Optimal Probability Distribution and Applicable Minimum Time-Scale for Daily Standardized Precipitation Index Time Series in South Korea

Round 1

Reviewer 1 Report

The research entitled “Optimal Probability Distribution and Applicable Mini- 2 mum Time-Scale for Daily SPI Time Series in Korea” is interesting, well written and organized research with a good potential of publication consideration in this journal. The authors tested the applicability of six probability distributions namely Gumbel, Gamma, Log-Logistic, Weibull, Log-normal, and GEV on 10-time scales for 56 stations in Korea. Research design is convincing and conducted well. Methodology is well-explanatory. Results are well reported. Discussion is fine. I recommend the paper after minor revisions

1-      Study area map is not attractive. It can be improved by adding coordinates and also in coordinates (longitude and latitude) extent must be mentioned in describing study area. The small data frame should be enlarged to the extent of SE Asia.

2-      The frequency fig of best performed highlighted PDFs can be added for different time-scales for more clearer understanding of results.

3-      Line 215-216 “Finally, from the perspective 215 of ?, the best fit is shown in the order of WEB, GAM, and GEV” can be like WEB > GAM > GEV.

 

4-      Introduction and discussion can be enriched by citing more literature. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

July 27, 2023

Manuscript: Optimal Probability Distribution and Applicable Minimum Time-Scale for Daily SPI Time Series in Korea.

The paper investigates the best probability distribution function (PDF) and the shortest time-scale for calculating the Standardized Precipitation Index (SPI), a drought assessment index, using daily precipitation data from 56 sites in six regions of Korea. The paper estimates the daily SPI by using 10 time-scales (5-day to 365-day) and six candidate PDFs (Gumbel, Gamma, GEV, Log-logistic, Log-normal, Weibull). The paper applies the Chi-square test and the Akaike Information Criterion (AIC) to evaluate the goodness of fit of the PDFs, and the Anderson-Darling test and the deviation from N0,1 to assess the normality of the SPI. I think it is a relevant topic within the scope of the MDPI Atmosphere journal. The article is well organized and well written, with enough scientific content. However, I find that the manuscript misses some key points that need to be reviewed before publication (minor revision).

Review Report:

- Abstract: The abstract is clear and concise.

- Introduction: The introduction provides a good background, motivation for the study, and research gap.

- Data and Methods: The data and methods section is well organized and detailed, but it could be improved by providing some descriptive statistics or plots of the daily precipitation data, such as mean, standard deviation, skewness, kurtosis, or histograms (lines 101-105). This point would help the readers understand the data's characteristics and variability (It could be included in the Supplementary Materials). Also, the authors should explain why they chose six candidates PDFs and 10 time-scales for their analysis (lines 114-117), and how they determined the number of class bins for the Chi-square test (line 138). They should also cite the sources of these methods (lines 135-140 and 169-172).

- Results: The results section is clear and comprehensive, but it could be improved by using more figures or tables to summarize or visualize the main results. For example, instead of listing all the mean p-values, accept rates, and win rates for each PDF, time-scale, region, and season in Table 2 (line 217), the authors could use a heatmap or a bar chart to show the relative performance of each PDF across different scenarios (The ggplot2 R package might be useful for this task.). Also, instead of showing all the AIC frequencies for each PDF in Figure 2, the authors could use a boxplot or a violin plot to show the distribution of AIC differences for each PDF across different scenarios. This aspect would make the results more concise and easier to compare (lines 249-250).

- Discussion: The discussion section is well written and logical, but it could be improved by providing more interpretation and implications of the results. For example, the authors should discuss in more detail how their findings compare or contrast with previous studies in other regions or climates (lines 424-432; 442-446) and what are the implications or applications of their findings for drought monitoring or assessment in Korea (lines 467-476).

- Conclusion: The conclusion section is brief and concise, but it could be improved by highlighting the main contributions and limitations of the study and suggesting some directions for future research (lines 499-502). The authors should also acknowledge some limitations or uncertainties of their study, such as data quality or availability, parameter estimation methods, or spatial variability of precipitation (lines 492-494). Finally, the authors should suggest possible extensions or improvements to their study, such as using other PDFs or indices, applying different transformation methods, or incorporating other climatic factors (lines 499-502).

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

I was kindly asked to review the manuscript: Optimal Probability Distribution and Applicable Minimum Time-Scale for Daily SPI Time Series in Korea. This is a well written and well detailed manuscript. Such study is highly relevant as it could add to the scientific body of knowledge, but I would encourage them to consider the following:

1.       Abstract: The abstract needs more of a ‘hook’ to engage the reader and establish the novelty of the analysis in the scientific literature. I recommend them to focus on the conclusions of the study.

2.       The introduction section should be reorganized, provide additional information related to the methods followed in previous studies.

3.       Could the authors provide more information on the control the validity of the results?

4. Overall, the discussion of findings is complete and concise. Please, explain in the discussion the limitations of the work.

Overall, it is a solid article and, methodologically, it is well planned and fits within the journal's objectives.

 

 Minor editing of English language required

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 4 Report

The article is interesting and deals with a widely used measure of rainfall variability. In the context of observed droughts and the increase of these events in the future, the research on the optimization of this indicator should be considered valuable. The article has a clear and, in addition, is enriched with numerous valuable figures to facilitate the interpretation of the results. I recommend the article for publication after taking into account the following comments.

What is the criterion for selecting the six regions?

Please provide more details on data acquisition. The indicated address https://data.kma.go.kr/ refers the reader to a general page. This is inaccurate in the context of the article's content.

How was the horizontal precipitation (dew, glaze, frost) measured. What is their influence on the results obtained?

The paper lacks a deeper spatial interpretation of the variation in results.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 5 Report

The article is titled “Optimal Probability Distribution and Applicable Minimum Time-Scale for Daily SPI Time Series in Korea”. In this study, the authors examined the optimal probability distribution and the minimum time scale used to calculate the daily SPI using a time series of daily precipitation observed over 42 years at 56 sites in Korea. The study used six candidate probability distributions (Gumbel, Gamma, GEV, Log-logistic, Log-normal, Weibull) and 10 time scales (5-day, 10-day, 15-day, 21-day, 30-day, 60-day -day-day, 90-day, 180-day, -270-day, 365-day) were used to calculate the daily SPI. I believe that the article can be published with minor additions.

 

Detailed notes:

- the introduction needs to be supplemented, the authors should refer to more recent research and the use of the SPI in different climatic zones

- subsection 2.1 needs to be supplemented. The authors did not describe the research area. They only indicated that the study area was divided into six regions (Fig. 1).

- the discussion lacks the link between the obtained test results and the location of the measurement data.

Technical Notes

- Fig. 1 needs improvement, insert geographic coordinates, north direction, scale on the general map

- What do the colors in table 2 mean ?

- literature needs to be supplemented with new items

​

 

 

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx