Remote Sensing and GIS Assessment of Drought Dynamics in the Ukrina River Basin, Bosnia and Herzegovina
, Davorin Bajić 1, Slobodan B. Marković 2,3,4, Dragutin Adžić 1, Velibor Spalevic 5, Paul Sestraș 6,7, Dragoslav Pavić 8 and Tin Lukić 8,9,*
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsTitle: Consider removing "Comprehensive" and "A Case Study of" as redundant and GPT possible authorship
The introduction is too extensive and could be reduced.
L 123. Most droughts of recent years are not included in the study.
L 126. Define “hot/tropical days”.
L 155. The authors comment on drought in Europe; the case of the Horn of Africa should be removed, or the paragraph reorganized
Fig 1: The authors should use consistent geographic data; in the map, the Ukrina River is not flowing at the lowest points. The map needs geographical references (graticule)
L 229. Citation 74 is too general; there should be a brief description of the data used
L231-233 and Fig. 2: The statistical significance of the precipitation and temperature change would be more useful than R2
L 259. Which version of CHIRPS was used? v2, v3? monthly data?
L 268. R = 0.875 gives a 76.6% agreement
L 297. SPI calculation is not only z-score computation (McKee 1993) but also includes distribution fitting. Also, the SPI calculation appears to use a 1-month timescale exclusively, but multiple timescales (3, 6, 12 months) are recommended for comprehensive drought assessment
L 387-402. Crop data are only available 2015-2019, not extended to the full study period (2015-2024); this is a critical issue and invalidates conclusions.
Indeed, 2020, 2021 (severe drought years), 2022 (most extreme drought year), 2023, 2024: have no socio-economic validation
The results section is too extensive and tedious to read; it could be condensed by 50-70% using tabular summaries.
L1024-1036 What is the proportion of the study area below 400 m? There should be a robust statistical analysis. Indeed, the author's conclusions do not seem relevant (Fig. 14), the persistent drought area zone should be represented with a line hash to avoid confusion with higher elevations.
Author Response
Dear Editor and Reviewers,
We would like to sincerely thank the Editor and the anonymous Reviewers for their careful evaluation of our manuscript and for the constructive comments and suggestions provided. The feedback was highly valuable and has substantially improved the clarity, methodological rigor, and overall quality of the manuscript.
A revised version of the manuscript has been prepared and is submitted together with this response. For transparency, the revised manuscript is provided in a tracked-changes format, enabling direct inspection of all modifications. In addition, comments within the tracked document are used to indicate where specific reviewer suggestions were addressed and to briefly describe the corresponding changes.
Below, we provide a point-by-point response to each reviewer comment. Reviewer comments are reproduced first, followed by our responses and a description of the revisions implemented. For each change, the location in the revised manuscript is indicated by section and line numbers (as shown in the tracked-changes version). Where applicable, the revised or newly added text is also quoted in the responses.
Reviewer 1
Comment:
Title: Consider removing "Comprehensive" and "A Case Study of" as redundant and GPT possible authorship
Answer:
Thank you for the suggestion. We have revised the title by removing the term “Comprehensive” and the phrase “A Case Study of”, as these may be perceived as redundant and formulaic. The revised title is more concise and directly reflects the scope and content of the manuscript: “Remote Sensing and GIS Assessment of Drought Dynamics in the Ukrina River Basin, Bosnia and Herzegovina”.
Revised text can be found on the lines: 4-6 (with tracking changes enabled)
Comment:
The introduction is too extensive and could be reduced.
Answer:
We thank the reviewer for the comment. In line with this suggestion and the related remark raised by Reviewer 3, we reduced and refocused the Introduction to improve conciseness and relevance to the manuscript’s scope. Specifically, we condensed the opening sections by removing overly general background statements and limiting the literature review to studies directly supporting the rationale and methodological approach of this work. We also streamlined the text by removing geographically less relevant examples and by emphasizing the study context (Europe/Bosnia and Herzegovina) and the key drought types and indicators used in the manuscript. As a result, the revised Introduction is shorter and more clearly aligned with the objectives of the study.
Revised text can be found on the lines: 87-98; 134-154; 161-166; 198-205. (with tracking changes enabled)
Comment:
L 123. Most droughts of recent years are not included in the study.
Answer:
Thank you for the comment. We acknowledge that the sentence in the Introduction lists drought years documented up to 2017; this reflects the temporal coverage of the national synthesis we cited (Trbić et al., 2022) and was included to provide background on drought occurrence in Bosnia and Herzegovina. The aim of the manuscript is not to reproduce a complete national chronology of drought years after 2017, but to deliver a basin-scale, integrated drought assessment for the Ukrina River Basin using a consistent set of remotely sensed and gridded inputs. We intentionally focus on 2015–2024 because this decade best represents the contemporary climate regime and recent extreme conditions relevant for current drought-risk management, while also ensuring methodological consistency and comparability of the satellite-based indicators used in the study (avoiding potential inhomogeneities related to changes in data availability, sensors, and processing over longer periods). To prevent any misinterpretation, we revised the Introduction sentence to explicitly note that the synthesis by Trbić et al. reports documented drought years up to 2017.
Revised text can be found on the line: 170 (with tracking changes enabled)
Comment:
L 126. Define “hot/tropical days”.
Answer:
Thank you for the comment. We have clarified the term “hot/tropical days” in the manuscript by adding a brief definition, i.e., days with daily maximum air temperature exceeding 30 °C (TR30), and cited the corresponding definition source where appropriate.
Revised text can be found on the lines: 173-174 (with tracking changes enabled)
Comment:
L 155. The authors comment on drought in Europe; the case of the Horn of Africa should be removed, or the paragraph reorganized
Answer:
Thank you for the comment. We agree that the previous version of this paragraph included geographically less relevant examples. Accordingly, we revised and reorganized the paragraph by removing the Horn of Africa case (as well as the U.S. damage estimate) and refocusing the discussion on Europe and Bosnia and Herzegovina, in line with the study scope. The revised paragraph now presents European impacts with specific examples from Germany and Poland, and concludes with an example from Bosnia and Herzegovina to strengthen the regional context.
Revised text can be found on the lines: 198-205 (with tracking changes enabled)
Comment:
Fig 1: The authors should use consistent geographic data; in the map, the Ukrina River is not flowing at the lowest points. The map needs geographical references (graticule)
Answer:
Thank you for this comment. We revised Figure 1 to improve cartographic consistency and geographic referencing. Specifically, we added a geographic graticule (latitude/longitude grid with coordinate labels). In addition, the basin boundary in the manuscript has been defined using the HydroSHEDS dataset (and this boundary was used throughout the subsequent analyses), and the river network was standardized to this framework by replacing the previous hydrography layer with HydroRIVERS, ensuring harmonized and consistent hydrographic data sources for the study area. Minor visual mismatches between the river polylines and the lowest terrain positions in the shaded relief may persist due to differences in spatial resolution and generalization between the hydrographic vector products and the elevation raster used for background visualization; however, the hydrographic representation is now consistent with the HydroSHEDS-defined basin framework.
Revised Figure can be found on the line: 298 (with tracking changes enabled)
Comment:
L 229. Citation 74 is too general; there should be a brief description of the data used
Answer:
Thank you for the comment. We agree that the ERA5 citation should be accompanied by a brief description of the dataset. We have therefore added a short clarification at its first mention, stating that ERA5 is a globally complete reanalysis dataset that combines model output with observations and provides temporally consistent climate variables at ~0.25° spatial resolution, which were used here to derive long-term temperature and precipitation trends (1984–2024) for the Ukrina River Basin.
Revised text can be found on the lines: 301-303 (with tracking changes enabled)
Comment:
L231-233 and Fig. 2: The statistical significance of the precipitation and temperature change would be more useful than R2
Answer:
Thank you for the suggestion. We agree that reporting statistical significance is more informative than R² alone. Accordingly, we complemented the Figure 2 description by reporting the statistical significance of the temperature and precipitation trends based on the Pearson correlation with time. The warming trend is statistically significant (p = 1.10 × 10⁻⁸), whereas the precipitation trend is not statistically significant (p = 0.479). R² values were retained as descriptive measures of the linear fit.
Revised text can be found on the lines: 305-308; 314-315. (with tracking changes enabled)
Comment:
L 259. Which version of CHIRPS was used? v2, v3? monthly data?
Answer:
Thank you for the comment. We have clarified the CHIRPS dataset specification in the manuscript. Specifically, we used CHIRPS v3 precipitation data and applied the pentadal (5-day) product (pentad = ~5-day precipitation totals) rather than monthly data, and this has now been explicitly stated at the first mention of CHIRPS in the text.
Revised text can be found on the lines: 352-354 (with tracking changes enabled)
Comment:
L 268. R = 0.875 gives a 76.6% agreement
Answer:
Thank you for the comment. We agree that the Pearson correlation coefficient should not be interpreted as a “percentage agreement”. In our manuscript, 94.18% refers to the overall percentage agreement between CHIRPS and meteorological station (MS) data reported in Sabljić et al. [8], while r = 0.875 represents the strength of the linear association (covariation) between the monthly CHIRPS estimates and MS observations. We have revised the sentence accordingly to clearly distinguish between these two validation metrics and to avoid any potential misinterpretation.
Revised text can be found on the lines: 362-367 (with tracking changes enabled)
Comment:
L 297. SPI calculation is not only z-score computation (McKee 1993) but also includes distribution fitting. Also, the SPI calculation appears to use a 1-month timescale exclusively, but multiple timescales (3, 6, 12 months) are recommended for comprehensive drought assessment
Answer:
Thank you for this constructive comment. We agree that SPI computation should not be presented as a simple z-score standardization only, but should include probabilistic distribution fitting. Accordingly, we revised the SPI methodology in the 2.2 Methodology section and implemented an SPI computation framework based on Gamma distribution fitting (with parameter estimation as described in recent literature), following the approach applied by Jamalzi et al. (2025).
In addition, we agree that using multiple accumulation timescales provides a more comprehensive characterization of meteorological drought conditions. Therefore, beyond the original monthly scale, we recalculated SPI at multiple timescales (SPI-1, SPI-3, SPI-6, and SPI-12), and updated the methodological description and the results accordingly. To reflect these revisions, we also replaced the previous SPI presentation with a new Figure 5, consisting of panels for SPI-1, SPI-3, SPI-6, and SPI-12, and expanded the results interpretation to explicitly discuss drought signals across each timescale.
Revised text can be found on the lines: 401-437 (for methodology) and 778-889 (for results); with tracking changes enabled
Comment:
L 387-402. Crop data are only available 2015-2019, not extended to the full study period (2015-2024); this is a critical issue and invalidates conclusions. Indeed, 2020, 2021 (severe drought years), 2022 (most extreme drought year), 2023, 2024: have no socio-economic validation
Answer:
We thank the Reviewer for raising this important point. We have clarified in the Methods section that municipal/city-level crop production data are only available for 2015–2019 due to official data availability constraints (Republic of Srpska Institute of Statistics, official reply No. 06.3/060-451/25, 15 August 2025). Following this, we revised the Methodology and Results/Discussion to avoid drawing socio-economic drought conclusions for 2020–2024 and to present the crop-yield analysis as a partial socio-economic validation for the period where consistent municipal-level statistics exist.
Revised text can be found on the lines: 571-590 (for methodology) and 1386-1394 (for results); with tracking changes enabled
Comment:
The results section is too extensive and tedious to read; it could be condensed by 50-70% using tabular summaries.
Answer:
Thank you for this helpful suggestion. In response, we have substantially revised the manuscript to improve readability and reduce repetitiveness by condensing the Results section overall, aiming for the level of reduction you recommended (approximately 50–70%, depending on the subsection). This was achieved through (i) replacing long year-by-year narrative descriptions with more systematic, synthesis-oriented reporting, and (ii) moving detailed quantitative breakdowns (e.g., extended category-by-category statistics, error matrices, and supporting figures/tables) from the main text to the Supplementary Materials, where they remain fully available for transparency and reproducibility without interrupting the flow of the core Results.
The most extensive condensation was implemented in Section 3.3 (Results of agricultural drought) and Section 3.5 (Results of the assessment of the negative impacts of drought). In these sections, we streamlined the narrative by integrating key quantitative information directly into the interpretation of the mapped patterns, while transferring the more detailed year-specific/tabular documentation and supporting outputs to the Supplementary Materials. As a result, these results are now presented in a more concise and structured way, while still preserving all essential information required to interpret the findings and evaluate the robustness of the analysis.
Sections 3.1 (Results of meteorological drought) and 3.2 (Results of hydrological drought) were also revised to reduce redundancy and improve clarity. In addition to selective shortening, these sections were partly expanded where necessary to reflect methodological improvements requested during the review process—most notably, upgrading the SPI analysis from a single monthly scale to multiple accumulation timescales (e.g., SPI-3, SPI-6, SPI-12). While this required adding some methodological and interpretative detail, we took care to present the results more systematically and clearly, focusing on the key patterns and implications rather than lengthy descriptive reporting. Overall, these revisions substantially enhance the coherence, readability, and structure of the Results section, while maintaining full access to supporting details through the Supplementary Materials.
Revised text can be found on the lines: 778-890 (for results of meteorological drought); 975-1008 (for results of hydrological drought); 1048-1074, 1168-1190, 1281-1307 (for results of agricultural drought); 1467-1492 (for results of the assessment of the negative impact of drought); with tracking changes enabled
Comment:
L1024-1036 What is the proportion of the study area below 400 m? There should be a robust statistical analysis. Indeed, the author's conclusions do not seem relevant (Fig. 14), the persistent drought area zone should be represented with a line hash to avoid confusion with higher elevations.
Answer:
Thank you for this helpful comment. Following your suggestion, we addressed this issue in a stepwise manner by updating the methodological framework, applying the additional analysis in the Results, and finally improving the visual presentation in Figure 14.
First, we updated the methodology (Section 2.2.4) to explicitly include a hypsometric baseline assessment and a more robust statistical evaluation of the elevation dependence of drought persistence. Specifically, the persistence core was intersected with SRTM-derived elevation belts and statistically compared with the basin-wide hypsometric structure using a chi-square (χ²) goodness-of-fit test.
Second, in the Results section (Section 3.5), we now report the proportion of the study area below 400 m a.s.l. (89.09% of the basin area) to provide an objective baseline for interpretation. Using the updated methodological approach, we further demonstrate that the persistent drought zone is disproportionately concentrated at lower elevations (96.13% of the persistent drought area occurs below 400 m a.s.l.), and that the elevation-belt distribution of the persistent drought zone differs significantly from the basin-wide distribution (χ² = 49.58, df = 8, p < 0.001), supporting the relevance of the conclusions.
Finally, we revised Figure 14 by representing the persistent drought zone with a hatched (line-hash) overlay, which improves visual clarity and avoids confusion with higher elevation classes shown by the hypsometric color ramp.
Revised text can be found on the lines: 614-618 (for methodology); 1442-1449 (for results); with tracking changes enabled
Revised Figure 14 (now Figure 12) can be found on the line: 1464.
In Novi Sad, On behalf of the co-author team,
14-01-2026 Prof. Tin Lukić,
Atmosphere Editorial Board Member
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsI have written some advice and opinions below to help you develop this manuscript. I hope these comments are helpful to you.
Comment 1.
From Fig. 2 and L229-231, you reported significant local warming, with a 2.37-degree increase over 40 years. As you know, drought indices based on standardization (e.g., SPI) are rooted in the stationary assumption. I can understand that it is difficult to apply any non-stationary drought indices in this study. In this regard, I suggest you mention this limitation specifically in your manuscript (i.e., Section 2.2.1 and discussion section).
Comment 2.
I'm not sure why the study basin should be the Ukraina River Basin. It may be more rational to study the basin, which includes the HS Srbac station. The authors should provide a more robust justification for why a large-scale river like the Sava reflects the hydrological flashiness of a smaller basin like the Ukrina. A simple correlation analysis between historical Ukrina discharge (if any exists) and Sava levels would be ideal.
Comment 3.
Even though the merged and simplified calculation process of the SPI is consequently the Z-score standardization, I feel the explanation of the definition should include distribution fitting using Gamma or another extreme distribution.
Comment 4.
I hope you provide additional support for using a=0.5 in L377-379.
Comment 5.
The caption of Table 3 can use "drought severity" or "class of droughts" rather than "drought types". The existing expression, "drought types", usually describes different droughts such as meteorological, hydrological, and agricultural droughts.
Comment 6.
From L400-402, the authors transparently acknowledge the lack of recent socio-economic data (post-2019). While understandable, this creates a gap in validating the 2021 and 2022 extreme drought events. I hope you mention your efforts to overcome this issue under L402.
Author Response
Dear Editor and Reviewers,
We would like to sincerely thank the Editor and the anonymous Reviewers for their careful evaluation of our manuscript and for the constructive comments and suggestions provided. The feedback was highly valuable and has substantially improved the clarity, methodological rigor, and overall quality of the manuscript.
A revised version of the manuscript has been prepared and is submitted together with this response. For transparency, the revised manuscript is provided in a tracked-changes format, enabling direct inspection of all modifications. In addition, comments within the tracked document are used to indicate where specific reviewer suggestions were addressed and to briefly describe the corresponding changes.
Below, we provide a point-by-point response to each reviewer comment. Reviewer comments are reproduced first, followed by our responses and a description of the revisions implemented. For each change, the location in the revised manuscript is indicated by section and line numbers (as shown in the tracked-changes version). Where applicable, the revised or newly added text is also quoted in the responses.
Reviewer 2
Comment:
From Fig. 2 and L229-231, you reported significant local warming, with a 2.37-degree increase over 40 years. As you know, drought indices based on standardization (e.g., SPI) are rooted in the stationary assumption. I can understand that it is difficult to apply any non-stationary drought indices in this study. In this regard, I suggest you mention this limitation specifically in your manuscript (i.e., Section 2.2.1 and discussion section).
Answer:
Thank you for this important remark. We agree that standardized drought indices such as SPI are typically computed with respect to a fixed reference climatology and therefore rely on the stationarity assumption. Following your suggestion, we have explicitly acknowledged this limitation in the revised manuscript. Specifically, in Section 2.2.1 we added a brief statement clarifying that SPI values for 2015–2024 are interpreted as standardized anomalies relative to the fixed long-term calibration period (1981–2024), and that potential non-stationarity under the observed warming trend may affect the probabilistic interpretation of standardized values. In addition, we have highlighted this issue in the Discussion section, noting that non-stationary climate conditions may influence long-term comparability of standardized drought metrics and that non-stationary or moving-baseline formulations represent a direction for future work.
Revised text can be found on the lines: 432-437 (with tracking changes enabled)
Comment:
I'm not sure why the study basin should be the Ukraina River Basin. It may be more rational to study the basin, which includes the HS Srbac station. The authors should provide a more robust justification for why a large-scale river like the Sava reflects the hydrological flashiness of a smaller basin like the Ukrina. A simple correlation analysis between historical Ukrina discharge (if any exists) and Sava levels would be ideal.
Answer:
We thank the Reviewer for this constructive comment and for highlighting the need for a stronger justification of the hydrological-drought indicator and the choice of the Ukrina River Basin as the study unit. The Ukrina River Basin was selected because it represents the target agro-ecological system analyzed throughout the manuscript (meteorological drought, vegetation response, and impacts), while hydrological data availability constitutes a key limitation within the basin.
As there are currently no active gauging stations within the Ukrina River Basin and no historical discharge or stage measurements available for Ukrina that would allow a direct comparison with the Sava River record, we used the nearest long-term station (HS Srbac on the Sava River) as a proxy indicator of the broader low-water signal relevant to the receiving Sava–Ukrina system at the monthly scale. Importantly, we do not interpret HS Srbac as a measure of short-term hydrological flashiness within the Ukrina Basin; rather, it is used to support hydrological drought chronology and to indicate broader phases of reduced water availability under data constraints.
To address the Reviewer’s request for a quantitative justification, we added an additional supporting analysis based on ERA5-Land monthly runoff reanalysis for the Ukrina River Basin (1997–2024). Basin-mean runoff anomalies (total runoff, surface runoff, and subsurface runoff) were calculated over the study area and compared with standardized monthly anomalies of HS Srbac water levels. The results show a statistically significant and coherent relationship between the two series at lag 0 (e.g., Pearson r = 0.492, p = 7.18 × 10⁻²² for total runoff), while lagged correlations (1–2 months) are lower, indicating that the shared signal is predominantly synchronous at the monthly scale. This confirms that the HS Srbac record captures a basin-relevant regional low-water signal that can be interpreted as an indirect indicator of hydrological drought conditions affecting the Ukrina River Basin.
Following this improvement, the methodology section was revised to clearly describe this additional step, and the corresponding results were added to the hydrological drought results subsection (with supporting figures provided in the Supplementary Materials).
Revised text can be found on the lines: 456-482 (for methodology); 908-920 (for results); with tracking chanes enabled
Supplementary Figure S1 can be found on the lines: 1862-1868 (with tracking changes enabled)
Comment:
Even though the merged and simplified calculation process of the SPI is consequently the Z-score standardization, I feel the explanation of the definition should include distribution fitting using Gamma or another extreme distribution.
Answer:
Thank you for this insightful comment. We agree that, although SPI values ultimately correspond to a standardized normal variate, the methodological definition should explicitly include the probabilistic step–i.e., fitting the precipitation series to an appropriate distribution prior to standardization.
Following this recommendation (and consistent with the related remark by Reviewer 1), we revised the SPI methodology in the 2.2 Methodology section and implemented a probabilistic SPI formulation based on Gamma distribution fitting (with parameter estimation as described in recent literature), following the approach applied by Jamalzi et al. (2025). The revised manuscript now clearly describes the distribution-fitting step, the treatment of zero-precipitation, and the subsequent transformation to standardized SPI values.
Revised text can be found on the lines: 401-437 (for methodology) and 778-889 (for results); with tracking changes enabled.
Comment:
I hope you provide additional support for using a=0.5 in L377-379.
Answer:
Thank you for this comment. We agree that the choice of α = 0.5 should be better supported. We have therefore expanded the explanation and added additional references indicating that VHI is commonly computed using equal weighting (α = 0.5) for VCI and TCI, as the relative contribution of moisture- and temperature-related vegetation stress is uncertain and varies across location and time (e.g., Kogan et al., 2012; Monteleone et al., 2020; Jiang et al., 2021; Mathbout et al., 2025).
Revised text can be found on the lines: 556-559 (with tracking change enabled).
Comment:
The caption of Table 3 can use "drought severity" or "class of droughts" rather than "drought types". The existing expression, "drought types", usually describes different droughts such as meteorological, hydrological, and agricultural droughts.
Answer:
Thank you for this comment. We agree that the term “drought types” may be confusing in this context. We have therefore revised the caption and column heading to use “drought severity classes” instead of “drought types”, to clearly indicate that Table 3 refers to intensity categories derived from index values rather than drought typology (meteorological/hydrological/agricultural).
Revised Table 3 can be found on the lines: 567-568 (with tracking changes enabled).
Comment:
From L400-402, the authors transparently acknowledge the lack of recent socio-economic data (post-2019). While understandable, this creates a gap in validating the 2021 and 2022 extreme drought events. I hope you mention your efforts to overcome this issue under L402.
Answer:
Thank you for this comment. We agree that the lack of municipal/city-level crop production data after 2019 limits the socio-economic validation of the extreme drought events in 2021 and 2022. To address this point transparently, we clarified our efforts to obtain recent socio-economic data and strengthened the limitation statement in the Methods section and we also revised parts od Results and Discussion section. Specifically, we note that we formally requested post-2019 municipal/city-level production statistics from the responsible institution; however, according to the official reply from the Republic of Srpska Institute of Statistics (No. 06.3/060-451/25, 15 August 2025), such detailed data have not been available since 2020 (only higher-level aggregates are reported). Therefore, the socio-economic drought assessment is explicitly limited to 2015–2019, and this constraint is acknowledged as a key limitation regarding the validation of the 2021–2022 drought events.
Revised text can be found on the lines: 571-590 (for methodology) and 1386-1394 (for results); with tracking change enabled.
In Novi Sad, On behalf of the co-author team,
14-01-2026 Prof. Tin Lukić,
Atmosphere Editorial Board Member
Author Response File: Author Response.docx
Reviewer 3 Report
Comments and Suggestions for AuthorsThe article is titled "Comprehensive Remote Sensing and GIS-Based Analysis of Drought Dynamics: A Case Study of the Ukrina River Basin, Bosnia and Herzegovina." This study focuses on improving the evidence base for drought monitoring by combining complementary drought indicators into a coherent assessment framework at the river basin level.
The article is valuable, but requires significant improvement.
My comments are as follows:
- The title of the article does not reflect its content.
- The article is too long and needs to be shortened. Some points are redundant, and there are many repetitions.
- The introduction is too long and needs to be shortened; it should be closely related to the article's topic. The first page of the introduction is unnecessary, discusses too obvious issues, and unnecessarily generates a literature review. The issues addressed in the article should be addressed.
- The methodological section requires improvement; the authors refer to paper by Bajić, D. Potential of Remote Sensing Techniques for Integrated Spatio-Temporal Monitoring and Analysis of Drought in the Sana River Basin, Bosnia and Herzegovina. Időjárás 2024, 128, 399–423. 1305 https://doi.org/10.28974/idojaras.2024.4. However, this work does not describe the sequence of activities. Figure 3 does not precisely describe the workflow. The authors determined various droughts using different methods but did not explain how the data were integrated. This section focuses on how the four types of droughts calculated using different methods were combined.
- The objectives of the work require modification; it should be consistent with what is described in the article.
- Some information is repeated, e.g., subsection 2.2.2.
- Water levels were used to assess hydrological drought, but standardized values were not used; there is a lack of literature on this topic.
- The authors state that pre-processing included filtering images based on percentage cloud cover and spatial extent. There is no information on the cloud cover level at which images were included and what percentage of satellite scenes were rejected.
- It should be explained why land use was analyzed and what impact it has on the occurrence of various types of drought.
- The discussion requires improvement and should focus solely on comparing the obtained results with studies in other parts of the country. Some information should be moved to the introduction, and here, the shortcomings and therefore the methodology used could be addressed (some information should be moved from the conclusions).
- The conclusions need to be shortened and improved.
Technical Notes:
- Figure 1 is missing geographic coordinates on the maps.
- The tables should be adapted to the journal's requirements.
Author Response
Dear Editor and Reviewers,
We would like to sincerely thank the Editor and the anonymous Reviewers for their careful evaluation of our manuscript and for the constructive comments and suggestions provided. The feedback was highly valuable and has substantially improved the clarity, methodological rigor, and overall quality of the manuscript.
A revised version of the manuscript has been prepared and is submitted together with this response. For transparency, the revised manuscript is provided in a tracked-changes format, enabling direct inspection of all modifications. In addition, comments within the tracked document are used to indicate where specific reviewer suggestions were addressed and to briefly describe the corresponding changes.
Below, we provide a point-by-point response to each reviewer comment. Reviewer comments are reproduced first, followed by our responses and a description of the revisions implemented. For each change, the location in the revised manuscript is indicated by section and line numbers (as shown in the tracked-changes version). Where applicable, the revised or newly added text is also quoted in the responses.
Reviewer 3
Comment:
- The title of the article does not reflect its content.
Answer:
Thank you for this comment. Taking into account the revisions implemented in response to the reviewers’ suggestions, we believe the revised manuscript is now more clearly focused and that the title should accurately reflect its scope and key contribution. Accordingly, we have revised the title to better represent the manuscript’s content and the remote-sensing and GIS-based drought assessment framework applied to the Ukrina River Basin. The new title emphasizes the core contribution—spatio-temporal assessment of drought dynamics using remote sensing and GIS—while remaining concise: Remote Sensing and GIS Assessment of Drought Dynamics in the Ukrina River Basin, Bosnia and Herzegovina.
Revised text can be found on the lines: 4-6 (with tracking changes enabled)
Comment:
- The article is too long and needs to be shortened. Some points are redundant, and there are many repetitions.
Answer:
Thank you for this important comment. We agree that the initial version of the manuscript was overly long and contained redundant statements and repetitions. In response – and in line with this reviewer’s recommendation as well as overlapping suggestions from the other reviewers—we have substantially shortened and tightened the manuscript overall.
To achieve this, we (i) removed repetitive explanations and merged overlapping statements across sections, (ii) replaced lengthy narrative descriptions with more concise, synthesis-oriented reporting, and (iii) moved detailed supporting materials (extended tables/figures and auxiliary outputs) to the Supplementary Materials, so that the main text remains focused while full documentation is still available for transparency.
A major part of the condensation was implemented in the Results section, particularly in Section 3.3 (Results of agricultural drought) and Section 3.5 (Results of the assessment of the negative impacts of drought), where year-by-year descriptions and extensive quantitative listings were streamlined and reorganized. Sections 3.1 (Results of meteorological drought) and 3.2 (Results of hydrological drought) were also edited to remove redundancies and improve clarity, while only adding essential text where methodological improvements required it (e.g., reporting SPI at multiple accumulation timescales such as SPI-3, SPI-6, and SPI-12) and keeping the presentation as systematic and concise as possible.
In addition, the Introduction was significantly shortened by removing non-essential background details, reducing repetition, and sharpening the focus on the research gap, objectives, and the rationale for the selected approach. Overall, these revisions reduced the manuscript length, improved readability, and ensured a clearer and more coherent structure throughout.
Revised text can be found on the lines: 87-98, 134-154 (for introduction); 778-890 (for results of meteorological drought); 975-1008 (for results of hydrological drought); 1048-1074, 1168-1190, 1281-1307 (for results of agricultural drought); 1467-1492 (for results of the assessment of the negative impact of drought); 1643-1733 (for discussion); 1795-1835 (for conclusion); with tracking changes enabled.
Comment:
- The introduction is too long and needs to be shortened; it should be closely related to the article's topic. The first page of the introduction is unnecessary, discusses too obvious issues, and unnecessarily generates a literature review. The issues addressed in the article should be addressed.
Answer:
We thank the reviewer for this valuable comment. We agree that the Introduction was overly long and contained general background statements and an unnecessarily broad literature review, particularly in the first page. Accordingly, we substantially shortened and refocused the Introduction to align more closely with the topic and scope of this study. Specifically, the first and second paragraphs were condensed and rewritten to remove obvious, overly general statements and to retain only the most relevant literature directly supporting the research rationale, the multi-indicator framework, and the study context. Overall, the revised Introduction is more concise and explicitly oriented toward the issues addressed in the manuscript.
Revised text can be found on the lines: 87-98 and 134-154 (with tracking changes enabled).
Comment:
- The methodological section requires improvement; the authors refer to paper by Bajić, D. Potential of Remote Sensing Techniques for Integrated Spatio-Temporal Monitoring and Analysis of Drought in the Sana River Basin, Bosnia and Herzegovina. Időjárás 2024, 128, 399–423. 1305 https://doi.org/10.28974/idojaras.2024.4. However, this work does not describe the sequence of activities. Figure 3 does not precisely describe the workflow. The authors determined various droughts using different methods but did not explain how the data were integrated. This section focuses on how the four types of droughts calculated using different methods were combined.
Answer:
Thank you for this valuable comment. We have substantially improved the Methodology section to clearly describe the sequence of activities and the integration of drought information. Specifically, the meteorological component was upgraded by implementing the standard SPI procedure using Gamma fitting and computing SPI at multiple accumulation scales (1-, 3-, 6-, and 12-month). The hydrological component was strengthened by introducing a standardized water-level indicator (SWLI) and by providing an explicit justification for the use of HS Srbac as a proxy through comparison with standardized ERA5-Land runoff anomalies. In addition, the interpretation of agricultural drought indicators (MODIS-based TCI, VCI, and VHI) was systematized, and the socio-economic assessment was clarified, including data-availability constraints limiting municipal crop-production statistics to 2015–2019. To directly address the integration issue, we added a concise workflow paragraph immediately after Figure 3 explaining that each drought type is first quantified independently and then integrated through a stepwise procedure that combines temporal consistency across indicators with GIS-based spatial overlay, including the delineation of persistent drought-prone areas and their interpretation with topography, land use, and administrative units. Finally, Figure 3 has been revised to more precisely represent the overall workflow and the integration steps.
Revised text can be found on the lines: 320-332; 341-349; 352-367; 401-437; 456-482; 501-514; 557-559; 571-590; 614-628; 640-643; 645-655; 662-664; 669-673 (with tracking changes enabled).
Revised Figure 3 can be found on the line: 334 (with tracking changes enabled).
Comment:
- The objectives of the work require modification; it should be consistent with what is described in the article.
Answer:
Thank you for this comment. We have revised the study objectives to ensure full consistency with the content of the revised manuscript. The updated objectives now explicitly reflect (i) the assessment of meteorological, hydrological, agricultural, and socio-economic drought dimensions, (ii) the use of an independent runoff-based proxy check to support the hydrological component, and (iii) the integration of these drought signals through GIS-based overlay to identify major drought episodes and delineate a persistent drought-prone zone, followed by interpretation in relation to topography, land use, and administrative units
Revised text can be found on the lines: 245-253 (with tracking changes enabled).
Comment:
- Some information is repeated, e.g., subsection 2.2.2.
Answer:
Thank you for this comment. We agree that parts of subsection 2.2.2 contained repetitive background statements. The subsection has been revised and condensed by removing redundant general descriptions and retaining only the information necessary to justify the data choice (HS Srbac as a proxy under data limitations) and to describe the analytical procedure.
Revised text can be found on the lines: 456-464 (with tracking changes enabled).
Comment:
- Water levels were used to assess hydrological drought, but standardized values were not used; there is a lack of literature on this topic.
Answer:
Thank you for this important comment. We agree that hydrological drought assessment should include standardized values supported by relevant literature. Accordingly, we updated the hydrological drought methodology by introducing a standardized indicator – the Standardized Water Level Index (SWLI) – as an additional component based on published methodology Nazarenko et al. (2023). SWLI was computed from monthly mean water levels at HS Srbac by standardizing water-level departures relative to the long-term monthly baseline, and its interpretation was aligned with the SPI intensity classes adopted in the manuscript. The SWLI calculation procedure is now explicitly described in the revised Methods section, and the corresponding SWLI results are presented and discussed in the revised Results section.
With this revision, both meteorological and hydrological drought are now evaluated through (i) an intuitive percent-based representation of precipitation and water-level departures from the long-term mean and (ii) standardized indices. Specifically, meteorological drought is quantified using the updated multi-timescale SPI framework (SPI-1/3/6/12), while hydrological drought is complemented by SWLI-based standardized values. These changes strengthen the comparability of drought signals across components and provide a more robust, literature-supported basis for linking meteorological drought triggers to hydrological constraints relevant for subsequent agricultural drought assessment.
Revised text can be found on the lines: 501-514 (for methodology) and 975-1010 (for results); with tracking change enabled.
Comment:
- The authors state that pre-processing included filtering images based on percentage cloud cover and spatial extent. There is no information on the cloud cover level at which images were included and what percentage of satellite scenes were rejected.
Answer:
Thank you for this comment. We agree that the Sentinel-2 pre-processing description should be more explicit. The manuscript has been revised to specify the scene-level cloud-cover threshold used for image selection (CLOUDY_PIXEL_PERCENTAGE < 6%) and to clarify that spatial filtering retained only scenes intersecting the study area (ROI). In addition, we now report the number of scenes intersecting the ROI, the number of scenes retained after cloud filtering, and the corresponding rejection percentages for each of the three sub-periods used to build the multi-temporal composite.
Revised text can be found on the lines: 645-655 (with tracking changes enabled).
Comment:
- It should be explained why land use was analyzed and what impact it has on the occurrence of various types of drought.
Answer:
Thank you for this comment. We have clarified in Section 2.2.4 why land use was analyzed and how it supports the interpretation of drought impacts. Specifically, land use was included to contextualize the adverse effects of drought by identifying which surface types are repeatedly affected within the persistent drought-prone zone and to quantify exposure of key classes, particularly agricultural areas, which are directly linked to vegetation stress and potential socio-economic consequences. We also note that the multi-temporal Sentinel-2 composite (three sub-periods) was used to improve class separability (e.g., agricultural areas vs. grasslands/meadows) under seasonal variability, thereby strengthening the reliability of the land-use overlay and subsequent impact interpretation.
Revised text can be found on the lines: 619-628 (with tracking changes enabled).
Comment:
- The discussion requires improvement and should focus solely on comparing the obtained results with studies in other parts of the country. Some information should be moved to the introduction, and here, the shortcomings and therefore the methodology used could be addressed (some information should be moved from the conclusions).
Answer:
Thank you for this constructive comment. In response, we substantially revised the Discussion to improve its focus and structure. The revised section is now organized primarily around direct comparisons between our findings and available studies from Bosnia and Herzegovina, covering the meteorological and hydrological drought components in a more systematic manner.
For the agricultural drought component based on remote sensing indices (TCI/VCI/VHI), we note that basin-scale applications of this approach in Bosnia and Herzegovina remain very limited (to our knowledge, beyond a single recent case-study), which constrains strict country-specific comparisons. Therefore, in this part of the Discussion, the interpretation is complemented by relevant broader literature that has established the applicability and robustness of these indices, while keeping the focus on how the observed patterns align with the drought years identified in the basin.
In line with your recommendation, we also moved general background information (on drought trends and climate context) from the Discussion to the Introduction, and we relocated the methodological limitations/shortcomings from the Conclusion to the Discussion. This restructuring improves the logical flow of the manuscript by keeping the Discussion centered on interpretation and comparison, while explicitly addressing methodological constraints where they affect the interpretation of results.
Revised text can be found on the lines: 1643-1733 (with tracking changes enabled).
Comment:
- The conclusions need to be shortened and improved.
Answer:
The Conclusions section has been substantially shortened and rewritten to improve clarity and focus. Redundant statements and repeated discussion points were removed, and the revised text now concisely summarizes the key findings and main methodological advancements (SPI based on Gamma fitting across multiple accumulation scales, SWLI-based hydrological assessment supported by ERA5-Land runoff correspondence, and delineation of the persistent drought-prone zone). Methodological limitations are addressed in the Discussion section, in line with the reviewer’s request.
Revised text can be found on the lines: 1795-1835 (with tracking changes enabled).
Comment:
- Figure 1 is missing geographic coordinates on the maps.
Answer:
Thank you for this technical note. We have revised Figure 1 by adding geographic coordinates (a latitude/longitude graticule with coordinate labels) to provide clear spatial referencing. This change has been implemented in the updated manuscript.
Revised Figure 1 can be found on the line: 298 (with tracking change enabled).
Comment:
- The tables should be adapted to the journal's requirements
Answer:
Thank you for this comment. All tables have been revised and formatted according to the journal’s requirements and template (layout, captions, and overall styling), and the updated versions are included in the revised manuscript.
In Novi Sad, On behalf of the co-author team,
14-01-2026 Prof. Tin Lukić,
Atmosphere Editorial Board Member
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe text quality has been upgraded, the authors replied to the comments.
- authors should replace very low p values by p < 0.001
- excel graphs have , instead of . for decimal separation
Author Response
Dear Editor and Reviewers,
We would like to sincerely thank the Editor and all Reviewers for the careful evaluation of our manuscript and for the constructive comments and suggestions provided in the second round of review.
A revised version of the manuscript has been prepared and is resubmitted together with this response. For transparency, the revised manuscript is provided in a tracked-changes format, enabling direct inspection of all modifications. Where helpful, we also inserted brief comments in the tracked document to indicate where specific reviewer suggestions were addressed and to summarize the corresponding revisions.
Below, we provide a point-by-point response to each reviewer comment. Reviewer comments are reproduced first, followed by our responses and a concise description of the revisions implemented. The location of each change is indicated by section and line numbers in the tracked-changes version and, where applicable, by reference to the updated Supplementary Materials (including Figures S2–S4 and the corrected supplementary dataset).
Reviewer 1
Comment:
The text quality has been upgraded, the authors replied to the comments.
-authors should replace very low p values by p < 0.001
Answer:
Thank you for your positive assessment and for noting that our previous revisions appropriately addressed your earlier comments and suggestions.
In the revised manuscript, we standardized the reporting of very small p-values and replaced extremely low numerical values with the journal-recommended format p < 0.001 throughout the manuscript.
Revised text can be found on the lines: 273, 282, 860, 862 and 864 (with tracking changes enabled).
Comment:
- excel graphs have , instead of . for decimal separation
Answer:
Thank you for pointing this out. We have updated the entire Supplementary dataset to ensure consistent decimal formatting by replacing decimal commas (",") with decimal points (".") throughout.
In Novi Sad, On behalf of the co-author team,
18-01-2026 Prof. Tin Lukić,
Atmosphere Editorial Board Member
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsI appreciated finding that my previous comments and suggestions were appropriately addressed in the revised manuscript.
My additional comment is about the Figs. S2-4, revising these figures more useful. In these three histograms, many subfigures include 'actually no value' columns. If you want to keep using these histograms, it would be more informative if you add percentages or other numbers to these tiny columns, like we generally do in the pie or donut charts. Otherwise, you can use tables. Actually, Fig. S3 includes only one piece of information, 'there are generally no droughts.'
Author Response
Dear Editor and Reviewers,
We would like to sincerely thank the Editor and all Reviewers for the careful evaluation of our manuscript and for the constructive comments and suggestions provided in the second round of review.
A revised version of the manuscript has been prepared and is resubmitted together with this response. For transparency, the revised manuscript is provided in a tracked-changes format, enabling direct inspection of all modifications. Where helpful, we also inserted brief comments in the tracked document to indicate where specific reviewer suggestions were addressed and to summarize the corresponding revisions.
Below, we provide a point-by-point response to each reviewer comment. Reviewer comments are reproduced first, followed by our responses and a concise description of the revisions implemented. The location of each change is indicated by section and line numbers in the tracked-changes version and, where applicable, by reference to the updated Supplementary Materials (including Figures S2–S4 and the corrected supplementary dataset).
Reviewer 2
Comment:
I appreciated finding that my previous comments and suggestions were appropriately addressed in the revised manuscript. My additional comment is about the Figs. S2-4, revising these figures more useful. In these three histograms, many subfigures include 'actually no value' columns. If you want to keep using these histograms, it would be more informative if you add percentages or other numbers to these tiny columns, like we generally do in the pie or donut charts. Otherwise, you can use tables. Actually, Fig. S3 includes only one piece of information, 'there are generally no droughts.'
Answer:
Thank you for your positive assessment and for noting that our previous revisions appropriately addressed your earlier comments and suggestions. We also appreciate your additional recommendation regarding Figures S2–S4.
In accordance with your guidance, we have updated Figures S2, S3, and S4 to improve their interpretability. Specifically, we added numeric values (data labels) above each bar/column so that even very small (or near-zero) categories can be clearly identified and compared. The revised versions of Figures S2–S4 are now included as an integral part of the Supplementary Materials in the revised manuscript.
Revised Figures can be found on the lines: 1492, 1497 and 1502 (with tracking changes enabled).
In Novi Sad, On behalf of the co-author team,
18-01-2026 Prof. Tin Lukić,
Atmosphere Editorial Board Member
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThe authors have largely improved the article. The authors did not use the change tracking feature, making it difficult to identify what has been removed and what has been added. There are still important issues that need improvement: the article is too long and should be shortened, for example, by shortening the introduction; the methodological section, which is obvious (SPI, ERA); special attention should be paid to integrating different drought types using different SPI description indices; detailed descriptions should be avoided; the results section should be shortened, for example, subsections 3.1 and 3.2. Detailed descriptions of all months, along with detailed data, are required – this is meant to be an analysis, not a description and listing of what is visible in the figure.
Author Response
Dear Editor and Reviewers,
We would like to sincerely thank the Editor and all Reviewers for the careful evaluation of our manuscript and for the constructive comments and suggestions provided in the second round of review.
A revised version of the manuscript has been prepared and is resubmitted together with this response. For transparency, the revised manuscript is provided in a tracked-changes format, enabling direct inspection of all modifications. Where helpful, we also inserted brief comments in the tracked document to indicate where specific reviewer suggestions were addressed and to summarize the corresponding revisions.
Below, we provide a point-by-point response to each reviewer comment. Reviewer comments are reproduced first, followed by our responses and a concise description of the revisions implemented. The location of each change is indicated by section and line numbers in the tracked-changes version and, where applicable, by reference to the updated Supplementary Materials (including Figures S2–S4 and the corrected supplementary dataset).
Reviewer 3:
Comment:
The authors have largely improved the article. The authors did not use the change tracking feature, making it difficult to identify what has been removed and what has been added. There are still important issues that need improvement: the article is too long and should be shortened, for example, by shortening the introduction; the methodological section, which is obvious (SPI, ERA); special attention should be paid to integrating different drought types using different SPI description indices; detailed descriptions should be avoided; the results section should be shortened, for example, subsections 3.1 and 3.2. Detailed descriptions of all months, along with detailed data, are required – this is meant to be an analysis, not a description and listing of what is visible in the figure.
Answer:
Thank you for your positive evaluation of the revised manuscript and for these clear and constructive recommendations for further improvement. We fully agree with the need to (i) facilitate inspection of revisions, (ii) reduce overall length, (iii) avoid overly descriptive, figure-driven listing, and (iv) strengthen the analytical integration of drought dimensions through the interpretation of SPI at multiple accumulation timescales.
First, to address your concern regarding revision transparency, we confirm that the resubmitted manuscript is provided with tracked changes enabled throughout, allowing direct identification of all deletions and additions. Where helpful, brief in-text comments were also used to indicate where specific reviewer suggestions were implemented.
Second, following your recommendation to shorten the manuscript, we substantially condensed the Introduction while retaining its core message. After the first review round, the Introduction contained 1,562 words, whereas the current version contains 1,027 words, representing a reduction of 535 words (−35.25%). Considering that the Introduction had already been shortened by ~10% during the initial revision stage, the Introduction has now been reduced by ~45% compared with the original submission. In the tracked-changes file, the removed and newly added paragraphs are clearly visible, and the shortening was implemented paragraph-by-paragraph to preserve coherence.
Revised text of Introduction can be found on the lines: 70-81, 102-116, 138-149, 180-191 (with tracking change enabled).
Third, with respect to the Methods section (SPI, ERA5/ERA5-Land), we acknowledge your observation that this part of the manuscript is relatively detailed. At the same time, we would like to emphasize that, in the first review round, both Reviewer 1 and Reviewer 2 explicitly requested methodological improvements and clarifications, which is fully consistent with your general remark that further strengthening of this section is important. In response, and following the reviewers’ recommendations, we updated and expanded the methodological framework by adopting more rigorous approaches and by improving the documentation of all key inputs and processing steps. Specifically, we (i) added a clearer description of the ERA5/ERA5-Land dataset and strengthened statistical reporting where relevant (Lines 268–275, tracked changes enabled); (ii) explicitly specified the CHIRPS product used and its temporal resolution (Lines 310–311); (iii) upgraded the SPI computation from a simplified monthly standardization to a probabilistic framework with distribution fitting and multi-timescale SPI (SPI-1/3/6/12) to support robust drought characterization (Lines 345–376); (iv) clarified limitations and data-availability constraints related to the socio-economic drought component (Lines 511–517); and (v) introduced a quantitative, proxy-based justification of using HS Srbac under data limitations by comparing standardized monthly anomalies of ERA5-Land basin-mean runoff with standardized HS Srbac water-level anomalies, including correlation and lag testing (Lines 400–417). In addition, to standardize the hydrological drought component, we introduced the SWLI index and briefly documented its purpose and computation (Lines 436–446).
As a result, the Methods section was expanded in the previous revision round primarily because these upgrades were requested by the reviewers and were necessary to ensure methodological rigor. We therefore retained this revised methodological content in the current version to preserve transparency, reproducibility, and usability of the workflow, enabling other researchers to replicate the procedure in their own study areas. Consistent with your recommendation to reduce overly descriptive text, we concentrated the shortening effort on sections where narrative listing was most pronounced (i.e., the Introduction and Results), while maintaining a method description that supports repeatability and clear interpretation of the subsequent analyses.
Finally, in full agreement with your recommendation to avoid month-by-month listing in the Results, we revised the Results section – especially Sections 3.1 and 3.2 – to emphasize interpretation and cross-timescale/seasonal patterns rather than reproducing values visible in the figures. In Section 3.1, detailed month-by-month narratives of CHIRPS anomalies were replaced by a concise synthesis of interannual patterns, and SPI interpretation was explicitly reorganized around SPI-1/3/6/12 as complementary “description indices” supporting process-based integration across drought dimensions (meteorological shocks, seasonal deficits relevant to vegetation/agriculture, medium-term persistence linked to hydrological response, and long-term anomalies relevant to water-resource and potential socio-economic implications). As a result, Section 3.1 was reduced by 988 words (−53.53%) while preserving its scientific content and conclusions. In Section 3.2, the paragraph describing monthly HS Srbac water-level anomalies was rewritten to highlight interpretation and interannual/seasonal patterns (with emphasis on the vegetation season and drought propagation) rather than listing individual months and values shown in Figure 6; this paragraph was reduced by 213 words (−38.31%), while maintaining the quantitative basis and scientific validity of the hydrological drought assessment.
Revised text can be found on the lines: 650-675, 784-835, 916-947 (with tracking changes enabled).
Overall, considering the revisions and shortening implemented across the manuscript, the text from the Introduction through the Conclusions has been reduced by 1,687 words (−11.73%) compared with the previous version, while preserving the scientific substance and validity of the study.
In Novi Sad, On behalf of the co-author team,
18-01-2026 Prof. Tin Lukić,
Atmosphere Editorial Board Member
Author Response File: Author Response.pdf
