Effects of Habitat Differences on Microbial Communities during Litter Decomposing in a Subtropical Forest

Round 1

Reviewer 1 Report

The manuscript “Effects of habitat differences on microbial communities during litter decomposing in a subtropical forest" by Guo et al. studies how leaf litter decomposition rate and microbial community on the decomposing leaves differ between three different habitats - headwater and intermittent streams, and forest floor - in subtropical mountain forest ecosystem in China. Currently, comprehensive understanding about microbial composition dynamics during the decomposition processes of leaf litter lacks, or it is focused separately on aquatic and terrestrial ecosystems that limits understanding about material migration and nutrient circulation between the ecosystems. To fill the gap in knowledge, leaves of a common specie were placed to the three habitats and harvested in one-month intervals during the four-month experiment (conducted during the rainy season) to determine the mass loss and change in the microbial community using PLFA analysis. The study reaches its hypothesis by providing new detailed information about the development of microbial communities in relation to decomposition rate of the leaves in these three separate ecosystems. Unfortunately, the manuscript has technical and statistical issues that concerns me, and the story narrative did not meet the outcomes of the study set-up. Therefore, I suggest rejecting the manuscript. See below some common as well as more detailed comments that I hope the authors find useful to improve the manuscript.

Overall, the narrative of the manuscript and the study set-up do not meet. The story is built over the aim of the study, which is to provide new viewpoints about material migration and nutrient circulation process of the subtropical mountain forest ecosystem. Yet, the methodological part and the outcomes provide information about the habitats separately without any links between them, and thus do not support reaching the aim of the study. Furthermore, the differences between the habitats were not clearly described and discussed. For example, in Lines 47-48: The paragraph is focused on comparing the different stream parts but does not really do this. To build a stronger story that can be answered by the outcomes, I would recommend to refocus the narrative – i.e., how do the three habitats differ and why it is important to understand the differences between these habitats. Alternatively, clearly explain how these habitats links to each other, though, avoid speculating.

Secondly, I am concerned about the statistical part and interpretation. The provided graphical information and the interpretation does not always meet (e.g., In line 153 it is reclaimed that no statistical differences were between the habitats despite in Fig 1a there are statistical differences. Furthermore, it seems that the intermittent stream and forest have similar trends that differ from headwater stream, which is not addressed properly). This makes me wonder, does the data meet the assumptions for the conducted statistical tests?  Was the data exploration done (normality, outliers, homogeneity, variance)? In Fig 3, there are relatively large standard errors in some of the samples (especially the ones that are reduced for the graph). If data exploration was not done, then some of the large standard errors could also be caused by outliers. I recommend to redo the data exploration, and based on the outcomes, either redo the stats or carefully return to the interpretation of the original stats in case no violations of the assumptions were noticed.

Thirdly, the experiment lasted only four months and it was conducted during the rainy season (from the early summer to mid-summer). How was the experimental length and time of the year selected? Why was it during the rainy season? What kind of information was expected and how would it differ if the experiment would have been conducted for example in the dry season, or it had lasted throughout the summer or year? There is some discussion already within lines 275-289, but a more detailed information and reflecting the results into the larger temporal scale would help with the story.

During the experiment, there were long periods when there was no water flow in the intermittent stream (21 out of 114 days. Knowing the stream dynamics there were more low-flow days). How does this place the intermittent stream when comparing between the forest floor and the headwater? The authors should discuss more about the similarities/differences of the intermittent stream between the forest and the headwater stream. Does the intermittent stream remind more forest or headwater stream during the dry season? Discuss this in a wider picture.

Lines 41-42: Need to highlight better how this paper addresses this. The paper, despite having headwater, intermittent stream and forest sites, still focuses on comparing the three different habitats. The link between the habitats does not exist methodologically nor is discussed in detail. 

Lines 47-48: The paragraph is focused on comparing the different stream parts. How do the forest and the stream parts relate, for example, does the intermittent stream remind more headwater forest remind each other during the dry phases?

Line 52-53: “When entering aquatic ecosystems, leaf litter will be promptly colonized by microbial decomposers, concomitantly with leaf litter leaching”. How about in the forest? One could assume the leaves to be colonialized as fast if not faster in the forest. What do the results of this study show?

Lines 73-74: The study focuses on the separate habitats and not the migration of the matter between them. Either change the narrative or reflect the outcomes of the study to already existing publications in discussion.

Lines 79-81: Were no differences between the intermittent and headwater stream expected?  Despite their differences were discussed between the lines 50-57?

Lines 84-85: Unfortunately, the study does not reach the aim.

Lines 93: So, the experiment captures the time when the streams are more flooded.

How realistically does the litter bag experiment, where the leaves are bound to the bottom describe the real-life conditions? One could assume that the leaves are rather mobile and washed away during the rainy season when the water flow is higher.

Lines 106: How far were the forest/stream plots from each other? Provide some distances in text or a map.

Lines 107-108: Removal of the litter and plants alters the microbiota. Why were they removed? How were the effects of the removal addressed?

Lines 108-109: How many litter bags? How many replicates in total? How many were removed at each collection time? This affects the N, which is important in the statistical part.

Line 115: pots = plots?

Line 115: Why were the leaves placed on the sites only for four months, which does not even cover the summer season? How does the length of the experiment affect the decomposition process and the results?

Line 119: ”The samples divided into two parts” à Each sample was divided into two subsamples (e.g. half of the same sample was dried and the half other half of the same sample was freeze-dried) OR some of the replicates were dried and others were used for PLFA analysis?

Line 122: How long were the samples frozen?

Lines 122-126: Please follow the journals guidelines. Dates in form day/month/year.  

Lines 122.126: This part could be written under a subtitle "Monitoring environmental conditions", where the monitoring system would be described in more details.  Was also the flow speed monitored? Was anything else than the water flow monitored? As the author's points earlier, besides the availability of oxygen also temperature affects on the decomposition rate. Was the temperature monitored?  Was anything monitored at the forest plots? If nothing else was monitored, cloud there be climate data available to describe the weather patterns at the sites? This could be useful information to as the water flow in the streams can be caused by rains in the higher elevations, in which case the forest site could have been dry. If more variables were monitored and presented, consider presenting the outcomes in the results.

Lines 145-146: The influence of habitat on what? (I know it is PLFA groups, but please mention it here). Also mention what method you used (RDA, PCA, CCA...)

Line 146: How did you control the decomposition time and habitat type? What analyses did you conduct?

Lines 149-150: “Leaf litter quality”? If the authors do not mean the leaf litter mass loss, please include to the Methods section how the quality was addressed.

Results: Either the statistical analysis, their interpretation or the graphics are not correct. Please, reconduct the data exploration, if needed redo the stats and pay attention how you interpret the outcomes.

Lines 153 & Fig 1.: Not correct. Based on Fig 1a, the Total PLFA is significantly higher in the headwater stream on samples collected on the day 62. Furthermore, the intermittent stream of reminds the forest rather than the headwater stream.

Fig 1. Perhaps add blocks to the graphs to indicate the no-flow dates. This could be interesting information to better understand the flow-dynamics of the streams.

Figure 2. The analysis includes multiple parameters and their ratios. While the ratios can cause collinearity/redundancy, they are often added to PCA analysis. Yet, in this case the number of the ratios. Did the authors investigate how influential the ratios were? Were the analyses done without the ratios to avoid autocorrelation issues? Furthermore, despite collinearity/redundancy would not be detected, the authors should consider if adding the ratios add extra information or not. If not, then they could be removed from the analysis to make it more readable.

Figure 3 and elsewhere suited: “Individual PLFAs were reduced by 10 times their true value, for better readability. “  This is a misleading way to present the data. Please use the generally accepted ways (e.g.  log transformation, break on the y-axis).  Additionally, the error bars are relatively high when high values reported. Was the data exploration done (normality, outliers, homogeneity, variance)? This could also explain the wrong statistical outcomes in Fig 1.

Line 231: Why it the PLFA expected to correlate with the mass loss? The sampling is just snapshots of the decomposer communities. If PLFA degrades fast, doesn’t that mean that the communities can change (seen in the samples), but their biomass is not linearly increasing (new mass is created, old is degraded).

Lines 248-250: Is this because the samples became invaded by different microbes or the biomass of the same microbial groups increased?

Line 250: How about leaching, could that influence on the leaf mass? How? Are there any publications about that and what is the importance in mass loss?

Line 254: This study did not focus on the shift of microbial community because the day 0 PLFA analysis were not done or at least presented in the manuscript. The study focused on development of microbial populations in three different habitats.

Line 257: Was the water depth measured in the streams? How deep was the maximum depth, and did it vary between the headwater and intermittent stream? How fast is the water flow in the streams --> how realistic it is to expect that the leaves would actually stay in one place like in litter bag studies? Please, add any available information to the manuscript (perhaps Materials & methods -section where the monitoring is discussed) This would be an interesting topic to study: leaf decomposition in streams with and without water transportation.

Line 259-260: So, the colonialization does not happen in the forest floor? I would expect that to happen in the forest floor as well.

Line 262: Are there differences in the leaching and oxygen conditions between the streams? E.g. how do the dry periods effect on these conditions, and thus to the decomposition processes? How about during high flow, does the high waterflow cause more leaching and perhaps even make it harder for microbes to colonize the leaf?

Lines 283, 288: Check the guidelines for citing

Lines 301-304 & 310-312: This is the first time when the authors discuss about the differences between the intermittent and headwater streams. This should have been done much earlier and in much more detailed.

321-323: The study focused on the three different habitats independently. The study did not include any parameters that would give any suggestions about any interaction between them. Please, revise the sentense

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

-The results were very interesting and corroborated with the literature. However, the author (s) left something to be desired in the discussion of the results. They could have explored the discussion better, emphasizing the importance of this study, for example.

-The work is interesting and compatible with the level of the journal. -All information on the work was understandable and consistent with each other.

-Work more on the discussion part

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

No comments. The experiment design, data treatment, and discussion/conclusion are correct.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The manuscript has improved greatly after the revisions. The authors gave throughout responses on the comments and addressed the issues related on the data analysis. The methods and the narrative  did not tell the same story in the previous version, but meet in the revised version.

While the story works in its current stage, it feels that the data is not used in full capacity and it is not allowed to tell its full story. I still miss more detailed comparisons between the habitats in the introduction. Additionally, throughout the manuscript the forest parts seems to be the secondary, terrestrial evil, which shows as short superficial description and lack of references. While neither of these changes is not necessary, considering them could improve the quality of the manuscript from nice habitat comparison paper to next level. Of course, these modifications take time, so consider this comment as suggestion that you can leave or take.

While the manuscript is now technically more sound, there are still major issues related on the text flow, writing style and language. In some parts the text is good, but unfortunately in other parts it is still hard to follow. Additionally, the story can be shortened (remove some repetition and extra words) and sharpened (=I see this as instead of thinking "how to write this" it is better to think "why to write this"). I would recommend the authors to sit down and read the text one more time while keeping the focus on editing of the text and maintaining its flow. This could perhaps help to reach the point of the previous comment.

Below my minor comments:

Line 48-49: The study is still focusing on litter decomposition separately in the three habitats, i.e., the litter bags were placed to each habitat and kept in the same habitat throughout the experiment. Wonder if the authors are trying to highlight that usually the studies focus only on one ecosystem/habitat only (the study either focus on terrestrial/aquatic habitat)?

Line 52-53: References. Throughout the manuscript the forest parts seems to be the secondary, terrestrial evil, which shows as short superficial description and lack of references.

Line 90-91: This is nicely written! Please, modify the text in the rows 48-49 to match this part.

Line 114-115: "Litter bags were placed on the forest floor after removing plants and litter from the soil surface of the quadrat."

Please, specify the reasoning for removing the plants and litter also in the manuscript. Also, discuss about the potential effects. This was addressed in the responde to reviewer:
Response 14: We investigated the annual accumulation of litter per unit area one year before the beginning of the experiment. The litter of the required weight is loaded according to the area of the litter bag and then placed in three habitats. Therefore, we removed the existing litter and plants in the sample site in this experiment, or there will be more litter existing compared with natural decomposition conditions.

Author Response

Please see the attachment

Author Response File: Author Response.pdf