A Comparative Study of Forest Fire Mapping Using GIS-Based Data Mining Approaches in Western Iran
, Mehdi Mirzaei Kurdalivand 3, Sabri Rasooli 4, Chaolong Yao 5 and Tongxin Hu 6,*
Round 1
Reviewer 1 Report (Previous Reviewer 1)
The manuscript still lacks a clear focus. The title and the introduction section indicates that it is going to develop a new methodology, however, the results, dicussion and conclustion sections shift their focuses more to applications. It is strongly suggested that the text must be reorganized accordingly to focus on either methodolgy or a scientific question.
Author Response
The manuscript still lacks a clear focus. The title and the introduction section indicates that it is going to develop a new methodology, however, the results, dicussion and conclustion sections shift their focuses more to applications. It is strongly suggested that the text must be reorganized accordingly to focus on either methodolgy or a scientific question.
Response: Thank you for providing constructive comments that have helped to present the focus of our study more clearly. We have changed the title to “A comparative study of forest fire mapping using GIS-based data mining approaches in western Iran”. We think it is much better to tell the story of our work clearly. If necessary, we would be willing to make changes following the suggestions from the reviewers. In addition, we have revised the Introduction Section accordingly to make it coincide with the title. Thank you again.
Author Response File: 
Author Response.pdf
Reviewer 2 Report (Previous Reviewer 3)
Unfortunately the authors' comments sent ina separate file do not include my review points, it like like they confused some of the reviewers' comments.
Hpwever in the present form the paper is ok, I had minor revision advised to the authors, they worked on the paper and the overall merit is enough to accept it
Author Response
Reviewer 2：
Unfortunately the authors' comments sent in a separate file do not include my review points, it like they confused some of the reviewers' comments.
However in the present form the paper is ok, I had minor revision advised to the authors, they worked on the paper and the overall merit is enough to accept it
Response: We apologize for not including your review points in the responses for resubmission. Thank you very much for taking the time and the efforts in reviewing our manuscript. We really appreciate it.
Author Response File: Author Response.pdf
Reviewer 3 Report (New Reviewer)
The subject of the peer-reviewed article is related to fire risk mapping. For this purpose, the authors used two known models, EBF and WOE. For the risk analysis, 284 examples of fires that took place in Marivan County of Kurdistan province, Iran, were adopted. They were divided into two groups, one for testing and the other for validation. The influence of various factors such as distance from the river, roads and settlements, maximum annual temperatures etc. on the zone fire risk was investigated. The results of this work may be useful in preventing and forecasting fires in a selected area.
I have some comments, suggestions and remarks about the work. They are listed below:
Figure 1 – Marivan City is hardly visible
Figure 3 - Letter markings a, b, c etc. and explanations of individual drawings are hardly visible. Also, instead of the first “h”, there should be “f”. The title of the drawing should be on the same page as the drawing
Line 131 - The word "It" is missing at the beginning of the sentence
Equation (1), (2) etc. – numbers of equations should be aligned to the right margin
Line 135 - no explanation of what classes are meant here
Lines 143 and 144 - explanations of the quantities of Wi+ and Wi- should be moved before the quantities of Npix1, Npix2 ...
Line 146 - one word "the" should be removed
Equation (4) - The sum in the numerator can be abbreviated as follows . The distance between the sum sign and the expression in parentheses is too great
Equation (5) - The sum in the numerator can be abbreviated as follows . The distance between the sum sign and the expression in parentheses is too great
Equation (6) - There is no closing parenthesis in the numerator. The value of "Unc" on the left side of the equal sign should be without parentheses and be on the same level as the equal sign
Section 2.4. – There is no information on how the values of Dis, Bel and Unc are determined for individual factors
Section 2.5. This subsection can be deleted and the text moved to the appropriate place in chapters 3 and 4
Table 1 - The words in the first column should not be carried over
Line 310 - The font for the region name is too small
Author Response
Reviewer 3：
The subject of the peer-reviewed article is related to fire risk mapping. For this purpose, the authors used two known models, EBF and WOE. For the risk analysis, 284 examples of fires that took place in Marivan County of Kurdistan province, Iran, were adopted. They were divided into two groups, one for testing and the other for validation. The influence of various factors such as distance from the river, roads and settlements, maximum annual temperatures etc. on the zone fire risk was investigated. The results of this work may be useful in preventing and forecasting fires in a selected area.
Response: Thank you for providing constructive comments that have helped substantially in improving the quality of the paper. We appreciate it.
I have some comments, suggestions and remarks about the work. They are listed below:
Figure 1 – Marivan City is hardly visible
Response: The figure 1 has been modified.
Figure 3 - Letter markings a, b, c etc. and explanations of individual drawings are hardly visible. Also, instead of the first “h”, there should be “f”. The title of the drawing should be on the same page as the drawing
Response: The figure a to J have been modified.
Line 131 - The word "It" is missing at the beginning of the sentence
Response: Thank you. We have added it in the revised manuscript.
Equation (1), (2) etc. – numbers of equations should be aligned to the right margin
Response: We have corrected in the revised manuscript.
Line 135 - no explanation of what classes are meant here
Response: This comment is not clear and I do not understand the purpose of the reviewers.
The classification of the forest fire ignition factors were done according to expert opinion, similar studies and regional conditions.
Lines 143 and 144 - explanations of the quantities of Wi+ and Wi- should be moved before the quantities of Npix1, Npix2 ...
Response: Thank you for the comment. The suggested correction has been made in the revised manuscript.
Line 146 - one word "the" should be removed
Response: We have removed it in the revised manuscript.
Equation (4) - The sum in the numerator can be abbreviated as follows . The distance between the sum sign and the expression in parentheses is too great
Response: The equations have been modified.
Equation (5) - The sum in the numerator can be abbreviated as follows . The distance between the sum sign and the expression in parentheses is too great
Response: The equations have been modified.
Equation (6) - There is no closing parenthesis in the numerator. The value of "Unc" on the left side of the equal sign should be without parentheses and be on the same level as the equal sign
Response: The equations have been modified.
Section 2.4. – There is no information on how the values of Dis, Bel and Unc are determined for individual factors
Response: First, the equations related to the EBF model (Dis, Bel and Unc) are formulated in Excel software and the value related to the classes of each factor will be determined.
Section 2.5. This subsection can be deleted and the text moved to the appropriate place in chapters 3 and 4
Response: According to similar studies, section 2.5 is necessary and should be a distinct section.
Table 1 - The words in the first column should not be carried over
Response: The standard method of presenting information in the form of a table in these models is as mentioned.
Line 310 - The font for the region name is too small
Response: The font of the desired word is similar to the font of other parts of the article.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report (Previous Reviewer 1)
It seems that the manuscript has been improved. Please accept it in its current form.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
The manuscript lacks of creativity. Methods applied in this study are basically established ones in the literature. Applying them in a specific area shows very low originality. Instead, I woud suggest the authors to reorganize the text from a different perspective, emphasing scientific questions in the specific region. Another point is that the manuscript needs a clear statement of research question. The title states forest fire is the main focus, while in the main text rangeland fire appears here and there.
Reviewer 2 Report
I have reviewed the paper titled “Forest fire forecasting using data mining approaches and meteorological data, a case study”. The paper's main aim is to use data mining methods for forest fire risk mapping. The article's title is misleading since it was about forecasting, but the main of the paper is a kind of susceptibility mapping, which is different from forecasting. Therefore, I advise the authors to change the article's title. I have the following concerns about the paper
1. Authors have used EBF and WOE methods, but these are not state-of-the-art data mining methods. Instead, I would suggest authors use models like the random forest or support vector machines.
2. Authors have used fire incidences as point data, but what is the justification for using these point data? Are these points the source of the fire?
3. The introduction of the paper lacks references that will justify the use of factors and the methods of probability mapping.
4. The authors failed to justify the use of the method as well as the importance of the paper in the introduction section.
5. It is not clear why the authors have chosen their study area.
6. The authors have not discussed the fire occurrences in the study area.
I think the paper is well written, but the problem is it is not an excellent way to prepare a forest fire potential map using the method of landslide susceptibility mapping. Forest fire is not like source-based problems like landslides. Moreover, the introduction section lacks the justification of the paper. I suggest authors spend more time reading more works in the field and make their work justifiable and essential to the scientific community.
I have the following concerns as well
1. Table 3. Row 1 and 2 are misleading.
2. Please improve the graphics of figure 5.
3. Why are the legends of the maps larger or almost similar in size of the map?
4. Please provide a workflow, and give a detailed justification for using the factors or predictors.
5. Authors also have to justify the use of EBF and WOE.
Reviewer 3 Report
Introduction is well-structured, with proper referencing the previous existing research. The case-study approach gives the idea a significant level of contribution and novelty. The use of the methods is well-grounded.
Section for methods and materials includes all the description of the design of the experiment. I suggest that the Methods are described in a general way while a new section is introduced for the experiment design and all the steps the authors have undertaken to prepare the data and the analysis.
The results section is well-structured, the figures are clear. The table 1 is overwhelmed with data, the authors might think to divide it to shorter tables labelling each, however it is just a thought.
The Discussion section is good with relevant points and references, however some analysis of the limitations of the study would be good here to strengthen the critical view on the proposed model.
Conclusion section would benefit from some bullet-points statements highlighting the main contributions
The sources referenced in the paper are good, however some more up-to-date work would benefit the theoretical background of the paper, there few references "younger" than 2019.
The style of the paper is good and easy to follow, however some minor revision is required, for example as in the sentence below:
In these methods, IT IS assumed that conditions that ....
