Effects of Post-Fire Management on a Mediterranean Small Mammal Community

Round 1

Reviewer 1 Report

Please see this article:

"PRESENCE OF LARGE AND MEDIUM-SIZED MAMMALS IN A BURNED PINE FOREST IN SOUTHWESTERN TURKEY "

Author Response

REVIEWER1

"PRESENCE OF LARGE AND MEDIUM-SIZED MAMMALS IN A BURNED PINE FOREST IN SOUTHWESTERN TURKEY "

Authors:  Thanks for the reference, but we did not use it because it was not related to small mammals.

Reviewer 2 Report

The authors discuss the effects of post-fire management on a Mediterranean small mammal community. Given the increasing drought conditions and changes in fire regimes, investigating the responses of animals to fire effects are crucial, especially when considering that these responses may be context-dependent. Although the manuscript is well-written and contributes to our understanding of faunal responses to fire and post-fire management, there are shortcomings that should be addressed. Below, I am listing several comments and questions regarding specific issues of the manuscript.

 1: Introduction: Usually, even in fire prone ecosystems, small mammal responses to fire are context-dependent, varying according to fire attributes (severity, extension, season, recurrence) and to species characteristics (Griffiths & Brook 2014). Hence, while some species may benefit from fire and the changes in vegetation structure promoted by this disturbance (mainly open habitat specialists), other species may suffer declines in species abundance following fire events (see Andersen 2021). I think you should take these divergent responses into account when proposing  your hypotheses. I am not sure that burnt habitats will hold more small mammals (abundance and species richness). I believe that the response of each species to fire may vary, and while some species may thrive in recently burnt habitats, others may decline. If you intend to maintain your hypotheses, be more specific that your study will focus only on the responses of open-habitat specialists. Either way, I think you ought to discuss the fact that responses to fire are context-dependent. Not all small mammals behave in the same way.

 

2. Considering that small mammals usually respond quickly to disturbances and environmental changes (Andersen et al. 2012), due to their short life cycles, why did you choose to sample small mammal communities only 2 years after fire? In several fire-prone ecosystems, open-habitat specialists recover to their original abundances in less than a year (about 6 months) (Vieira and Briani 2013). Hence, I am not sure how sampling the community 2 years after the fire would affect the results obtained.

3. line 136: Vegetation cover was estimated on a visual basis? The same observed estimated vegetation cover, to reduce errors? There are several quantitative methods to estimate vegetation cover (see Freitas et al. 2002).  It is very subjective, and impossible to replicate, an estimate of vegetation cover based on a visual observer. I think the authors need to use a more adequate method

4. Could you specify the vegetation strata used? How did you categorized each strata (tree; shurb; herb; branch)?

5. Data analysis – how did you performed model selection after the construction of models in the R environment? How many models were built in each analysis?

6. Table 1 and Table 2 – this is not a result of GLMM or GLM (model selection approaches).  In my opinion, there is something wrong with the result presentation in the Table 1 and 2. How did you get these parameters ? What estimate in table 1 and Table 2 stands for?

Besides, in table 2, could you explain if you compared species richness or species density between treatments? Density and richness are very different concepts. Moreover, how did you compare species richness if you had 3 species in all treatments? I am not sure if it is possible to run a GLMM with such a low sample size.

Figure 4 – Improve the legend, specifying which colors represent each species

Are you sure GLMMS and GLMs are the best analysis to test your hypothesis? I think you could answer the same questions with more basic analysis.

Discussion

It would be interesting to point out pre-fire community composition and pre-fire abundance patterns. Is there any studies in this sense?

The authors need to connect the discussion with their hypothesis.  Did you corroborated your hypothesis or refuted them? Discuss

 

Again, I think the authors need to take into account the fact that small mammals' responses to fire vary between species. Although this study focused on open-habitat specialists, what happens to forest-dependent species? The discussion and implications of the results are very limited when focused only on open-habitat specialists, and when the authors leave the impression that all small mammals in fire-prone ecosystems will respond positively to fire. 

Author Response

REVIEWER2

The authors discuss the effects of post-fire management on a Mediterranean small mammal community. Given the increasing drought conditions and changes in fire regimes, investigating the responses of animals to fire effects are crucial, especially when considering that these responses may be context-dependent. Although the manuscript is well-written and contributes to our understanding of faunal responses to fire and post-fire management, there are shortcomings that should be addressed. Below, I am listing several comments and questions regarding specific issues of the manuscript.

1: Introduction: Usually, even in fire prone ecosystems, small mammal responses to fire are context-dependent, varying according to fire attributes (severity, extension, season, recurrence) and to species characteristics (Griffiths & Brook 2014). Hence, while some species may benefit from fire and the changes in vegetation structure promoted by this disturbance (mainly open habitat specialists), other species may suffer declines in species abundance following fire events (see Andersen 2021). I think you should take these divergent responses into account when proposing your hypotheses. I am not sure that burnt habitats will hold more small mammals (abundance and species richness). I believe that the response of each species to fire may vary, and while some species may thrive in recently burnt habitats, others may decline. If you intend to maintain your hypotheses, be more specific that your study will focus only on the responses of open-habitat specialists. Either way, I think you ought to discuss the fact that responses to fire are context-dependent. Not all small mammals behave in the same way.

Authors: We agree that responses can be species-dependent, but the three species detected in the study are the bulk of small mammal communities, representing the 95% of individuals trapped in Mediterranean habitats of NE Spain (Torre et al. 2018), and between 75-90% of small mammal consumed by generalist predators (e.g. Torre et al. 2013, 2018). Our paper is centred on these species owing to that reason, and based to our already developed hypothesis about the role of predators along vegetation structure gradients in Mediterranean landscapes (Torre and Díaz 2004, Torre et al. 2022a,b). This hypothesis states that recently burnt areas can be good for small mammals due to less predation pressure (less species of predators) and less predation risk (less exposure under resprouting shrubby vegetation). This can be a reason why recently burnt areas hold a high number of individuals. We will try to refer to that hypothesis when formulating it in the Introduction.  

2. Considering that small mammals usually respond quickly to disturbances and environmental changes (Andersen et al. 2012), due to their short life cycles, why did you choose to sample small mammal communities only 2 years after fire? In several fire-prone ecosystems, open-habitat specialists recover to their original abundances in less than a year (about 6 months) (Vieira and Briani 2013). Hence, I am not sure how sampling the community 2 years after the fire would affect the results obtained.

Authors: We sampled immediately after logging, not after the fire. The objective of analysing these two years is to capture the variations immediately after logging. Most papers do not usually attend to this immediacy. As the reviewer rightly says, we would expect to see some recovery immediately after the fire, but habitat modification due to logging can alter the recovery. Therefore, we sampled two years after the fire when the intervention was recently done.

3. line 136: Vegetation cover was estimated on a visual basis? The same observed estimated vegetation cover, to reduce errors? There are several quantitative methods to estimate vegetation cover (see Freitas et al. 2002).  It is very subjective, and impossible to replicate, an estimate of vegetation cover based on a visual observer. I think the authors need to use a more adequate method

Authors. We agree about the vegetation estimate was poorly described, and we included more information. We know that there are different methods for the vegetation cover estimation, such as those derived from remote sensing that we used in other contexts (e.g. LiDAR), but classical vegetation sampling by visual estimation is the one with more tradition and with strong support, even its subjectivity. Recent investigations pointed out the concordance of classical visual estimates with other objective techniques (drone flights, Pérez-Luque et al. 2022). We stated that vegetation cover was estimated by the same person (Alexis Ribas), and this reduced the potential biases in estimations.

4. Could you specify the vegetation strata used? How did you categorized each strata (tree; shurb; herb; branch)?

Authors: We included further information on the vegetation strata estimated.

5. Data analysis – how did you performed model selection after the construction of models in the R environment? How many models were built in each analysis?

Authors: We did not use a model selection procedure, as this was not the objective. We wanted to know the effects of post-fire management on a Mediterranean small mammal community. Consequently, we only tested a single model with the three variables considered relevant to be tested: Treatment, time since fire, and season. e.g., we wanted to know the effect of each variable on the relative abundance of three species of small mammals, the total relative abundance, and the species density. Although this could have been done with simpler statistics, the GLMM allowed us to include random variables that also minimize spatial and temporal pseudoreplication.

6. Table 1 and Table 2 – this is not a result of GLMM or GLM (model selection approaches).  In my opinion, there is something wrong with the result presentation in the Table 1 and 2. How did you get these parameters ? What estimate in table 1 and Table 2 stands for?

Authors: As mentioned above, this is not a model selection approach. The results presented here are the model coefficients of the three fixed effects and their corresponding SE and signification. The parameters are those given by the lme4 package (glmer).

Besides, in table 2, could you explain if you compared species richness or species density between treatments? Density and richness are very different concepts. Moreover, how did you compare species richness if you had 3 species in all treatments? I am not sure if it is possible to run a GLMM with such a low sample size.

Authors: We compared species density, that is, the number of species found in a plot/session. We followed the term defined by Gotelli and Colwell (2001), and included that reference. Species density oscillates between 0 and 3 species within a plot, so it can be treated as a Poisson variable.

Figure 4 – Improve the legend, specifying which colours represent each species

Authors: Thanks, we included that colours were the same as the fig.3

Are you sure GLMMS and GLMs are the best analysis to test your hypothesis? I think you could answer the same questions with more basic analysis.

Authors: We think that considering three fixed factors and the random factor plot, the use of GLMMs is the best way to analyse the data.

Discussion

It would be interesting to point out pre-fire community composition and pre-fire abundance patterns. Is there any studies in this sense?

Authors: Unfortunately, we have no information on the small mammal community before the fire in the study area. However, we have information on nearby areas (e.g. Garraf Natural park, Torre and Díaz 2004), and we suppose that responses can be rather similar. Indeed, almost during the last two decades, we have developed (and confirmed) a theory about the abundance patterns of small mammals along vegetation gradients created by wildfires. Our results suggested that communities are poor (fewer species) and with lower densities (fewer individuals) in unburnt areas, or areas burnt a lot of years ago. We included some of these comments in the discussion.

The authors need to connect the discussion with their hypothesis.  Did you corroborated your hypothesis or refuted them? Discuss

Authors: We included in the discussion

 Again, I think the authors need to take into account the fact that small mammals' responses to fire vary between species. Although this study focused on open-habitat specialists, what happens to forest-dependent species? The discussion and implications of the results are very limited when focused only on open-habitat specialists, and when the authors leave the impression that all small mammals in fire-prone ecosystems will respond positively to fire. 

Authors: We agree, and we need to put in context our findings regarding the lack of information on other small mammal species (forest dwellers) which are very scarce in unburnt forests and absent in burnt forests. But as previously stated, the three small mammal species captured during this study represent the bulk of communities in Mediterranean habitats (>95% of live captures) and are key preys for generalist open and forest predators. We included the possible effects of post-fire management strategies on these forest species. 

Reviewer 3 Report

Manuscript ID: fire-2052756

Title: Effects of post-fire management on a Mediterranean small mammal community

Authors: Ignasi Torre *, Alexis Ribas, Roger Puig-Gironès

 

Review

 

This paper focuses on the analysis of small mammal community formation in Mediterranean habitats after fires. As there is still a debate on whether fires have a positive impact on biodiversity, the article is of interest to a wide readership. It is based on appropriate hypotheses and is well developed. Most of my comments are not critical, although the author should clarify a few points before accepting.

 

Critical comments

1.       check references: [48] in Line 201 is an obvious mistake, therefore I expect there are more inconsistencies in reference numbers.

2.       Previous publications by author(s) should be acknowledged in the Introduction, explaining what is new in the current manuscript.

 

General comments:

1.       Please stick to the Template. E.g.: [8–10], not [8-10], check throughout the text.

2.       I suppose there is “relative abundance”, not “abundance” analysed, check throughout if correct. If abundance is /per plot, this is also relative abundance index.

3.       Do not use hyphen for periods, use en dash, e.g., Line 153 should be 8–10%, not 8-10%; check throughout.

4.       In figures and in captions, (a), (b) and so on, should be placed in parentheses, and in bold.

5.       For readers, who are not familiar with Mediterranean-specific patterns of small mammal communities, explanation about low species richness and seasonal changes in abundance should be added to Introduction.

 

Language

1.       Line 41: suggested use “chance to thrive in those structurally simple new habitats” instead of “chance to thrive in those new – structurally simple – habitats”

2.       Lines 75–77: please simplify sentence structure

3.       Line 92: do you mean “species richness”?

4.       Line 106: please put “investigation plots with different treatment” instead of “eight”

5.       Lines 171–172: mistypes

 

Introduction

If any other publications from the same area exist, e.g., pre-fire investigations, please mention these in the Introduction.

 

Material and Methods

Figure 1: it seems that burn area is delineated by hatching, not by shading; there are also few grey areas – what are they?

Figure 2: (c) is missing in the caption.

 

Results

Line 168: greater? Usage of capital letters in species names is not consistent. See also captions.

Figure 4: please repeat what colors are for different species; figure must be self-sustainable.

Lines 181–182: please add 95Į CI for these proportions

Line 191: is there such thing as species density?

 

Discussion

Line 201: “consistent decreasing trend of species density”. Hardly believable, if there are three species only in the community, how many will remain “with time elapsed since the last fire”?

Line 235: mistype

Lines 255–256: can a paragraph about the seasonal dynamics be added to Results?

 

Back matter

Conflicts of Interest – please add explanation, if funders had any role in the choice of research project; design of the study; in the collection, analyses or interpretation of data; in the writing of the manuscript; or in the decision to publish the results

 

References

Page numbers must be separated by en dash

[12] – add doi number

Abbreviate journal names – see Template

[41] – I think there are mistakes in the provided reference

 

Author Response

REVIEWER3

This paper focuses on the analysis of small mammal community formation in Mediterranean habitats after fires. As there is still a debate on whether fires have a positive impact on biodiversity, the article is of interest to a wide readership. It is based on appropriate hypotheses and is well developed. Most of my comments are not critical, although the author should clarify a few points before accepting.

 Critical comments

1. check references: [48] in Line 201 is an obvious mistake, therefore I expect there are more inconsistencies in reference numbers.

Authors: We checked all references throughout the manuscript

2. Previous publications by author(s) should be acknowledged in the Introduction, explaining what is new in the current manuscript.

 Authors: We consider that this is already given in the last two paragraphs of the introduction, as well as in the (new) last paragraph of the discussion.

General comments:

1. Please stick to the Template. E.g.: [8–10], not [8-10], check throughout the text.

Authors: We checked it throughout the manuscript

2. I suppose there is “relative abundance”, not “abundance” analysed, check throughout if correct. If abundance is /per plot, this is also relative abundance index.
3. Do not use hyphen for periods, use en dash, e.g., Line 153 should be 8–10%, not 8-10%; check throughout.

Authors: We checked it throughout the manuscript

4. In figures and in captions, (a), (b) and so on, should be placed in parentheses, and in bold.

Authors: Done

5. For readers, who are not familiar with Mediterranean-specific patterns of small mammal communities, explanation about low species richness and seasonal changes in abundance should be added to Introduction.

 Authors: We included some reference about the Mediterranean small mammal communities and their seasonality

Language

1. Line 41: suggested use “chance to thrive in those structurally simple new habitats” instead of “chance to thrive in those new – structurally simple – habitats”

Authors: Done

2. Lines 75–77: please simplify sentence structure

Authors: Done

3. Line 92: do you mean “species richness”?

Authors: We changed with species density, the appropriate term

4. Line 106: please put “investigation plots with different treatment” instead of “eight”

Authors: Done

5. Lines 171–172: mistypes

Authors: Done

Introduction

If any other publications from the same area exist, e.g., pre-fire investigations, please mention these in the Introduction.

Authors: As commented to reviewer 2, there are no previous or nearby studies in the study area.

Material and Methods

Figure 1: it seems that burn area is delineated by hatching, not by shading; there are also few grey areas – what are they?

Authors: Thanks for the observation. We changed the word “shading” by “hatching” and included the meaning of grey areas.

Figure 2: (c) is missing in the caption.

 Authors: Done

Results

Line 168: greater? Usage of capital letters in species names is not consistent. See also captions.

 Authors: Done

Figure 4: please repeat what colors are for different species; figure must be self-sustainable.

 Authors: Done

Lines 181–182: please add 95Į CI for these proportions

 Authors: These are not mean values but frequencies of occurrence collapsed for the whole control and post-fire habitats.

Line 191: is there such thing as species density?

 Authors: we omitted the term richness to avoid confusion.

 Discussion

Line 201: “consistent decreasing trend of species density”. Hardly believable, if there are three species only in the community, how many will remain “with time elapsed since the last fire”?

 Authors: The number of species per plot (density) normally decreases from 3 to 1 because M.spretus disappears and C.russula becomes scarce, and A.sylvaticus is almost the only species in wooded landscapes

Line 235: mistype

 Authors: Done

Lines 255–256: can a paragraph about the seasonal dynamics be added to Results?

 Back matter

Conflicts of Interest – please add explanation, if funders had any role in the choice of research project; design of the study; in the collection, analyses or interpretation of data; in the writing of the manuscript; or in the decision to publish the results

  Authors: We added a sentence explaining the lack of a role of the funders in the study.

References

Page numbers must be separated by en dash

Authors: We checked it throughout the manuscript

 [12] – add doi number

Authors: Done

Abbreviate journal names – see Template

Authors: Done

[41] – I think there are mistakes in the provided reference

Authors: This is how the citation is shown in the publication: Sikes, R.S.; and the Animal Care and Use Committee of the American Society of Mammalogists. American Society of Mammalogists guidelines for the use of wild mammals in research and education. J mammal 2016, 97, 663-688, doi: 10.1093/jmammal/gyw078.

 

 

Round 2

Reviewer 3 Report

I have more comments for the Round 2, and not all previous comments were acknowledged, see attached,

Comments for author File: Comments.pdf

Author Response

Dear reviewer,

thanks for the comments to improve the final version of our article.

In figures and in captions, (a), (b) and so on, should be placed in parentheses, and in bold.

Authors: We used bold in all the letters of the figures,

Page numbers must be separated by en dash

Authors: we used en dash for page separation

Abbreviate journal names – see Template

Authors: We formatted all the references as requested, we had problems with our reference manager version with Fire template, we hope it was completely solved

I do not see where is word “and” after Sikes

Authors: It was deleted

Line 141: font differs

Authors: we are sorry but do not know which font differs

Line 146: please use “it was suggested” instead of some authors – you have one reference for this

Authors: done

Line 196: suggested caption is 

Authors: Done

Figure 4 – use (a) … (d) in both the figure and caption

Authors: Done