A New Multi-Objective Genetic Programming Model for Meteorological Drought Forecasting
, Ali Danandeh Mehr 3,*, Rifat Tur 4, Abdelkader T. Ahmed 5, Laith Abualigah 6,7,8,9,10,11,12 and Dominika Dąbrowska 13
Round 1
Reviewer 1 Report
The submitted paper titled “A new multi objective genetic programming model for meteorological drought forecasting” provides an application of a drought model using SPI data in a region of Turkey. The idea is interesting and up to date. However, I would like to mention some comments for the improvements of the text.
My recommendation is major revisions.
General comments
- Avoid using keywords that are already in the title.
-Abstract: Please in the beginning add the importance of the current application. I suggest rephrasing the last sentence.
-Introduction: This chapter is poor.
(i) Please be more specific about the problems caused by droughts events, the gaps in literature-previous works, the importance of the current work and the innovation.
(ii) Why was this model chosen?
(iii) Mention the connection between the meteorological and hydrological drought. Please check the suggested literature.
-Line 80: “employs pareto front” please explain.
-Change chapters. 2 material and methods, 2.1 study area, 2.2 Data collection.
-Add figure of study area. Show the lake and rivers.
-Add information about the area. For instance, land use, min-max temperature.
-why SPI was chosen rather than SPEI?
-Figure 1. I suggest to change the color for the negative values. The same in figure 6. Try to improve the figures.
-chapter 3.1 . Why 6-month time scale? Please explain in the text.
-Figure 2. I suggest eliminating it. If you would like to remain, please explain the figure in the text.
-Figure 3. Avoid using mathematical equations in the caption.
-First mention a figure in the text and the provide it.
-Please eliminate RMSE and NSE equations.
-Please current in the text “meteorological station”.
-Mention in the discussion the drawbacks of using only one station in the area for the spatial distribution of the results.
-line 234. Eliminate “two”.
-Line 242-243. Lags 1-12?
-Lines 274-257. Please add the very big figures and equations in the supplementary materials.
-Avoid using “in this study” and “thus” a lot of times.
-Figure 11 is of utmost importance. Provide the diagrams vertical (one by one) to change the size.
-The conclusion is chapter 5.
-Academic writing is missing.
-Lines 346-348. How to overcome this drawback? Any suggestions?
Suggested literature
The impacts of drought on groundwater resources in the Upper Volturno basin, Southern Italy. 16th International Congress of Geological Society of Greece, Patra, 17-19 October 2022.
Investigating effect of climate change on drought propagation from meteorological to hydrological drought using multi-model ensemble projections (2019). Stochastic Environmental Research and Risk Assessment.
From meteorological to hydrological drought using standardised indicators (2016). Hydrology and Earth System Sciences, 20(6), 2483–2505.
Grammar and syntax issues have to be handled. Academic writing is missing.
Author Response
Reviewer: The submitted paper titled “A new multi-objective genetic programming model for meteorological drought forecasting” provides an application of a drought model using SPI data in a region of Turkey. The idea is interesting and up to date. However, I would like to mention some comments for the improvements of the text. My recommendation is major revisions.
Response: The authors are grateful for the reviewer's constructive comments and positive opinions on our work. We addressed all the comments and improved our paper substantially.
General comments:
- Avoid using keywords that are already in the title.
Response: Upon the comment, we replaced "genetic programming" with "SPI"
-Abstract: Please in the beginning add the importance of the current application. I suggest rephrasing the last sentence.
Response: Upon the comment, we revised the abstract substantially. Please the new Abstract.
-Introduction: This chapter is poor. (i) Please be more specific about the problems caused by droughts events, the gaps in literature-previous works, the importance of the current work, and the innovation. (ii) Why was this model chosen? (iii) Mention the connection between the meteorological and hydrological drought. Please check the suggested literature.
Response: Upon the comment, we revised the Introduction section substantially. Please the new Introduction Section.
-Line 80: “employs pareto front” please explain.
Response: Pareto-front plot was shown and explained in Figure 9 and Section 3.3
-Change chapters. 2 material and methods, 2.1 Study area, 2.2 Data collection.
Response: As we explained all the methods in Chapter 3, upon the comment we changed the Chapter 2 title to "study area and data collection" as recommended.
-Add figure of study area. Show the lake and rivers.
Response: Added. Please see Figure 1 in the revised version.
-Add information about the area. For instance, land use, min-max temperature.
Response: Upon the comment, the following information was added to the manuscript. Information about land use is out of the scope of our study.
"According to long-term observations, the minimum temperature in Burdur was determined as -16.7°C in January, and the maximum temperature 41.0°C in July and August."
-why SPI was chosen rather than SPEI?
Response: SPI can be simply calculated and suggested by WMO. But the proposed methodology can be tested for any other indices such as SPEI, RDI, ...
-Figure 1. I suggest changing the color for the negative values. The same in Figure 6. Try to improve the figures.
Response: Figure 6 was removed. We also replaced Figure 1 with a new one (Now Figure 2) having a red line to separate negative and positive SPI events.
-Chapter 3.1. Why 6-month time scale? Please explain in the text.
Response: Upon the comment, the following sentence was added to section 3.1.
"The underlying reason for selecting SPI-6 is that a medium-term accumulation period is more appropriate for measuring drought impacts on soil moisture, thus could be considered as both meteorological and agricultural drought signal"
-Figure 2. I suggest eliminating it. If you would like to remain, please explain the figure in the text.
Response: The figure was explained as below:
"For instance, Figure 3 illustrates a gene with a maximum depth of seven, a root node of multiplication function, and three variables (i.e., inputs; x1, x2, x3). Inner nodes are randomly filled by a user-defined function (here Log, sin, cos, addition, subtraction, and addition). Terminal nodes are those that only can assign a variable or a constant value (here a random value equal to 0.213). Overall, the gene expresses Eq. (1) in the tree form.
- Figure 3. Avoid using mathematical equations in the caption.
Response: According to the comment, all the equations were moved to the text body.
-First, mention a figure in the text and then provide it.
Response: We revised the manuscript accordingly.
-Please eliminate RMSE and NSE equations.
Response: We request to keep them as they might be new for young researchers.
-Please current in the text “meteorological station”.
Response: Upon the comment, we replaced "meteorology station" with "meteorological station" in the entire text.
-Mention in the discussion the drawbacks of using only one station in the area for the spatial distribution of the results.
Response: Upon the comment, we added the following paragraph to the revised manuscript.
"Inasmuch as the proposed MOMGGP model was trained and tested solely for Burdur station, the point forecasts cannot be transferred to locations far from the weather station. Following [39, 40], topographic, meteorologic, and other environmental attributes together with SPI forecasts at all stations available in the basin could be utilized to develop spatial distribution map of SPI across the study area. In ungauged catchments, multi-source precipitation products could be used for SPI monitoring and prediction [41]."
-line 234. Eliminate “two”.
Response: Removed.
-Line 242-243. Lags 1-12?
Response: Corrected.
-Lines 274-257. Please add the very big figures and equations in the supplementary materials.
Response: As the resolution of the big figures is well enough, they are readable. Referring to supplementary figures could be difficult for hardcopy readers. Thus, if the reviewer agrees, we would like to have the figures in the main text.
-Avoid using “in this study” and “thus” a lot of times.
Response: Done.
-Figure 11 is of utmost importance. Provide the diagrams vertical (one by one) to change the size.
Response: The figure was revised as commented.
-The conclusion is chapter 5.
Response: Corrected.
-Academic writing is missing.
Response: The English Language Edit service was taken from Native English. The attained certificate is attached.
-Lines 346-348. How to overcome this drawback? Any suggestions?
Response: Upon the comment, we added the following sentences to the revised manuscript.
"The relevant literature showed that data preprocessing techniques such as wavelet or variational mode decomposition may increase the accuracy of vanilla models [18,33]. However, their inclusion in the modeling process increases the solution's complexity markedly. Further studies are required to quantify the hybrid models' complexity that is raised through the decomposition process."
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This manuscript explores the Genetic programming (GP) in the forecasting of drought for a Turkish meteorological station in the Mediterranean basin. The drought is characterized by SPI (standardized precipitation index). The authors also used multi-gene GP (MGGP) to improve the methodology and results. And finally, they proposed to use the Pareto-optimality to take into account multi-objective constraints (MO-MGGP, trade-off between complexity and performance). The paper is generally well written with good illustrations for their methodology. I would recommend its acceptance with a few minor revisions.
1. There are some confusions for the number of Figures in Section 3 (Methods)
2. Figure 1 and Figure 6 are also confusing. I think Fig. 6 is almost Fig. 1, but with positive SPI values removed. Such cases should be left blank in Figure 6. The legend uses “time frequency”. I can’t understand it.
Author Response
Reviewer: This manuscript explores the Genetic programming (GP) in the forecasting of drought for a Turkish meteorological station in the Mediterranean basin. The drought is characterized by SPI (standardized precipitation index). The authors also used multi-gene GP (MGGP) to improve the methodology and results. And finally, they proposed to use the Pareto-optimality to take into account multi-objective constraints (MO-MGGP, trade-off between complexity and performance). The paper is generally well written with good illustrations for their methodology. I would recommend its acceptance with a few minor revisions.
Response: The authors appreciate the reviewer's time and positive feedback about our manuscript. Regarding their comments, we modified our manuscript.
Reviewer: 1. There are some confusions for the number of Figures in Section 3 (Methods)
Response: Upon the comment, we checked the numbers and corrected them.
Reviewer: 2. Figure 1 and Figure 6 are also confusing. I think Fig. 6 is almost Fig. 1, but with positive SPI values removed. Such cases should be left blank in Figure 6. The legend uses “time frequency”.I can’t understand it.
Response: As the reviewer mentioned, the figure only shows negative SPI values. Our goal was to show the count of (we called frequency) dry spells (SPI<0). Upon the comment, we removed the figure and modified the associated paragraph as given below:
" Frequency of the observed dry spells (i.e., SPI-6 < 0.0) at the station revealed 295 dry events with a long-term mean of – 0.81. This indicates that the station has a mild drought condition in general."
Reviewer 3 Report
Comments for author File: Comments.pdf
Moderate editing of English language required.
Author Response
Reviewer: The manuscript "A new multi-objective genetic programming model for meteorological drought forecasting" (water-2651361), is a work that addresses promising results, in which the authors present and apply a new explicit model for predicting meteorological droughts, which in my opinion will be a work that is widely explored and cited in the literature. However, before recommending this manuscript for publication, there are some points that concern me. Attached, I am also sending some notes highlighted in the work. Therefore, I am recommending this work for major revisions.
Response: The authors appreciate the reviewer's comments and positive feedback on our work. We addressed all of them to improve the manuscript.
Reviewer: In a preliminary analysis of the similarity of works with the present manuscript (water-10 2651361), I detected similarity with the present works:
Yaseen, Z. M., Ali, M., Sharafati, A., Al-Ansari, N., & Shahid, S. (2021). Forecasting standardized precipitation index using data intelligence models: regional investigations of Bangladesh. Scientific reports, 11(1), 3435. doi: 15 https://doi.org/10.1038/s41598-021-82977-9
Achite, M., Bazrafshan, O., Azhdari, Z., Wałęga, A., Krakauer, N., & Caloiero, T. (2022). Forecasting of SPI and SRI using multiplicative ARIMA under climate variability in a Mediterranean Region: Wadi Ouahrane Basin, Algeria. Climate, 10(3), 36. doi: 19 https://doi.org/10.3390/cli10030036
Response: Upon the comment, we substantially revised the Introduction section of our manuscript and reduced the similarity rate.
As small observations, which must be attended to, I highlight:
1 – I noticed some grammatical errors in writing, therefore, I suggest the revision of English by a native speaker.
Response: Upon the comment, an English language Edit was taken from Native English, and the certificate was attached.
2- The authors did not follow Water's formatting guidelines. References and citations also need to be revised, I suggest doing them in the Mendeley Reference Manager.
Response: References were revised according to the Journal format.
3-The abstract of the work has 126 words, the authors could explore the abstract further (Water limit of 200 words), as I miss the presentation of the main results and the conclusion of the main conclusion of this study in accordance with the objective of the work.
Response: The abstract section was revised substantially.
4 – The SPI in table 2 was based on which methodology? I suggest the authors cite the work of McKee et al. [21] in this table. Suggestion:
• Table 2. Classifications of drought states using the SPI according to the methodology of McKee et al. [21].
Response: According to the comment we cited McKee et al. [26] in the caption.
5 – Lines 188-191, 192-195, and 206-208 are "sentence paragraphs." It is important that authors punctuate these texts, with at least two full stops.
Response: Corrected
6 – In the conclusions, the authors include a citation. This citation should be removed, it is not an interesting citation in conclusion, as it only presents the conclusion of the work based on the results obtained, in accordance with the general objective of the study.
Response: The cited work was moved to the end of the results and discussion section.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The manuscript has been improved according to the reviewers' comments. In my opinion, can be published in the Journal of Water.
Reviewer 3 Report
In view of the authors' reviews and compliance with my suggestions, I am accepting this manuscript for publication.
