An Assessment of Fire Refugia Importance Criteria Ranked by Land Managers
Round 1
Reviewer 1 Report
This is a well executed, succinct and clearly reported study of the criteria that land managers use to rank the importance of fire refugia. Although the findings are somewhat inconclusive (possibly due to low sample size), it represents a valuable first study on this topic and in my opinion is worth publishing for this reason.
Abstract: Effective and well written summary.
Introduction: A succinct coverage of relevant background, but also quite wide ranging in scope and geography. Perhaps more detail on how fire refuges can arise and where in the landscape they tend to be found (i.e. in addition to topography) could be given. Clear study objectives developed.
Materials & Methods: I am satisfied with the methodology used to recruit participants (good representation) and for data collation and analysis. However I couldn’t see the specific criteria (Table 1) in the document I downloaded (although they are generally described in Table A2).
Results: Well presented, but I wonder if there are any statistical tests that can be employed to test for differences between groups (i.e. groups from both k means clustering and the predefined groups)?
Discussion: It looks like only 24 respondents were included in the final analysis, with some groups of respondents being quite small (<5). This is identified as a short-coming but what would be a suitable (and realistic) sample size, and how can this be achieved?
Results were fairly and impartially evaluated in the discussion. The finding of two main groups of respondents (USFS and others) is probably the most striking one and is nicely interpreted.
Ethical approvals etc need to be added to paper.
Author Response
Please see attached file.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Title: Land manager’s rankings of fire refugia importance criteria
Authors: Martinez, Meddens, Kolden and Hudak
Fire refugia are of high ecological value notably under the current change in fire regime observed in many areas of North-America. Due to increases in burn severity and fire sizes, ecosystem resilience is being challenged. This research idea of ranking fire refugia importance criteria certainly has merit in order to identify on-going and future managing challenges in either protecting or encouraging the creation of fire refugia during a fire, and to also identify the types of fuel management tactics required to achieve such goals.
General comments
As the authors point out, the sample size is too small and their research objectives were not fully met. Overall, I find that even as a Research Note, this paper unfortunately falls short of expectations. I did not learn anything and with common sense the outcome of your survey results was easily predictable. My background is in wildfire science and it is a well-established knowledge that values at risk of greatest importance are always lives, then supporting community infrastructure. These have been long-standing practices, well before the publications by Calkin. You would be hard press to justify that someone’s home cannot be protected because there is a fire refugium nearby.
Due to the lack of respondents, you had to lump everything “ecological” under one umbrella, which undermines the paper title for ranking importance criteria and leaves the reader with the already established knowledge that fire refugia, by definition, offer ecological benefits.
My personal expectations of this paper when I read the title was that a group of experts (fire ecologists and managers) had determined a scoring system on what makes a valuable or critical fire refugia (length of survival, critical habitat for sensitive species, etc). And that expert opinion had relied on an extensive literature review to determine scores. For land management, I was expecting a scoring system based on the vulnerability of refugia patches. What are their sizes? Are they clustered or widely distributed? What are the probabilities of ignitions, what is the fire threat? Are they vulnerable to harvesting or other disturbances? Are there topographic elements that can enhance their protection from burning? What is the distance to a community or other values at risk receiving a higher value ranking? These are all questions that land managers could answer through a scoring system and such scoring system can be applied across regions or agencies. But all this paper managed to demonstrate is that managers from different agencies have different priorities in what they value. And like I said, I believe this was easily predictable.
What is not clear in the paper is your definition of fire refugia and if the respondents had been educated on the types of fire refugia (island skips in recent burns, or more persistent ones) and which criteria were considered for a patch of forest to be labelled as a fire refugium. It would also be of benefit to include your survey in the appendix.
This paper does not have any clear value at this stage, and I cannot recommend its publication.
Specific comments
L59 remove “the” after “as well as”
L92: Add “A” to Table 1 (Table A1)
L117 I don’t understand the “(two)” in brackets.
L155 I don’t see what this graphs achieves when all variables are located at the 0,0 axes.
L267 There are two “i” in Climate.
Suggested reading that may help with a scoring system
Evaluating the persistence of post-fire residual patches: Ouarmin et al. 2015
Effect of topography on persistent fire refugia: Rogeau et al. 2018
Author Response
Please see attached file.
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors of the manuscript have undertaken a project that is ultimately important in working to reconcile natural landscape variability with the impressions and priorities of managers. As fire refugia are important structural components of fire-adapted landscapes, studies that investigate the perceptions of their importance by those in the research and management community can be very useful. The survey methods seem sound, but my major concerns are related to the analyses of responses and the lack of connections to the overall relevance. I feel both of these issues can be addressed to make the paper more robust. Attached are detailed comments on the content.
Comments for author File: Comments.pdf
Author Response
Please see attached document.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Dear Authors. Thank you for taking the time to address my concerns. While I still don’t completely agree with the relevance of this process, I will go with the majority and accept the Research Note for publication.
Please note that while you state you’ve made this correction, you have omitted the correction originally on L59 to remove the extra “the”, which is now on L65.
If I can offer a few words of advice. Your research team is indeed well-respected in this field of study. It troubles me to read that you wish to develop a model based on what land managers believe are the criteria of importance to manage fire refugia. You are the authority. Why would you rely on others’ perception of what refugia are and what makes them important. Many land managers do not have the depth of knowledge that you have. To me, it seems like backward thinking. I also believe that developing different models based on different criteria derived from what agencies think are important criteria is also a mistake. In the future when you’ll wish to compare the vulnerability of fire refugia for a region, regardless of the managing agency, it will be like comparing apples to oranges. You need a baseline model across all agencies that first rank refugia strictly for their ecological values using a combination of spatial metrics, such as size, number of patches, clustering, as well as environmental values based on ecosystems, habitats and species. It should not matter if a region sees species habitat “x” or “y” as being most important. What matters in your scoring is if the species is threatened, at risk or vulnerable. You could also have a separate attribute for species of local interest in your ranking system. In a separate attribute you can still name the species so that it can be cross-referenced with other regions for further spatial analyses.
Your second model layer is to rank the vulnerability of refugia patches. Are they threatened by human disturbances, by a high probability of ignition, a high probability of burning, or by fire management practices? And this is on this last factor that I see your results as being relevant. Your results are also relevant to identify which agency may require more awareness in the matter.
I will be looking forward to reading about this future model. Best wishes.
Author Response
We fixed the error on line 65. Thank you for catching our mistake.
Additionally, I want to sincerely thank you for all your criticisms. Your concerns have not fallen on deaf ears and we are honestly taking them to heart. While I am moving out of a research role, I can assure you that the rest of the team will address some of these concerns as they plan future research.
All the best,
Anthony Martinez
(on behalf of the authors)
Reviewer 3 Report
The authors have added more details about the scope and importance of the study, their sampling methods, and their analyses. These additions have satisfied my previous reservations, and I have no objections to the publication of the research note. In fact I think the topic is important and I appreciate the authors' work to investigate attitudes of managers with respect to fire refugia.
Author Response
We appreciate your feedback. It truly improved the quality of our paper!
All the best,
Anthony Martinez (on behalf of the authors)
