Analysis of the Space–Temporal Trends of Wet Conditions in the Different Rainy Seasons of Brazilian Northeast by Quantile Regression and Bootstrap Test

Round 1

Reviewer 1 Report

This study concerns the space-time trends in wetness conditions in the North-East part of Brazil (NEB). It is based on 3-month time scale Standardized Precipitation Evapotranspiration Index (SPEI-3) calculated from precipitation and potential evapotranspiration data at the resolution of 0.25° x 0.25° for the period between 1981 and 2013. The study aims to identify the SPEI trends estimated by quantile regression (QR). While the subject is of great importance for detecting the temporal regularities in the wetness conditions at regional scale, I believe that the current version of the manuscript lacks clarity and precision in statements, and should be improved before considering for publication. Below are my comments:

1. The "Abstract" does not fully reflect the content of the study. From the sentence "The purpose of this research was to investigate drought trends in Northeast Brazil (NEB) in the quarters of highest precipitation, from a new drought index, considering an alternative statistical approach" it might be understood that a new drought index is developed (which is not the case). Moreover, it seems that it would be more adequate to mention "trends in wetness conditions" instead of "drought trends", as in the manuscript also wet conditions are analyzed. Here, the grid analysis is not conducted, so mentioning "...from a high quality grid analysis" is misleading. Besides, it is not clear what exactly is meant by "...coherence with future NEB scenarios". Please do not use the abbreviations that are not explained.

2. The title (Space-Time Trends of Dry and Wet Atmospheric  Conditions in Northeast Brazil) seems to be too general. Neither name of the wetness index nor specific method is mentioned. Moreover, by using SPEI index, not only atmospheric conditions can be characterized. Actually only temporal trends of a regional space seem to be  analyzed.

3. The "Introduction" section should be improved.  Line 46: The symbol of ET is not explained when used for the first time. As in Line 85 (next section) another symbol (PET) is used for potential evapotranspiration, it is suggested to unify symbols in the whole manuscript. Lines 56-61: Exact description of the bootstrap method should be moved to the "Materials and Methods" section. In line 66, the symbol of ET is used again, and not explained that You use potential evapotranspiration for the calculation of  SPEI. Leaving symbols like it is, one could have wrong impression that ET is just evapotranspiration and not potential evapotranspiration (PET).

4. Figure 1: Here, You present the distribution of  1,794 NEB municipalities for which analysis is conducted. Please explain in the methods section how do You obtain  SPEI values for municipalities as the original set of weather grid data has a resolution of (0.25x0.25), so rather for regular grid network. An explanation would be useful if any extraction of SPEI grid data was done to obtain values for these municipalities. Moreover, please add a scale bar to the figure 1.

5. The "Materials and Methods" section should be improved, especially this concerns the subsection "2.3. Quantile Regression (QR)".

Lines 95-97: The abbreviations (FMA and MJJ) should be consequently explained, as You did it for NDJ.

Lines 98: what is meant by "...quarterly or SPEI-3"?

Lines 99-101: the sentence " For each of these areas, in addition to the spatial analysis, were studied the trends of a significant municipality: Barreiras, in the Midwest area; Joăo Pessoa, in the East area; and Fortaleza, in the North of NEB (Figure 1a)." is not clear.

Lines 110-111: The Authors mentioned that "The method is briefly explained here for illustrative purposes." Although, an explanation is not exact. Line 112: what is meant by Y as a random variable; please indicate that Y indicates the SPEI... Please explain Xi, T, m. The text given in lines 115-116, and eq. 5 should improved considering the merit.

6. The "Results and Discussion" section should be improved.

Line 140: "municipalities" or rather centers of municipalities??

Figures 3, 5 & 7: Axes should be given the titles. Please use symbols which You use in equations or in the text of Manuscript. Please add explanation for the black solid line. The time series seems to be of 36 years, and not 34 years (1980-2013, lines 70-71), please clear it up.

Figures 2, 4 & 6: Please add scale bar, and enlarge the text.

Lines 197-204: This text better fits for the Introduction.

Lines 205-206: This conclusion seems to be trivial.

7. The "Conclusions" section needs improvements. I suggest to rework the conclusions. Such formulations as e.g.: "...easier to interpret from the social and environmental impact point of view", or "The choice of the 3-months (SPEI-3) time scale was an important consideration, as meteorological drought generally responds on this scale." are too general, and need re-formulation.

Author Response

Dear Reviewer, We are very grateful for the considerations on the article. We believe we have been able to answer most of the suggestions for manuscript improvement. The title has been redefined incorporating important details that allow the reader to be aware of what has been worked on in the article, as well as the abstract has been rewritten in order to tend his suggestions. The Introduction has been improved as suggested. Every time evapotranspiration is written, the acronym PET has been corrected. Regarding the grid used, the form of data extraction for the geographic coordinates of the municipalities was explained, as well as the analysis period was corrected, from 1980 to 2015, instead of from 1980 to 2013. The requested scale bar was added to Figure 1 in the framework of municipalities. In the Materials and Methods section, in addition to these adjustments already mentioned, the explanation of the item "2.3. Quantum Regression (QR)" was improved, as suggested. All abbreviations were properly explained, as well as the meaning of the SPI index and its time scale 3, used in the research. The results and discussions were improved by adding more discussion text about the implication of the results, the figures were also redone, adding subtitles on the axes and improving the resolution for better visualization. Regarding the conclusions, the authors believe that what was written is sufficient to synthesize the results obtained. In short, everything that was suggested was accepted, making the quality of the article much better.

Reviewer 2 Report

Abstract:

L18: Did not define NDJ anywhere earlier. 

Introduction:

L33: You mean surface water TEMPERATURE anomalies? 

LL33-36: It's confusing. The sentence does not quite make sense "NEB is also influenced by rainfall associated with large-scale dynamics..." 

- If starting a sentence with a citation then it's better to refer to the author, date as well, such as in L41, it could read as Mckee et al. (1993) [21] developed ... This is the case at many other places in the article. 

 - There is no reference to any study that used SPEI. It would give more weights to this work if you cite some papers that used SPEI. There are plenty.

 - I do not understand why 'humidity' is used to explain the results. Humidity is not the same as SPEI. 

 - There are no x and y labels on Figures 3, 5 and 7. 

 Conclusion:

LL209-210: there is a mention of social and environmental impact but in the results and earlier in the paper there are no such discussion on social and environmental impacts.

- The authors did not discuss the implications of this work.  

Author Response

Dear Reviewer, We are very grateful for the considerations on the article. We believe we can meet the recommendations. All acronyms were well defined in the text. In the Introduction, all items were taken care of, new references were included, and the whole part concerning rainfall dynamics was rewritten, as suggested. References from classic and recent works using SPEI were added, as defined in the quantile regression equation. The axes of Figures 3, 5 and 7 were placed. The implications for the social and environmental impacts of droughts were placed at the end of the discussion of the results, as well as the implications of the results obtained in the research. In short, everything that was suggested was accepted, making the quality of the article much better.

Round 2

Reviewer 1 Report

1. Please correct the period of analysis mentioned in the Abstract.

It should be "...between 1980 and 2015" instead of 

                  "...between 1980 and 2013.

2. Lines 35-37: Please correct the sentence: "Droughts are natural phenomena, but studies indicate that for positive trends in the intensity and magnitude of events, which may be associated with accelerated deforestation of the caatinga, which contributes to increased temperatures, and consequently PET [7]."

3. Line 53: Please use capital letters when mentioning "...north atlantic oscillation, pacific decadal oscillation, atlantic multidecadal oscillation..."

4. Lines 128-130: when using "Altitude", please add "m a.s.l."

Author Response

Dear Reviewer, Thank you for submitting your reviews. All have been answered and are highlighted in red in the text.

Regards,

Fabricio

Corresponding author