The Response of Dung Beetle Communities to Land Use Change in the Brazilian Cerrado

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript contains a comparative of diversity and abundance of dung beetles across different plant systems. The manuscript presents the main results using sound methods. However, some points need to be addressed to improve the paper. Response variables are classified as local or landscape-related, however, they are not listed in the methodology one way or the other; also, other variables seems to be excluded from this classification, as suggested in the summary. Therefore, this issue needs to be clarified. The manuscript need to be checked by a native english speaker. It has portuguese lines inter mixed in tables, figures.

Materials and Methods
Include references for the equations used.
Use p value for statistical tests in a consistent manner (lowercase), sometimes they put as P and in other as p.

Results and Discussion
Results are incomplete, Table 4 is not included.

Figures
Use Figure in the text in full.
Describe individual images for composite figures.
Describe line types and colors in a legend (Figure 3)
Change portuguese writing for english.

Tables
Use Table in the text in full
Table 4 is not presented.
Change portuguese writing for english.

Other suggestions in the enclosed manuscript.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

Nee to check figures and tables, portuguese is mixed with english.

Author Response

Thank you to the reviewer for all the care and attention given to the revision of the work.

We would like to apologize for the details in the English language that were present in the final version submitted to the esteemed periodical. All changes regarding language and spelling errors have been made directly in the text and/or figures and tables. (See the attachment for confirmation).

Comments 1: Aren´t density and arboreal basal area landscape-related variables? what is the difference?

Response 1: Density and arboreal basal area are actually considered local environmental variables, as they refer specifically to the conditions within a particular area (e.g., the density of trees in a given patch of forest). These factors directly influence the habitat characteristics and resource availability that dung beetle communities rely on.

On the other hand, landscape-related variables pertain to the broader ecological context surrounding a specific area, including the overall composition and configuration of the landscape (such as the proportion of forest versus agricultural land across a larger region). Landscape-related variables might include factors like land use types, the presence of habitat corridors, and the degree of fragmentation.

Comments 2: in what units are expressed?

Response 2: The analysis of the fractal dimension using the Sidelook program typically expresses the results in dimensionless units. This means that the obtained values do not have a specific unit of measurement, such as meters or seconds. The fractal dimension (Df) is a numerical value that describes the complexity of a fractal object. It indicates how the object fills the space.

Comments 3: cite reference for the equation

Response 3: We used the references closest to the equation to facilitate the understanding of the origin of each one.

Comments 4: DV cannot be both, density and height, confuse.

Response 4: We rewrote the section to clarify the origin of the data used for DV: 

Where DV refers to the vegetation density, and the height and width are derived from the black cloth used as the background for the photos. The cloth has a size of 1 m², with the height and width representing its dimensions. Both fractal dimension and vegetation density were used as proxies for the complexity of the vegetation.

Comments 5: Looks line a roman numeral, use another symbol

Response 5: We would like to thank you again for the comment, but it refers to an acronym for Indicator Value (IV), which originates from the cited and referenced work in the text.

Comments 6: list or annotate which are local and which are environmental variables

Response 6: We made a new paragraph in order to assist in understanding the categorization of each variable in the work, ensuring that both the local variables and the landscape variable are clearly understood.

Other minor corrections have been made directly in the attached text. Once again, we would like to thank you for your very helpful suggestions, which significantly contribute to improving the readability of the manuscript.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Below are the detailed review suggestions for the manuscript land-3497271-The manuscript presents an insightful study on the response of dung beetle communities to land use changes in the Cerrado biome. Below are my comments and suggestions to improve the clarity, rigor, and impact of your manuscript.

 

Major Comments

1. Hypothesis and Justification:
• The hypotheses are well-defined but should be better justified in the introduction. The expected mechanisms through which land-use change influences dung beetle communities should be elaborated more clearly.
• For example, how exactly does resource availability differ across land-use types, and how might this drive community shifts?
2. Methodology Refinement:
• Pitfall Trap Bait: The use of human feces as bait should be further discussed. Given that dung beetles respond to various dung types, was there a rationale for not using a mix of dung types (e.g., cattle or native mammal dung)?
• Sampling Design: The study uses a distance of 2 km between sampling areas, but this distance might still lead to spatial autocorrelation effects. Have you tested for spatial dependence in the dataset?
• Statistical Analyses: The choice of PERMANOVA is appropriate, but additional details on data transformation and model assumptions should be provided. How were normality and homogeneity of variance assessed before applying Kruskal-Wallis and Dunn’s post-hoc tests?
3. Results and Interpretation:
• Community Structure: While differences in community composition between open and closed habitats are statistically significant, the effect size of these differences should be discussed. Are the shifts in beetle assemblages strong enough to be ecologically meaningful?
• Indicator Species Analysis: The IV threshold for indicator species is set at 70%, yet species with IV below 70% (e.g., Agamopus viridis and Ontherus sp. 3) are discussed. It would be helpful to clarify how "detector species" contribute to ecological interpretation.
4. Discussion Enhancement:
• The discussion should link findings more explicitly to conservation implications. For instance:
• What do the results suggest about restoration strategies for degraded Cerrado landscapes?
• How might management practices in soybean and pasture systems be modified to mitigate biodiversity loss?
• The importance of landscape connectivity is briefly mentioned but should be expanded—how does fragmentation impact dung beetle movements and metapopulation dynamics?
5. Figures and Tables:
• Figures 3, 5, and 6 effectively illustrate trends but should be more clearly annotated. Axes should be labeled with full variable names rather than abbreviations.
• The Venn diagram (Figure 2) should include sample sizes for each category to help readers interpret species overlap.

 

Minor Comments

• Line 44-46: The statement "ecological studies on the impacts of different land uses are fundamental to understanding this dynamic" is too broad. Consider specifying what aspects of land-use impact remain understudied.
• Line 77-79: "Sensitivity of beetles to environmental changes makes them ideal"—add a citation to support this claim.
• Line 87-89: The phrase "species that are more sensitive to environmental change" should be more specific—what traits make them sensitive (e.g., body size, feeding specialization)?

Author Response

Major: Comments 1: Hypothesis and Justification:

The hypotheses are well-defined but should be better justified in the introduction. The expected mechanisms through which land-use change influences dung beetle communities should be elaborated more clearly.

For example, how exactly does resource availability differ across land-use types, and how might this drive community shifts?

Response 1: I would like to thank you for the comment. Indeed, after reading again, we build two new paragraphs, the work seems much better contextualized regarding the issue addressed (Please check the attached file).

Comments 2: Pitfall Trap Bait: The use of human feces as bait should be further discussed. Given that dung beetles respond to various dung types, was there a rationale for not using a mix of dung types (e.g., cattle or native mammal dung)?

Response 2: We included a sentence to explain the reason for choosing human feces as bait, as a previous study revealed that baits composed of human feces have a greater potential to attract a higher diversity of dung beetles. (Please check the attached file).

Comments 3: Community Structure: While differences in community composition between open and closed habitats are statistically significant, the effect size of these differences should be discussed. Are the shifts in beetle assemblages strong enough to be ecologically meaningful?

Response 3: This is an important observation. Although PERMANOVA indicates statistically significant differences in community composition between open and closed habitats, and PERMDISP suggests that these differences are not driven solely by variations in data dispersion, it is important to highlight that there is relevant ecological significance in these patterns.

When we analyze other results, such as the Venn diagram, we see differences in the number of species exclusive to each system, for example when comparing forest and soy. In addition, the species rarefaction curve reinforces these distinctions by showing variations in the richness and diversity of the communities between the habitats. These findings suggest that the changes in beetle composition are not only statistically significant, but also reflect real ecological impacts, possibly related to the loss of microhabitats, variations in resource availability and alterations in the fundamental ecological processes in these environments.

Comments 4: Indicator Species Analysis: The IV threshold for indicator species is set at 70%, yet species with IV below 70% (e.g., Agamopus viridis and Ontherus sp. 3) are discussed. It would be helpful to clarify how "detector species" contribute to ecological interpretation.

Response 4: We appreciate the observation regarding the classification of species with Indicator Value (IV) below 70% and their role in ecological interpretation. This is an important point that we can clarify further in our discussion. Therefore, we have included two new paragraphs (marked for easier understanding).

Although the 70% threshold is often used for the strict classification of species as indicators, species with IV below this value can still provide valuable information about environmental conditions and community changes. These species, referred to as "detector species," are sensitive to variations in habitat conditions and generally respond predictably to environmental changes, even if they do not reach the 70% IV threshold. Exploring these data has indeed contributed to a better understanding of what we intended to express in our conclusions. (Please check the attached file).

Comments 5, 6, and 7: What do the results suggest about restoration strategies for degraded Cerrado landscapes?

The importance of landscape connectivity is briefly mentioned but should be expanded—how does fragmentation impact dung beetle movements and metapopulation dynamics?

How might management practices in soybean and pasture systems be modified to mitigate biodiversity loss?

Response 5, 6, and 7: We appreciate your comments. We used your suggestions to draft two paragraphs to clarify the study's impacts on Cerrado conservation and improve agricultural practices in these systems. (Please check the attached file).

Comment 8: Sampling Design: The study uses a distance of 2 km between sampling areas, but this distance might still lead to spatial autocorrelation effects. Have you tested for spatial dependence in the dataset?

Response 8: The selection of 2 km buffers in dung beetle studies is grounded in principles of spatial ecology and the species' dispersal patterns. While dung beetles exhibit diverse home range sizes and movement capabilities, research provides specific guidance on minimizing interference. Larsen and Forsyth (2005) determined that a 50-meter distance effectively reduces trap interference for Canthon acutus. Similarly, mark-recapture studies by Silva et al. (2015) suggest a 100-meter threshold. Therefore, the 2 km separation employed in this study significantly exceeds these recommendations, effectively ensuring sampling independence. This is particularly relevant considering the study area's landscape, which is characterized by an agricultural matrix interspersed with limited forest fragments.

Comment 9: Statistical Analyses: The choice of PERMANOVA is appropriate, but additional details on data transformation and model assumptions should be provided. How were normality and homogeneity of variance assessed before applying Kruskal-Wallis and Dunn’s post-hoc tests?

Response 8: The choice of PERMANOVA is appropriate, but additional details on data transformation and model assumptions should be provided. The analysis was based on a similarity matrix constructed using the Bray-Curtis index, calculated from standardized data and square-root transformed. Additionally, normality and homogeneity of variance were assessed before applying the statistical tests. The Shapiro-Wilk test indicated that the data did not meet the normality assumption, and Levene’s test revealed the absence of homoscedasticity in the variances. Consequently, the authors chose to use non-parametric tests, such as Kruskal-Wallis and Dunn’s post-hoc test, as they are more suitable for data that do not meet normality and homogeneity of variance assumptions. These tests, based on medians and data ranking, provide a robust analysis, capturing differences and similarities between groups more accurately. Furthermore, the use of non-parametric tests avoided the need for data transformation, preserving relevant ecological information that could have been distorted. Details on data transformation have been included in the text.

Minor:

Comments 1: Line 44-46: The statement "ecological studies on the impacts of different land uses are fundamental to understanding this dynamic" is too broad. Consider specifying what aspects of land-use impact remain understudied.

Response 1: We appreciate the comment regarding the statement suggesting that ecological studies on the impacts of different land uses are fundamental to understanding this dynamic. We agree that this statement may be considered broad and that we need to specify which aspects of land use impacts require more attention in the scientific literature.

In the revised version of the manuscript, we have included the key aspects identified to better guide the discussion.

Comments 2: R2 -Line 77-79: "Sensitivity of beetles to environmental changes makes them ideal"—add a citation to support this claim.

Response 2: We appreciate the comments and have included a citation within the sentence to make it clear that there are clear indications of the use of dung beetles as sensitive indicators of environmental changes.

Comments 3: Line 87-89: The phrase "species that are more sensitive to environmental change" should be more specific—what traits make them sensitive (e.g., body size, feeding specialization)?

Response 2: We did not conduct a traits analysis; we only used the presence or absence of species in the systems. Additionally, as can be noted throughout the text, a comparison was made with the natural system (forest). Therefore, we believe that the passage was indeed vague in this regard, and we suggest this correction.

To check the minor adjustments in the figure, please see the attached file.

You will see in the document 

Author Response File: Author Response.docx