Atmospheric Blocking Events over the Southeast Pacific and Southwest Atlantic Oceans in the CMIP6 Present-Day Climate

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Overview and General Impression

The presented manuscript (MS) is dedicated to the evaluation of the global circulation model’s (GCM) capability to represent atmospheric blocking events in the Southeast Pacific and Southwest Atlantic regions. The selected GCMs are sixteen and are from Phase 6 of the Coupled Model Intercomparison Project (CMIP6). ECMWF’s ERA5 reanalysis is utilized as a reference. Different aspects of the blocking events historical period (1985-2014) are investigated: total number in both sectors, multi-year seasonal as well as longitudinal distribution, and place of the maximum. The key message confirms the already known fact that despite the ongoing improvements, the CMIP6 GCMs still underestimate the number of occurrences of the blocking events in the considered domain (similarly to the other parts of the globe). Although the physical and synoptic aspect of the blocking phenomena is completely outside of the scope of the MS (which, strictly speaking, is a certain weakness. The Southern hemisphere specifics in this regard deserve separate publication), the present material is reasonable, keeping in mind the profound importance of the phenomena and its far-reaching climatic consequences. The detected non-negligible underestimation should be taken into consideration by the investigators before the realization of the tempting idea to use the GCMs to project the long-term change of the blocking phenomena metrics up to the end of the century.

Most generally, the material is well conceptualized and, as a result, well written and illustrated. The length is suitable. The bibliography is adequate and reveals the author’s expertise in the area. In my opinion, the sober, clear, and concise style of presentation should be admired. The MS fits well in the thematic scope of ‘Climate’ and has a significant potential to attract readers.

Although a relatively big number of my remarks none of them do not reflect any general flaws or principal caveats. Nevertheless, they should be addressed before the material becomes publishable.

Remarks:

-r52: Hence the considered number of the CMIP6 GCMs is significantly smaller than the total one, the question(s) arise(s) why exactly these models are selected? Where are taken from? There is neither a link to the access point nor the mandatory in such cases acknowledgment to the principal data vendors.

 -r56: Neither reference nor acknowledgment to the ECMWF’s ERA5.

 -r64: Reformulate ‘...from the blocking events..’ to, say, ‘...from the daily values…’.

 -Table 1: Caption is missing.

 -Figure 1: ‘Blocking Area’ → ‘Considered Area/Domain’. An important question that should be clarified: Why the border between the Atlantic and Pacific sectors is along the meridian 80 deg. W rather than 70 deg. W which seems the correct one from the geographical point of view?

-r65-83: About the methodology of the blocking index: It is longitude-dependent, i.e. one index is calculated for each meridional strip, i.e. in the selected 2.5-degree spacing |140-80|/2.5+1 = 25 for the Pacific and 27 for the Atlantic sector, right (Figure 4 seems obtained so) If yes, how is the total number for the sector obtained? Mean, sum, something other. This aspect of the methodology should be explained.

 -r80: Threshold units are missing. Perhaps geopotential meters (gm)?

 -r102: Which reanalysis is mind at all? The official name and reference are missing.

 -r136: About the seasons: What is the segmentation of the year? I suppose opposite of the Northern hemisphere, i.e. DJF Summer, MAM autumn, JJA winter, and SON spring, right? It should be stated explicitly in the text.

-r138: Winter/Summer → winter/summer

-Figure 4: About the longitudinal/seasonal distribution: A not very well expressed minimum appears around 60 deg. W – some comments and explanations?

-Spatial patterns of the blocking events, Figures 6-7: Due to the selected visualization method (smoothed field with shading) and color key (visually close color bins) all figures are hardly distinguishable, i.e. these figures can not illustrate the main intended goal – to demonstrate the differences between the models. Probably mosaic type (i.e. ‘one grid cell-one value’) could reveal at least center position displacement (in grid cell units, relative to the ERA5-based pattern). Reconsider figures 6-7!

Comments on the Quality of English Language

The quality of the English Language is acceptable, the style is clear.

Author Response

The authors are very grateful for the additional comments and suggestions from the Reviewer. Find below our point-by-point responses. In addition to the point-by-point responses, the reviewer will find a tracked-changes version of the manuscript with the revisions. In this version, removed text is marked in red-crossed format, while new discussions are highlighted in blue text.

Replies to Reviewer:

1 -r52: Hence the considered number of the CMIP6 GCMs is significantly smaller than the total one, the question(s) arise(s) why exactly these models are selected? Where are taken from? There is neither a link to the access point nor the mandatory in such cases acknowledgment to the principal data vendors.

Response 1: From the array of CMIP6 models available, we specifically chose those that provided daily data for geopotential height at the 500 hPa level, covering both the historical period and future projections for the SSP1-2.6, SSP2-4.5, and SSP5-8.5 scenarios. This selection was made as part of a broader research project analyzing future projections of blockings in the region. Additionally, we limited our selection to models offering the 'r1i1p1f1' run, which represents a commonly accessible simulation configuration, characterized by specific initial conditions, initialization methods, physical schemes, and forcing configurations. These criteria led us to include a total of 13 CMIP6 models in our analysis, which were available at the time we generated the results for the manuscript.

The research center responsible for developing each CMIP6 model used in this study is listed in Table 1. All data was downloaded from the ESGF (Earth System Grid Federation) portal, accessible at https://esgf.github.io/nodes.html Additionally, in the revised version of the manuscript, we have included this information in the caption of Table 1. 

2 -r56: Neither reference nor acknowledgment to the ECMWF’s ERA5.

Response 2: We added a reference for ERA5. Thank you.

3 -r64: Reformulate ‘...from the blocking events..’ to, say, ‘...from the daily values…’.

Response 3: As suggested, we reformulate the phrase to make it more clear. Thank you.

4 -Table 1: Caption is missing.

Response 4: We apologize for the oversight; the table caption has now been included in the revised version of the manuscript. Thank you.

5 -Figure 1: ‘Blocking Area’ → ‘Considered Area/Domain’. An important question that should be clarified: Why the border between the Atlantic and Pacific sectors is along the meridian 80 deg. W rather than 70 deg. W which seems the correct one from the geographical point of view?

Response 5: Other studies addressing atmospheric blocking in the study region have also used 80° as the border between the Atlantic and Pacific sectors. For example, the studies by Oliveira and Ambrizzi (2016): "The effects of ENSO-types and SAM on the large-scale southern blockings," and Mendes et al. (2022): "Synoptic-dynamic indicators associated with blocking events over the Southeastern Pacific and South Atlantic oceans."

Additionally, in our study, we observed that when the index identified blocked days in the 70-80° longitude range, the blocking system was generally located over the Atlantic Ocean rather than the Pacific. Therefore, we also decided to use 80° as the border between these two sectors.

6-r65-83: About the methodology of the blocking index: It is longitude-dependent, i.e. one index is calculated for each meridional strip, i.e. in the selected 2.5-degree spacing |140-80|/2.5+1 = 25 for the Pacific and 27 for the Atlantic sector, right (Figure 4 seems obtained so) If yes, how is the total number for the sector obtained? Mean, sum, something other. This aspect of the methodology should be explained.

Response 6: Actually, the index checks each 2.5° latitudinal strip to verify if the following conditions are met: (a) GHGN > 0 and (b) GHGS < -10 m. If these criteria are met, a given longitude is defined as 'blocked' at a specific instant in time. However, we considered a sector blocked on a particular day if three or more adjacent longitudes within the Pacific or Atlantic domain are blocked.Additionally, to be considered a blocking event, there must be a series of blocking days, with a minimum of five consecutive days. For both the Atlantic and Pacific sectors, the total number of blocking events is given by the sum of the identified blocking events during the entire period 1985-2014. Our blocking index follows the criteria used by Mendes et al. (2022): "Synoptic-dynamic indicators associated with blocking events over the Southeastern Pacific and South Atlantic oceans." 

7 -r80: Threshold units are missing. Perhaps geopotential meters (gm)?

Response 7: Thank you for pointing that out. The units were indeed missing, and we have added them in the revised version.

8 -r102: Which reanalysis is mind at all? The official name and reference are missing.

Response 8: The studies mentioned used the NCEP-NCAR and CFSR reanalyses. In the revised version, we have added the references for both the NCEP-NCAR and CFSR reanalysis.

9 -r136: About the seasons: What is the segmentation of the year? I suppose opposite of the Northern hemisphere, i.e. DJF Summer, MAM autumn, JJA winter, and SON spring, right? It should be stated explicitly in the text.

Response 9: Yes, your assumption is correct; the seasons are opposite to those in the Northern Hemisphere. We have added this clarification to the legend of Figure 3.

10 -r138: Winter/Summer → winter/summer

Response 10: Thank you for pointing it out, it is now corrected. 

11-Figure 4: About the longitudinal/seasonal distribution: A not very well expressed minimum appears around 60 deg. W – some comments and explanations?

Response 11: This is an interesting observation and a brief discussion of this was added  in the revised manuscript. The minimum around 70-60°W seems to be related to the climatological position of blockings in the region. The Pacific is the most active region, with the highest frequency of blockings, followed by secondary maxima over the South Atlantic and the southern Indian Ocean. Studies, mainly for the Northern Hemisphere, have shown that blocking occurs primarily in locations that correspond to the end of the climatological storm tracks or near stationary or standing high-pressure anomalies.

12-Spatial patterns of the blocking events, Figures 6-7: Due to the selected visualization method (smoothed field with shading) and color key (visually close color bins) all figures are hardly distinguishable, i.e. these figures can not illustrate the main intended goal – to demonstrate the differences between the models. Probably mosaic type (i.e. ‘one grid cell-one value’) could reveal at least center position displacement (in grid cell units, relative to the ERA5-based pattern). Reconsider figures 6-7!

Response 12: The main reason the figures are not very distinguishable is that there is not much difference between the models. Since the models had difficulty capturing the total number of blockings and their duration, particularly in the Atlantic sector, we decided to analyze the Z500 anomaly fields to see if the overall geopotential field was also poorly represented in some models. However, the majority of models were able to represent the overall Z500 field of the blockings, with anomaly intensities and the position of the blocking high not differing significantly from ERA5 and among the models.

We tried various visualization methods to highlight differences and used the same anomaly scale for all the models to facilitate comparison. We also attempted plotting in grid cell units as you suggested, but the visualization did not improve.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Detailed comments are in the PDF attached.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Quality of English is fine overall. Some section in the results need editing. Please see my detailed comments.

Author Response

The authors are very grateful for the additional comments and suggestions from the Reviewer. Find below our point-by-point responses. In addition to the point-by-point responses, the reviewer will find a tracked-changes version of the manuscript with the revisions. In this version, removed text is marked in red-crossed format, while new discussions are highlighted in blue text.

Major issues

1: Plagiarism

• L36 – L 39 are exactly copied from Patterson et al. 2019. The authors haven’t cited them in those lines but have cited them elsewhere. Even after citing a paper, the text cannot be copied word to word unless specifically quoting the paper.

Response 1: Indeed, the phrase was very similar to Patterson et al. 2019. We read their paper and cited them in the discussion of the results. However, we did not realize that the phrase was so similar, so we have removed it from the manuscript.

2: Introduction needs more literature review

• The authors should add content on why CMIP models struggle to identify blocking. Has there been any improvement from CMIP5 to CMIP6 shown in the existing studies e.g., in the Northern Hemisphere (Schiemann et al. 2020, Liu et al. 2022, etc)? Liu et al. 2022 also analyzed blocking in the southwestern Pacific as mentioned in L43. What does this study adds compared to their analysis? This should be clearly motivated.

Response 2: 

We have tried to add discussion regarding the difficulties CMIP models face in representing blocking events. Please also see our Response 5.

Liu et al. (2022) did a broader and more detailed analysis, including a discussion on blocking in the Northern Hemisphere. They focused their analysis on JJA and DJF and used a different blocking identification method. Additionally, they included a set of models from both CMIP5 and CMIP6. For the Pacific sector, they analyzed a larger longitude band than our study, we focus only on the southeast Pacific. They found that the GCMs overestimated blocking frequency in the Pacific sector, whereas we found an overall underestimation of blocking in this region. We have added more discussion to compare our results with theirs.

3: More details needed about the blocking index in the Methods section

• Since the use of a blocking index is central to theme of this study, there needs to be more explanation on the Tibaldi index. The authors can also consider adding a paragraph in the introduction or methods on the broad themes behind the different blocking index that exists.

Response 3: The blocking index we used is based on reference 7 (Mendes et al. (2022)), which in turn is based on the Tibaldi index. Mendes et al. provide a detailed explanation of this index. We acknowledge that the method or index used to identify blocking events is important and can influence the results. However, in this manuscript, we did not explore different methods and their impact on the blocking climatology, so we did not include a discussion on this topic.

Mendes, M. C. D., da Silva Aragão, M. R., Mendes, D., Mesquita, M. D., Correia, M. D. F., & Cavalcanti, E. P. (2022). Synoptic–dynamic indicators associated with blocking events over the Southeastern Pacific and South Atlantic oceans. Climate Dynamics, 60(7), 2285-2301.

4: Figure 6 and Figure 7

• Seasonal shift in the jet will affect the location of the blocking centres. It is better to

present these results for DJF and JJA seasons separately.

Response 4: First, we tried to separate the composites into DJF and JJA. However, since the number of blocking events is smaller in DJF and the Z500 anomalies are weaker, the plot was not suitable for visualization, especially in the Atlantic sector. Therefore, we decided to show only the composite with all events. 

5: Discussion of the results is missing

The authors should discuss their results in the light of the existing literature as part of the results section or in a separate section. Some points that can be discussed are:

• Why do CMIP models generally underestimate blocking? What does existing literature say about the missing processes needed for capturing blocking?

• Do they find any improvements from CMIP5 to CMIP6 in any of the models in the representation of blocking?

Response 5: 

Point 1: As discussed in Lupo (2021) and Davini and D’Andrea (2020), and as recommended by Schiemann et al. (2020), problems in representing blocking in GCMs mainly arise from an incorrect representation of the mean state, as well as model parameters such as horizontal and vertical resolution, and parameterization schemes. Additionally, an incorrect representation of the jet stream and storm tracks can also negatively affect blocking simulations. The majority of studies discuss blocking in the Northern Hemisphere.

Point 2: Overall, there seem to be improvements from CMIP5 to CMIP6 in the representation of blocking events, as discussed in Davini and D’Andrea (2020). However, we did not delve into this topic in our study because we are exclusively focusing on CMIP6 data.

These are important questions. However, for this study, our main goal was to quantitatively assess whether the CMIP6 models could reproduce the overall climatology of blocking events in the two sectors close to South America. For now, we have not delved into the detailed physical explanations.

For a future manuscript, we are starting to examine the upper-level atmospheric circulation patterns and will also assess how well the models represent the planetary-scale jet stream. This will allow us to start to delve deeper into the physical explanations.

Lupo, A. R. (2021). Atmospheric blocking events: a review. Annals of the New York Academy of sciences, 1504(1), 5-24.

Davini, P., & d’Andrea, F. (2020). From CMIP3 to CMIP6: Northern Hemisphere atmospheric blocking simulation in present and future climate. Journal of Climate, 33(23), 10021-10038.

Other issues

6: L 16: Define blocking? Not all readers of the Climate journal will be familiar with Atmospheric blocking.

Response 6: We added a definition of blocking at the beginning of the introduction section.

7: L44-L45: What does the cited studies say about the reason behind blocks being underestimated?

Response 7:

[14] Liu attributed the main reason to be dependent on the GCMs mean state, suggesting that a biased climatology of geopotential heights and zonal eddies can result in biased representations of blocking events. Therefore, improving the mean state would improve the representation of blocks. 

[15] Petterson does not discuss in detail the reasons for blocks underestimation in the Australia-New Zealand sector, but they mention that the highest blocking biases is related to bias and difficulty in representing the wind field and jet position. 

8: L 55: Why only 1985-2014 period was used? Is it because CMIP6 data is only available for that period?

Response 8: We were interested in blocking events that occurred more recently. We opted for using a period of 30 years, and chose the most recent so this is why we started in 1985 since CMIP6 historical period ends in 2014.  However, we are aware that ERA5 and CMIP6 data is available for a longer period.   

9: Table 1. caption missing

Response 9: We apologize for the oversight; the table caption has now been included in the revised version of the manuscript. Thank you.

10: L 82: To satisfy the synoptic definition of a block ([? ])

Response 10: We included the reference, thank you.

11: L89: “significantly smaller” How’s the statistical significance assessed here? Else, remove “significantly”.

Response 11: We removed the word “significantly”. Thank you.

12: L95: Which algorithm did [7] use? Mention that for clarity.

Response 12: We mentioned in the methodology that the algorithm used by [7] is based on the Tibaldi index, and we based our blocking detection method on theirs. However, as you suggested, we have also added a brief mention of this in this section.

13: L99-104: Mention the blocks identified in number of blocks per year so it is easier to compare.

Response 13: As suggested, we added the blocks per year in the discussion. 

14: L128: Why do some models overestimate blocking events over the Atlantic? Can the authors discuss that?

Response 14: For now we do not have an explanation.This is a region when blocks are not that frequent, so we did not find discussion on blocks in CMIP for specific for this region in the literature.  However, we think that maybe it can be because CMIP6 models did not represent the mean state and the jet stream well. 

15: L158-L161: Remind the readers which latitude band was considered?

Response 15: We added the longitude band, thank you. 

 

 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This manuscript documents the capability of 13 CMIP6 models in representing atmospheric blocking events over Southeast Pacific and Southwest Atlantic during 1985-2014 period as compared with ERA5 reanalysis. The study adopts a simple blocking index based on daily z500 values, and examines the frequency, duration, and spatial patterns of blocking events in these models, and reveals the generally poor performance of CMIP6 models regarding this phenomenon. 

 

The study is quite straightforward, and the writing is easy to read and understand.  There are several editorial issues that can be resolved by a careful proofread.  The more concerning issue is the lack of insights in this study.  Most of the results are based on simple comparison, and most of the text and conclusion focus on describing the differences.  It's certainly useful to document the behaviors of these models, but the study does not provide any substantial insights on why these models behaved this way and how they may be improved. So it feels quite weak in scientific contribution.  I encourage the authors to provide some discussion on the scientific contribution of this research to strengthen the manuscript.  Some specific comments are listed below. 

 

- Line 82: missing reference

- Line 88-90: are these results from ERA5?  It needs to be clearly stated. 

- Figure 3 and related text, it would be useful to state if these are boral summer or austral summer.  And it is weird that the four seasons are not in the right order.

- Table 1 caption is missing, instead it has text from a template, I assume. 

- Line 206 and many other places:  Fig. Figure 6a. Repeated "Fig." and "Figure"

- Line 206: "an" area, not "a" area.

- Figure 6 and 7:  It's not clear how these tow figures were constructed.  Are this composite of periods when a blocking is identified? If so, I assume their sample sizes are quite different from each other as the blocking frequency and duration were underestimated by most models.  So, I am not sure if this is a fair comparison. But at least, more details should be provided on how these maps are constructed.

 

Comments on the Quality of English Language

Overall good writing, but a more careful proofread can eliminate some obvious errors. 

Author Response

The authors are very grateful for the additional comments and suggestions from the Reviewer. Find below our point-by-point responses. In addition to the point-by-point responses, the reviewer will find a tracked-changes version of the manuscript with the revisions. In this version, removed text is marked in red-crossed format, while new discussions are highlighted in blue text.

Replies to Reviewer:

1 - Line 82: missing reference

Response 1: Thank you, we corrected the reference in the revised version of the manuscript.

2- Line 88-90: are these results from ERA5?  It needs to be clearly stated. 

Response 2: Yes, these results are from ERA5. We have added a clarification in the revised version. Thank you.

3 - Figure 3 and related text, it would be useful to state if these are boral summer or austral summer.  And it is weird that the four seasons are not in the right order.

Response 3: We have added a clarification of each season to the legend of Figure 3. Regarding the order of the seasons, we chose to place summer and winter next to each other because they represent the extremes with the lowest and highest number of events. Autumn and spring, having an intermediate and similar number of events, follow. We think this arrangement facilitates comparison.

4 - Table 1 caption is missing, instead it has text from a template, I assume. 

Response 4: We apologize for the oversight; the table caption has now been included in the revised version of the manuscript. Thank you.

5- Line 206 and many other places:  Fig. Figure 6a. Repeated "Fig." and "Figure"

Response 5: Thank you for pointing that out. It was an oversight during editing. We have corrected it throughout the manuscript.

6- Line 206: "an" area, not "a" area.

Response 6: Thank you, it is now corrected.

7- Figure 6 and 7:  It's not clear how these tow figures were constructed.  Are this composite of periods when a blocking is identified? If so, I assume their sample sizes are quite different from each other as the blocking frequency and duration were underestimated by most models.  So, I am not sure if this is a fair comparison. But at least, more details should be provided on how these maps are constructed.

Response 7: The anomalies were obtained by considering the daily climatology of geopotential height at 500 hPa (Z500) for the 30-year period from 1985 to 2014, which is considered the mean state. This climatology was subtracted from the daily Z500 values during each day with blocking events, and the results were averaged to form the composite field of Z500 anomalies for blocking events. This procedure was applied to each model and ERA5. We added information to the methodology to provide more details on how these figures were constructed.

And yes, you are correct. The sample sizes are different because, as you mentioned, the frequency and duration of blocking events vary among the models. Consequently, the periods of blocking are not the same for all CMIP6 models and ERA5.

We computed the Z500 anomalies to observe the general spatial characteristics of blocking events, such as the intensity of Z500 anomalies and the overall position of the blocking high, and to compare the composites from CMIP6 models with those from ERA5. Since the models had difficulty capturing the total number of blockings and their duration, particularly in the Atlantic sector, we decided to analyze the Z500 anomaly fields to see if the overall geopotential field was also poorly represented in some models. However, the majority of models were able to represent the overall Z500 field of the blockings, with anomalies intensities and the position of the blocking high not differing significantly from ERA5.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

The manuscript by Ferreira et al. presents the simulation of blocking events in the southern hemisphere from a suite of CMIP6 models. The research topic is highly significant since it deals with southern hemisphere blocking events, an understudied research area.

I suggest the authors to give possible physical explanation for the over/under estimation of blocking events in the Southeast Pacific and Southwest Atlantic regions. This can be included while discussing the results.

Lines 116-117: What might be the possible causes for the variation in performance of CMIP6 models in the two domains? Does it depend on the model mean state?

Did you get a chance to check the upper level atmospheric circulation pattern in the two cases? Can one conclude that the frequency of blocking events in the southeast Pacific and southwest Atlantic depends on how well the models represent the planetary scale jet stream (polar jet stream)/? This will in turn decide the scale interaction (planetary scale-synoptic scale ) and its role in the occurrence of blocking events . 

Line 122: The majority of the CMPI6 models closely resembled ERA5 blocking patterns.... CMIP6??

Line 267: Rephrase Line 267.

Author Response

The authors are very grateful for the additional comments and suggestions from the Reviewer. Find below our point-by-point responses. In addition to the point-by-point responses, the reviewer will find attached a tracked-changes version of the manuscript with the revisions. In this version, removed text is marked in red-crossed format, while new discussions are highlighted in blue text.

1: I suggest the authors to give possible physical explanation for the over/under estimation of blocking events in the Southeast Pacific and Southwest Atlantic regions. This can be included while discussing the results.

Response 1: We have tried to incorporate discussion based on other studies, but there is limited research addressing detailed blocking in these sectors, particularly in the SW Atlantic. We are beginning to examine the physical aspects and plan to publish a future study on this topic.

2: Lines 116-117: What might be the possible causes for the variation in performance of CMIP6 models in the two domains? Does it depend on the model mean state?

Response 2: As discussed in Lupo (2021) and Davini and D’Andrea (2020), and other studies mainly for the Northern Hemisphere, problems in representing blocking in GCMs mainly arise from an incorrect representation of the mean state, as well as model parameters such as horizontal and vertical resolution, and parameterization schemes. Additionally, an incorrect representation of the jet stream and storm tracks can also negatively affect blocking simulations. 

3: Did you get a chance to check the upper level atmospheric circulation pattern in the two cases? Can one conclude that the frequency of blocking events in the southeast Pacific and southwest Atlantic depends on how well the models represent the planetary scale jet stream (polar jet stream)/? This will in turn decide the scale interaction (planetary scale-synoptic scale ) and its role in the occurrence of blocking events . 

Response 3: These are all interesting questions and points raised. For this study, our main goal was to just quantitatively assess whether the CMIP6 models could reproduce the overall climatology of blocking events in the two sectors close to South America. For now, we have not delved into the detailed physical explanations.

However, we are just starting to examine atmospheric circulation patterns, such as Z500 and other levels and the wind field. We also want to assess how well the models represent the base state and the jet stream. So, we plan to write another manuscript with this analisis and also try to have more physical explanations.

4: Line 122: The majority of the CMPI6 models closely resembled ERA5 blocking patterns.... CMIP6??

Response 4:  Yes, it was a typo, it is now corrected. Thank you.

5: Line 267: Rephrase Line 267.

Response 5: We corrected the phrase, thank you. 

 

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addressed most of my issues and my recommendation is that the revised manuscript can be accepted.

Comments on the Quality of English Language

I did not check specifically for the quality of English and leave it to the editorial team.