Impact of Vertical Atmospheric Structure on an Atypical Fire in a Mountain Valley

Round 1

Reviewer 1 Report

The manuscript has been greatly improved. Below are minor suggestions:

 

- Line 121: You don't state a question as you say, so either replace 'question' or use another word such as hypothesis (and rephrase the sentence accordingly).

- Lines 178-182: Sounds like cross sections (in general) are for the BARRA-TA data specifically and the sentence is very long. Re-phrase this. 

- Lines 427-428: Include the sentence in the main text rather than the Table caption since it's a result. 

- Lines 437-438 and 442-443: Same as above comment.

Author Response

To,
The Reviewer 1,
Fire Editorial Office,
MDPI, St. Alban-Anlage 66, 4052 Basel, Switzerland.

Subject: Impact of vertically atmospheric phenomena on an atypical fire in mountain valley.

Dear Reviewer 1,
We appreciate your advice and have made corrections by following your advice.

Advice 1: You don't state a question as you say, so either replace 'question' or use another word such as hypothesis (and rephrase the sentence accordingly). (Line 121)
Response: We rephrased this part as “Our research aims to determine whether upper air winds had greater impacts on fire simulation in rugged terrain than other, less topographically complex areas of the fire.”

Advice 2: Sounds like cross sections (in general) are for the BARRA-TA data specifically and the sentence is very long. Re-phrase this (Line Lines 178-182)
Response: We made this sentence more concise.

Advice 3: Include the sentence in the main text rather than the Table caption since it's a result. (Lines 427-428)
Response: We modified it as advised.

Advice 4: Same as above comment (Lines 437-438 and 442-443)

Response: We modified it as advised.

 

 

Please find a copy of our research manuscript ‘OzakiM_2021_Wind_effect_on_Riveaux_Road_Fire’ for consideration for publication in Fire. 
We thank you for your time and consideration of our manuscript. 
Sincerely,
Mitsuhiro Ozaki, Rebecca Harris, Peter Love, Jagannath Aryal, Paul Fox-Hughes and Grant Williamson
Discipline of Geography and Spatial Sciences, 
School of Geography, Planning, and Spatial Sciences,
University of Tasmania, 
Private Bag 79, Hobart, 
Tasmania 7001, Australia
Email: [email protected]
Tel.: +61-4-02773191 (mobile)
 

Author Response File: Author Response.docx

Reviewer 2 Report

The use of demolished at line 51 seems odd. Can we just say destroyed, which seems more accurate. Some minor tidy up of headings and spelling is needed and clarity over US spelling vs UK spelling is needed. Eg behavor v behaviour. Neighbor v Neighbour. 

Author Response

To,
The Reviewer 2,
Fire Editorial Office,
MDPI, St. Alban-Anlage 66, 4052 Basel, Switzerland.

Subject: Impact of vertically atmospheric phenomena on an atypical fire in mountain valley.

Dear Reviewer 2,
We appreciate your advice and have made corrections by following your advice.

Advice 1: The use of demolished at line 51 seems odd. Can we just say destroyed, which seems more accurate. 
Response: “Demolished” is now replaced with destroyed as advised.

Advice 2: Some minor tidy up of headings and spelling is needed and clarity over US spelling vs UK spelling is needed. Eg behavor v behaviour. Neighbor v Neighbour ()
Response: We use US English for this journal

Please find a copy of our research manuscript ‘OzakiM_2021_Wind_effect_on_Riveaux_Road_Fire’ for consideration for publication in Fire. 
We thank you for your time and consideration of our manuscript. 
Sincerely,
Mitsuhiro Ozaki, Rebecca Harris, Peter Love, Jagannath Aryal, Paul Fox-Hughes and Grant Williamson
Discipline of Geography and Spatial Sciences, 
School of Geography, Planning, and Spatial Sciences,
University of Tasmania, 
Private Bag 79, Hobart, 
Tasmania 7001, Australia
Email: [email protected]
Tel.: +61-4-02773191 (mobile)
 

Author Response File: Author Response.docx

Reviewer 3 Report

This one is significantly improved compared to the previous version of the manuscript. In my opinion, the text is a valuable contribution to the general understanding of fire behaviour in a particular type of terrain. Now it can be published.

Author Response

To,
The Reviewer 3,
Fire Editorial Office,
MDPI, St. Alban-Anlage 66, 4052 Basel, Switzerland.

Subject: Impact of vertically atmospheric phenomena on an atypical fire in mountain valley.

Dear Reviewer 3,
We appreciate your review. In order to apply advice from other reviewers. We made some minor corrections.
Please find a copy of our research manuscript ‘OzakiM_2021_Wind_effect_on_Riveaux_Road_Fire’ for consideration for publication in Fire. 
We thank you for your time and consideration of our manuscript. 
Sincerely,
Mitsuhiro Ozaki, Rebecca Harris, Peter Love, Jagannath Aryal, Paul Fox-Hughes and Grant Williamson
Discipline of Geography and Spatial Sciences, 
School of Geography, Planning, and Spatial Sciences,
University of Tasmania, 
Private Bag 79, Hobart, 
Tasmania 7001, Australia
Email: [email protected]
Tel.: +61-4-02773191 (mobile)

 

Author Response File: Author Response.docx

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

General comments:   The study demonstrated the difficulties of simulating wildfires in complex terrain and with unstable and spatially diverse weather conditions during a wildfire in Australia. Multiple wind inputs were used in the simulations to attempt to improve results as compared to the observed fire perimeters, and various analyses of wind profiles showed the intricate conditions observed and simulated during the fires. The most well simulated fires were relatively small and occurred in regions with flat or mild terrain. This study uses fairly new models and techniques to simulate the fire. There are no major weaknesses or flaws with this manuscript.    The paper has been greatly improved compared to the previous submitted version. The suggestions below are generally to improve the paper grammatically and the flow of the paper rather than ask for clarification or request complete restructuring. The figures are all significantly improved as well.    I recommend that this paper be accepted into Fire after very minor revisions. The authors have taken all of the feedback previously submitted and have greatly improved the paper. It will fit well into the current body of literature.      Specific recommendations:  

Line 93: What does 'Australian conditions' mean? The weather in Australia, or the topography, or something else? You can drop 'conditions' unless you're referring to something specifically.

Line 108: Comma after 'topography'

  Lines 117-119: It's debatable whether to include results of your simulations in the introduction. I would remove the last part of this sentence; keep the reader in suspense for the results!   Lines 156-157 (Caption for Fig. 2): Put the information regarding how the sensitivity predicts vegetation recovery in the main paragraph and not in the caption.   Lines 349-350 (Caption for Fig. 4): The blue line is the #2, right? If so, add that in the caption.   Fig.s 5, 6: Can the isochrone be another color like white, so it does not blend in with the red grid cells? Also could you remove the 'Completed' option in the legend since there are no purple grid cells?   Lines 429-430: What do you mean by 'valid runs'?    Table 5: Do you have the exact wind direction in degrees for this?   Fig. 8. You can keep as-is if desired, but do you need all three Skew-T diagrams? Would one work and you can explain conditions were similar for that time period? I was expecting conditions to change since there were three diagrams but they look very similar.   Line 547 (Caption for Fig. 11): Remove either 'higher' or 'elevated'.   Lines 614-623: I would refrain from numbering within this paragraph, since you are already numbering each paragraph in this section separately.

Author Response

We appreciate your advice and have made corrections by following your advice. Following careful consideration of the reviewer concerns, we have substantially revised this paper, undertaking a new and simplified analysis that shifts the focus to upper-air interactions. As such there are significant changes to the manuscript. Although we have corrected some parts by following your advice, other parts were not reflected due to the overall change in manuscript structure and methods.

Advice 1: What does 'Australian conditions' mean? The weather in Australia, or the topography, or something else? You can drop 'conditions' unless you're referring to something specifically. (Line 93)
Response: “condition” has been removed.

Advice 2: Comma after 'topography' (Line 108)
Response: Due to the structure change, this advice is not reflected.

Advice 3: It's debatable whether to include results of your simulations in the introduction. I would remove the last part of this sentence; keep the reader in suspense for the results! (Line 117-9)
Response: The sentence has already been changed as advised.

Advice 4: Put the information regarding how the sensitivity predicts vegetation recovery in the main paragraph and not in the caption (Lines 156-157 (Caption for Fig. 2))
Response: Now the sensitivities are addressed in the main text as advised.

Advice 5: The blue line is the #2, right? If so, add that in the caption. (Lines 349-350 (Caption for Fig. 4))
Response: Fire isochrones are addressed in the caption and the legend as advised.

Advice 6:  Can the isochrone be another color like white, so it does not blend in with the red grid cells? Also could you remove the 'Completed' option in the legend since there are no purple grid cells? (Fig.s 5, 6)
Response: Now the isochrone color is more emphasized than before. Complete option has been removed as advised.

Advice 7: What do you mean by 'valid runs'?  (Lines 429-430)
Response: They meant the simulations which reach the threshold. Now they are explained so. 

Advice 8: Do you have the exact wind direction in degrees for this? (Table 5)
Response: Because we have changed the strategy, we do not focus on the wind direction at ignition anymore. Instead, we measured conducive structures during the fire period. 

Advice 9: You can keep as-is if desired, but do you need all three Skew-T diagrams? Would one work and you can explain conditions were similar for that time period? I was expecting conditions to change since there were three diagrams, but they look very similar. (Fig. 8)
Response: We have removed Skew-T diagrams. Instead, we counted conducive structures not only at ignition but also the period of fire propagation by using cross-section.

Advice 10: Remove either 'higher' or 'elevated'. (Line 547 (Caption for Fig. 11))
Response: This figure has been removed as advised.

Advice 11: I would refrain from numbering within this paragraph, since you are already numbering each paragraph in this section separately (Lines 614-623)
Response:  The nested numbering has been removed as advised.

 

 

Please find a copy of our research manuscript ‘OzakiM_2021_Wind_effect_on_Riveaux_Road_Fire’ for consideration for publication in Fire. 
We thank you for your time and consideration of our manuscript. 
Sincerely,
Mitsuhiro Ozaki, Rebecca Harris, Peter Love, Jagannath Aryal, Paul Fox-Hughes and Grant Williamson
Discipline of Geography and Spatial Sciences, 
School of Geography, Planning, and Spatial Sciences,
University of Tasmania, 
Private Bag 79, Hobart, 
Tasmania 7001, Australia
Email: [email protected]
Tel.: +61-4-02773191 (mobile)

Author Response File: Author Response.docx

Reviewer 2 Report

Summary
The manuscript claims to present a numerical investigation of the influence of atmospheric structure on fire behavior. However, the presented work is presumably characterised by significant methodological flaws that need to be addressed. In addition, the presentation quality is very low that makes it really difficult to follow the text. Overall, my suggestion is to reject the manuscript. 

Major comments
1. The use of language in the manuscript is relatively poor with many grammatical and other errors, hindering the reader to follow the text. If the manuscript is to be considered for publication, the authors need to thoroughly revise the text. 
2. The last paragraph of the Introduction is truly very hard to follow. While the authors state the purpose of their study, the added value (accounting for the influence of atmosphere's vertical state) is not properly highlighted. The inclusion of text describing the model used (see also minor comments) distracts the reader from understanding why the presented study is an important one. This whole paragraph needs to be rewritten from scratch.
3. L160-168: I had a very hard time to follow this part of the text. Please, revise.   
4. Sect. 2.2.1: I did not manage to figure out if this section provides a general overview of the WindNinja and BARRA, or provides useful information on the setup used the study. The text is very complex to follow. Further, WindNinja is a software package (a model), whereas BARRA is a reanalysis product. Both are mixed in the text, making it really really hard to follow. 
5. I do not see any table that summarises the numerical experiments conducted. This would be useful to get an overview of the implemented methods. 
6. Sect. 2.2.5: The methodology described here is very unclear. I understand that two sets of simulations were performed; one driven by CoM-WindNinja and another one driven by CoMM-WindNinja. I also understand that the two simulations have different grid spacings. Could this be the reason for the differences seen between simulations and observations? Also, the authors say that the examine upper-air meteorology only when the high-resolution simulation differs from the observations. So, what is the reason for carrying out the first simulations (CoM-WindNinja). I suspect that there are serious methodological flows here that need to be addressed by the authors. 
7. Sect. 3: The presentation of the results does not allow the reader to draw his/her conclusions. The text is extremely hard to follow and I cannot verify that the conclusions are supported by the analysis. This also relates to my previous comment about the methodology followed. This issue needs to be addressed. 

Minor comments
1. Several times throughout the text, the authors refer to simulating "fire behaviours". In my view, it would more appropriate to use the singular, i.e., "fire behaviour". This is just and example of poor language use. 
2. L99-113: This part provides information on the modeling system used and should be merged with Sect. 2.2.2. 
3. What is the value that Figure 2 adds to the manuscript? It can be removed. 

Author Response

To,
The Reviewer 2,
Fire Editorial Office,
MDPI, St. Alban-Anlage 66, 4052 Basel, Switzerland.

Subject: Impact of vertically atmospheric phenomena on an atypical fire in mountain valley.

Dear Reviewer 2,
We appreciate your advice and have made corrections by following your advice. Following careful consideration of the reviewer concerns, we have substantially revised this paper, undertaking a new and simplified analysis that shifts the focus to upper-air interactions. As such there are significant changes to the manuscript. Although we have corrected some parts by following your advice, other parts were not reflected due to the overall change in manuscript structure and methods.

Advice 1: The manuscript claims to present a numerical investigation of the influence of atmospheric structure on fire behavior. However, the presented work is presumably characterised by significant methodological flaws that need to be addressed. In addition, the presentation quality is very low that makes it really difficult to follow the text. Overall, my suggestion is to reject the manuscript.
Response: The document structure has been changed. To be clear contrast, there are three groups with different wind datasets for experiments. One is experiment group and other two are control groups. Methods have been simpler than before. We counted the turbulent wind not only at ignition but also during the fire spreading. Then we conducted the fire simulations again. Now we address the association of ruggedness, upper air interaction and quality of fire simulation. We hope it is helpful to follow the text.

Advice 2: The use of language in the manuscript is relatively poor with many grammatical and other errors, hindering the reader to follow the text. If the manuscript is to be considered for publication, the authors need to thoroughly revise the text.
Response: The language has been checked.

Advice 3: The last paragraph of the Introduction is truly very hard to follow. While the authors state the purpose of their study, the added value (accounting for the influence of atmosphere's vertical state) is not properly highlighted. The inclusion of text describing the model used (see also minor comments) distracts the reader from understanding why the presented study is an important one. This whole paragraph needs to be rewritten from scratch
Response: This paragraph has been changed.

Advice 4: I had a very hard time to follow this part of the text. Please, revise (L160-168)
Response: The paragraph has been updated as advised.

Advice 5: I did not manage to figure out if this section provides a general overview of the WindNinja and BARRA, or provides useful information on the setup used the study. The text is very complex to follow. Further, WindNinja is a software package (a model), whereas BARRA is a reanalysis product. Both are mixed in the text, making it really really hard to follow. (Sect. 2.2.1)
Response: Yes, WindNinja is a software. We focus on not WindNinja but data resampled by WindNinja. Therefore, comparison of resampled data and BARRA wind are addressed in table 5. However, some explanation of WindNinja remains because we still think that it is necessary to describe how to setup downscaled data by WindNinja. 

Advice 6: I do not see any table that summarises the numerical experiments conducted. This would be useful to get an overview of the implemented methods
Response: Now a ground summary is available in Table 10, the result section, as advised.

Advice 7: The methodology described here is very unclear. I understand that two sets of simulations were performed; one driven by CoM-WindNinja and another one driven by CoMM-WindNinja. I also understand that the two simulations have different grid spacings. Could this be the reason for the differences seen between simulations and observations? (Sect. 2.2.5)
Response: The reason of using different types of wind is to see the impact of upper air interaction. To do so, we have counted the conducive structures during the fire propagation by cross section. Then we simulated the fire by using different types of winds. BARRA wind contains more upper air interactions than resampled winds. As a results, simulation with BARRA winds were better than the others in rugged terrain.

Advice 8: Also, the authors say that the examine upper-air meteorology only when the high-resolution simulation differs from the observations. So, what is the reason for carrying out the first simulations (CoM-WindNinja). I suspect that there are serious methodological flows here that need to be addressed by the authors. (Sect. 2.2.5)
Response: We have changed the strategy. We used to focus on the wind direction at ignition only. This time, we have measured the turbulent wind during the fire period to prove the upper air interaction affected the fire simulation. We have created three groups for simulation. Two are control groups with resampled winds which do not include upper air interaction much. The other is the experiment group with BARRA-TA which contains more upper air interaction than others.

Advice 9: The presentation of the results does not allow the reader to draw his/her conclusions. The text is extremely hard to follow and I cannot verify that the conclusions are supported by the analysis. This also relates to my previous comment about the methodology followed. This issue needs to be addressed (Sect. 3)
Response: We have changed the section structure. We counted the number of conducive structures. Then we simulated fire and described the association of wind changes and the quality of simulation. In Table 10, these results are summarized.

Advice 10: Several times throughout the text, the authors refer to simulating "fire behaviours". In my view, it would more appropriate to use the singular, i.e., "fire behaviour". This is just and example of poor language use. 
Response: We have checked the language.

Advice 11: This part provides information on the modeling system used and should be merged with Sect. 2.2.2. (L99-113)
Response: We diminished the description of wind data. However, some still remain in the introduction in order to address a research question and hypotheses.

Advice 12: What is the value that Figure 2 adds to the manuscript? It can be removed
Response: This figure has been removed as advised.

Please find a copy of our research manuscript ‘OzakiM_2021_Wind_effect_on_Riveaux_Road_Fire’ for consideration for publication in Fire. 
We thank you for your time and consideration of our manuscript. 
Sincerely,
Mitsuhiro Ozaki, Rebecca Harris, Peter Love, Jagannath Aryal, Paul Fox-Hughes and Grant Williamson
Discipline of Geography and Spatial Sciences, 
School of Geography, Planning, and Spatial Sciences,
University of Tasmania, 
Private Bag 79, Hobart, 
Tasmania 7001, Australia
Email: [email protected]
Tel.: +61-4-02773191 (mobile)

Author Response File: Author Response.docx

Reviewer 3 Report

The manuscript has been improved significantly, therefore, it can be published in the journal.

Author Response

To,
The Reviewer 3,
Fire Editorial Office,
MDPI, St. Alban-Anlage 66, 4052 Basel, Switzerland.

Subject: Impact of vertically atmospheric phenomena on an atypical fire in mountain valley.

Dear Reviewer 3,
Following careful consideration of the reviewer concerns, we have substantially revised this paper, undertaking a new and simplified analysis. As such there are significant changes to the manuscript. The below shows the main changes.
1.    We have shifted the focus towards upper air interactions during the fire period, rather than wind direction at ignition only.
2.    We have found associations between frequency of upper wind changes and ruggedness of terrain.
3.    We also found that the simulation with BARRA wind showed better than with other downscaled wind data, because BARRA-TA wind better includes and represents upper air interactions.
Please find a copy of our research manuscript ‘OzakiM_2021_Wind_effect_on_Riveaux_Road_Fire’ for consideration for publication in Fire. 
We thank you for your time and consideration of our manuscript. 
Sincerely,
Mitsuhiro Ozaki, Rebecca Harris, Peter Love, Jagannath Aryal, Paul Fox-Hughes and Grant Williamson
Discipline of Geography and Spatial Sciences, 
School of Geography, Planning, and Spatial Sciences,
University of Tasmania, 
Private Bag 79, Hobart, 
Tasmania 7001, Australia
Email: [email protected]
Tel.: +61-4-02773191 (mobile)

Author Response File: Author Response.docx