You are currently viewing a new version of our website. To view the old version click .
MDPIMDPI
Search all articles
by
  • Ilya Serykh1,2,
  • Svetlana Krasheninnikova1 and
  • Mariia Safonova1
  • et al.

Reviewer 1: Rizwan Karim Reviewer 2: Zeinab SALAH Reviewer 3: Nítalo Farias Machado Reviewer 4: Anonymous

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Based on the review of the work titled "Analysis of Near-Surface Air Temperature Trends in Brazil Region Using Meteorological Station Data, ERA5 Reanalysis, and CMIP6 Models", the following suggestions and recommendations have been penned down for the authors:

  1. line 18: Add a one or two sentences regarding the importance and context of this work over the Study Region.
  2. line 20: Please provide a short description of analyses conducted in this work.
  3. line 22: please use a short annotation like t2m for abbreviations throughout the manuscript.
  4. Line 40-42: Rephrase this sentence
  5. Line 45: Introduction : The introduction seems to lack flow of ideas, with missing important information in different sections provided here. Also there is very limited information provided for all of the sections whether its climate models, evaluation of temperature or projections of temperature across the region and beyond. This sort of information gives the idea and the gaps left in those studies and provides a basis for your study's importance overall. Therefore, the authors are advised to add more information and in a proper flow, thus revise the overall introduction section please.
  6. line 50-51: Add the information about the GCMs background (line 85-88) here. Also add information on why they are created and what purpose they serve to show their importance.
  7. Line 78-83: Please try to merge and connect this with above paragraph and show that why these historical an future t2m are necessary across the Brazilian Regions.
  8. Line 66-77: Please try to create a flow of idea in the whole introduction part. The projections seems to be written apart from the previous paragraph about coastal climate changes assessment, while this jumps directly to projections. there is a gap of why, where and using what means this future climate could be assessed. Revise these two paragraphs carefully.
  9. Line 78-83: Please try to merge and connect this with above paragraph and show that why these historical an future t2m are necessary across the Brazilian Regions.
  10. Line 95-99: Place this paragraph before the above (89-94) so as to show a flow of information on global to regional/ study area scales. Also expand this information about the regional projections studies around the study area such as in South America etc. Further, you have to mention the gaps in previous studies which provided the thought of designing this study. Remember that any study is designed to fill the gap left by previous studies in case of climate sciences. The authors should go through the studies conducted previously to find a gap and mention it here.
  11. Line 114-118: This is a a big caveat in your work if observational data is missing. So you need to provide the procedure of how you filled in the missing gaps from the reanalyses dataset (ERA5). Should provide some preliminary results of correlation between the observed and ERA5 to show tiedness between the two at the feasible timescales (e.g. daily, monthly, sesonal as per your data usage in this work). This is suggested to show the readers about sanctity of the proxy datasets (ERA5) you are using.
  12. Line 133-137: Since the authors are using ERA5 data to fill in missing values for temperature which has good spatial resolution, wheareas MERRA2 with better resolution but short temporal coverage is used to validate the ERA5. Later the NCEP/NCAR reanalyses dataset is used to validate the above all datasets? Isn’t it better to use a one dataset like ERA5 to validate or evaluate the CMIP6 models using the 1970s or 1980s and onwards data for ERA5 as the new normal in temperature is considered after 1990s so it will cover the pre and post new normal temperature change era.
  13. Line 141-145: please mention the name of ensemble technique and its description
  14. Line 175-193: It is suggested that most of this text should be put in the introduction part and only retain the relevant literature here. This is the methods sections so it would be better to state the methods/procedures here and cite or relate literature from introduction here for reference.
  15. Line 213-215: It is strongly suggested to use the Mannkendall Trends tests to draw correct trends since the linear trends may not give the true signal of natural variability as proven in several studies.
  16. Figure 2: It is strongly suggested to improve the quality of figures as the timeseries lines seem blurred, as well as the text. the text and graphics should be clear to readers.
  17. Line 242-246: As the results reveal difference in NSAT for 200-2023 period trends in ERA5 and meteorological stations, doesn't it shows inconsistency in values distribution in both datasets, while the correlation values (as you claim) show consistency of values across the entire timeperiod. It is also suggested to provide the correlation tables and figures to prove your results as well. You can refer to other studies for relevant figures to be drawn.
  18. Figure 3: For the boundaries of variability (S.D), the region in between could be filled in with a suitable color shades for each SSP scenario. There are plenty of examples for this figure avaiable in projections studies. For reference you can follow the figures from this study "Bias‐Corrected Climate Projections for Xinjiang: Decomposing Future Trends and Uncertainties in Temperature and Precipitation" rencntly published for the referenced figures.
  19. Line 310-212: There is a difference of NSAT b/w land and ocean area across southern Brazil with stronger values of NSAT at ocean than on the land in Figure 4, a and b here. What could be the reason. Could provide any literature which provides the reason behind this phenomenon here
  20. Line 323-324: It is suggested to change the name " CMIP6 ESMs en- 323semble (Historical and SSP2-4.5 experiments)" to simply " CMIP6 MME" throughout the manuscript.
  21. Line 336-338: For the projected changes distributions across the study region, why the periods such as 2070-2099 and 2024-2053 are chosen and in this process, around 16 years of missing information (2054-2069) is missing. This period falls into mid century and has same importance in context of climate information as for the other periods delineated here. It is suggested to redefine the periods to include these missing years. You can choose like 35 years period for the future and historical periods which is fine as climate change signal is usually established in min 30 years.
  22. Line 356-357: It is suggested that the scalebar/colorbar needs to be fixed for the values present in the maps as the values exceed in scalebar/colorbar please fix them. Follow the same for all figures above and below
  23. Line 390: Do you mean projected changes here or increase here?. Is should be changes written technically speaking.
  24. Line 418-419: Always remember that you can never start a sentence/phrase using a word like 'and'. Please rephrase this sentence and go through the manuscript for similar errors and fix them. There are plenty of grammatical errors as well.
  25. Line 427-440: This discussion paragraph highlights the supression of climate signal upon ensemble approach. Is this a caveat of this study or explaining something else here. The purpose of this paragraph should be clear with counter evidence provided to defend the procedure you chose in your work. Addidionally, you have not given a description of the factors aiding to rising temperatures across the study region and beyond. This will strengthen your results. You can provide the references to the atmospheric circulations studies, anthropogenic impacts studies etc to provide your narrative of why your results are right and valuable.
  26. Line 490-493: there is no need of mentioning the account of volcanic eruptions for better modeling as you already provided it in discussions. If you feel it necessary, give a short one to two sentences here. Further also give a recommendation for policy makers and other stakeholders and scientific community contextualing your work to collectvie efforts to counter changing climate in the study region.
  27. Line 514: In references, please provide the DOIs for available literature given below if it is available. "

Comments for author File: Comments.pdf

Comments on the Quality of English Language

The manuscript provides valuable scientific insights; however, the quality of English requires attention in some places. There are some grammatical errors, inconsistent phrasing, and sentence structures throughout. I recommend a comprehensive language revision to improve clarity and readability.

Author Response

Response to Reviewer 1 Comments

The authors are grateful to the reviewer for the correct comments that allowed us to improve the text of the manuscript.

Suggestions and recommendations from Reviewer 1:

  1. line 18: Add a one or two sentences regarding the importance and context of this work over the Study Region.

Response 1: Added.

  1. line 20: Please provide a short description of analyses conducted in this work.

Response 2: Added.

  1. line 22: please use a short annotation like t2m for abbreviations throughout the manuscript.

Response 3: A short annotation of NSAT was used.

  1. Line 40-42: Rephrase this sentence

Response 4: Rephrased.

  1. Line 45: Introduction: The introduction seems to lack flow of ideas, with missing important information in different sections provided here. Also there is very limited information provided for all of the sections whether its climate models, evaluation of temperature or projections of temperature across the region and beyond. This sort of information gives the idea and the gaps left in those studies and provides a basis for your study's importance overall. Therefore, the authors are advised to add more information and in a proper flow, thus revise the overall introduction section please.

Response 5: The Introduction has been significantly supplemented and modified.

  1. line 50-51: Add the information about the GCMs background (line 85-88) here. Also add information on why they are created and what purpose they serve to show their importance.

Response 6: Added.

  1. Line 78-83: Please try to merge and connect this with above paragraph and show that why these historical an future t2m are necessary across the Brazilian Regions.

Response 7: Added.

  1. Line 66-77: Please try to create a flow of idea in the whole introduction part. The projections seems to be written apart from the previous paragraph about coastal climate changes assessment, while this jumps directly to projections. there is a gap of why, where and using what means this future climate could be assessed. Revise these two paragraphs carefully.

Response 8: Added.

  1. Line 78-83: Please try to merge and connect this with above paragraph and show that why these historical an future t2m are necessary across the Brazilian Regions.

Response 9: Added.

  1. Line 95-99: Place this paragraph before the above (89-94) so as to show a flow of information on global to regional/ study area scales. Also expand this information about the regional projections studies around the study area such as in South America etc. Further, you have to mention the gaps in previous studies which provided the thought of designing this study. Remember that any study is designed to fill the gap left by previous studies in case of climate sciences. The authors should go through the studies conducted previously to find a gap and mention it here.

Response 10: Added.

  1. Line 114-118: This is a a big caveat in your work if observational data is missing. So you need to provide the procedure of how you filled in the missing gaps from the reanalyses dataset (ERA5). Should provide some preliminary results of correlation between the observed and ERA5 to show tiedness between the two at the feasible timescales (e.g. daily, monthly, sesonal as per your data usage in this work). This is suggested to show the readers about sanctity of the proxy datasets (ERA5) you are using.

Response 11: Added.

  1. Line 133-137: Since the authors are using ERA5 data to fill in missing values for temperature which has good spatial resolution, wheareas MERRA2 with better resolution but short temporal coverage is used to validate the ERA5. Later the NCEP/NCAR reanalyses dataset is used to validate the above all datasets? Isn’t it better to use a one dataset like ERA5 to validate or evaluate the CMIP6 models using the 1970s or 1980s and onwards data for ERA5 as the new normal in temperature is considered after 1990s so it will cover the pre and post new normal temperature change era.

Response 12: We left this comparison of different reanalyses in the methodology to show that the ERA5 reanalysis is still better suited for solving the problems of assessing long-term changes in NSAT.

  1. Line 141-145: please mention the name of ensemble technique and its description

Response 13: The description was added.

  1. Line 175-193: It is suggested that most of this text should be put in the introduction part and only retain the relevant literature here. This is the methods sections so it would be better to state the methods/procedures here and cite or relate literature from introduction here for reference.

Response 14: The Introduction is already quite long. We think that this text focuses more on the data and methodology.

  1. Line 213-215: It is strongly suggested to use the Mannkendall Trends tests to draw correct trends since the linear trends may not give the true signal of natural variability as proven in several studies.

Response 15: Added.

  1. Figure 2: It is strongly suggested to improve the quality of figures as the timeseries lines seem blurred, as well as the text. the text and graphics should be clear to readers.

Response 16: Corrected.

  1. Line 242-246: As the results reveal difference in NSAT for 200-2023 period trends in ERA5 and meteorological stations, doesn't it shows inconsistency in values distribution in both datasets, while the correlation values (as you claim) show consistency of values across the entire time period. It is also suggested to provide the correlation tables and figures to prove your results as well. You can refer to other studies for relevant figures to be drawn.

Response 17: The work has been updated with clarifications regarding the direction of trends and correlations.

  1. Figure 3: For the boundaries of variability (S.D), the region in between could be filled in with a suitable color shades for each SSP scenario. There are plenty of examples for this figure avaiable in projections studies. For reference you can follow the figures from this study "Bias‐Corrected Climate Projections for Xinjiang: Decomposing Future Trends and Uncertainties in Temperature and Precipitation" rencntly published for the referenced figures.

Response 18: If we fill these ranges with color, it will merge and be difficult to see.

  1. Line 310-212: There is a difference of NSAT b/w land and ocean area across southern Brazil with stronger values of NSAT at ocean than on the land in Figure 4, a and b here. What could be the reason. Could provide any literature which provides the reason behind this phenomenon here

Response 19: This is explained by the stabilizing role of the ocean.

  1. Line 323-324: It is suggested to change the name " CMIP6 ESMs en- 323semble (Historical and SSP2-4.5 experiments)" to simply " CMIP6 MME" throughout the manuscript.

Response 20: Replaced.

  1. Line 336-338: For the projected changes distributions across the study region, why the periods such as 2070-2099 and 2024-2053 are chosen and in this process, around 16 years of missing information (2054-2069) is missing. This period falls into mid century and has same importance in context of climate information as for the other periods delineated here. It is suggested to redefine the periods to include these missing years. You can choose like 35 years period for the future and historical periods which is fine as climate change signal is usually established in min 30 years.

Response 21: Our goal was to obtain a predictive estimate of NSAT changes over 50 and 100 years, so these periods were chosen for analysis. The intermediate period was not considered separately in the analysis. The 30-year periods are chosen based on WMO recommendations.

  1. Line 356-357: It is suggested that the scalebar/colorbar needs to be fixed for the values present in the maps as the values exceed in scalebar/colorbar please fix them. Follow the same for all figures above and below

Response 22: The colors on the scales were chosen to be consistent across the different Figures being compared during analysis. Changing these colors would make it more difficult to compare these Figures.

  1. Line 390: Do you mean projected changes here or increase here?. Is should be changes written technically speaking.

Response 23: Adjusted.

  1. Line 418-419: Always remember that you can never start a sentence/phrase using a word like 'and'. Please rephrase this sentence and go through the manuscript for similar errors and fix them. There are plenty of grammatical errors as well.

Response 24: Adjusted.

  1. Line 427-440: This discussion paragraph highlights the supression of climate signal upon ensemble approach. Is this a caveat of this study or explaining something else here. The purpose of this paragraph should be clear with counter evidence provided to defend the procedure you chose in your work. Addidionally, you have not given a description of the factors aiding to rising temperatures across the study region and beyond. This will strengthen your results. You can provide the references to the atmospheric circulations studies, anthropogenic impacts studies etc to provide your narrative of why your results are right and valuable.

Response 25:

We write that: “The observed weak multi-decadal changes in average NSAT in the study region in 1940–1975, followed by a gradually increasing grow, qualitatively repeat the changes in global NSAT during this period [IPCC, 2023] and, apparently, are closely related to them.”. We make a link to IPCC. Since the factor of natural variability in the ensemble of models is suppressed, only anthropogenic global warming remains as a cause of long-term temperature increase in the Brazilian region.

  1. Line 490-493: there is no need of mentioning the account of volcanic eruptions for better modeling as you already provided it in discussions. If you feel it necessary, give a short one to two sentences here. Further also give a recommendation for policy makers and other stakeholders and scientific community contextualing your work to collectvie efforts to counter changing climate in the study region.

Response 26: Partially removed.

  1. Line 514: In references, please provide the DOIs for available literature given below if it is available. "

Response 27: Added

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents a regional analysis of near-surface air temperature changes across Brazil, combining weather station data, the ERA5 reanalysis, and 33 CMIP6 models with different scenarios. This work is methodologically sound and provides useful regional insights. However, it needs a thorough review due to the lack of a clear statement of its novelty. The introduction also requires significant improvements to reflect previous work in Brazil and Latin America.

Introduction

  • The introduction is very general and repetitive. It includes many basic explanations about greenhouse gases, ESM models, and the history of CMIP, which are well-known.
  • It does not include a climatic description of Brazil (such as major climate zones or temperature gradients).
  • Previous studies on temperature trends in Brazil or Latin America are missing; only the global context is discussed.
  • The knowledge gap is implicitly mentioned, but not explicitly stated.
  • The authors state that one of the aims of the research is to " By leveraging the output of CMIP6 models, we seek to elucidate the underlying drivers of observed NSAT changes through the deconvolution of processes operating at diverse scales and physical mechanisms within the climate system." but the research does not actually analyze the physical mechanisms behind these changes.
  • Recommendation: provide a brief background, add a brief summary of previous regional studies, describe the Brazilian climate context, and explain the novelty of this research.

Methodology

  • The authors noted the use of MERRA2 and NCEP/NCAR to validate ERA5, this is acceptable for consistency, but not a true validation, which is achieved using meteorological observations.

Author Response

Response to Reviewer 2 Comments

The authors are grateful to the reviewer for the correct comments that allowed us to improve the text of the manuscript.

Introduction

  1. The introduction is very general and repetitive. It includes many basic explanations about greenhouse gases, ESM models, and the history of CMIP, which are well-known.

Response 1: Changed.

  1. It does not include a climatic description of Brazil (such as major climate zones or temperature gradients).

Response 2: Added.

  1. Previous studies on temperature trends in Brazil or Latin America are missing; only the global context is discussed. The knowledge gap is implicitly mentioned, but not explicitly stated.

Response 3: Added.

  1. The authors state that one of the aims of the research is to " By leveraging the output of CMIP6 models, we seek to elucidate the underlying drivers of observed NSAT changes through the deconvolution of processes operating at diverse scales and physical mechanisms within the climate system." but the research does not actually analyze the physical mechanisms behind these changes.

Response 4: removed information about mechanisms.

  1. Recommendation: provide a brief background, add a brief summary of previous regional studies, describe the Brazilian climate context, and explain the novelty of this research.

Response 5: Added.

 

Methodology

  1. The authors noted the use of MERRA2 and NCEP/NCAR to validate ERA5, this is acceptable for consistency, but not a true validation, which is achieved using meteorological observations.

Response 6: The study verifies the ERA 5 reanalysis with meteorological station data using Pearson correlation analysis. Trends across different observation periods are also compared across the meteorological stations, reanalysis, and model ensemble.

Reviewer 3 Report

Comments and Suggestions for Authors

This article aims to assess near-surface air temperature (NSAT) trends in the Brazil region (10°N–40°S; 75–25°W) from 1940 to 2099 using data from 39 meteorological stations, ERA5 reanalysis, and an ensemble of 33 CMIP6 models under four SSP scenarios. The main contributions include a multi-source validation showing a NSAT increase of ~0.3–0.4°C from 1964–2023, with projections indicating 1.05–3.53°C warming by 2070–2099 depending on emissions, and faster warming over land than ocean. Strengths lie in the robust ensemble approach, clear spatial analyses, and relevance to regional climate projections. In general, these results are interesting and guide adaptation policies. Furthermore, the manuscript has a theme within the scope of the journal.
The manuscript is clear, relevant to climate science, and well-structured, with a logical flow from data description to projections. The references are mostly recent and relevant, with no excessive self-citations. The study is scientifically sound, employing appropriate methods like ensemble averaging to isolate anthropogenic forcing from natural variability, and the experimental design suits testing the hypothesis that NSAT warming accelerates under higher emissions. Results appear reproducible based on detailed Methods, including data sources and anomaly calculations. Figures and tables effectively present data, with spatial maps aiding interpretation, and statistical analyses (e.g., correlations, trends) are consistent. Conclusions align with evidence. However, weaknesses include uneven meteorological station distribution and short time series (many starting ~2000), potentially biasing regional trends; what do the authors think about this? The hypothesis on ocean stabilization could be more testable with quantitative metrics; and missing controls for urban heat island effects in station data.

Line 194: "Monthly mean NSAT data were selected" – justify why not seasonal breakdowns, as tropical regions like Brazil show strong wet/dry season variability.

Line 242: Figure 2 trends – unclear if trends are statistically significant; add p-values or confidence intervals. Would it be possible to add a dispersion measure?

It would be interesting to include a discussion of the practical applications of the results obtained, such as their implications for public climate adaptation policies in Brazil (e.g., in agricultural planning, water resource management, and mitigation of extreme events such as heat waves and droughts, which are exacerbated by faster warming over land compared to the ocean). Furthermore, highlighting the importance of this study in the context of global climate change, such as its contribution to understanding the stabilizing role of the ocean and to regional projections based on CMIP6, would strengthen the manuscript.

The figures could have better resolution.

Level of English: Appropriate and comprehensible; minor phrasing issues, but no major errors.

Author Response

Response to Reviewer 3 Comments

The authors are grateful to the reviewer for the correct comments that allowed us to improve the text of the manuscript.

Comments 1: However, weaknesses include uneven meteorological station distribution and short time series (many starting ~2000), potentially biasing regional trends; what do the authors think about this? The hypothesis on ocean stabilization could be more testable with quantitative metrics; and missing controls for urban heat island effects in station data.

Response 1: For those meteorological stations for which it was possible, an analysis was carried out for an earlier period. We also used several long-term weather stations Natal1, Porto Alegre, which allowed comparison with the ERA5 reanalysis for 1964–2023. Differences were identified between temperature data from the ERA5 reanalysis at weather stations in the southern, northern, and northeastern regions. Temperatures at weather stations in the southern and northern regions show insignificant or negative correlations, while strong correlations were found between temperatures at weather stations in the northern and northeastern regions. Temperature trends at weather stations in the northern and northeastern regions also coincide, with a negative trend observed in the southern region.

The hypothesis of a stabilizing role for the ocean is supported by the results of CMIP6 models, which do not use observational data from stations. Accordingly, they are not affected by urban heat islands. Quantitative metrics are illustrated in the form of figures and are given in the text of the manuscript.

Comments 2: Line 194: "Monthly mean NSAT data were selected" – justify why not seasonal breakdowns, as tropical regions like Brazil show strong wet/dry season variability.

Response 2: We are planning to explore seasonality in future studies.

Comments 3: Line 242: Figure 2 trends – unclear if trends are statistically significant; add p-values or confidence intervals. Would it be possible to add a dispersion measure?

Response 3: We've added a trend significance assessment using the Mann-Kendall test. We calculated the standard deviation based on average monthly temperature values at weather stations and the ERA5 reanalysis. However, adding ±σ to the figures significantly clutters them.

Comments 4: It would be interesting to include a discussion of the practical applications of the results obtained, such as their implications for public climate adaptation policies in Brazil (e.g., in agricultural planning, water resource management, and mitigation of extreme events such as heat waves and droughts, which are exacerbated by faster warming over land compared to the ocean). Furthermore, highlighting the importance of this study in the context of global climate change, such as its contribution to understanding the stabilizing role of the ocean and to regional projections based on CMIP6, would strengthen the manuscript.

Response 4: Added.

Comments 5: The figures could have better resolution.

Response 5: Figures have been improved.

Comments 6: Level of English: Appropriate and comprehensible; minor phrasing issues, but no major errors.

Response 6: We have improved the English of the manuscript.

Reviewer 4 Report

Comments and Suggestions for Authors

The article addresses a relevant issue of recent decades in the very large and important region. The overall design of the article is good. Nevertheless, here is a number of concrete points which could be further amended or at least explained:

-row 80-81. [15] is bringing the investigation over Northeastern Brazil. How far can be its results extended to other regions?

-since many results are presented at the regional level, brief climatic characteristics of the defined regions may provide the reader with better orientation.

-it is not fully clear what data in the analysis were exactly used. There are only 9 stations with full 2000-2023 data sets listed in Tab. 1.  Were the data gaps in the data sets filled out or the temperature averages at the stations were rather calculated from existing data? Were the station data homogenized? It seems that the station data were used only to illustrate the results of the reanalysis in a certain period of the overall assessment. This should be stated if truth.

-rows 198-210. WMO advocates 30year intervals for establishing climate normal. Nevertheless, what was the reason to select 1940-1969, 1994-2023 for the respective analysis? 30year WMO normals used by WMO are usually 1961-90 and 1991-2020.

-row 213. [44] does not clearly state the quasi 60 year climate oscillations. 60-80 year period is suggested.

-row 231-235. There is a statistically significant positive relationship between NSAT data obtained from the majority of meteorological stations and the ERA5 reanalysis. Nevertheless, 10 stations show the difference more than 0,8 °C (Teresopoli even 3.4 °C). This fact needs to be discussed in terms of its possible impact on the results of further analyses.

-Fig. 2. What was the criteria for the station selection? Are the stations with no trend or negative trend representative enough to be considered in the analysis? What are the trends in the stations not showing the trend lines?

-Fig. 3 shows annual anomalies for Brazil. Which ERA5 data were aggregated in this analysis? All ERA5 grids or grids containing the station location? Other?

-There are 7 stations listed in Table 2 with mean NSAT in the period 1964-93 while only 2 of them show the means in Fig 2. from 1964. The datasets of the rest of stations are shorter in Fig. 2. What is the reason for this difference? Which stations do have the real data from 1964?

-Table 2. It is not clear what the station data in the period 1964-93 represent. Northeastern region is represented by Pao de Acucar station in Table 2 while “meteorological stations” are mentioned in the text (row 254). One station per region or more stations were used in this analysis?

-the legenda over column 3 in Table 3 is confusing. What does “Historical and SSP2-4.5 mean/represent?

-row 417-418. The maxima in 1987 are not expressed in all stations. It is probably better to list only 1998.

Overall text, though clearly expressed, suffers with stressing very many details which is in some cases confusing. The analysis is performed in a good way, it states the facts found but it does not take a clear stance towards some “exceptions” from the general trend (negative trend of NSAT in southern part of Brasilia based on meteorological station data). Further to that it is not clear if this finding is based on one station or on the analysis of a set of stations. The title of the article announces the NSAT analysis in Brazil regions but there is no definition of the regions. Official geographic regions are used in some cases (Table 2) while for example Amazon region is used to show the region with most pronounced warming. Beside this fact, Amazon region includes vast areas of wetlands which should rather indicate higher thermal stability. Such findings should be commented.

The conclusions are quite wordy, often stating things already stated in the discussion. The first part of the conclusions brings relatively few cumulative conclusions. Nevertheless, the last two paragraphs of the conclusions bring a valuable cumulative summary.

Formal mistakes

-Pan-di-Asukare is listed in Fig. 2 is listed under different name in Tab. 1.

-There are not trend lines in Fig. 2b in manuscript provided to reviewers and even not all stations in Fig. 2a show the trend line.

-row 286-287. This statement in this sentence can`t be derived from Fig. 1.

-please check the language for typos (although there are very few, e.g. line 107 … to or…)

Author Response

Response to Reviewer 3 Comments

The authors are grateful to the reviewer for the correct comments that allowed us to improve the text of the manuscript.

Comments 1: row 80-81. [15] is bringing the investigation over Northeastern Brazil. How far can be its results extended to other regions?

Response 1: We believe that the results of work 15 can be extended to the entire Brazilian region.

Comments 2: since many results are presented at the regional level, brief climatic characteristics of the defined regions may provide the reader with better orientation.

Response 2: Added a description of the region.

Comments 3: it is not fully clear what data in the analysis were exactly used. There are only 9 stations with full 2000-2023 data sets listed in Tab. 1.  Were the data gaps in the data sets filled out or the temperature averages at the stations were rather calculated from existing data? Were the station data homogenized? It seems that the station data were used only to illustrate the results of the reanalysis in a certain period of the overall assessment. This should be stated if truth.

Response 3: The temperature averages at the stations were rather calculated from existing data.

Comments 4: rows 198-210. WMO advocates 30year intervals for establishing climate normal. Nevertheless, what was the reason to select 1940-1969, 1994-2023 for the respective analysis? 30year WMO normals used by WMO are usually 1961-90 and 1991-2020.

Response 4: Yes, the norms 1961-90 and 1991-2020are usually used, but we decided to use all available data starting from 1940, so we took the first 30 as the norm.

Comments 5: row 213. [44] does not clearly state the quasi 60 year climate oscillations. 60-80 year period is suggested.

Response 5: Changed.

Comments 6: row 231-235. There is a statistically significant positive relationship between NSAT data obtained from the majority of meteorological stations and the ERA5 reanalysis. Nevertheless, 10 stations show the difference more than 0,8 °C (Teresopoli even 3.4 °C). This fact needs to be discussed in terms of its possible impact on the results of further analyses.

Response 6: Yes, it is observed. Described in the text.

Comments 7: Fig. 2. What was the criteria for the station selection? Are the stations with no trend or negative trend representative enough to be considered in the analysis? What are the trends in the stations not showing the trend lines?

Response 7: The main criterion for selecting stations was the longest data series.

Comments 8: Fig. 3 shows annual anomalies for Brazil. Which ERA5 data were aggregated in this analysis? All ERA5 grids or grids containing the station location? Other?

Response 8: All ERA5 grids or grids containing the station location

Comments 9: There are 7 stations listed in Table 2 with mean NSAT in the period 1964-93 while only 2 of them show the means in Fig 2. from 1964. The datasets of the rest of stations are shorter in Fig. 2. What is the reason for this difference? Which stations do have the real data from 1964?

Response 9: Natal 1, Porto Alegre.

Comments 10: Table 2. It is not clear what the station data in the period 1964-93 represent. Northeastern region is represented by Pao de Acucar station in Table 2 while “meteorological stations” are mentioned in the text (row 254). One station per region or more stations were used in this analysis?

Response 10: Corrected.

Comments 11: the legenda over column 3 in Table 3 is confusing. What does “Historical and SSP2-4.5 mean/represent?

Response 11: This means that data before 2014 is taken from Historical, and data after 2015 is taken from SSP2-4.5.

Comments 12: row 417-418. The maxima in 1987 are not expressed in all stations. It is probably better to list only 1998.

Response 12: Removed.

Comments 13: Overall text, though clearly expressed, suffers with stressing very many details which is in some cases confusing. The analysis is performed in a good way, it states the facts found but it does not take a clear stance towards some “exceptions” from the general trend (negative trend of NSAT in southern part of Brasilia based on meteorological station data). Further to that it is not clear if this finding is based on one station or on the analysis of a set of stations. The title of the article announces the NSAT analysis in Brazil regions but there is no definition of the regions. Official geographic regions are used in some cases (Table 2) while for example Amazon region is used to show the region with most pronounced warming. Beside this fact, Amazon region includes vast areas of wetlands which should rather indicate higher thermal stability. Such findings should be commented.

Response 13: In conclusion, a phrase was added about temperature differences in different regions. Apparently, the Amazon River basin's strongest warming, as observed in the models, is due to its distance from the ocean, which plays a stabilizing role. Since moisture in the Amazon region is brought from the ocean, and its condensation releases latent heat, which also comes from the ocean, the high humidity and marshy conditions of this region do not provide the same strong stabilizing factor as the heat capacity of the upper active layer of the ocean.

Comments 14: The conclusions are quite wordy, often stating things already stated in the discussion. The first part of the conclusions brings relatively few cumulative conclusions. Nevertheless, the last two paragraphs of the conclusions bring a valuable cumulative summary.

Response 14: The conclusions are abbreviated.

Formal mistakes

Comments 15: Pan-di-Asukare is listed in Fig. 2 is listed under different name in Tab. 1.

Response 15: Added

Comments 16: There are not trend lines in Fig. 2b in manuscript provided to reviewers and even not all stations in Fig. 2a show the trend line.

Response 16: Added

Comments 17: row 286-287. This statement in this sentence can`t be derived from Fig. 1.

Response 17: Corrected

Comments 18: please check the language for typos (although there are very few, e.g. line 107 … to or…)

Response 18: Checked

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors responded to all comments and made the recommended revisions.

Reviewer 3 Report

Comments and Suggestions for Authors

The current version of the manuscript has been substantially improved, with significant enhancements in scientific clarity, analytical rigor, and structural coherence. The manuscript fully meets the publication standards of Climate and is recommended for acceptance in its present form, following only minor copyediting and final formatting by the editorial team.