Decadal Fire Effects on the Structure, Composition, Diversity, and Aboveground Carbon Stocks of a Neotropical Savanna

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The authors aims to study the effect of repeated  fire on a cerrado forest. They measure several relevant metrics for biodiversity and above ground carbon stock in three different types of plots: no fire in the last 30 years, once 15 years ago and finally plots with fire every year since 15 years.
Unfortunately the sample design seems biased, the treatment is not well described and error bar/statistical significance are not always reported

Major comments
1) the plot under fire regime seems to be systematically different from control for reason beyond the fire treatment.
In fact from figure 1, the fire treated plots are all on the border of the forest bounded by two streets, while control plot are well embeded in the forrest. This could be a sufficient reason to observe the difference reported in the results.  Here the authors need to add some other data, some possibilities could be: a) a survey of the border forest 15 years ago before the start of the treatments b) add some plot of the border forest, enclosed by the two road that are not treated with annual fire. See also minor comment 2)

2) the actual regime of annual fire is not very precisly defined. The only description is:
"with a history of recurrent annual burning for at least 15 years, with the use of black firebreaks in the months of June or July"
the plot with fire are bounded by black firebreaks and another fire treatment is performed within it, or the plot  are black firebreaks for the rest of the forest?  

3) The actual number of measures are unclear, given that each of the 15 plots is then subdivied in 10 subplots, and is unclear if the measure reported are an average across subplot  or directly measured summing observations across subplots. This is especially true for figure 2. Some error bar or standard deviation are not reported ( figure 3-4)

Minor Comments

1) the MS lack of line number
2) "Fifteen plots (20 m x 50 m) were randomly allocated". The design from figure 1 do not seems random at all, probably authors should explicit there stratified sample. see also major comment1
3) given that the authors have the estimated biomass of each individual plant observed they could add to the panel of measurments also a shannon weighted by biomass and not number of individuals, this should reflect better the biodiversity functioning of the ecosystem.
4) figure 4 should be shown with individual dots, given that for each category there are only five observations. With so few values,  the boxplot hide more that showing.

Author Response

 "Please see the attachment." 

Thank you very much for handling our manuscript and allowing us to make the necessary changes to improve the quality of our work. As you will see in the revised manuscript and response to reviewers, we have addressed their comments and explained why our sampling approach is robust, providing further clarification.

Please feel free to get in touch in case further clarification is required.

Best,
Prof. Reginaldo Sergio Pereira
On behalf of all authors

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

General comments

The manuscript contains important results about fire effects on the structure, diversity, composition, and carbon stocks. I see the potential of this document because it is a great contribution to the knowledge of fire ecology in tropical savannas and results support findings of other similar studies.

 

However, there are several problems that I saw in the manuscript. The Methods section is not well organized. For that reason, there are several words repeated in different parts of this section. Also, you should explain better the design of the study. To avoid confusion and redundancy, I suggest dividing it into a) study site, b) sample design, c) data collection, and d) data analysis. The manuscript would be better organized if you do not mix methods or results with discussion or conclusions. Finally, you run ANOVA, but you did not make post-hoc analysis. Also, I was asking: what was the criteria to run ANOVA and not non-parametric Kruskal-Wallis? Finally, there are several problems in the REFERENCES section.

 

Please look at the specific comments

 

Specific comments

 

Title

In your title, you do not include “diversity” being one of the most crucial subject described in your objectives, methods, and results.

 

2.1. Study site

Even though many of us know Cerrado, other people do not know. I suggest describing the floristic composition of the studied vegetation.

 

2.2. Fire experiment

What other criteria are considered to allocate your plots? In the map, the study area seems greater, but your plots are concentrated only on one site. Plots randomly allocated should be distributed in all study sites.

 

Black firebreak. - You dedicated one paragraph to black firebreak, so you should include it in the map.

 

2.3. Field inventory survey

Several aspects of the Methods are repeated in different subsections. I recommend reviewing redundancy.

 

What is “uma suta”?

 

2.4. Aboveground carbon stock estimation

The first part of the paragraph seems not part of the methods. It seems part of the discussion.

The last part of the paragraph seems to be part of the statistical analysis; in that case, you should move to section 2.5.

 

2.5 Statistical analysis

Floristic composition and structure

First paragraph: You mention dominance by the basal area and frequency as methods to measure structure, however, I could not see the results using this analysis. At the end of the paragraph, you are repeating the use of Shannon and Pielou analysis.

 

Second paragraph: What is the difference between abundance and density?

 

Carbon stocks

The last two sentences of the paragraph do seem not a statistical analysis.

 

RESULTS

3.1. Species composition across treatments

 

Figure 1. Need to move the title above of figure, to the Figure legend.

The first sentence of the second paragraph seems to be a conclusion or part of the discussion section.

Species name: I could not find in Tropicos the species Camisala montana. In Figure 1 it appears as Roupala montana. Review in the entire document. I could see in Appendix A only R. montana.

 

3.2. Diversity metrics across treatments

Figure 2. What figure is representing the structure? I suggest making a Table with the results of ANOVA for Figure 2. ANOVA only says, in a general way, that there is a difference between treatments, but does not compare. You should make a post hoc analysis for all parameters measured in the three treatments.

 

First paragraph. - You should move to the Method section

Second paragraph. - To see the difference between control, legacy, and annual fire treatment, you need to run posthoc analysis

The fourth sentence for the second paragraph seems discussion.

 

Figure 3.

(a)   These results should be extrapolated to 1 ha. In addition, you have 5 plots by treatment. So, you should calculate MEAN and standard error for each treatment. You can also run ANOVA and post-hoc analysis.

 

 3.4. Aboveground carbon stocks across treatments

 

Figure 4.- This figure should show MEAN and standard error, considering that the study had 5 plots by treatment. In addition, should run ANOVA and posthoc analysis to compare between treatments.

 

First paragraph: Lines 5-7 seem to be part of the Discussion.

The second paragraph is part of the Discussion.

 

4. DISCUSSION

4.1. Relationship between fire and Cerrado species composition

Third paragraph. - You mentioned mechanisms to survive after the fire. You only mentioned bark thickness. However, resprout capacity is another mechanism to survive. I guess you need to review about resprouts.

 

4.3. Fire effect on vegetation diversity

A total of 473 live species??? I guess it is “individuals”. I think it is better to convert to stem density per area (i. e. 1 ha).

 

4.4. Fire effects on aboveground carbon stocks across treatments

The first paragraph, lines 8-11.- Use mean results. I think the results will be similar.

 

Appendix A

I suggest including Family names for all species (it should be only the first words of every family). In addition, you should take out dots that are in the scientific names. Some scientific names are incomplete.

 

REFERENCES

References are in different formats. Review the case of authors, bold or not bold in the name of journals, and revise author separations with commas or semicolons. Some references are incomplete. Are you going to put DOI in the references that they have? Please look at the comments in the manuscript.

Comments for author File: Comments.pdf

Author Response

 "Please see the attachment." 

Thank you very much for handling our manuscript and allowing us to make the necessary changes to improve the quality of our work. As you will see in the revised manuscript and response to reviewers, we have addressed their comments and explained why our sampling approach is robust, providing further clarification.   Please feel free to get in touch in case further clarification is required.   Best, Prof. Reginaldo Sergio Pereira On behalf of all authors

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

In Appendix A, you should put scientific names in italics. Also, some family names are not in the same format. For example, in Dalbergia miscolobium you put Mimosideae, but in Entelerolobium gummiferum you put Fabaceae-Mimosoideae.

Author Response

 "Please see the attachment." 

Thank you very much for handling our manuscript and allowing us to make the necessary changes to improve the quality of our work. As you will see in the revised manuscript, we have italicized the scientific names of the species in Appendix A and standardized the family names. Please feel free to get in touch in case further clarification is required.

Best,

Prof. Reginaldo Sérgio Pereira

On behalf of all authors

Author Response File: Author Response.pdf