Spatial Dependencies and Neighbour Interactions of Wildfire Patterns in Galician Mountain Areas (NW Spain)
Round 1
Reviewer 1 Report
The paper presents spatial influences on wildfire patterns in Galicia.
It is well written, with good structure and without extraneous material.
I would like to recommend publication with minor revision.
The findings are very well supported by presented data. Nevertheless, I would like to clarify just some aspects:
- What does k represent in equation (3)?
- In line 292 it is mentioned that lag 1 is the optimum spatial lag and, thus, this is the one used to measure correlation with the dependent variables. As is possible to observe in figure 6, for all variables, there are common values in the confidence intervals of Moran’s I for lag-1 and lag-2. Have the authors made any test to analyze whether higher values are reached for lag-1? This idea is clearer in the conclusions’ section (lines 456 -460).
- Line 295 – a correlation test could also be addressed to prove the significance of positive correlation between variables.
- Line 353-363: it seems a little weird that shrub cleanings could increase the frequency of fires. Though, the reasons presented by the authors seem plausible.
- Line 393: why do common lands increase the affected area of wildfires?
- It is mentioned along the manuscript that other variables may also have an influence on wildfires behavior. Even though it is generally mentioned at the end of the manuscript, perhaps, some specific possibilities could be referred (for instance, wind, slope, existence of paths, among others).
Author Response
Please see the attachment
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This manuscript proposes an analysis of the spatial wildfire patterns in the Galician mountain area in terms of frequency, ratio of affected area, suppression time and extension. Policy improvements are given to take into account neighbouring interconnections of wildfire risk and impact.
The manuscript is written clearly and concisely.
The approach is original and interesting.
The questions, parameters, methods and statistic approach are clearly stated and put in evidence the effect of spatial heterogeneity and land use in relation with wildfire risk and impact.
The spatial statistical and econometric models described lead to results and conclusions.
The results and discussion aim to bring a better understanding of the impact of the relation between dependent and independent variables (frequency, ratio of affected area, ratio of land use etc).
The results are presented in many maps that are well suitable explicit of the impact of variables.
The interest of the proposed paper is the exhaustive approach of the subject.
The subject is logically presented and developed. The flow of the paper is logical and clear, it includes sufficient details on the background importance.
Tables and figures are well presented sufficient and necessary. The review of the literature is adequate.
The length of the presentation can be considered as appropriate in order to provide enough elements in the description of the methods.
Conclusions are adequate.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Reviewer 3 Report
The article presents an economic and statistical approach to fire issues in the Spanish mountains of Galicia.
I have a few comments on the article:
- It should definitely be supplemented with an environmental background. In the first two chapters there is no information about the type of environment, forests, in which the discussed fires occur.
- Figures with maps need to be improved.
- References are missing in some places in the text.
I have put much more comments in the uploaded file.
Comments for author File: Comments.pdf
Author Response
Please see the attachment
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
I am satisfied with the changes introduced by the authors. Changes have been made to all parts that I have marked as questionable or incorrect. The text has been supplemented with additional references and their explanations. The figures have been improved. Two minor comments:
- in figure 3, Ancares-Caurel: Without zooming, the scale is illegible (font size too small).
- intervals: I didn't know that the statistics assume that the interval is closed on the left and open on the right. In some fields, this conjecture is not used, so I'd be careful with that. Perhaps it is better to consider a more universal and unambiguous notation.
Author Response
Referee comment 1: in figure 3, Ancares-Caurel: Without zooming, the scale is illegible (font size too small).
Authors' answer: We have redone figure 3 with a larger font for the scale.
Referee comment 2: intervals: I didn't know that the statistics assume that the interval is closed on the left and open on the right. In some fields, this conjecture is not used, so I'd be careful with that. Perhaps it is better to consider a more universal and unambiguous notation.
Authors' answer: We have added a comment in the manuscript (marked in red), namely "Note that this figure uses the usual $[a,b)$ convention for intervals", with a reference to the tmap package with which this map and the intervals were obtained. The tmap package, as well as all the statistical software that we are aware of, uses the [a,b) convention by default.
