The Power Grid/Wildfire Nexus: Using GIS and Satellite Remote Sensing to Identify Vulnerabilities

Round 1

Reviewer 1 Report

This work examines potential wildfire to occur in areas with critical power grid infrastructure by considering biomass production, seasonal precipitation, wind speed, and lightning strikes at Idaho in USA. This is a well written manuscript and is worth to be published. Yet, it seems that the following advancements should be taken into consideration.

Fig.1: The grid lines are all assumed to be of the same significance. But in most places these lines should be differentiated according to the level of their utilization in terms of time and power as well as to the width of the clear area maintained in each side of these lines. In addition, it is common practice that the annual frequency of their maintenance with inert gases is also quoted because all these changes drastically the likelihood of fire potential. The caption of this figure and the text should be modified accordingly.

Table 1: Not all the nine-web links quoted here seem to work or be openly available.

Lines 151-156: This section need to be enhanced with the mitigation actions that could be taken locally in reducing the overall burn probability over the 2018-2021 study period.

Lines 159-160: The overall burn probability model as developed by Reference 12 should be described earlier in this paper (e.g. in section 3) and the authors should highlight why this model is suitable for this work.

Lines 168-170: It is essential to explain more why the authors think that “it is likely that a broader study would reveal a parabolic curve best describes the relationship between NDVI and wildfire vulnerability” and to what extend local preparedness with mitigation actions as those mentioned at the comments for Fig.1 might change this conclusion.

Lines 171-172: The web link for the supplementary material is not functioning.

Following the implementation of all aforementioned changes I think this work should be ready for publication.

Author Response

Thank you for your valuable time and feedback.

• Fig 1: Yes, in this study, the power transmission lines are assumed to be of the same significance. This study focuses on the presence/absence of power grid infrastructure. Your suggestions provide for an excellent scope for future work. I updated the figure 1 caption reflecting that in the revised document.
• Table 1: I have confirmed that all nine web links in Table 1 are working and openly available.
• Lines 151-156: Recommendations for specific mitigation actions during the study period are beyond the scope of this paper, which focuses on providing the raw data and statistics on select variables that may indicate a natural area's vulnerability to wildfire. Mitigation action recommendations are a natural next step and needed area of research, but beyond the scope of the study. We addressed this comment in lines 52-54 in the revised document.
• Lines 159-160: I reference the probability model in Section 2 as "Scott, et al [12]." I highlighted why this model is suitable for this work on lines 65-67 in the revised document.
• Lines 168-170: This paragraph focuses on potential future work based on an observed relationship rather than a conclusion. I added phrases to make this more clear in lines 173-179 in the revised document.
• Lines 171-172: I have confirmed that this link is functioning, this is now on lines 192-194 in the revised document.

Reviewer 2 Report

The length of this study and the overall analysis justify a research note, at best. Certainly not a research article.

The variable values are taken at a different resolution and, therefore, are not suitable for statistical analysis.

Statistical analysis alone does not explain the 'cause and effect' relationship between fire occurence and the variables used in this study. By the same token, other variables could have a stronger statistical correlation with fire activity without necessarily exerting an influence on fires.

The distinction between low- and high- fire occurrence areas is, bibliographically at least, meningless.

Author Response

Thank you for your valuable time and feedback.

• This paper has been changed to a research note.
• I updated the Table 1 figure caption to reflect this comment.
• The scope of this study is not to establish through statistical analysis the cause-and-effect relationships, but rather to examine, in general, vulnerabilities in two areas with previously established differences in risk and to identify probable correlations between burn probabilities. Your suggestions are excellent next steps for much needed future work. I have included this response in the revised paper on lines 173-177.

Reviewer 3 Report

The article has an overview character. Perhaps it would be good to present more examples and specific solutions.

Author Response

Thank you for your valuable time and feedback.

Recommendations for specific mitigation actions during the study period are beyond the scope of this paper, which focuses on providing the raw data and statistics on select variables that may indicate a natural area's vulnerability to wildfire. Mitigation action recommendations are a natural next step and needed area of research, but beyond the scope of the study. This is addressed in lines 52-54 in the revised document.

Reviewer 4 Report

I believe this study is incomplete. While it could provide additional insight on the relationship among fire occurrence, fire drivers and the impact on grid infrastructure, the overall design is somehow lacking required depth.

The analysis is an ANOVA implementation between selected drivers for two areas. I do not see the connection to the power grid. Burn probability is affected by many factors in general, these drivers could be implemented in any case of fire risk calculation. 

One way to improve the analysis would be to study the fire location and occurrence in the two areas and also their distance to power grid. That would be a meaningful approach, for example.

The discussion and conclusions are too general.

Author Response

Thank you for your valuable time and feedback.

• This paper has been changed to a research note.
• This paper focuses on landscape vulnerability to fire so that infrastructure that resides in these vulnerable landscapes can be managed and prioritized accordingly. I have clarified this in the introduction on lines 52-54 in the revised document.
• The fire location, occurrence, and distance to the power grid are all important variables for future research, but beyond the scope of this study.
• Recommendations for specific mitigation actions during the study period are beyond the scope of this paper, which focuses on providing the raw data and statistics on select variables that may indicate a natural area's vulnerability to wildfire. Mitigation action recommendations are a natural next step and needed area of research, but beyond the scope of the study. This comment is addressed in lines 52-54 in the revised document.

Round 2

Reviewer 2 Report

The revised version is improved and appropriate for publication.

Author Response

Thank you kindly for the feedback and approval.

Reviewer 4 Report

The authors have made some changes to the manuscript, mostly related to its scope and corresponding length, as a research note. This justifies to an extent its content, but i still have the same concerns regarding the research design. I believe that this could be a very interesting contribution if the appropriate time is devoted and necessary data and variables are analyzed. The distinction to low and high fire occurrence alone cannot be the basis for meaningful results.

Author Response

Hello, thank you for your additional comments. I can see where the misunderstanding arose as we did not fully explain the basis for identifying the high versus low burn probability areas used in this study.  We have corrected this in the revised manuscript on lines 47-58, 77, 146-154, and 192-199 to make clear the High BP and Low BP areas were based on predicted future burn probability as well as document past fire occurrence.