Prediction of the Potential Distribution of Drosophila suzukii on Madeira Island Using the Maximum Entropy Modeling
, Carla Ragonezi 1,2,3,*, Fábio Reis 1, José G. R. de Freitas 1, David Horta Lopes 4, António Miguel Franquinho Aguiar 5, Délia Cravo 5 and Miguel A. A. Pinheiro de Carvalho 1,2,3
Round 1
Reviewer 1 Report
1.The background of the study is not clearly stated and lacks sufficient background support. For example: the invasion of Drosophila Susukui has caused particular harm to the island
2.The overall logic of this article is confusing, for example. Paragraph 5 of the introduction describes the biology of D. In the discussion, Susuk II should be used as an explanation of the model results.。
3.The distribution sample is too small, based on only one data source, and the result is a model without enough training samples
4. There are 97 sample points involved in the methodology, but there are fewer than 97 sample points shown in Table 1, which data are used for the model running.
5. The grammar of the paper tends to be colloquial and the presentation is not concise.
6. The climatic variables used were average, maximum and minimum temperatures (°C), accumulated precipitation(mm), and average humidity (%). Why chose them and is there a strong correlation between the data e.g., average, maximum and minimum temperatures (°C).
7. The model results were not optimized using the ENMeval package, and the parameter settings are not referenced or based on the literature.
8. There is no need to cite other people's results in the results, I wasn't able to separate what was in the results from the discussion, and no specific discussion section appeared in the article.
9. The article is too one-dimensional, and lacks a discussion section.
Moderate editing of English language required
Author Response
Response to Reviewer 1 Comments
Manuscript ID: agriculture-2569861
Type of manuscript: Communication
Title: Prediction of the potential distribution of Drosophila suzukii on Madeira Island using the Maximum Entropy Modeling
Authors: Fabrício Lopes de Macedo, Carla Ragonezi *, Fábio Reis, José G. R. de Freitas, David Horta Lopes, António Miguel Franquinho Aguiar, Délia Cravo, Miguel Â. A. Pinheiro de Carvalho
The authors appreciate the reviewer’s comments about the manuscript. The comments were addressed properly below and can be visualized in the resubmitted version.
Point 1: The background of the study is not clearly stated and lacks sufficient background support. For example: the invasion of Drosophila Susukui has caused particular harm to the island.
Response 1: The authors revised the article throughout which can be visualized in the resubmitted version.
Point 2: The overall logic of this article is confusing, for example. Paragraph 5 of the introduction describes the biology of D. In the discussion, Susuk II should be used as an explanation of the model results.
Response 2: This was corrected in the resubmitted version.
Point 3: The distribution sample is too small, based on only one data source, and the result is a model without enough training samples.
Response 3: Thank you for your comment, but that was all the data we had available for this study. We agree that a larger number of samples would provide a more robust model, and this is something that we will consider in the future.
Point 4: There are 97 sample points involved in the methodology, but there are fewer than 97 sample points shown in Table 1, which data are used for the model running.
Response 4: The 97 collection points represent the places where the presence of the insect was identified. Table 1 shows the weather stations used. The data are different from each other.
Point 5: The grammar of the paper tends to be colloquial and the presentation is not concise.
Response 5: The authors appreciate the comment and tried at best to revise the article throughout taking the comments into consideration.
Point 6: The climatic variables used were average, maximum, and minimum temperatures (°C), accumulated precipitation(mm), and average humidity (%). Why chose them and is there a strong correlation between the data e.g., average, maximum, and minimum temperatures (°C).
Response 6: The choice of environmental variables was based on an article currently under review for publication entitled: "FRUITFLYRISKMANAGE: a Euphresco project for Ceratitis capitata Wiedemann (Diptera: Tephritidae) risk management applied in some European countries" in Eppo Buletin. We tested the same variables that the authors tested for Ceratitis capitata. Yes, there is a strong correlation between climate variables and altitude, except for relative humidity.
Point 7: The model results were not optimized using the ENMeval package, and the parameter settings are not referenced or based on the literature.
Response 7: We don't use it, the data presented relates only to that obtained by using Maxent, but we appreciate the comment.
Point 8: There is no need to cite other people's results in the results, I wasn't able to separate what was in the results from the discussion, and no specific discussion section appeared in the article.
Response 8: By mistake, the results and discussion were not fully written, meaning that section 3 is results and discussion, and not just results. This was corrected in the resubmitted version. The authors chose to write the results and discussion in one section, instead of apart.
Point 9: The article is too one-dimensional, and lacks a discussion section.
Response 9: By mistake, the results and discussion were not fully written, meaning that section 3 is results and discussion, and not just results. This was corrected in the resubmitted version. The authors chose to write the results and discussion in one section, instead of apart.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments for author File: Comments.pdf
Author Response
Response to Reviewer 2 Comments
Manuscript ID: agriculture-2569861
Type of manuscript: Communication
Title: Prediction of the potential distribution of Drosophila suzukii on Madeira Island using the Maximum Entropy Modeling
Authors: Fabrício Lopes de Macedo, Carla Ragonezi *, Fábio Reis, José G. R. de Freitas, David Horta Lopes, António Miguel Franquinho Aguiar, Délia Cravo, Miguel Â. A. Pinheiro de Carvalho
The authors appreciate the reviewer’s comments about the manuscript. The comments were addressed properly below and can be visualized in resubmitted version.
Point 1: Object, methods, and results are simple and clear. However, surprisingly, there is no part of reference. I hope that this part may be missed during submission process. In addition, there are errors, questions, and suggestions.
Response 1: Regarding the reference, the authors already contacted the Managing Editor to know why the reference section was not in the version sent to the reviewers. Other corrections can be seen in the corrected version.
Point 2: Line 74: Scientific name should be italic. Drosophila suzukii-→Drosophila suzukii This error is found in other lines (175, 186, and 196).
Response 2: The information was correct and can be seen in the resubmitted version.
Point 3: Line 154-162: Climate factors such as temperature and precipitation are highly correlated with altitude. The low influences of climatic factors in this study may be due to influence of altitude. Modelling using only climatic factors (Figure 3 tells that a model using only climatic factors is similar to a model using only altitude) are necessary to compare influences of climatic factors with other studies. Correlation between altitude and climatic factors are also required.
Response 3: The information was added and can be seen in the resubmitted version.
Point 4: 3.3. Individual response curves
Suitability should be defined by a consistent standard such as 0.4 of occurrence probability. “species is located between 0 -1200 m. It should be noted that Madeira has mountainous”. In 1200m of altitude, logistic output (occurrence probability?) is below 0.05. Why do you insist that the fly is suitable in such high altitude on what base? Figure 4 shows that 0-500m is suitable on the base of 0.4 of occurrence probability. Part of conclusion (line 222-223) should be revised. On this concept, suitable humidity ranges from 60 to 90.
Response 4: The information was added and can be seen in the resubmitted version.
Point 5: Figure 4: What is 500mof altitude?
Response 5: We do not understand what the symbol before 500 means, but we presume that is the minus sign (?). If yes, we corrected the figure where the negative altitude values were retrieved.
Author Response File: Author Response.pdf
Reviewer 3 Report
General comments
The paper provides some useful information regarding the potential distribution of D. suzukii in the Madeira island, some minor changes should be made to publish the paper.
Abstract
It has general content, should describe main results regarding importance of predictor variables, distribution range. Recommendations regaarding pestt management go beyond the results obtained. Should restrict to main findings.
Introduction
Fails to cite published works related to SDM of D suzukii in different parts of the world (e.g. Castro -Sosa et al 2017, Fla. Ent., 100,4; de la Vega & Korley, 2019. Int. J. Pest Mngmt., DOI:
10.1080/09670874.2018.1547460; dos Santos et al., 2017, PLOS ONE, https://doi.org/10.1371/journal.
pone.0174318; Vejrum Ørsted & Ørsted 2018, J. Appl. Ecol., DOI: 10.1111/1365-2664.13285). Also, use these cites to compare with results.
Should compare with areas of crop hosts.
Additional coments in the revised manuscript.
Comments for author File: Comments.pdf
Author Response
Response to Reviewer 3 Comments
Manuscript ID: agriculture-2569861
Type of manuscript: Communication
Title: Prediction of the potential distribution of Drosophila suzukii on Madeira Island using the Maximum Entropy Modeling
Authors: Fabrício Lopes de Macedo, Carla Ragonezi *, Fábio Reis, José G. R. de Freitas, David Horta Lopes, António Miguel Franquinho Aguiar, Délia Cravo, Miguel Â. A. Pinheiro de Carvalho
The authors appreciate the reviewer’s comments about the manuscript. The comments were addressed properly below and can be visualized in the resubmitted version.
Point 1: The paper provides some useful information regarding the potential distribution of D. suzukii in the Madeira island, some minor changes should be made to publish the paper.
Response 1: The authors appreciate the reviewer´s comments.
Point 2: Abstract - It has general content, should describe main results regarding importance of predictor variables, distribution range. Recommendations regaarding pestt management go beyond the results obtained. Should restrict to main findings.
Response 2: The authors corrected the information in the resubmitted version.
Point 3: Introduction - Fails to cite published works related to SDM of D suzukii in different parts of the world (e.g. Castro -Sosa et al 2017, Fla. Ent., 100,4; de la Vega & Korley, 2019. Int. J. Pest Mngmt., DOI:10.1080/09670874.2018.1547460; dos Santos et al., 2017, PLOS ONE, https://doi.org/10.1371/journal. pone.0174318; Vejrum Ørsted & Ørsted 2018, J. Appl. Ecol., DOI: 10.1111/1365-2664.13285). Also, use these sites to compare with results.
Response 3: Since the reference section did not appear in the submitted version, some of the reference examples recommended here, are already cited in the manuscript (Ref 31. Castro -Sosa et al 2017, Ref 30. Vejrum Ørsted & Ørsted 2018). The other references were added in the introduction and results sections.
Point 4: Should compare with areas of crop hosts.
Response 4: Thank you for your comment, but unfortunately, we don't currently have the data to make this kind of comparison. This is something that we will take in consideration.
Author Response File: Author Response.pdf
Reviewer 4 Report
Review
Manuscript Number: agriculture-2569861
The study entitled as “Prediction of the potential distribution of Drosophila suzukii on Madeira Island using the Maximum Entropy Modeling” established a Maxent model to predict the suitable areas in Madeira Island and analyzed the main influencing factors. In the current form, I do not recommend publishing, there are several issues - major issues are listed below.
1). In section 2.2, the authors considered the issue of matching the resolution of distribution points and bioclimatic data, but the multicollinearity between the bioclimatic data was not taken into account, which may be one of the reasons for the low AUC values of the models.
2). In section 2.3, Data Modeling. The choice of feature combination and regularization multiplier determine the performance of the model, and I'm curious how the authors determined this, as it usually requires further analysis.
3). Line 145. According to the authors' classification, the model performance is regular (0.7-0.8) and the model settings should be improved to obtain better performance, otherwise the predictions are less credible.
4). The results indicate that elevation is the main factor affecting this species, and the authors are advised to put elevation changes and distribution points together for a better visualization.
5). There are many similar studies on the distribution of Drosophila suzukii, and the authors do not discuss it accordingly.
5). It's unbelievable that the article is missing the references section.
Author Response
Response to Reviewer 4 Comments
Manuscript ID: agriculture-2569861
Type of manuscript: Communication
Title: Prediction of the potential distribution of Drosophila suzukii on Madeira Island using the Maximum Entropy Modeling
Authors: Fabrício Lopes de Macedo, Carla Ragonezi *, Fábio Reis, José G. R. de Freitas, David Horta Lopes, António Miguel Franquinho Aguiar, Délia Cravo, Miguel Â. A. Pinheiro de Carvalho
The authors appreciate the reviewer’s comments about the manuscript. The comments were addressed properly below and can be visualized in the resubmitted version.
Point 1: In section 2.2, the authors considered the issue of matching the resolution of distribution points and bioclimatic data, but the multicollinearity between the bioclimatic data was not taken into account, which may be one of the reasons for the low AUC values of the models.
Response 1: Thank you very much for your comments. We didn't take the issue of multicollinearity into account. We ran the MaxEnt process again, removing the highly correlated variables, and found that the AUC value changed slightly. As a result, we decided to leave this article in its current format. The authors appreciate the reviewer’s comments.
Point 2: In section 2.3, Data Modeling. The choice of feature combination and regularization multiplier determines the performance of the model, and I'm curious how the authors determined this, as it usually requires further analysis.
Response 2: As this is our initial work, we took into account the parameters used in other scientific articles and tested them to see which would be the best fit. We know we need to improve on this for future work and thank you in advance for your attention.
Point 3: Line 145. According to the authors' classification, the model performance is regular (0.7-0.8) and the model settings should be improved to obtain better performance, otherwise the predictions are less credible.
Response 3: We are in accordance with this comment and will consider this in the future.
Point 4: The results indicate that elevation is the main factor affecting this species, and the authors are advised to put elevation changes and distribution points together for better visualization.
Response 4: The information was added and can be seen in the resubmitted version.
Point 5: There are many similar studies on the distribution of Drosophila suzukii, and the authors do not discuss it accordingly.
Response 5: The authors add new references in the Introduction Results and Discussion sections. Please see the resubmitted version.
Point 6: It's unbelievable that the article is missing the references section.
Response 6: Regarding the reference, the authors already contacted the Managing Editor to know why the reference section was not in the version sent to the reviewers.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
This paper can be accepted
Author Response
Response to Reviewer 1 Comments (2º round)
Manuscript ID: agriculture-2569861
Type of manuscript: Communication
Title: Prediction of the potential distribution of Drosophila suzukii on Madeira Island using the Maximum Entropy Modeling
Authors: Fabrício Lopes de Macedo, Carla Ragonezi *, Fábio Reis, José G. R. de Freitas, David Horta Lopes, António Miguel Franquinho Aguiar, Délia Cravo, Miguel Â. A. Pinheiro de Carvalho
The authors appreciate the reviewer’s comments about the manuscript and the authors are grateful for the review.
Reviewer 4 Report
The authors improved their paper significantly. In 2.3, if the authors cite modeling parameters from other articles, please indicate and cite them.
Author Response
Response to Reviewer 4 Comments (2º round)
Manuscript ID: agriculture-2569861
Type of manuscript: Communication
Title: Prediction of the potential distribution of Drosophila suzukii on Madeira Island using the Maximum Entropy Modeling
Authors: Fabrício Lopes de Macedo, Carla Ragonezi *, Fábio Reis, José G. R. de Freitas, David Horta Lopes, António Miguel Franquinho Aguiar, Délia Cravo, Miguel Â. A. Pinheiro de Carvalho
The authors appreciate the reviewer’s comments about the manuscript.
Point 1: The authors improved their paper significantly. In 2.3, if the authors cite modeling parameters from other articles, please indicate and cite them.
Response 1: The comment was addressed properly and can be visualized in the resubmitted version.
