Exploration of Soil Microbial Diversity and Community Structure along Mid-Subtropical Elevation Gradients in Southeast China

Round 1

Reviewer 1 Report

In the manuscript entitled "Exploration of soil microbial diversity and community structure along mid-subtropical elevation gradient in Southeast China: What are the driving factors?", the authors present an altitudinal study of the microbial ecology of a transect in the Wuyi Mountain, and the drivers of this ecology. While the science presented in this manuscript is sound and well discussed, I fail to grasp the novelty and relevance of the study, considering there are already similar studies performed in the same region. Therefore, the authors need to justify more clearly what sets this study apart from previous studies. In addition, there are several issues with the text (including grammatical errors) that need to be addressed. See below for a list of these: 

 

Introduction:

Lines 43-44: What to the authors mean by “strategy of space for time substitution”? Unclear, please clarify or re-phrase.

Line 45: “decisive” is not the correct term, please re-phrase.

Lines 53-56: In my opinion, the authors have not adequately introduced the concept of the high altitudinal mid-subtropical zones as being ecologically important hotspots. Please expand on this point in the introduction section.

Lines 57-59: Unclear what the authors mean by humpbacked or hollow trends – please clarify. In addition, the studies referenced in the sentence are not necessarily comparable, as they were conducted in different years and different biomes. Please see more recent global examples such as  https://doi.org/10.1890/10-1170.1 for appropriate references.

Lines 64-70: This is a very long sentence with multiple ideas that should be re-written into multiple smaller sentences.

Lines 70-73: In my opinion, reference 19 is not sufficient to back up the claim that fungi have a higher carbon and nitrogen demand than bacteria. Additionally, the authors should be specific when talking about altitudinal transects, as the diversity trends are different depending on the altitudinal transect being studied.

Lines 77-79: There are no references to this statement – please include appropriate references.

Lines 90-91: Is this a true statement? I google scholar search using microbial diversity in the Wuyi mountains outputs several studies dating back to 2013. Please discuss some of these studies and highlight how your study is unique and relevant.

Lines 95-98: References missing for this statement.

Lines 98-101: As stated above, the authors should highlight more clearly the relevance and novelty of this study, taking into consideration the studies conducted previously in the same region.

Materials and Methods:

Lines 160: In my opinion, the section “Illumina sequencing” deals with the quality control and processing of the raw reads, rather than the sequencing methodology. Therefore I would suggest renaming it to “Illumina sequence processing”.

Lines 170-172: Current microbial ecology studies use DADA2 to process reads into ASVs, which have been shown to be more accurate and reproducible than OTUs (https://www.nature.com/articles/ismej2017119). Amplicon processing pipelines such as QIIME2, which can be run as a virtual box in a laptop, implement DADA2 by default, as is easily accessible to run. Can the authors justify why an outdated OUT-based methodology was used instead?

Line 172: Please specify the version of the databases used.

Line 179: Please specify the version of database used.

Lines 182-203: References missing for several of the statistical methods employed – please correct. Examples include refs for Shapiro-Wilk test, ANOSIM, etc.

 

Line 206: In my opinion, the authors should include a description or supplementary table with the overall species count for each of the plots.  

Line 226: Considering the sequencing method used and read processing pipeline, one would expect all high-quality reads to have the same length (corresponding to the distance between primers used). Can the authors explain the variation in read lengths observed? Were the reads truncated according to a quality score? If so, this should be described in methods section.  

Line 230: Do you mean “not significant”?

Figure 1: Please label the Y axes. In addition, there is no description of what A and B mean in the legend to the figure.

Lines 257-264: Please specify what the percentages within the brackets refer to. In addition, correct grammatical errors throughout this paragraph (for instance, “which together accounting” is the wrong expression).

Lines 269-279: The terminology “Phyla” and “Phylum” are used incorrectly. Phylum should be used when referring to a single clade, and Phyla should be used when referring to multiple. In addition, please specify what the percentages within the brackets refer to.

Line 277: Please correct “from those in the low elevation forests from those” to “between those in the low elevation forests and those”

Line 288: The Mantel test by itself is not sufficient to identify drivers of community structure ( a RDA or CCA would be more appropriate). Rather, the Mantel test highlights significant correlations between the environmental and compositional matrices. Therefore, the authors should refrain from specifying the significant variables as “drivers” of community structure.

Table 5. I think the legend for this table is wrong, as the table shows relative abundance of the different clades, and not the community structure. In addition, please specify what the letter codes represent in terms of significance values.

Lines 305 -309: A common practice of representing CCA results is to plot them as an ordination with arrows showing the influence of each significant factor on the community structure. By comparison, the table shown in the manuscript is more akin to the results one gets from a variation partitioning analysis. Can the authors clarify the exact method used to calculate these values? What step-wise model did they use to calculate significance, etc? In addition, I would suggest for the authors to include the ordination plots of the CCA analysis, so that the readers have an idea of what the distribution of different clades according to community structure looks like.

Line 315: Please specify the type of correlation used.

Figure 5. For clarity, I would suggest that the authors re-draw the figure with a clear separation between the fungal and bacterial phyla.

Lines 338-339: I am a bit confused by this statement – what environmental changes are you comparing your data to? Please clarify.

Line 247: “inconclusive” is not the correct term to use here. Correct to “differing”.

 

Lines 426-428: I find it surprising, considering the ranges of pH recorded in the study, that Acidobacteria were not the most dominant bacterial taxa identified. Can the authors expand on why this was not the case? Are the Proteobacterial clades identified particularly acidophilic – if so, please include this in the discussion.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Dear colleagues! Thank you so much for the opportunity to get acquainted with the manuscript of the article “Exploration of soil microbial diversity and community structure along mid-subtropical elevation gradient in Southeast China: What are the driving factors?” A very interesting article. The topic that the authors are considering is very relevant in connection with global climate change. The latter can trigger changes in the biotic component of soils at the lowest level of its organization. The data obtained are of interest to anyone who deals with the problem of soil indication in mountain landscapes. I hope to see the authors' articles on this topic again soon.

In the process of reviewing the manuscript, a number of questions arose, to which, I hope, the authors know the answer.

line 19-24-Soils (according to the CCI), type of forest

line 68- how trace elements affect the microbial community

119 - What is the peculiarity of these soils?Please provide a morphological formula.

133- samples were mixed, selected according to genetic horizons?

179-Pathotroph and others...for what purpose were they examined?...please specify the offer, check

216 and 303- did not understand Low altitude(S?) and Low elevationS...these are synonyms...please make it uniform if synonyms

line

32-Tablet S7? Why didn't you mention the information about Zoopagomycota?

377- what exudates of plants affect microbiocenosis?

413- maybe...it is not clear whether you are writing it or other authors.

404-439- How does the acidity of coniferous litter change during its mineralization?

The succession of microbocenosis during decomposition of the litter depends on its chemical composition from the composition of plants entering the ecosystem?

443- Does the type of soil not affect the microbiota? It is necessary to write one sentence about the soil.

441-455-I didn't see the answer to the question in What are the driving factors?, which was included in the title of the article. Somehow very vague, not clear. Can I remove it from the title of the article?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Exploration of soil microbial diversity and community structure along mid-subtropical elevation gradient in Southeast China: What are the driving factors?

Summary

The authors aimed to study the variations in soil microbial (bacteria and fungi) communities and their driving factors along elevation gradients at high (1038m) and low (550m) elevations in a mid-subtropical mountainous zone in Wuyi Mountain in Southeast China. Overall, the authors did not detect significant differences except for significant differences in the bacteria and fungi community structure using ANOSM but performed several correlations with few turning out to significant. Plant diversity and pH were the strongest predictors of the bacterial and fungi communities, respectively.

 

Overall, I enjoyed reading this paper but have concerns about the statistical tests performed and the sample selection. Often parametric tests such as ANOVA, T-tests and Pearson’s correlation were used throughout this manuscript, but I doubt if the assumptions of these tests were met as microbiome datasets are notoriously known to deviate from the assumptions of normality hence non-parametric test such as Kruskal Wallis, Wilcoxon and Spearman test should have been used instead. I am also of the opinion that the inability of the authors to detect significant differences was most likely due to the fact that they pooled 24 disparate samples to for either the low or high-altitude group. Soil is a highly heterogenous matrix hence sampling from very different locations will result in high variations in the results that will make it difficult to detect significant differences.

 

Abstract

Line 16: Climate warming sound obscure.  Perhaps global warming or climate change would be better.

Line 20: change “Shifts” to “Shift”

Line 24: change “within” to “between”

 

Line 25: Delete “The shifts in”

Line 26: along “the” two altitudes ….

 

Introduction

Line 37, 43 and 55: Climate warming sound a little off here. Perhaps global warming or climate change would be better.

 

Methods

Line 109: the study area was in ….

Line 128: delete “the”

Line 149: delete “The”

 

Line 151: change “isolated” to extracted

 

Line 161: change “by the” to “with” or “using”

 

Line 164: delete “of” and change “was” to “were”, and delete “The”.

 

Lines 167-169: does not make any sense hence it is difficult to comprehend what the authors are writing about. Please make this sentence comprehensible.

 

Results

Line 208 – 210: contradicts the following sentences in lines 210 – 214. Table 2 clearly shows no significant difference between low and high altitudes for all soil physicochemical characteristics as rightly stated by the authors in Lines 208 – 210. However, the authors begin to talk about significant differences in lines 210 – 214. Could the authors clarify what they mean? In addition, the authors have used t-tests for pairwise comparison between low and high altitudes which is fine if the assumptions of a t-test are met. However, the dataset does not seem normally distributed to me as the standard errors are pretty large sometimes even larger than the mean. Please see MC, NO3, NH4 and DOC in Table 2.

 

Line 230: change “less” to “not”

Table S1: Do the authors mean significant differences were detected using T-tests not ANOVA because it is possible to run pairwise comparison with ANOVA but it is incorrect and will output the same results as a T-test. Hence, I recommend that the authors changed ANOVA to T-test throughout the manuscript (Text and tables) wherever it was used to compare between two groups (Low vs High altitudes).

 

Line 233: Delete “The” before Pearson’s and “analyses” to “analysis”.

 

Line 247: soil pH looks like it was significantly positive and not negatively correlated with chao1 (Table 3).

 

Line 262: change “was” to “were”

Line 263: delete “which”

Line 269: change “Phylum” to “Phyla”.

Line 276-277: should read like so “The soil bacterial and fungal community structures of the low elevation forests were clearly different from those in the high elevation level forests (Pbacteria = 0.001, Pfungi = 0.001, Fig. 3),”…

 

Line 284: Change “of” to “between”

Line 287-289: Figure 4 is very difficult to understand but Table S4 is very clear and easy to understand. It might be a good idea to move Table S4 into the manuscript then add figure 4 as a supplementary figure for easy comprehension by the reader.

Line 297: delete “nearly”

 

Lines 315 – 326: Why was Person correlation used? In my experience, amplicon sequencing datasets are not normally distributed.

 

 

Discussion

Line 347 – 349: I agree. This might be due to the fact that 24 disparate samples were pooled together for each elevation level (Low and High) thus resulting in a great variation within each level. This can easily be seen in the high variation between samples on the NMDS plot particularly within the low altitude samples (Figure 3). I wonder what the plot would look like if colored by the actual Forests (L1, L2, L3, L4, H1, H2, H3, H4) within each altitude. My guess is that the samples will cluster based on those forests because soil is a highly heterogenous matrix.

 

Line 419: delete “to”

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

In the revised manuscript “Exploration of soil microbial diversity and community structure along mid-subtropical elevation gradient in Southeast China”, the authors have addressed most of my previous concerns with the study and text and have produced a much-improved manuscript. However, the authors should still take care with the style of English and grammatical structures used in the text. Please find below a list of corrections that the authors should address:

Line 40: Change “are” with “is”.

Lines 41-47: While the addition of these lines helps contextualize the importance of the study, the grammatical structure needs to be refined, as there are several instances of repeated words within the same sentence, and commas on the wrong part of the sentence. Please revise.

Line 72-73: This line is not scientifically accurate. There are several studies showing how fungal and bacterial communities respond differently to soil chemistry, as you explore in the next sentences. Please rephrase to reflect this.

Line 76-79: Incorrect phrase structure, please re-phrase.

Line 79: “While” is not a suitable manner to start this sentence.

Line 81: Change “of” to “for”. In addition, please give examples of which fungi rely of decomposition of SOM – isn’t this the case for most fungi and bacteria?

Lines 108-115: This sentence is too long and includes several ideas that could be more easily separated into individual sentences. Please re-structure.

Line 117: Unclear, please re-write.

Line 123: Change “span” to “spanning”.

Lines 124-125: You repeated “Wuyi Mountains” earlier in the sentence. Additionally what are you referring to when you say “their regulating factors”? Please clarify.

Line 127: It is not clear what “future” and “current situation” scenarios you are trying to study. Please specify.

Line 242: I would recommend not starting the section with “Although, …” as it puts the reader on a negative frame of mind. I would start by writing “ A sharp decline was recorded….”.

Line 300: Delete “While”.

Line 308: Suggest adding a “and” before Mortierellomycota.

Line 313: Delete “While” and “the relative abundance of”.

Line 315: Change “was” to “were”.

Line 332: Again, I don’t think you can identify drivers of community distribution using a mantel test (for that you would need to run an RDA or CCA). Please re-phrase.

Lines335-339: Please revise the English in this section. In addition, in the Figure I can see that the fungal CCA also shows some arrows with significant correlations between microbial distribution and the geochemical variables. This contradicts the text. If the arrows in the figure are not significant, you should highlight this point in the legend to the figure, and make sure that it is also visualised in the figure (eg. significant arrows appearing in a different color to the non-significant).

Lines 361-362: These lines seem to contradict the previous statement in lines 335-339, where you claimed that no determinants of fungal distribution were identified using the CCA analysis. Please revise.

Lines 438-447: I fail to see how this information is relevant to the discussion of your results . Please specify in the text how this information might explain your results, otherwise please delete.

Line 455: Delete “also that”.

Line 464: I don’t understand how this statement is contrary to the previous paragraph.

Conclusions: In my opinion, there is a statement (and discussion) missing regarding the importance of these results in predicting the impact of climate change on these important biodiversity hotspots. It would of relevance to future studies if the authors could extrapolate how microbial diversity in the study region will be affected by changing climate and plant patterns, based on the CCA and driver analysis.

 

Author Response

Please see the attachment

Author Response File: Author Response.docx