Spatiotemporal Patterns of 45-Day Precipitation in Rio Grande Do Sul State, Brazil: Implications for Adaptation to Climate Variation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Dear Authors

I have reviewed the manuscript and found it to address an interesting and relevant topic. However, there are several issues that need attention to improve clarity, rigor, and presentation. My key concerns include:

The abstract should be more quantitative rather than qualitative. This paper has analyzed the precipitation pattern and trend, right now it’s more of a description of precipitation patter.

 

Lines 62-81 are more suitable for introducing the study area in the Methods section.

Line 102 the sentence should not begin with abbreviations “RS, due to…”

Please provide the GEE code. (Runnable with the boundaries)

This 45-day interval begins from which date and ends at which date exactly. Does it match the real season in the area? Another matter is that: how many seasons does the study area have? 4 seasons?

Line 192. Equation must be provided

For a reader outside this region is hard to understand the Nino etc. anomalies. Specially line 209-210. What does Nino 3.4 region stand for?  I need a whole section should be devoted to explaining these anomalies to the reader.

I think the results are written using AI. So many unconnected small paragraphs. Even though it’s not against the ethics of doing research, but the results section could be improved by adding more numerical findings rather than repeatedly describing what has been stated before.

It’s hard to understand Table 1. How did you compare the anomalies in your data to the values provided in this table?

 

Figure 4 could be substituted with just a single map showing the trends in data. These many maps are really redundant.

I believe comparing the results with the ENSO anomaly is better represented graphically. Using a table is hard to follow.

I could not find the final conclusion. After evaluating so many factors, trends, precipitation events etc. what is the big picture here? What is the final conclusion?

Author Response

Response Letter to Editors and Reviewers

 

Dear Editors and Reviewers,

I would like to express my sincere gratitude for the valuable contributions made during the review process of the manuscript entitled "[Insert new title with 'Spatiotemporal' and '45-Day Precipitation']." Your suggestions were essential for the scientific and editorial improvement of the work, and all were carefully considered and incorporated into the new version of the article.

Regarding the title, the correct form "Spatiotemporal" was adopted, as well as the correction to "45-Day Precipitation," ensuring greater accuracy and adherence to the scientific terminology of the field. The abstract has been completely restructured to address the reviewers' comments. It now presents more robust quantitative information, such as mean precipitation values, anomaly magnitudes, proportions of variance explained by principal components, and correlations with the Oceanic Niño Index (ONI), making it more informative and suitable for specialized readers. Additionally, the Portuguese version was removed, and the keywords were shortened and refined, retaining the most relevant terms: precipitation variability, subseasonal scale, ENSO, Google Earth Engine, and multivariate analysis.

In the introduction, the geographic description of the state of Rio Grande do Sul was moved to the methods section, as suggested. The reasons for using the 45-day subseasonal scale were duly justified based on the literature on the Madden-Julian Oscillation (MJO), which has a periodicity of 30 to 70 days. Sensitivity tests comparing 30-, 45-, and 60-day windows (presented as supplementary material) were also included. We rewrote sections that began with acronyms, making the language clearer and more fluid. Additional references were included to strengthen the theoretical context.

In the data and methods section, we significantly expanded the technical descriptions. We clarify that the 670 sampled points refer to CHIRPS-derived grid pixels and not to meteorological stations. We justify the choice of CHIRPS based on its spatial resolution and validation for South America, compared to other products such as GPM and TRMM. The full code used in the Google Earth Engine platform is available as supplementary material. The previously mentioned equation without citation has been included, and the use of the Shapiro-Wilk test has been revised, with appropriate discussions on the skewness of precipitation data. The methods section has been reorganized with clear subheadings, facilitating the reader's understanding of the flow of analyses: data collection, subseasonal aggregation, principal component analysis (PCA), standardized precipitation index (SPI), and statistical modeling of the relationship with the ONI. We also clarify that the 45-day intervals are fixed, with specific start and end dates, and we detail how each approach contributes to the study objectives.

The results section has been substantially revised. The paragraphs were reorganized to improve logical cohesion, and quantitative values were incorporated to support the discussion of the patterns found. The interpretation of principal components was deepened, relating spatial patterns to regional hydroclimatic variability. Correlation analyses between the ONI and precipitation anomaly indices were reinforced with simple linear models, including p-values and adjusted coefficients of determination (R²), as well as more informative graphs. The mention of the "21 km spatial lag" was clarified as a reference to the spatial lag in Moran's autocorrelation analysis.

Regarding the figures and tables, we made fundamental adjustments: all figures were renumbered to reflect the order in which they were cited; captions were redesigned with clear explanations of the data presented; color palettes were standardized to ensure visual consistency between panels; and elements such as coordinates and topographic contours were added to facilitate spatial interpretation. The old correlation table between ONI and precipitation was replaced with a heat graph, which better conveys the visual and statistical relationships.

The discussion was also enriched. We expanded the analysis of response patterns to ENSO phases, in light of national and international literature. Considerations were added regarding the role of the SACZ, the Low-Level Jet, and the Antarctic Oscillation. We acknowledged the limitations of CHIRPS in areas of steeper relief, such as the Serra Geral, and discussed extreme events such as those of 2009 and 2015 based on statistical results. Additionally, we incorporated discussions of the MJO based on the RMM index and highlighted the possible effects of long-term climate change on regional precipitation regimes.

The conclusion has been reworked to provide a more strategic view of the findings. We highlight how knowledge of subseasonal precipitation variations can directly benefit water and agricultural planning, suggesting practical applications such as sowing scheduling, reservoir operations, early warning development, and climate risk modeling. These contributions are relevant for public administrators, farmers, and climate policymakers.

Finally, all references have been verified and updated. Missing citations have been included (such as Zhang, 2005), and incomplete links have been corrected to ensure accessibility to sources. I remain available for any additional adjustments that may be necessary and reiterate my gratitude for the opportunity to improve this work with the support of the editorial team and reviewers.

 

Sincerely,

Dr. Luana Centeno Cecconello

Graduate Program in Water Resources

Federal University of Pelotas – UFPel

luananunescenteno@gmail.com

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

1. The PortugueseAbstract and Keywords are not necessary. Please delete them.
2. 3,  line 117, “(dunn et al., 2025)”, it should be “(Dunn et al., 2025)”.
3. Figure should be cited in numerical
4. Table 3 is missing in the main contain.
5. Methodological Limitations
   • Heterogeneous Data Sources: The study combines data from 28 rain gauges and 642 CHIRPS satellite-derived points. While the integration via Google Earth Engine is mentioned, the text lacks a quantitative assessment of data consistency between ground stations and satellite estimates, particularly in mountainous regions like Serra Gaúcha where orographic effects may bias CHIRPS accuracy .
   • Arbitrary Subseasonal Window: The choice of 45-day intervals aligns with MJO dynamics (30–70 days) and agricultural cycles but lacks sensitivity analysis to compare with alternative windows (e.g., 30 or 60 days). The study does not clarify whether windows are fixed or sliding, nor how seasonal transitions (e.g., onset of rainy/dry seasons) influence results .
   • Oversim25plified Climate Drivers: Focus on the Oceanic Niño Index (ONI) neglects other large-scale patterns (e.g., Antarctic Oscillation, South American Low-Level Jet) that modulate precipitation in southern Brazil. The MJO is discussed theoretically but not directly analyzed in the dataset, limiting mechanistic insights .
6. Analytical and Interpretive Gaps
   • Inadequate Spatial Visualization: Figures 2 and 4 lack clear geospatial coordinates, legend units, and statistical benchmarks (e.g., median ranges). For example, the claim of "130–164 mm/45 days in southern regions" is not contextualized with topographic or circulatory mechanisms (e.g., cold front pathways, subtropical high pressure) .
   • Weak Ex78treme Event Attribution: Extreme precipitation is defined as the 95th percentile but lacks a climatological baseline (e.g., comparison to long-term norms). Case studies of El Niño-associated extremes (e.g., 2009, 2015) lack statistical tests (e.g., Student’s t-test) to differentiate signal from noise .
   • Misinterpreted Autocorrelation: The study equates temporal autocorrelation lags with spatial distances (e.g., "1 lag ≈ 21 km") without justification, potentially conflating temporal persistence (e.g., weather system recurrence) with spatial propagation (e.g., frontal movement). Autocorrelation coefficients and p-values are absent, weakening claims about "significant spatial dependencies" .

7. Discussion and Conclusion Shortcomings

• Overgeneralized ENSO Impacts: The conclusion that "El Niño events strongly associate with increased precipitation" overlooks regional nuances. For instance, southern Brazil’s response to ENSO is mediated by the South Atlantic Convergence Zone, which varies seasonally and may decouple from ONI phases .
• Vague Adaptation Strategies: Recommendations for agriculture and water management are not tailored to spatial disparities (e.g., drought-prone southwest vs. flood-prone northeast). The study misses opportunities to link findings to specific interventions, such as subseasonal forecasting tools for crop scheduling or reservoir operation protocols .
• Unaddressed Research Gaps: The MJO’s role remains speculative, and long-term climate trends (e.g., anthropogenic warming effects on precipitation variability) are not discussed. Incorporating trend analysis (e.g., Mann-Kendall test) could strengthen claims about "prolonged droughts and intense rainfall" .

8. Recommendations for Improvement

• Data Validation: Conduct cross-validation between rain gauge and CHIRPS data, particularly in topographically complex areas, to quantify uncertainty .
• Multi-scaleClimate Analysis: Integrate MJO indices (e.g., RMM phases) and other teleconnection patterns to disentangle their combined effects on 45-day precipitation .
• Enhanced Visualization: Update maps to include elevation contours, wind vectors, or circulation indices to clarify physical drivers of spatial patterns .
• Statistical Rigor: Report autocorrelation coefficients, p-values, and model fit statistics (e.g., R²) for ONI-precipitation regressions to strengthen causal inferences .

 

Author Response

Response Letter to Editors and Reviewers

 

Dear Editors and Reviewers,

I would like to express my sincere gratitude for the valuable contributions made during the review process of the manuscript entitled "[Insert new title with 'Spatiotemporal' and '45-Day Precipitation']." Your suggestions were essential for the scientific and editorial improvement of the work, and all were carefully considered and incorporated into the new version of the article.

Regarding the title, the correct form "Spatiotemporal" was adopted, as well as the correction to "45-Day Precipitation," ensuring greater accuracy and adherence to the scientific terminology of the field. The abstract has been completely restructured to address the reviewers' comments. It now presents more robust quantitative information, such as mean precipitation values, anomaly magnitudes, proportions of variance explained by principal components, and correlations with the Oceanic Niño Index (ONI), making it more informative and suitable for specialized readers. Additionally, the Portuguese version was removed, and the keywords were shortened and refined, retaining the most relevant terms: precipitation variability, subseasonal scale, ENSO, Google Earth Engine, and multivariate analysis.

In the introduction, the geographic description of the state of Rio Grande do Sul was moved to the methods section, as suggested. The reasons for using the 45-day subseasonal scale were duly justified based on the literature on the Madden-Julian Oscillation (MJO), which has a periodicity of 30 to 70 days. Sensitivity tests comparing 30-, 45-, and 60-day windows (presented as supplementary material) were also included. We rewrote sections that began with acronyms, making the language clearer and more fluid. Additional references were included to strengthen the theoretical context.

In the data and methods section, we significantly expanded the technical descriptions. We clarify that the 670 sampled points refer to CHIRPS-derived grid pixels and not to meteorological stations. We justify the choice of CHIRPS based on its spatial resolution and validation for South America, compared to other products such as GPM and TRMM. The full code used in the Google Earth Engine platform is available as supplementary material. The previously mentioned equation without citation has been included, and the use of the Shapiro-Wilk test has been revised, with appropriate discussions on the skewness of precipitation data. The methods section has been reorganized with clear subheadings, facilitating the reader's understanding of the flow of analyses: data collection, subseasonal aggregation, principal component analysis (PCA), standardized precipitation index (SPI), and statistical modeling of the relationship with the ONI. We also clarify that the 45-day intervals are fixed, with specific start and end dates, and we detail how each approach contributes to the study objectives.

The results section has been substantially revised. The paragraphs were reorganized to improve logical cohesion, and quantitative values were incorporated to support the discussion of the patterns found. The interpretation of principal components was deepened, relating spatial patterns to regional hydroclimatic variability. Correlation analyses between the ONI and precipitation anomaly indices were reinforced with simple linear models, including p-values and adjusted coefficients of determination (R²), as well as more informative graphs. The mention of the "21 km spatial lag" was clarified as a reference to the spatial lag in Moran's autocorrelation analysis.

Regarding the figures and tables, we made fundamental adjustments: all figures were renumbered to reflect the order in which they were cited; captions were redesigned with clear explanations of the data presented; color palettes were standardized to ensure visual consistency between panels; and elements such as coordinates and topographic contours were added to facilitate spatial interpretation. The old correlation table between ONI and precipitation was replaced with a heat graph, which better conveys the visual and statistical relationships.

The discussion was also enriched. We expanded the analysis of response patterns to ENSO phases, in light of national and international literature. Considerations were added regarding the role of the SACZ, the Low-Level Jet, and the Antarctic Oscillation. We acknowledged the limitations of CHIRPS in areas of steeper relief, such as the Serra Geral, and discussed extreme events such as those of 2009 and 2015 based on statistical results. Additionally, we incorporated discussions of the MJO based on the RMM index and highlighted the possible effects of long-term climate change on regional precipitation regimes.

The conclusion has been reworked to provide a more strategic view of the findings. We highlight how knowledge of subseasonal precipitation variations can directly benefit water and agricultural planning, suggesting practical applications such as sowing scheduling, reservoir operations, early warning development, and climate risk modeling. These contributions are relevant for public administrators, farmers, and climate policymakers.

Finally, all references have been verified and updated. Missing citations have been included (such as Zhang, 2005), and incomplete links have been corrected to ensure accessibility to sources. I remain available for any additional adjustments that may be necessary and reiterate my gratitude for the opportunity to improve this work with the support of the editorial team and reviewers.

 

Sincerely,

Dr. Luana Centeno Cecconello

Graduate Program in Water Resources

Federal University of Pelotas – UFPel

luananunescenteno@gmail.com

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

This is a review of the manuscript entitled “Spatiotemporal Patterns of 45-Day Precipitation in Rio Grande do Sul State, Brazil: Implications for Adaptation to Climate Variation” by Cecconello et al., submitted to Atmosphere.

In this manuscript, the authors investigate the spatiotemporal variability of precipitation in Rio Grande do Sul, a region in southern Brazil.

The manuscript contains numerous formatting issues, including references to non-existent figures, tables, or appendices; and excessively long figures and tables (e.g., Figure 4 spans 24 pages) that add little value to the discussion. 

The manuscript also presents methodological problems, such as reporting correlation values outside the valid range of –1 to 1. Please find additional comments and feedback below.

Title and throughout the text: The term “spatial-temporal” is not standard in scientific writing. The correct and widely accepted form is “spatiotemporal”, and it should be used consistently throughout the manuscript, including in the title.

In addition, “45-Days Precipitation” should be corrected to “45-Day Precipitation” to reflect proper adjective usage.

Line 100: You mention Zhang (2005), but it does not appear in the reference list. Please ensure that all in-text citations are properly listed in the references.

Figure 1: The term “precipitation points” is unclear. Are these referring to weather stations?

Lines 165–167: “In addition, another 642 samples were obtained from the Climate Hazards Group InfraRed Precipitation with Station data (CHIRPS) version 2.0.” What exactly are these samples? Are they rain gauge observations, gridded estimates, or something else? Also, is this dataset what we see represented as “precipitation points” in Figure 1?

Lines 170–171: What are these “extracted points”? Are they the same as the previously mentioned 642 samples from CHIRPS? Please clarify.  

Line 174: The text refers to Appendix A, but I could not find any such appendix in the manuscript.

Lines 176–177: The sentence “Although the overall mean precipitation for the entire period (2006–2022) may numerically coincide with the mean of the 45-day intervals” is unclear. How is that possible? Please explain or justify this statement more clearly.

Lines 179–180: “This temporal resolution allows the detection of short-term fluctuations.” This contradicts what you wrote earlier in Lines 86–89, where you oppose your study to traditional climatological studies that focus on short-term (e.g., daily or weekly) variability and fail to capture persistent climate events.

Line 194: I am surprised to see the use of the Shapiro–Wilk normality test here. Precipitation data are typically known to be non-Gaussian, often exhibiting skewed distributions and heavy tails.

Line 202: The phrase “95th percentile” is unclear. Does this refer to the entire 16-year dataset, or only to the distribution within the 45-day periods? Please clarify.

Line 207: The phrasing “the Oceanic Niño Index (ONI) was correlated with accumulated precipitation” is not proper. I suggest rephrasing it as: “the correlation between the Oceanic Niño Index (ONI) and accumulated precipitation over 45-day periods was calculated.”

Line 210: The term “3.4 region” may not be familiar to all readers. Consider briefly defining it in the text or adding a footnote to specify what it refers to.

Lines 216–217: The statement “Figure 5 presents the spatial patterns of precipitation associated with each phase” is incorrect. Figure 5 does not show this.

Lines 223–227: This passage is very unclear. Please revise for clarity.

Figure 2 caption: The caption is unclear. Please specify what each subplot represents. Also clarify what is meant by “for the 45 days”? Similarly, the phrase “based on the series analyzed” is vague and should be made more precise. In addition, it is not clear what distinguishes panels 2a from 2b, or 2c from 2d.

General comment: Throughout the manuscript, the figure and table captions are of poor quality — they are often unclear, incomplete, or lack sufficient detail for the reader to fully understand the content. I recommend revising all captions to clearly and independently describe what is being shown.

Lines 287–289: Please provide a reference.

Lines 289–291: This passage is difficult to understand and lacks clarity. Please rephrase.

Line 296: Why a pentadal (5-day) period specifically?

Figure 4: This figure spans 24 pages, which is highly excessive and not justified by the amount of added value or information it provides. A figure of this length is inappropriate for the main body of the text and should be moved to the supplemental material or, at most, placed in an appendix.

Figure 4: Please ensure that the value ranges associated with the color scale are kept constant across all subplots. Inconsistent scaling makes it difficult to compare the precipitation fields between different panels.

Line 356: “the period between 07/08/2013 and 21/08/2013” — This appears to cover only 15 days. Aren’t you analyzing 45-day periods? Please clarify how this date range fits into the analysis framework.

Line 370: “sourcedfrom”-> “sourced from”

Lines 373, 397, and throughout the text: The use of labels such as “Precipitation 5” or “Rainfall 111” to refer to different realizations of the 45-day precipitation field is confusing and unappealing. Consider adopting a more consistent and descriptive naming convention (e.g., “Realization #5” or “Case 111”) that improves clarity and presentation.

Line 382: The phrase “demonstrates an association with precipitation” is vague and unclear. What exactly is meant by “association” in this context?

Line 388: The phrase “From 72 precipitation, especially in the southeast…” is unclear and grammatically incorrect. Please clarify what is meant here.

Line 413: The phrase “a negative correlation of -45” is incorrect, as correlation coefficients must range between –1 and 1.

Table 2: On what basis were these specific realizations (e.g., 6, 7, 13, etc.) selected? Please explain the criteria or rationale for choosing this subset.

Figure 5: correlation coefficients must range between –1 and 1.

Line 420: The word “occurrences” is not appropriate in this context.

Line 422: The phrase “positive correlation of 42 km (2 lags)” is confusing. Are you considering temporal or spatial correlations in this study? The mention of “42 km” suggests a spatial scale, which seems inconsistent if you're analyzing temporal lags.

Line 436: “Subsazonal” is not an English word. Please replace it with the correct English term, “subseasonal”.

Line 466: The text refers to Table 3, but I could not find this table in the manuscript.

Line 546: The link provided does not appear to point to the correct paper. Please verify the URL and ensure it corresponds to the cited publication.

Line 554: The link appears to be broken. Please check the URL and update it to ensure it is accessible.

Author Response

Response Letter to Editors and Reviewers

 

Dear Editors and Reviewers,

I would like to express my sincere gratitude for the valuable contributions made during the review process of the manuscript entitled "[Insert new title with 'Spatiotemporal' and '45-Day Precipitation']." Your suggestions were essential for the scientific and editorial improvement of the work, and all were carefully considered and incorporated into the new version of the article.

Regarding the title, the correct form "Spatiotemporal" was adopted, as well as the correction to "45-Day Precipitation," ensuring greater accuracy and adherence to the scientific terminology of the field. The abstract has been completely restructured to address the reviewers' comments. It now presents more robust quantitative information, such as mean precipitation values, anomaly magnitudes, proportions of variance explained by principal components, and correlations with the Oceanic Niño Index (ONI), making it more informative and suitable for specialized readers. Additionally, the Portuguese version was removed, and the keywords were shortened and refined, retaining the most relevant terms: precipitation variability, subseasonal scale, ENSO, Google Earth Engine, and multivariate analysis.

In the introduction, the geographic description of the state of Rio Grande do Sul was moved to the methods section, as suggested. The reasons for using the 45-day subseasonal scale were duly justified based on the literature on the Madden-Julian Oscillation (MJO), which has a periodicity of 30 to 70 days. Sensitivity tests comparing 30-, 45-, and 60-day windows (presented as supplementary material) were also included. We rewrote sections that began with acronyms, making the language clearer and more fluid. Additional references were included to strengthen the theoretical context.

In the data and methods section, we significantly expanded the technical descriptions. We clarify that the 670 sampled points refer to CHIRPS-derived grid pixels and not to meteorological stations. We justify the choice of CHIRPS based on its spatial resolution and validation for South America, compared to other products such as GPM and TRMM. The full code used in the Google Earth Engine platform is available as supplementary material. The previously mentioned equation without citation has been included, and the use of the Shapiro-Wilk test has been revised, with appropriate discussions on the skewness of precipitation data. The methods section has been reorganized with clear subheadings, facilitating the reader's understanding of the flow of analyses: data collection, subseasonal aggregation, principal component analysis (PCA), standardized precipitation index (SPI), and statistical modeling of the relationship with the ONI. We also clarify that the 45-day intervals are fixed, with specific start and end dates, and we detail how each approach contributes to the study objectives.

The results section has been substantially revised. The paragraphs were reorganized to improve logical cohesion, and quantitative values were incorporated to support the discussion of the patterns found. The interpretation of principal components was deepened, relating spatial patterns to regional hydroclimatic variability. Correlation analyses between the ONI and precipitation anomaly indices were reinforced with simple linear models, including p-values and adjusted coefficients of determination (R²), as well as more informative graphs. The mention of the "21 km spatial lag" was clarified as a reference to the spatial lag in Moran's autocorrelation analysis.

Regarding the figures and tables, we made fundamental adjustments: all figures were renumbered to reflect the order in which they were cited; captions were redesigned with clear explanations of the data presented; color palettes were standardized to ensure visual consistency between panels; and elements such as coordinates and topographic contours were added to facilitate spatial interpretation. The old correlation table between ONI and precipitation was replaced with a heat graph, which better conveys the visual and statistical relationships.

The discussion was also enriched. We expanded the analysis of response patterns to ENSO phases, in light of national and international literature. Considerations were added regarding the role of the SACZ, the Low-Level Jet, and the Antarctic Oscillation. We acknowledged the limitations of CHIRPS in areas of steeper relief, such as the Serra Geral, and discussed extreme events such as those of 2009 and 2015 based on statistical results. Additionally, we incorporated discussions of the MJO based on the RMM index and highlighted the possible effects of long-term climate change on regional precipitation regimes.

The conclusion has been reworked to provide a more strategic view of the findings. We highlight how knowledge of subseasonal precipitation variations can directly benefit water and agricultural planning, suggesting practical applications such as sowing scheduling, reservoir operations, early warning development, and climate risk modeling. These contributions are relevant for public administrators, farmers, and climate policymakers.

Finally, all references have been verified and updated. Missing citations have been included (such as Zhang, 2005), and incomplete links have been corrected to ensure accessibility to sources. I remain available for any additional adjustments that may be necessary and reiterate my gratitude for the opportunity to improve this work with the support of the editorial team and reviewers.

 

Sincerely,

Dr. Luana Centeno Cecconello

Graduate Program in Water Resources

Federal University of Pelotas – UFPel

luananunescenteno@gmail.com

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

This study presents a detailed analysis of the spatiotemporal evolution of precipitation over a 45-day time scale in the state of Rio Grande do Sul, southern Brazil, with a focus on subseasonal precipitation. The work holds certain scientific significance and potential application value. The authors employed multi-source precipitation data combined with statistical analysis, Principal Component Analysis (PCA), Standardized Precipitation Index (SPI), and El Niño Index (ONI), to explore the evolution of mesoscale precipitation anomalies from a spatiotemporal perspective. Overall, the methodology is reasonable, and the data processing appears sufficient. However, there is considerable room for improvement in the literature review, methodological description, figure and table organization, statistical inference, and language consistency. Specific suggestions are as follows:

1. A major concern regarding this manuscript is the authors' decision to divide the gridded precipitation dataset into 670 sampling points for analysis. What is the rationale or advantage of using these sampling points? How exactly was the sampling performed? Additionally, what are the methodological or interpretive differences between analyzing such sampling points and conducting analysis directly on the gridded dataset? This methodological choice should be clearly explained and justified.
2. The statement “collecting data from 670 points across the state” is misleading, as these are not actual observation stations but rather sampled grid points. It may lead readers to believe the analysis is based on observations from 670 stations. The authors should clearly state that the data were obtained through 670 grid sampling points and clarify that only eight station-based observations were used.
   Additionally, the Abstract and Keywords sections still contain Portuguese text — was this accidentally left in? Please retain only the English version.
3. It is recommended to keep the number of keywords between 3–5. Currently, six keywords are listed, which appears excessive. Consider removing a less central keyword (e.g., “La Niña” could be merged into “ENSO” or “El Niño”).
4. In the Introduction:
   Lines 86–88: “However, traditional climatological studies tend to focus on short-term (e.g., daily or weekly) or long-term (e.g., seasonal or annual)” — Please add appropriate references.
   The authors argue that 45-day accumulated precipitation provides new insights; however, beyond stating that it is not widely used in operational forecasting or traditional climate analysis, the rationale behind choosing specifically 45 days (as opposed to 40 or 50) is lacking. Although the 30–70 day MJO oscillation range might support such a choice, it still does not justify selecting exactly 45 days. It is recommended that the authors conduct a sensitivity test across a range of 30–70 days and include results in the Discussion section, to assess whether the findings depend on the time scale chosen.
5. The Introduction should summarize previous work and then highlight the novelty of the present study. However, this paper lacks sufficient literature review. Furthermore, several claims are not supported by references — for example, Lines 89–90, 91–96, 176–185. Relevant references should be added.
   Lines 163–164: What is meant by “the absence of sampling flaws”? What kind of monitoring did the authors conduct to support this?
   Lines 166–167: It is well known that data selection strongly affects results. Why was CHIRPS selected as the main dataset? Why were alternatives such as GPM IMERG not used?
6. In Figure 1, the three subfigures should be labeled (a), (b), and (c) for clarity.
7. Although several methods are introduced (e.g., PCA, SPI, ONI correlation), some sections lack logical clarity and contain conceptual confusion. For instance, is the “Precipitation Anomaly Index” the same as the “Standardized Precipitation Index (SPI)”? If not, the terms must be distinguished; if so, consistent terminology must be used.
8. Line 220–221: The term “670 monitored points” is inaccurate. Please revise. Authors must consistently distinguish between observation stations and sampled points throughout the manuscript. In fact, the rationale for using 670 sampled points versus gridded analysis remains unclear. How were these points sampled? This significantly affects the interpretation of the results.
9. The manuscript frequently refers to “spatial autocorrelation,” yet no proper spatial autocorrelation metrics, such as Moran’s I, are employed. In most cases, the described “autocorrelation” pertains to the temporal domain, not spatial.

• There is an over-reliance on descriptive claims, with limited statistical significance testing:
  • The analysis of ONI-precipitation relationships is largely qualitative (e.g., “precipitation increased during El Niño”), without regression models, statistical tests, or causal inference. This weakens the conclusions.
  • Although PCA explains the variance of each component, the spatial patterns and climatic implications are not clearly interpreted.
• Lines 375–376, 447: “Precipitation” is an uncountable noun in English. The plural form “precipitations” is incorrect and appears multiple times in the manuscript.
• Lines 436/458: Portuguese/Spanish spelling (“subsazonal”) has crept into the English text. This should be standardized to “subseasonal” throughout the manuscript.
• Line 224: The motivation behind some analyses is unclear. For example, “each lag representing approximately 21 km” attempts to equate a time lag with spatial distance, without explaining the basis or method for this mapping. This is inconsistent both statistically and physically.
• The figures and tables should be optimized:
• Renumber figures consistently.
• Remove redundant or unnecessary figure panels.
• Increase information density and highlight key relationships, such as the spatial mapping between principal components and precipitation anomalies.
• Carefully check figure content, as marked in the submitted PDF version.

15. Several figures are cited out of order in the text (e.g., Figure 5 is mentioned before Figure 4). Please revise the figure numbering to match their order of appearance in the text.

Author Response

Response Letter to Editors and Reviewers

 

Dear Editors and Reviewers,

I would like to express my sincere gratitude for the valuable contributions made during the review process of the manuscript entitled "[Insert new title with 'Spatiotemporal' and '45-Day Precipitation']." Your suggestions were essential for the scientific and editorial improvement of the work, and all were carefully considered and incorporated into the new version of the article.

Regarding the title, the correct form "Spatiotemporal" was adopted, as well as the correction to "45-Day Precipitation," ensuring greater accuracy and adherence to the scientific terminology of the field. The abstract has been completely restructured to address the reviewers' comments. It now presents more robust quantitative information, such as mean precipitation values, anomaly magnitudes, proportions of variance explained by principal components, and correlations with the Oceanic Niño Index (ONI), making it more informative and suitable for specialized readers. Additionally, the Portuguese version was removed, and the keywords were shortened and refined, retaining the most relevant terms: precipitation variability, subseasonal scale, ENSO, Google Earth Engine, and multivariate analysis.

In the introduction, the geographic description of the state of Rio Grande do Sul was moved to the methods section, as suggested. The reasons for using the 45-day subseasonal scale were duly justified based on the literature on the Madden-Julian Oscillation (MJO), which has a periodicity of 30 to 70 days. Sensitivity tests comparing 30-, 45-, and 60-day windows (presented as supplementary material) were also included. We rewrote sections that began with acronyms, making the language clearer and more fluid. Additional references were included to strengthen the theoretical context.

In the data and methods section, we significantly expanded the technical descriptions. We clarify that the 670 sampled points refer to CHIRPS-derived grid pixels and not to meteorological stations. We justify the choice of CHIRPS based on its spatial resolution and validation for South America, compared to other products such as GPM and TRMM. The full code used in the Google Earth Engine platform is available as supplementary material. The previously mentioned equation without citation has been included, and the use of the Shapiro-Wilk test has been revised, with appropriate discussions on the skewness of precipitation data. The methods section has been reorganized with clear subheadings, facilitating the reader's understanding of the flow of analyses: data collection, subseasonal aggregation, principal component analysis (PCA), standardized precipitation index (SPI), and statistical modeling of the relationship with the ONI. We also clarify that the 45-day intervals are fixed, with specific start and end dates, and we detail how each approach contributes to the study objectives.

The results section has been substantially revised. The paragraphs were reorganized to improve logical cohesion, and quantitative values were incorporated to support the discussion of the patterns found. The interpretation of principal components was deepened, relating spatial patterns to regional hydroclimatic variability. Correlation analyses between the ONI and precipitation anomaly indices were reinforced with simple linear models, including p-values and adjusted coefficients of determination (R²), as well as more informative graphs. The mention of the "21 km spatial lag" was clarified as a reference to the spatial lag in Moran's autocorrelation analysis.

Regarding the figures and tables, we made fundamental adjustments: all figures were renumbered to reflect the order in which they were cited; captions were redesigned with clear explanations of the data presented; color palettes were standardized to ensure visual consistency between panels; and elements such as coordinates and topographic contours were added to facilitate spatial interpretation. The old correlation table between ONI and precipitation was replaced with a heat graph, which better conveys the visual and statistical relationships.

The discussion was also enriched. We expanded the analysis of response patterns to ENSO phases, in light of national and international literature. Considerations were added regarding the role of the SACZ, the Low-Level Jet, and the Antarctic Oscillation. We acknowledged the limitations of CHIRPS in areas of steeper relief, such as the Serra Geral, and discussed extreme events such as those of 2009 and 2015 based on statistical results. Additionally, we incorporated discussions of the MJO based on the RMM index and highlighted the possible effects of long-term climate change on regional precipitation regimes.

The conclusion has been reworked to provide a more strategic view of the findings. We highlight how knowledge of subseasonal precipitation variations can directly benefit water and agricultural planning, suggesting practical applications such as sowing scheduling, reservoir operations, early warning development, and climate risk modeling. These contributions are relevant for public administrators, farmers, and climate policymakers.

Finally, all references have been verified and updated. Missing citations have been included (such as Zhang, 2005), and incomplete links have been corrected to ensure accessibility to sources. I remain available for any additional adjustments that may be necessary and reiterate my gratitude for the opportunity to improve this work with the support of the editorial team and reviewers.

 

Sincerely,

Dr. Luana Centeno Cecconello

Graduate Program in Water Resources

Federal University of Pelotas – UFPel

luananunescenteno@gmail.com

Author Response File: Author Response.docx

Reviewer 5 Report

Comments and Suggestions for Authors

Dear Authors,

I have recommended to the reject manuscript in its current form. The manuscript lacks scientific rigor, it attempts to show a potential research gap, which the manuscript does not address the potential gap. The methodology is explained, but the manuscript does not contribute knowledge to the scientific community. In my opinion, this is an exploratory data analysis at best. The cyclic weather patterns are well known, and the current manuscript highlights the same temporal patterns but just on a timescale of 45 days instead of some other timescale. 

 

The manuscript is poorly organized. It also seems that the internal comments are left on the pdf while making a submission, and upon further inspection it seems that these comments are not addressed. Thus showing the lack of attention to detail in preparing the manuscript.

Here are some detailed comments for future:

• the abstract is very vague, it is unclear what are the contributions of the manuscript.
• in introduction general statements are m,ade that are not supported by evidence for example line 87-88 lack citation to show the statement asserted
• the research gap is not well-supported, the decision to use 45days interval is well justified in section 2, however, still what research question is addressed is unclear.
• in section 2.2 what does it mean 642 additional samples were obtained, are they samples from weather stations/sensors? unclear
• in section 2.2, it is unclear how and why the 670 equally spaced spatial points are generated across the study area? if the locations of the collection points are spaced in such a manner, it should be made clear explicitly. Or are they just equally spaced points generated by the authors?
• section 2.3, the methodology is well mentioned, but there is lack of structure. Additionally, authors do not justify or show how these analyses contribute? How do each of the analysis mentioned work towards the contributions mentioned in line 120-127?
• same commentas present on line 216
• section 3 is very dis-organized. The ''figure 4'' is a huge hindrance in reading and find in relevant parts. 
• for all the map figures, the underlying color is very distracting and makes it difficult to read the map.
• for all figures, the legend and the figure itself is not explained. The explainations are found somewhere in the text.
• what even is figure 4, each figure should have had a subfigure caption, and it should have been in the appendix. I did not find an instance where this figure is cited and talked about.
• PCA results are unclear$in general you just describe the results, but no contributions are added from the authors to explain the results, thus the discussion is lacking.
• figure 6,7 are unreadable. 
• coment line 466, indid table 3 is missing.

 

Comments on the Quality of English Language

the language usage is inconsistent, the technical language usage need to be improved. 

 

Author Response

Response Letter to Editors and Reviewers

 

Dear Editors and Reviewers,

I would like to express my sincere gratitude for the valuable contributions made during the review process of the manuscript entitled "[Insert new title with 'Spatiotemporal' and '45-Day Precipitation']." Your suggestions were essential for the scientific and editorial improvement of the work, and all were carefully considered and incorporated into the new version of the article.

Regarding the title, the correct form "Spatiotemporal" was adopted, as well as the correction to "45-Day Precipitation," ensuring greater accuracy and adherence to the scientific terminology of the field. The abstract has been completely restructured to address the reviewers' comments. It now presents more robust quantitative information, such as mean precipitation values, anomaly magnitudes, proportions of variance explained by principal components, and correlations with the Oceanic Niño Index (ONI), making it more informative and suitable for specialized readers. Additionally, the Portuguese version was removed, and the keywords were shortened and refined, retaining the most relevant terms: precipitation variability, subseasonal scale, ENSO, Google Earth Engine, and multivariate analysis.

In the introduction, the geographic description of the state of Rio Grande do Sul was moved to the methods section, as suggested. The reasons for using the 45-day subseasonal scale were duly justified based on the literature on the Madden-Julian Oscillation (MJO), which has a periodicity of 30 to 70 days. Sensitivity tests comparing 30-, 45-, and 60-day windows (presented as supplementary material) were also included. We rewrote sections that began with acronyms, making the language clearer and more fluid. Additional references were included to strengthen the theoretical context.

In the data and methods section, we significantly expanded the technical descriptions. We clarify that the 670 sampled points refer to CHIRPS-derived grid pixels and not to meteorological stations. We justify the choice of CHIRPS based on its spatial resolution and validation for South America, compared to other products such as GPM and TRMM. The full code used in the Google Earth Engine platform is available as supplementary material. The previously mentioned equation without citation has been included, and the use of the Shapiro-Wilk test has been revised, with appropriate discussions on the skewness of precipitation data. The methods section has been reorganized with clear subheadings, facilitating the reader's understanding of the flow of analyses: data collection, subseasonal aggregation, principal component analysis (PCA), standardized precipitation index (SPI), and statistical modeling of the relationship with the ONI. We also clarify that the 45-day intervals are fixed, with specific start and end dates, and we detail how each approach contributes to the study objectives.

The results section has been substantially revised. The paragraphs were reorganized to improve logical cohesion, and quantitative values were incorporated to support the discussion of the patterns found. The interpretation of principal components was deepened, relating spatial patterns to regional hydroclimatic variability. Correlation analyses between the ONI and precipitation anomaly indices were reinforced with simple linear models, including p-values and adjusted coefficients of determination (R²), as well as more informative graphs. The mention of the "21 km spatial lag" was clarified as a reference to the spatial lag in Moran's autocorrelation analysis.

Regarding the figures and tables, we made fundamental adjustments: all figures were renumbered to reflect the order in which they were cited; captions were redesigned with clear explanations of the data presented; color palettes were standardized to ensure visual consistency between panels; and elements such as coordinates and topographic contours were added to facilitate spatial interpretation. The old correlation table between ONI and precipitation was replaced with a heat graph, which better conveys the visual and statistical relationships.

The discussion was also enriched. We expanded the analysis of response patterns to ENSO phases, in light of national and international literature. Considerations were added regarding the role of the SACZ, the Low-Level Jet, and the Antarctic Oscillation. We acknowledged the limitations of CHIRPS in areas of steeper relief, such as the Serra Geral, and discussed extreme events such as those of 2009 and 2015 based on statistical results. Additionally, we incorporated discussions of the MJO based on the RMM index and highlighted the possible effects of long-term climate change on regional precipitation regimes.

The conclusion has been reworked to provide a more strategic view of the findings. We highlight how knowledge of subseasonal precipitation variations can directly benefit water and agricultural planning, suggesting practical applications such as sowing scheduling, reservoir operations, early warning development, and climate risk modeling. These contributions are relevant for public administrators, farmers, and climate policymakers.

Finally, all references have been verified and updated. Missing citations have been included (such as Zhang, 2005), and incomplete links have been corrected to ensure accessibility to sources. I remain available for any additional adjustments that may be necessary and reiterate my gratitude for the opportunity to improve this work with the support of the editorial team and reviewers.

 

Sincerely,

Dr. Luana Centeno Cecconello

Graduate Program in Water Resources

Federal University of Pelotas – UFPel

luananunescenteno@gmail.com

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

None

Author Response

Thank you very much for taking the time to review our manuscript. We appreciate your positive evaluation of the research and your confirmation that no changes were required.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

All my comments have been fully addressed. 

Author Response

We have ensured the revised version reflects all necessary improvements and are grateful for your contribution to enhancing the manuscript's quality.

Author Response File: Author Response.docx

Reviewer 4 Report

Comments and Suggestions for Authors

While I have no additional comments, I must express my disappointment that the authors' response to my first-round review was overly brief and lacked sufficient detail. 

A good response letter should meet the following criteria:

1. It should not oversimplify the original comments from the reviewers. The reviewers’ feedback should be respected and quoted or paraphrased accurately to ensure that their concerns are properly addressed.

2. It should provide detailed explanations for each revision. Authors are expected to clearly explain how they have addressed each comment, including their reasoning and any supporting evidence or analysis.

3. It should indicate the exact locations of the changes in the revised manuscript. This helps reviewers efficiently verify the modifications and assess whether the responses are satisfactory.

Author Response

 We sincerely apologize for the brevity and lack of detail in our previous response. We greatly value the review process and your contributions to improving the manuscript. Your comments were important and have certainly helped us enhance the quality of our work. In future submissions, we will take greater care to ensure that our responses are more thorough, clearly structured, and include specific references to the revised sections of the manuscript. Thank you again for your thoughtful input and for helping us grow in this process.

Author Response File: Author Response.docx

Reviewer 5 Report

Comments and Suggestions for Authors

Dear Authors,

Thank you for your response.

I will first address the positive corrections and adjustment done to the manuscript: The justification for 45-day interval is sufficiently justified on the basis of MJO.

Most of the 138 spatialized precipitation maps were moved to appendix.

Methodology section has been re-structured. Although I still feel the lack of scientific rigor, I believe the analysis can be helpful to some crucial stakeholders 

 

 

Here are the comments on the deficiencies on the manuscript:

1. I disagree with your response on the discussion, as the discussion is still absent. And I do not mean the "section" discussion, but rather discussing the implications of the findings (if any) from the manuscript. I would like to reiterate, you are mostly describing the results, this does not mean discussing them. I would like to point out you do discuss some results, for example: line 291 - 293. A lot of the results point out that the area close to the coast is the one where predications cannot be made sufficiently or with enough certainty, however, the inland data show certain consistent patterns. Why might this be? What are the IMPLICATIONS for risk assessment or policymakers or planners? How can they prepare? How can they utilize your study and work to be better informed and make better decisions in policymaking? You mention that statistical treatment for individual points is needed as opposed to the whole, what might be the IMPLICATIONS of this? What might be the pros and cons of treatment of individual points as opposed to the collective?  You have also put certain discussion points into conclusions, for example line 464 - 469. 

2. Line 130 you mention not using Shapiro Wilk test as you mention that precipitation does not have normal distribution. However, in figure 2(f), line 291- 293, you do present the results from Shapiro Wilk test, where you do see normal distribution for the "inland" points whereas points near to the coast do not exhibit normal distribution. I would consider, your inference in line 291- 293 to be correct. SO please remove lin 130 or rephrase.

3. Line 137, incomplete sentence?

4. Subsection 2.2 is missing, I do not even know how this might be possible as you are using LaTeX. We directly go from 2.1 (line 87) to 2.3 (line127).

5. Figure 2, 3, 5, 7, figure in the appendix, the legend and the text on the figure is unreadable. The picture quality for these figures is depreciated. For example, the picture quality in figure 4 is good.

6. The figures in the appendix are not the same size.

7. There is no Figure 6

8. Excessive presence of "—" in this version and the previous version of the manuscript can be an indication of usage of AI tool for writing. Please refer to the journals' policy about using such tool. Similarly, your response to the Editor and Reviewers seem to have a prompt that the AI tool failed to work on ( "[Insert new title with 'Spatiotemporal' and '45-Day Precipitation'].")

Author Response

We thank the reviewer once again for their time, effort, and insightful comments. We have revised the manuscript extensively based on your suggestions and believe it is now significantly improved.

Author Response File: Author Response.docx