Wildfire Susceptibility of Land Use and Topographic Features in the Western United States: Implications for the Landscape Management

Round 1

Reviewer 1 Report

Dear authors!

Thank you for your manuscript. I think your article is in the scope of Forests journal.

But current version has some disadvantages.

Title and concept of the article

I work in forest fire topic more the two last decades.

Always terminology "forest fire danger" was used in such works.

Why did you use terminology "wildfire susceptibility"? I think this is wrong concept and terminology adopted last years. Please, explain you vision of this problem within your article.

I think you should use following title: Forest fire danger caused by land use and topographic features in the western United States: Implications for the landscape management

Abstract

Please, rework abstract accodring commom structure: Introduction, Background, Methods, Results and Discussion, Conclusion makin emphasize on the novelty of your research.

Introduction

You provided Introduction with integrated Background part of manuscript. I found only 36 cited references in this section. It is too small for such complex problem of forest fires. You must provide separate section with Background taking into account at least 50-60 unique references. You should show international background of the problem because Forests is an international journal.

Materials and Methods

Please, increase the size of Figure 1 for readers suitability.

I think you must rework Figure 2 taking into account land use zoning over the considered territory. And you should increase the size of the picture 2.

You consider land use and topographic factors of the forest fires. As you knew, forest fire is a multistage phenomenon with following stages: inert heating, drying, pyrolysis, ignition, flame combustion, afterburning. I did not understand why did you not consider ignition sources within your research like human factors and lightnings. You can not eliminate ignition stage from your research. You must provide in Materials and Methods section some information about ignition sources. It is obligatery in case of forest fire susceptibility (I do not agree with this terminology. You should use forest fire danger terminology. In my opinion, susceptibility is non-scientific explanation of forest fire danger).

Lets see the formulae (1). In fact, this this part of burned area of forested territory. In terminology of risk-analysis wi is a damage to forested territory. But you also must provide the probability pi of forest fire. In this case, we can use pi multiply by wi. This will be pure risk analysis explanation of forest fire danger and risk over controlled territory. But you must use ignition sources to compute probability pi.

Results

Table 1 must be modified taking into account ignition sources.

Also, you must provide results og pi and wi computing over the considered territory.

Discussion

This section must be reworked taking into account changes in Results section.

Please, provide information on limitations of your research.

Conclusions

You must completely rework this section taling into account ignition sources.

You should mark out 3-4 key findings with corresponding conclusions.

Please, provide information on future researches within this section.

References

This section must be extended with new references for Background section.

Author Response

Response to Reviewer # 1

1. Why did you use terminology "wildfire susceptibility"? I think this is wrong concept and terminology adopted last years. Please, explain you vision of this problem within your article. I think you should use following title: Forest fire danger caused by land use and topographic features in the western United States: Implications for the landscape management

Reply:

Thank you very much for your insights. The “wildfire susceptibility” terminology has been widely used in many journals article and has been defined as “a measure of land propensity for the occurrences of wildfires based on terrain’s intrinsic” (Leuenberger et al., 2018). Forest fire danger is defined as “a general term used to express an assessment of both fixed and variable factors of the fire environment that determine the ease of initiation, rate of spread, difficulty of control, and fire impact” (Merrill and Alexander, 1987). In our manuscript, we focused on the ease of initiation. In the revised manuscript we think “wildfire occurrence” might be a better alternative terminology and change the title to “Wildfire occurrence of land use and topographic features in the western United States: Implications for the landscape management”.

2. Please, rework abstract according common structure: Introduction, Background, Methods, Results and Discussion, Conclusion making emphasize on the novelty of your research.

Reply:

We revised the abstract as suggested.

3. Introduction: You provided Introduction with integrated Background part of manuscript. I found only 36 cited references in this section. It is too small for such complex problem of forest fires. You must provide separate section with Background taking into account at least 50-60 unique references. You should show international background of the problem because Forests is an international journal.

Reply:

Thank you for the suggestions. We followed instruction for authors suggested by Forests in structuring our manuscript. Forest suggests the structure should include an Abstract, Keywords, Introduction, Materials and Methods, Results, Discussion, and Conclusion. We also checked previously published manuscript in the journal and therefore we integrated background in the introduction section.

As suggested, we have revised our introduction section citing current research article. We have added some of the international journal to make the section readable for international readers as well.

4. Materials and Methods: Please, increase the size of Figure 1 for readers suitability.

Reply:

The full-size map will be uploaded separately during the submission. For the online readers, they will be able to click on the map to see the full-size image.

5. I think you must rework Figure 2 taking into account land use zoning over the considered territory. And you should increase the size of the picture 2.

Reply:

Thank you for the suggestion. We have included land use/ land cover map in figure 4. To make it simple and readable we tried to separate figure 2 (ignition points and wildfire boundary) and figure 4 (land cover map).

The full-size map will be uploaded separately during the submission.

6. You consider land use and topographic factors of the forest fires. As you knew, forest fire is a multistage phenomenon with following stages: inert heating, drying, pyrolysis, ignition, flame combustion, afterburning. I did not understand why did you not consider ignition sources within your research like human factors and lightnings. You can not eliminate ignition stage from your research. You must provide in Materials and Methods section some information about ignition sources. It is obligatory in case of forest fire susceptibility (I do not agree with this terminology. You should use forest fire danger terminology. In my opinion, susceptibility is non-scientific explanation of forest fire danger).

Reply:

The reviewer is correct in this observation. We did not consider ignition sources in our study. We only considered larger wildfire following past literature (Moreira et al., 2001; Nunes, 2001; Moreira et al., 2009; Carmo et al., 2011; Oliveira et al., 2013) and how topographical feature and land cover is responsible in its occurrence. We assumed regardless of ignition sources, fire burn to certain degree (>405 ha) contribute most of the total area burned and investigating this area will be interesting to readers.

We replaced “fire susceptibility” with “fire occurrence” in the revised manuscript.

7. Lets see the formulae (1). In fact, this part of burned area of forested territory. In terminology of risk-analysis wi is a damage to forested territory. But you also must provide the probability pi of forest fire. In this case, we can use pi multiply by wi. This will be pure risk analysis explanation of forest fire danger and risk over controlled territory. But you must use ignition sources to compute probability pi.

Reply:

The reviewer is correct of having pi to compute fire risk. The study followed Manly et al. (1993) to compute wildfire proneness. For risk analysis we need so many other factors such as pi, ignition sources etc. as suggested by the reviewer which is beyond the research objective. In this manuscript we assessed topographical feature and land cover impact on wildfire occurrence of size greater than 405 ha.

8. Results: Table 1 must be modified taking into account ignition sources. Also, you must provide results og pi and wi computing over the considered territory.

Reply:

Please see response to comments 6 and 7. We also have added limitation of the study as ignition sources, starting by the sentence “In addition, this research did not distinguish between human-induced wildfires and lightning-induced wildfires.”

9. This section must be reworked taking into account changes in Results section.

Reply:

The discussion section has been revised.

10. Please, provide information on limitations of your research.

Reply:

We have added limitation of our research in the last paragraph of the discussion section.

11. You must completely rework this section taling into account ignition sources.

Reply:

We have added one of the limitations of our research as not accounting ignition sources and tailored future researcher to consider it for risk analysis.  

12. You should mark out 3-4 key findings with corresponding conclusions.

Please, provide information on future researches within this section.

Reply:

We have revised the conclusion section and provided what future researcher may consider in discussion section.

13. This section must be extended with new references for Background section.

Reply:

In the revised version we have added new citation and updated reference section accordingly.

Reviewer 2 Report

Dear authors
Can the following information be added to the introduction or discussion?
Grasses are combustible material, but are there other accumulations of fire-promoting material in the form of dead, standing or lying trees in the study areas? Is there an increased fire risk in such cases?
Is it known how fires affect biodiversity, including the microbiome or bacteriomes in the soils of the study areas? Are soils sterilised by temperature? Do pathogens that live a hidden life in the roots of trees survive? or should this aspect be studied in the future? I wonder how quickly insects and fungi return after fires? Are there perhaps organisms that take advantage of the opportunities that open up for them after fires? Such species can probably occur en masse?
Does the resulting ash fertilise the soil or is it quickly blown away or washed into the ground and the new forest generation does not use it? Will the species composition be appropriately modified in the planning of the new forest to reduce fire risk? Are these wilderness areas part of parks and are such preventive measures not implemented at all?

Author Response

1. Grasses are combustible material, but are there other accumulations of fire-promoting material in the form of dead, standing or lying trees in the study areas? Is there an increased fire risk in such cases?

Reply:

Thank you for your comments. The accumulations of fire-promoting materials in the form of dead, standing or lying trees in the study areas could be referred as fuel accumulation. We have revised the sentence in Line 316 to read “Land use influences fire occurrence directly through fuel accumulation (e.g., dead, standing or lying trees) and fuel continuity….”

2. Is it known how fires affect biodiversity, including the microbiome or bacteriomes in the soils of the study areas? Are soils sterilised by temperature? Do pathogens that live a hidden life in the roots of trees survive? or should this aspect be studied in the future? I wonder how quickly insects and fungi return after fires? Are there perhaps organisms that take advantage of the opportunities that open up for them after fires? Such species can probably occur en masse?
   Does the resulting ash fertilise the soil or is it quickly blown away or washed into the ground and the new forest generation does not use it? Will the species composition be appropriately modified in the planning of the new forest to reduce fire risk? Are these wilderness areas part of parks and are such preventive measures not implemented at all?

Reply:

Thank you for your insights. These ideas sound very interesting. However, these ecological aspects need to be studied separately, which is beyond our study objectives. The objective of this research is to examine how land cover types and topographic feature influence the wildfire occurrence in Washington, Oregon, and California.

Round 2

Reviewer 1 Report

Dear authors!

Thank you for your revision.

But I steel have to provide some comments on disadvantages of last version of manuscript.

Response 1: I agree with you.

Response 2: OK

Response 3: OK

Response 4: OK

Response 5: OK

Response 6: Sorry? I do not agree with you. The measure of "Fire Occurrence" is probability. In this case your research will sounds scientificly. You can suggest to readers data on most burned areas, but you must also provide the layer with probability of occurrence. The simplest way to take into account such factor is calculation of probability through the frequences. For example, probability of occurrence can be calculated like relationship of fire days to non fire days both for lightning activity and human activity factors. You can use data on forest fire ignitions to calculate this data.

Response 7: I do not agree with you again. Let's repeat, measure of occurrence is probability. You must take into account this data.

Response 8: I do not agree with you. You should at least put this reasons into Table with notation "Not taking into account".

Response 9: This section must be reworked with Results changes.

Response 10: OK

Response 11: Please, rework according to new Results

Response 12: Please, rework with new Results

Response 13: OK

Please, rework your manuscript.

Author Response

1. You consider land use and topographic factors of the forest fires. As you knew, forest fire is a multistage phenomenon with following stages: inert heating, drying, pyrolysis, ignition, flame combustion, afterburning. I did not understand why did you not consider ignition sources within your research like human factors and lightnings. You can not eliminate ignition stage from your research. You must provide in Materials and Methods section some information about ignition sources. It is obligatory in case of forest fire susceptibility (I do not agree with this terminology. You should use forest fire danger terminology. In my opinion, susceptibility is non-scientific explanation of forest fire danger).

Earlier Reply:

The reviewer is correct in this observation. We did not consider ignition sources in our study. We only considered larger wildfire following past literature (Moreira et al., 2001; Nunes, 2001; Moreira et al., 2009; Carmo et al., 2011; Oliveira et al., 2013) and how topographical feature and land cover is responsible in its occurrence. We assumed regardless of ignition sources, fire burn to certain degree (>405 ha) contribute most of the total area burned and investigating this area will be interesting to readers.

We replaced “fire susceptibility” with “fire occurrence” in the revised manuscript.

Reviewer’s further comments:

Response 6: Sorry? I do not agree with you. The measure of "Fire Occurrence" is probability. In this case your research will sounds scientifically. You can suggest to readers data on most burned areas, but you must also provide the layer with probability of occurrence. The simplest way to take into account such factor is calculation of probability through the frequencies. For example, probability of occurrence can be calculated like relationship of fire days to non fire days both for lightning activity and human activity factors. You can use data on forest fire ignitions to calculate this data.

Current Reply:

Thank you for additional insights. We agree to the reviewer to some extent that we did not investigate “fire occurrence” which refers to average number of fires in a specified area during a specified time period (or probability of fire occurrence) but “susceptibility” which refers to the propensity of an area to be damaged if a wildfire occurs. We investigated how topographical features and land use add as a catalyst to wildfire damage. In other word we investigated “fire susceptibility”. After careful consideration, we decided to keep “fire susceptibility” in our title. We came to this conclusion by reading number of manuscripts and interpreting following definition from different peer-reviewed literature:    

“The term fire occurrence is used to describe the presence and frequency of fires within a finite time and space. Fire occurrence statistics account for all reported ignitions regardless of the area they burn or damage they cause” (Plucinski, 2012).

The “wildfire susceptibility” terminology has been widely used in many journals article and has been defined as “a measure of land propensity for the occurrences of wildfires based on terrain’s intrinsic” (Leuenberger et al., 2018).

Forest fire danger is defined as “a general term used to express an assessment of both fixed and variable factors of the fire environment that determine the ease of initiation, rate of spread, difficulty of control, and fire impact” (Merrill and Alexander, 1987).

The objective of this paper is just to examine the land propensity for the occurrence of wildfires with respect to land cover and topographic features. In this manuscript we did not investigate what cause fire (source of ignition) or the probability of fire occurrence. The objective of this paper is entirely different. It would be however worth investigating probability in addition to what we did as reviewer suggested but because of funding, time, data availability, and other constraint we limit our research to the current version.

2. Lets see the formulae (1). In fact, this part of burned area of forested territory. In terminology of risk-analysis wi is a damage to forested territory. But you also must provide the probability pi of forest fire. In this case, we can use pi multiply by wi. This will be pure risk analysis explanation of forest fire danger and risk over controlled territory. But you must use ignition sources to compute probability pi.

Earlier Reply:

The reviewer is correct of having pi to compute fire risk. The study followed Manly et al. (1993) to compute wildfire proneness. For risk analysis we need so many other factors such as pi, ignition sources etc. as suggested by the reviewer which is beyond the research objective. In this manuscript we assessed topographical feature and land cover impact on wildfire occurrence of size greater than 405 ha.

Response 7: I do not agree with you again. Let's repeat, measure of occurrence is probability. You must take into account this data.

Current Reply:

Thank you for your suggestions. We agree with reviewer in investigating “fire probability” in addition to what we present in this manuscript will be interesting to readers. However, because of many constraints as discussed earlier we limit our research to current version. However, we direct future research to consider “fire probability” part in their research.

3. Results: Table 1 must be modified taking into account ignition sources. Also, you must provide results og pi and wi computing over the considered territory.

Earlier Reply:

Please see response to comments 6 and 7. We also have added limitation of the study as ignition sources to read “In addition, this research did not distinguish between human-induced wildfires and lightning-induced wildfires.”

Response 8: I do not agree with you. You should at least put this reasons into Table with notation "Not taking into account".

Current Reply:

Thank you for your suggestions. We have added “the ignition sources are not considered in this research” in the note of Figure 3.

4. This section must be reworked taking into account changes in Results section.

Earlier Reply:

The discussion section has been revised.

Response 9: This section must be reworked with Results changes.

Current Reply:

Thank you for your suggestions. We did not make significant changes in our results as we did not compute wildfire probability. However, discussion section has been revised with added limitation, direction to future researcher, and other minor changes.

5. You must completely rework this section taking into account ignition sources.

Earlier Reply:

We have added one of the limitations of our research as not accounting ignition sources and tailored future researcher to consider it for risk analysis. 

Response 11: Please, rework according to new Results

Current Reply:

We did not compute probability and did not account ignition sources. The objective of this research is just to investigate likelihood of wildfire caused by topographical feature and land use. We have revised the results based these two major factors in fire likelihood and added other aspects to consider in future research.  

6. You should mark out 3-4 key findings with corresponding conclusions.

Please, provide information on future researches within this section.

Earlier Reply:

We have revised the conclusion section and provided what future researcher may consider in discussion section.

Response 12: Please, rework with new Results

Current Reply:

Thank you for your suggestions. We have revised the results.