Forecasting of Standardized Precipitation Index Using Hybrid Models: A Case Study of Cape Town, South Africa
, Nkanyiso Mbatha 3, Shaun Ramroop 1, Sileshi Melesse 1 and Felix Silwimba 2
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors1. The mathematical description of SPI should be removed, and its detailed description and reference should be referred.
2. The description of ARIMA should be removed.
3. The description of performance metrics should be removed as these are widespread metrics found everywhere in the literature.
4. The caption of Figure 6 should be detailed. Also, the same may checked for all tables and figures.
5. The quality of the figures should be improved.
Author Response
Response to Reviewer 1 Comments
We would like to express our gratitude to the reviewer for their invaluable suggestions and comments.
- The mathematical description of SPI should be removed, and its detailed description and reference should be referred.
Reply: We would like to thank the reviewer for this comment. This was removed and detailed description and references were referred.
- The description of ARIMA should be removed.
Reply: We would like to thank the reviewer for this comment. This was removed and clearly explained using the Box-Jenkins methodology flowchart.
- The description of performance metrics should be removed as these are widespread metrics found everywhere in the literature.
Reply: We would like to thank the reviewer for this comment. The metrics has been removed.
- The caption of Figure 6 should be detailed. Also, the same may checked for all tables and figures.
Reply: We would like to thank the reviewer for this comment. The caption on Figure 6 has been improved. Also, in Figure 2 – 4, Figure 7 – 8.
- The quality of the figures should be improved.
Reply: We would like to thank the reviewer for this comment. The figures resolutions were improved.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsReviewer's Remarks
General observations.
The study employs a hybrid model that incorporates CEEMDAN, ARIMA, and LSTM to address the critical challenge of drought forecasting. This hybrid method is new, especially given Cape Town's history of severe droughts. The combination of these specific approaches to improving forecast accuracy is novel, and it constitutes a significant development in the field.
• The research cites the hybrid model's higher performance, but a comparative analysis with other advanced models (not simply ARIMA and LSTM) would provide a more comprehensive evaluation of its effectiveness.
• There are a few minor grammatical mistakes and confusing phrases that might be clarified.
• A more thorough examination of the constraints and any biases in the model's predictions is required.
• Future research directions could be mentioned by focusing on integrating more variables or utilizing machine learning approaches.
• There are a few minor grammatical mistakes and confusing phrases that might be clarified.
• While figures and tables are suitable, but improved labeling and captions will make them simpler to understand.
Specific comments
• Lines 92–104: Research problem and objectives should be stated more clearly.
• Lines 251-261: Provide an explanation of the hybrid CEEMDAN-ARIMA-LSTM model's selection, along with a comparison to alternative models.
• Lines 265-275: Provide thorough details on the preprocessing measures done to get the data ready for analysis.
• Lines 299–334: Go over how the LSTM and ARIMA models are implemented, including additional technical information and parameters.
• Lines 381-407: Go into more detail about the results' interpretation, especially with reference to the findings' importance and any ramifications for the field.
• Lines 408–419: Provide an analysis, backed by pertinent literature, that compares the performance of the suggested model with that of other models or approaches already in use.
• Lines 564-579: Strengthen the conclusion by providing an overview of the main conclusions.
Recommendation: I recommend minor revisions for this manuscript.
Comments on the Quality of English Language
Minor editing of English language required
Author Response
Response to Reviewer 2 Comments
We would like to express our gratitude to the reviewer for their invaluable suggestions and comments.
- Lines 92–104: Research problem and objectives should be stated more clearly.
Reply: We would like to thank the reviewer for this comment. The problem statement and objective are clearly stated.
Lines 251-261: Provide an explanation of the hybrid CEEMDAN-ARIMA-LSTM model's selection, along with a comparison to alternative models.
Reply: We would like to thank the reviewer for this comment. The explanation and comparison have been provided.
Lines 265-275: Provide thorough details on the preprocessing measures done to get the data ready for analysis.
Reply: We would like to thank the reviewer for this comment. This was included and explained under results section.
Lines 299–334: Go over how the LSTM and ARIMA models are implemented, including additional technical information and parameters.
Reply: We would like to thank the reviewer for this comment. The implementation parameters selection are include on Table 4 and Table 5.
Lines 381-407: Go into more detail about the results' interpretation, especially with reference to the findings' importance and any ramifications for the field.
Reply: We would like to thank the reviewer for this comment. References are included in the discussion Section.
Lines 408–419: Provide an analysis, backed by pertinent literature, that compares the performance of the suggested model with that of other models or approaches already in use.
Reply: We would like to thank the reviewer for this comment. Done.
Lines 564-579: Strengthen the conclusion by providing an overview of the main conclusions.
Reply: Reply: We would like to thank the reviewer for this comment. The conclusion has been rewritten and improved.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for Authors
· Abstract is too general, please add your results and main finding.
· Write more detail about SPI in introduction section.
· Fig. 1, study area map is incorrect, make new study area map.
· Figure 2 is missing, check it carefully and set all figure numbers.
· Resolution of all figures should be improved.
· Write results and discussion in separate sections.
· Fig. 11, is very low resolution, please improve it resolution.
· Please add more pervious research in discussion section and compare their research for scientific results.
· The authors should rewrite the conclusion with more clarity and suggest some mitigation measures to combat the problem.
· Overall, I recommend that the paper should be revised.
Author Response
Response to Reviewer 3 Comments
We would like to express our gratitude to the reviewer for their invaluable suggestions and comments.
Abstract is too general, please add your results and main finding.
Reply: We would like to thank the reviewer for this comment. The abstract has been modified, results and main findings are also included.
Write more detail about SPI in introduction section.
Reply: We would like to thank the reviewer for this comment. The detailed information about SPI is added in introduction section
Fig. 1, study area map is incorrect, make new study area map.
Reply: We would like to thank the reviewer for this comment. The map is corrected.
Figure 2 is missing, check it carefully and set all figure numbers.
Reply: We would like to thank the reviewer for this comment. Figures are checked and the numbering is corrected.
Resolution of all figures should be improved.
Reply: We would like to thank the reviewer for this comment. The resolution of Figures been improved.
Write results and discussion in separate sections.
Reply: We would like to thank the reviewer for this comment. The results and conclusion sections are separated.
Fig. 11, is very low resolution, please improve it resolution.
Reply: We would like to thank the reviewer for this comment. The resolution of Figures been improved.
Please add more pervious research in discussion section and compare their research for scientific results.
Reply: We would like to thank the reviewer for this comment. The results comparison has been done under discussion section.
The authors should rewrite the conclusion with more clarity and suggest some mitigation measures to combat the problem.
Reply: We would like to thank the reviewer for this comment. The conclusion has been rewritten and improved.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
Comments and Suggestions for AuthorsAll comments are completed, now it is ready for publication.
