to glance in to the future’ this is not accurate (or scientific enough) as to what scientists do when assessing the impact of Arctic on mid-latitudes. We assess the impact of Arctic on mid-latitudes to improve accuracy for forecasting or climate projections, we cannot glance in to the future but give a scenario based prediction of what may occur.
Lines 54- 62: I do not understand the relevance of this section. I suggest either removing it or being explicit about why the reader needs to know about the pace of climate change.
Line 66: This sounds like you have multiple models but with the CESM-LE you have one model that has multiple members.
Line 101: Can you give references to the classification of the AO, NAO and in particular the Barents Oscillation as it is less well known that the other two.
Line 104: Can you be more explicit here in your methods as to how the ‘correlations with and without the effect of teleconnection indices’ were analysed
Line 109: ‘were clearly weaker than with the remaining regions’ feels like there is a word(s) missing here.
Line 123: What do you mean by teleconnection transformations?
Line 126: ‘weaker correlations are supposedly not important’ – yes stronger correlations are more important
Line 137: This line is very confusing, please edit.
Line 147: Is row 1 or 2 the one that is not controlled by NAO and what is the sign of the strong teleconnection in the Atlantic to the east of Iceland, in my view the strongest correlations are off the coast of North America.
Table 2: Why did you separate East and West Greenland
Lines 181-185: I am not sure which figure you are referencing here, can you add this.
Line 202: The correlation becomes negligible off the coast of Siberia rather than fades away and Barents sea is not off the coast of Siberia but Russia. Also a positive correlation is found in the East Siberian Sea in 2080-2100 showing that not all correlation fades away.
Line 217-218: What rate are you talking about here and what is meant by incoming parameters?
Line 218-219: Revise this line starting ‘It turned out that quite common…’ as I am not sure what is meant in particular what is quite common?
Line 224: What is meant by CESM-LE self-consistent database?
Line 225: These correlations are very weak, plus this value differs from the one stated in the methods, are these values supposed to be different?
Line 230: Are there no more up to date papers than Hildebrandsson which is 109 years old?
Line 236: What differences in the model parameters and different periods from CESM-LE complicated a comparison with the Jacokson et al. 2017 paper? A lot of studies compare reanalysis and historical CESM simulations.
Line 237: The comparable T2m – comparable to what?
Line 253: What is natural SAT variability?
Line 260. This sentence is confusing, please reword.
Line 262: I am not sure what you mean here, can you reword.
Line 273: what local factors are you talking about?
Citation: https://doi.org/10.5194/egusphere-2023-1637-RC2 - AC2: 'Reply on RC2', Erko Jakobson, 13 Oct 2023
-
RC3: 'Comment on egusphere-2023-1637', Anonymous Referee #3, 14 Sep 2023
This submission follows up some earlier work by these authors which made use of reanalysis data (namely NCEP and ERA-I) for the period 1979-2015 to explore teleconnections between the Arctic and the Baltic. This present work attacks a similar issue using model data. The submission has the potential to make a significant contribution to the literature, but it is not quite there yet. Before I would be able to recommend acceptance, there are a number of issues which need to be addressed.
Lines 29-30: Also here to cite the recent analysis of
Mika Rantanen, Alexey Yu Karpechko, Antti Lipponen, Kalle Nordling, Otto Hyvärinen, Kimmo Ruosteenoja, Timo Vihma and Ari Laaksonen, 2022: The Arctic has warmed nearly four times faster than the globe since 1979. Communications Earth & Environment, 3, 168, doi: 10.1038/s43247-022-00498-3.
Line 30: What ‘average’ is referred to here?
Line 38: ‘Screens’ should be ‘Screen’
Line 42: Question raised here is to which specific region of the Arctic is of importance. Add to this reference to Wenqin Zhuo & co-authors, 2023: The key atmospheric drivers linking regional Arctic amplification with East Asian cold extremes. Atmospheric Research, 283, 106557, doi: 10.1016/j.atmosres.2022.106557.
Lines 44-45: Valuable here to cite the more recent analyses of
Overland, J. E., Ballinger, T. J., Cohen, J., Francis, J. A., Hanna, E., Jaiser, R., Kim, B.-M., Kim, S.-J., Ukita, J., Vihma, T., Wang, M. and Zhang, X. 2021. 'How do intermittency and simultaneous processes obfuscate the Arctic influence on midlatitude winter extreme weather events?', Env. Res. Lett. 16, 043002, doi: 10.1088/1748-9326/abdb5d,
Luo, D., X. Chen, J. Overland, et al., 2019: Weakened potential vorticity barrier linked to recent winter Arctic sea ice loss and midlatitude cold extremes. J. Climate, 32, 4235-4261, doi: 10.1175/JCLI-D-18-0449.1, and
Rudeva, & coauthors, 2021: Midlatitude winter extreme temperature events and connections with anomalies in the Arctic and tropics. J. Climate, 34, 3733-3749, doi: 10.1175/JCLI-D-20-0371.1.
Lines 79-80, …: The paper makes frequent allusions to ‘forecasting’, but it is not always clear what timescale is meant. Please make these parts more specific. One could argue that anything longer than 2-3 weeks is really an ‘outlook’.
Lines 121-129: The value of very long simulations as used here is that it is ‘easier’ to achieve statistical significance. However, such long integrations may not be able to reveal PHYSICAL significance. The threshold correlations here are very small and explain less than 1% of the variance. Strongly suggest the authors add some remarks to point out these issues.
Also, the time series considered here will possess considerable autocorrelation. This, in turn, with reduce the ‘effective’ degrees of freedom with which the tests for significance are conducted. Was allowance made for this effect (see, e.g., Bretherton C S et al 1999 The effective number of spatial degrees of freedom of a time-varying field J. Climate 12 1990-2009).
Lines 132-134: Not all readers will be familiar with Ridge regression. Helpful here to reference the recent (and accessible) monograph of Saleh, A. K. Md. Ehsanes; Arashi, Mohammad; Kibria, B. M. Golam (2019). Theory of Ridge Regression Estimation with Applications. New York: John Wiley & Sons. ISBN 978-1-118-64461-4.
Lines 146-150: The demonstrated link here to the NAO is interesting. A little more physics-based discussion is required on this. Make reference here to the investigation of
Luo, D., Y. Xiao, …, 2016: Impact of Ural Blocking on winter Warm Arctic–Cold Eurasian anomalies. Part II: The link to the North Atlantic Oscillation. J. Climate, 29, 3949-3971, doi: 10.1175/JCLI-D-15-0612.1.
That paper also implies a role of the Ural Mountains, and the atmospheric blocks situated over them. The Urals are only a short distance ‘downstream’ of the Baltic and would be expected to influence this local region. The paper would greatly benefit from some thoughts on this aspect, and of changes and variability in blocking. Beneficial in this to cite analysis of
Luo, Dehai & coauthors, 2017: Increased quasi-stationarity and persistence of winter Ural Blocking and Eurasian extreme cold events in response to Arctic warming. Part II: A theoretical explanation. J. Climate, 30, 3569–3587, doi: 10.1175/JCLI-D-16-0262.1.
Lines 181-188: The ‘local’ correlations are similar to what one would expect. It is not clear to me that they are ‘necessary to better understand the teleconnections …’ Please clarify.
Lines 193-194: See my earlier point on sample size and ‘statistically reliable results’. May wish to reword this.
Lines 195-206: I have trouble interpreting these results, as it is not made clear what the sea ice is doing in these simulations. The integrations to 2100 are performed with RCP8.5 forcing and the reader is entitled to some (limited) information as to how the Arctic ice is changing over the period. As the changes will almost certainly be large the correlations in the last epochs of this century will essentially refer to physical associations which are different from those in the earlier part of the 21st century. The interpretation here needs much more thought.
Of relevance to these considerations is the recent analysis of
Yeon-Hee Kim, Seung-Ki Min, Nathan P. Gillett, Dirk Notz and Elizaveta Malinina, 2023: Observationally-constrained projections of an ice-free Arctic even under a low emission scenario. Nature Communications, 14, 3139, doi: 10.1038/s41467-023-38511-8.
Line 216: Insert ‘of the variance’ after ‘10%’.
Line 228-233: An interesting point is made here in connection with the potential influence from the Greenland Sea. It would be worth mentioning in the text that Zhuo, Yao et al., 2023: The key atmospheric drivers linking regional Arctic amplification with East Asian cold extremes. Atmospheric Research, 283, 106557 found sea ice cover this Greenland region to be one of the most influential of all Arctic regions on teleconnections to, and conditions in, the Baltic.
Lines 243-245: What are the physics behind this correlation. Is it ‘on shore’ flow, Ekman effect, …?
Also here may be informative to reference the recent work of
Vavrus, S. J., and R. Alkama, 2022: Future trends of arctic surface wind speeds and their relationship with sea ice in CMIP5 climate model simulations. Climate Dyn., 59, 1833-1848, doi: 10.1007/s00382-021-06071-6.
Lines 263-265: Please to note that the year of Lantao Sun’s paper is ‘2018’, not ‘2016’. The reason(s) for this disagreement must be canvassed. Sun t al. used the GFDL model (with RCP8.5). Related to my earlier point part of this difference may be due to how the sea ice evolves over the century. A more critical examination is required here.
Line 411-416: These two reference are the same, except the dates are different. I strongly suspect that the authors meant the first of these to be
Lantao Sun, Judith Perlwitz and Martin Hoerling, 2016: What caused the recent "Warm Arctic, Cold Continents" trend pattern in winter temperatures? Geophysical Research Letters, 43, 5345-5352, doi: 10.1002/2016gl069024
But please check carefully.
Citation: https://doi.org/10.5194/egusphere-2023-1637-RC3 - AC3: 'Reply on RC3', Erko Jakobson, 13 Oct 2023
Peer review completion
Journal article(s) based on this preprint
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 220 | 72 | 24 | 316 | 12 | 11 |
- HTML: 220
- PDF: 72
- XML: 24
- Total: 316
- BibTeX: 12
- EndNote: 11
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
Erko Jakobson
Liisi Jakobson
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
- Preprint
(1713 KB) - Metadata XML