Effect of Habitat Disturbance on the Composition of Soil Nematode Functional Groups Associated with a Tropical Herb: Heliconia collinsiana

Round 1

Reviewer 1 Report

It can be published after some (very few) suggested corrections (see  attachment).

Comments for author File: Comments.pdf

Author Response

Comments from Reviewer 1

Line 119.

Thank you. We know the area quite well and we are using the term “secondary forest” as a synonym for “secondary vegetation”. The text was not modified.

Line 140

Thank you. The text has been modified according to the suggestion.

Line 148

sp. was added after the genus Trichodorus

Line 195

mm² was formatted accordingly

Line 266

In the figure, the legend µS cm^-1 was added

Line 319

(Fig. 1) was removed from the text

Line 415

“oh” was changed to “of”

Reviewer 2 Report

This article contains a discussion of an important scientific and practical issue and may be of interest to a wide range of readers of the journal. The article presents a literature review of sufficient length. The research methods are described quite fully. Overall, the article creates a pleasant impression. Some recommendations are presented below.

Of principle note, we should point out that the authors a priori consider the division of the nematode community into guilds as an axiom, which must necessarily be confirmed. However, one cannot exclude the alternative that the responses of individual species, which are not due to their trophic specificity, may be responsible for the responses that are attributed to the guild as a whole. Therefore, confirmation of the influence on the trophic structure is a trivial result. And evidence that it is trophic specialization that is key to understanding nematode community dynamics is of scientific interest.

Line 175: “the soil sample was comprised of 4 subsamples taken on four equidistant corners of the plant using a soil core at a depth  of 25 cm.” – 1) Was the sample obtained by mixing subsamples, or were the subsamples analyzed separately? 2) What is the area of the soil core ?

Line 182: “A 100-ml aliquot of the composite soil sample taken around each heliconia plant” – What was the volume of the composite sample and how was a 100-ml aliquot selected from it ?

Subsection 2.6 discusses methods for measuring soil properties. The procedure for soil sampling should also be specified. Were the samples collected for the nematode records used or were they collected separately?

Line 209: "we fitted two types of generalized linear mixed effect models, the first with absolute density as the response variable and the second with relative density (proportions)". It is necessary to clarify what we are talking about: the trophic guilds, total abundance or abundance of individual species?

Line 210: “In both models, guilds, soil properties, habitat and their interaction were included as predictor variables” – In order to include guilds as a predictor, the data for all individuals in a single sample must be presented consistently in the same column. However, it is then necessary to add the species belonging of the individuals as a predictor as well. What was the dimensionality of the matrix subjected to generalized linear mixed effect models ?

Line 234: “Figure 1” → Fig. 1

Figures 1 and 2 actually contain duplicate information: the abundance of trophic groups is presented in different units. Obviously, it is appropriate to present in Figure 1 the change in the total abundance of the community in absolute units and the variability in the proportions of trophic groups in relative units. Then there would be no need for Figure 2

Line 239: “Nematode density was only marginally affected by the interaction between habitat and guild” – At the level of mathematical terminology, the sentence is correct. But the article is about biology, not mathematics. If you look closely, it can also be improved from a mathematical point of view. – «The term that reflects the interaction of the predictors "habitat" and "guild" is statistically significant for explaining variability in nematode community density». The category of "explanation" is broader and does not exclude effect, but we have no reason to claim that there is exactly an effect. Also, the mathematical interaction of terms and the interaction of real biological processes are not the same thing. A mathematical interaction may be a reflection of a real biological interaction, and it may also be the result of other factors acting on the "habitat" and "guilds", resulting in a relationship between them.

Subsection 2.7. discusses the methods of statistical analysis and points out that generalized linear mixed effect models have been calculated. However, no such information is given in the results (information confirming individual effects does not give a complete picture of the model as a whole).

Comments for author File: Comments.docx

Author Response

Comments from Reviewer 2

Our answers are in italics

This article contains a discussion of an important scientific and practical issue and may be of interest to a wide range of readers of the journal. The article presents a literature review of sufficient length. The research methods are described quite fully. Overall, the article creates a pleasant impression. Some recommendations are presented below.

Of principle note, we should point out that the authors a priori consider the division of the nematode community into guilds as an axiom, which must necessarily be confirmed. However, one cannot exclude the alternative that the responses of individual species, which are not due to their trophic specificity, may be responsible for the responses that are attributed to the guild as a whole. Therefore, confirmation of the influence on the trophic structure is a trivial result. And evidence that it is trophic specialization that is key to understanding nematode community dynamics is of scientific interest.

Line 175: “the soil sample was comprised of 4 subsamples taken on four equidistant corners of the plant using a soil core at a depth  of 25 cm.” – 1) Was the sample obtained by mixing subsamples, or were the subsamples analyzed separately? 2) What is the area of the soil core?

 Thank you for the question. 1) Yes, the sample was a composite soil sample comprised of 4 subsamples. 2) A standard soil core usually holds 100 to 150 ml of soil depending on the density of the soil. The text has been modified in the manuscript and additional details were included.

Line 182: “A 100-ml aliquot of the composite soil sample taken around each heliconia plant” – What was the volume of the composite sample and how was a 100-ml aliquot selected from it?

Thank you for the question. The total volume of each composite sample was approximately 400 to 500 ml. It varies depending on the density of the soil. To collect the 100-ml aliquot, the soil sample was spread out on a bench and subsamples were taken from all corners and the center of the sample using a spatula. Enough random samples were taken to fill a 100ml beaker. The text has been modified in the manuscript and additional details were included.

Subsection 2.6 discusses methods for measuring soil properties. The procedure for soil sampling should also be specified. Were the samples collected for the nematode records used or were they collected separately?

Thank you for the question. We used an aliquot of the original composite soil sample to determine the other soil parameters. The text has been modified in the manuscript.

Line 209: "we fitted two types of generalized linear mixed effect models, the first with absolute density as the response variable and the second with relative density (proportions)". It is necessary to clarify what we are talking about: the trophic guilds, total abundance or abundance of individual species?

Thank you. We have modified the text to specify that both absolute and relative densities were analyzed at the guild level.

Line 210: “In both models, guilds, soil properties, habitat and their interaction were included as predictor variables” – In order to include guilds as a predictor, the data for all individuals in a single sample must be presented consistently in the same column. However, it is then necessary to add the species belonging of the individuals as a predictor as well. What was the dimensionality of the matrix subjected to generalized linear mixed effect models?

Thank you. We do not concur with the reviewer. It is not necessary to add species ID to assess the effect of guild in the analyses, since nematode densities were aggregated at the guild level. So, for the models we had a 80 (rows) x 5 (columns) matrix. The columns of the matrix were Site, Habitat, H. collinsiana ID, Guild and Density. We had 2 (habitat) x 10 (sites) x 4 (guilds) = 80 rows in the matrix. For the model assessing relative densities, we added in the matrix a column of the total densities within each sample (H. collinsiana ID), which was then included as weights in the model.

Line 234: “Figure 1” → Fig. 1

Figures 1 and 2 actually contain duplicate information: the abundance of trophic groups is presented in different units. Obviously, it is appropriate to present in Figure 1 the change in the total abundance of the community in absolute units and the variability in the proportions of trophic groups in relative units. Then there would be no need for Figure 2

Thank you. We have merged Figures 1 and 2 in a new Figure 1. In this new Figure, direct data on relative densities rather than model estimates are presented. Figure numbering has been changed in the text accordingly.

Line 239: “Nematode density was only marginally affected by the interaction between habitat and guild” – At the level of mathematical terminology, the sentence is correct. But the article is about biology, not mathematics. If you look closely, it can also be improved from a mathematical point of view. – «The term that reflects the interaction of the predictors "habitat" and "guild" is statistically significant for explaining variability in nematode community density». The category of "explanation" is broader and does not exclude effect, but we have no reason to claim that there is exactly an effect. Also, the mathematical interaction of terms and the interaction of real biological processes are not the same thing. A mathematical interaction may be a reflection of a real biological interaction, and it may also be the result of other factors acting on the "habitat" and "guilds", resulting in a relationship between them.

 Thank you. We have modified the sentence according to the reviewer´s suggestion: “The term that reflects the interaction of habitat and guild was statistically marginally significant for explaining variability in nematode community density”.

Subsection 2.7. discusses the methods of statistical analysis and points out that generalized linear mixed effect models have been calculated. However, no such information is given in the results (information confirming individual effects does not give a complete picture of the model as a whole).

To answer this comment we have:

• Added information on the percentage of variance explained by both fixed and random factors for each model in the main text.
• Provided the output of the model fitting and hypothesis testing procedures for both models (for relative and absolute abundance) in the supplementary information (See Annex).