the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 02 Jul 2026
Linking the environmental factors and heterotrophic bacteria to the variability of carbon isotope (δ13C) across crustacean zooplankton in shallow freshwater ecosystems
Abstract. Planktonic crustaceans are important part of carbon cycling in in shallow, temperate freshwater habitats. The objective of the study was to examine the link between selected environmental variables, densities of various groups of bacteria and δ13C of calanoid copepods, cyclopoid copepods and Daphnia sp. Generalized Linear Models (GLMs) with stepwise-selected environmental predictors showed that δ13C of calanoid copepods was positively correlated with the Day of the year and Other bacteria density and negatively correlated with NH₄⁺ and NO₃⁻ concentrations. Cyclopoid copepods’ δ13C was correlated exclusively with water temperature, which showed a positive effect on the stable carbon isotopes ratio. GLM for Daphnia sp. was the most complex and retained the Day of the year, Other bacteria density and PO43- as positively correlated predictors, while β-proteobacteria NH₄⁺ and NO₃⁻ were included in the model as negatively correlated predictors for δ13C. Our study highlights the complex impact of environmental conditions, microbial processes and functional trait-driven responses on δ13C values in planktonic crustaceans. Thanks to integration of plankton ecology, microbiology and the analysis of stable isotopes the study improves the understanding of carbon flow in shallow freshwater ecosystems.
- Preprint
(868 KB) - Metadata XML
-
Supplement
(215 KB) - BibTeX
- EndNote
Status: closed
-
RC1: 'Comment on egusphere-2026-3444', Anonymous Referee #1, 17 Jul 2026
-
AC1: 'Reply on RC1', Wojciech Krztoń, 14 Sep 2026
Reviewer 1:
General comments:
The manuscript presents a valuable and timely contribution to research on carbon cycling in shallow freshwater ecosystems. The authors address the important issue of identifying the environmental and microbial factors shaping carbon isotope (δ¹³C) variability in the major groups of crustacean zooplankton. A particular strength of the study is the integration of stable isotope analyses with bacterial community characterization and environmental parameters, providing insights into pelagic ecosystem functioning from both ecological and biogeochemical perspectives. The study is based on an extensive field dataset collected seasonally from several shallow eutrophic water bodies. The analytical methods and statistical models are appropriate for addressing the stated research objectives, and the results are presented in a clear and logical manner. The Discussion deserves particular recognition for placing the findings within the context of current knowledge and interpreting them in light of the functional differences among the investigated zooplankton groups. I also appreciate that the authors openly acknowledge the main limitation of the study, namely the lack of δ¹³C measurements of potential food sources. Recognizing this limitation and identifying directions for future research strengthens the credibility of the conclusions and highlights promising avenues for further investigation.
Overall, I believe that the manuscript provides novel insights into the functioning of freshwater food webs and is suitable for publication after minor revisions.
Response: We sincerely thank Reviewer 1 for the positive and constructive evaluation of our manuscript. We greatly appreciate the reviewer’s recognition of the value of integrating stable isotope analyses with environmental and microbial data, as well as the strengths of our field dataset and ecological interpretation. Please find responses to all of the comments below. We believe, that revisions that we will make will improve the presentation and interpretation of our results.
Specific comments:
- In the Introduction, the authors hypothesize that the relationship between δ¹³C values and bacterial abundance would be observed only in Daphnia, as copepods are not expected to feed directly on bacteria. However, the results indicate that bacterial-related variables ("Other bacteria" and, marginally, β-proteobacteria) were also retained in the final model for calanoid copepods. It would be worthwhile to discuss this observation in greater detail and propose possible explanations, for example by considering indirect effects of bacteria mediated through the food web or through changes in the quality of phytoplankton food resources. Such an addition would help explain why the obtained results partially differ from the original hypothesis.
Response: Thank you for this comment. We propose following explanation in the “Discussion”: “Calanoid copepods are able to feed predatory on ciliates, which was demonstrated e.g. by Kunzmann et al. (2019). Additionally, Kunzman et al. (2019) observed increases in bacteria abundance, which was most likely a cascading effect of ciliates removal by the predatory copepod. Although, calanoid copepods cannot feed directly on the bacteria, therefore positive effect of ‘Other bacteria’ on δ¹³C may indicate indirect effects. On the other hand, given findings of Kunzmann et al. (2019) this may be just a circumstantial correlation, suggesting increases in ‘Other bacteria’ abundance as an effect of grazing by the calanoid copepods”.
- The study focuses on the abundance of individual bacterial groups; however, it is worth noting that abundance does not necessarily reflect metabolic activity. Carbon cycling processes may be driven by the activity of only a small fraction of the bacterial community. A brief discussion of this limitation would provide a more balanced interpretation of the results.
Response: Thank you for pointing this out. We will provide a better explanation in the discussion: “Biogeochemical cycle of carbon are regulated by microbes (Landa et al. 2016). However, the efficiency of carbon transfer may depend on small fraction of the bacterial community that is metabolically active. It is worth to point out that this fraction can be different from total abundance of bacteria. Moreover, the activity may vary throughout the vegetation season (Kubera et al. 2025)”
- The analysis of bacterial communities represents one of the major strengths of the study and provides valuable information for interpreting the observed patterns. Nevertheless, it would be useful to discuss in greater detail the limitations of the FISH method, which enables the quantitative assessment of broad phylogenetic groups but does not provide information on species composition or the metabolic potential of individual taxa. This is particularly relevant for the "Other bacteria" group, which emerged as an important predictor in the statistical models while encompassing a highly diverse assemblage of microorganisms with different ecological functions. A brief discussion of this limitation would strengthen the manuscript. Furthermore, it could be noted that only high-throughput sequencing approaches, such as metagenomics or functional metagenomics, would provide a more comprehensive characterization of bacterial communities and enable the identification of specific taxa and metabolic pathways responsible for the observed relationships with δ¹³C values. Such a statement would not diminish the value of the present study but rather highlight an interesting direction for future research.
Response: We thank the reviewer for this valuable comment. We agree that a more detailed discussion of the methodological limitations of the FISH technique will improve the quality of the manuscript and provide a more adequate interpretation of the data regarding the bacterial community. We propose following clarification, in the revised version of the manuscript: “FISH is an effective method that allows for the direct visualization, counting, and spatial localization of microorganisms in their natural environment. Importantly, FISH probes can be designed to target phylogenetic groups, ranging from broad taxonomic levels down to the genus or species level under appropriate conditions. However, the taxonomic resolution of this method is inherently dependent on the specificity and range of the applied probes. In this study, the FISH method was primarily used to quantitatively determine broad phylogenetic groups; therefore, the obtained data should not be interpreted as a detailed characterization of the bacterial community at the species level.
FISH method provides information based on the detection of ribosomal RNA and thus does not directly reveal the genomic or metabolic potential of the individual detected taxa. Quantitative signals may also be influenced by differences in the range and specificity of probes, cell permeability, ribosomal RNA content, and the physiological state of the microorganisms. Consequently, differences in abundance determined by the FISH method between phylogenetic groups should be interpreted as differences in abundance/detectability of the studied groups, rather than as direct evidence of differences in their metabolic activity or ecological functions. These remarks are particularly important regarding the category "Other bacteria," which represents a heterogeneous residual group encompassing microorganisms that may differ significantly in terms of taxonomy, physiology, and ecological role. Therefore, although the "Other bacteria" group proved to be an important predictor in statistical models, this result should be interpreted at the level of the aggregated bacterial category detected using the applied FISH method. Unfortunately, the current data do not allow us to determine which specific taxa within this group contributed to the observed relationship with δ¹³C values, nor to establish what metabolic processes may underlie this relationship.
The application of high-throughput sequencing methods, such as 16SrRNA gene amplicon sequencing or shotgun metagenomics, could provide a much more comprehensive characterization of the structure and functional potential of the studied bacterial communities. High-throughput sequencing and integrated multi-omic approaches as an important direction for future research. Such approaches could help establish the taxonomic composition of the "Other bacteria" category and identify microbial genes and metabolic pathways potentially related to the observed variability in δ¹³C values.”.
We believe that the explanation will moderate the original formulations regarding functional interpretation in accordance with the obtained research results. The above explanation will strengthen, rather than diminish, the value of this study by defining the scope of conclusions that can be drawn from an assessment based on the FISH technique.
- The explanation of the negative effects of NH₄⁺ and NO₃⁻ on δ¹³C values is convincing and well supported by the literature. However, it would be beneficial to emphasize more explicitly that these relationships are correlative and do not necessarily imply direct cause-and-effect relationships. Such clarification would make the interpretation more balanced.
Response: We will include the following explanation in the Discussion: “The study was carried out as the fieldwork-based monitoring, therefore we can discuss only correlative effects”. Thank you for pointing this out.
- The number of isotope samples available for Daphnia is considerably smaller than for the two copepod groups. This is understandable given the field conditions; however, it would be useful to briefly acknowledge in the Discussion that the smaller sample size may have reduced the statistical power of the analyses for this taxon.
Response: We thank the reviewer for pointing this out. We apologize for the confusion regarding the sample sizes reported in the manuscript. The sample sizes stated in the previous version of the manuscript (36 calanoid copepod, 29 cyclopoid copepod, and 12 Daphnia samples) were inadvertently retained from an earlier version of the dataset used for a different set of analyses. In fact, the isotope dataset used for the analyses presented in the manuscript contains 45 calanoid copepod, 45 cyclopoid copepod, and 26 Daphnia samples. From this dataset, stepwise-selected models included slightly lower number of fitted observations (45 calanoid copepod samples, 45 cyclopoid copepod samples and 26 Daphnia samples). Thus, the sample sizes reported in the previous version of the manuscript were incorrect. We will correct this information in the revised manuscript.
We have carefully checked the dataset and revised the manuscript to report the correct sample sizes throughout. We also agree with the reviewer that the number of Daphnia samples is lower than that of the two copepod groups and therefore may provide lower statistical power for analyses involving Daphnia. However, the difference is considerably smaller than suggested by the sample sizes reported in the previous version (26 Daphnia samples compared with 45 samples for each copepod group). We will nevertheless add a following statement to the Discussion: “Considering the sample size for Daphnia (N = 26) and relatively large number of stepwise-selected environmental predictors (11), presented results for relationships of δ¹³C in this group should be interpreted with appropriate caution”.
Technical comments:
- Line 17: "important part of carbon cycling in in shallow..." – remove the duplicated word "in".
- Line 28: "Thanks to integration..." → "Thanks to the integration..."
- Throughout the manuscript, please ensure consistency in the formatting of units (e.g., "10 mL" instead of "10mL").
Response: Thank you. We will take these comments into account while revising the manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-3444-AC1
-
AC1: 'Reply on RC1', Wojciech Krztoń, 14 Sep 2026
-
RC2: 'Comment on egusphere-2026-3444', Anonymous Referee #2, 21 Jul 2026
Reviewer comments for Krzton et al. 2026, submitted to Biogeosciences
Krzton et al. attempt to understand the parameters affecting the carbon isotope composition (δ13C) of planktonic crustaceans. If I understand their introduction correctly, they are motivated by the larger topic of climate change impacting food webs and carbon flow within freshwater ecosystems. By correlating a variety of environmental parameters (e.g., pH, temperature, ammonium and nitrate, abundance of bacterial groups, etc.) with the δ13C of planktonic crustaceans using generalized linear models, they attempt to derive a mechanistic understanding of planktonic δ13C. This, in turn, would allow their results to be applied to freshwater ecosystems more broadly and not just their study sites. This is a topic of interest to many groups, as the δ13C of planktonic crustaceans (as well as other freshwater macro- and microorganisms) have been measured to disentangle the complex food webs of marine and freshwater ecosystems, and assess potential impacts of climate change. Therefore, the manuscript is of interest in that regard.
However, I have significant concerns about the manuscript as it is currently presented. My first concern regards readability; at times, it was difficult for me to understand what exactly the authors were trying to say. My second concern regards data availability. I was not able to fully independently evaluate the manuscript as the raw data measured by the authors was never presented, either in the main text or as a supplementary table. Instead, the authors only present averages of their data, or model outputs. For example, Figures 2-4 are just graphical renderings of Table 3, which are their modeled outputs. They also do not give the sequences used for their FISH probes. For the probes, this is extremely important as some FISH probes are outdated and will incorrectly classify certain bacteria in certain clades, largely because our knowledge of bacterial taxonomy & classification is rapidly changing as more genome-based approaches are used, and as sequence data repositories steadily grow. Instead, the authors state that “The dataset used within this study will be made available on a reasonable request” – I do not think this is sufficient (is it reasonable to expect a reviewer to ask for your data, when you send it in for review?) and wonder if it even violates the journal’s own policies regarding data availability (see more detail below). I also note that this is to the detriment to the authors – fewer people will cite your paper if the data is not readily available and they have to contact you (e.g. go through the trouble of finding your current institution, finding updated contact information, waiting for your response, potentially negotiating with you for what is a ‘reasonable’ request…). More often than not, people may just find a similar study and cite that instead.
Given all that, I had to take the authors at their word regarding what they found. In this regard, it does not appear that they can draw strong conclusions. They do not find significant differences in δ13C among the three planktonic crustacean groups they examined (Supplemental Table S2). Perhaps not surprisingly, they can therefore only conclude that there is a “complex interplay of environmental conditions, microbial processes and functional trait-driven response of δ13C values in planktonic crustaceans” (Line 315). They also note, correctly, that one of the weaknesses of their paper is that they don’t measure the δ13C of the actual dietary sources – this would be a much stronger paper if they had instead waited to publish after getting that data. Instead, this paper relies on correlation of many parameters with few data points. For example, only 12 Daphnia samples were measured for δ13C and the fitted model for Daphnia contains 11 predictors – the potential for model overfitting, and potential over-interpretation, is extremely high. And, as I noted above, since they do not include any of their actual measured data… I cannot independently confirm whether their data is indeed overfitted or not.
Therefore, I recommend that the manuscript should be rejected.
Major comments
Issues with manuscript readability
I would strongly recommend that the authors work with the editor or other English speaker to edit the writing of the final manuscript. Alternatively, it appears that Biogeosciences allows use of AI tools in manuscript writing, as long as the usage is either described in the Methods section or Acknowledgments: https://www.biogeosciences.net/submission.html
My concern is twofold. First, there are numerous small errors – e.g. grammar, typos – throughout the manuscript that need to be fixed. For example, just in the abstract: i) in the first sentence of the abstract, the word “in” appears twice; ii) “Day” and “Other” are capitalized in the abstract in Line 21, but should not be; iii) The last sentence of the abstract is missing a period at the end. These small errors make it difficult to read the manuscript smoothly. There are also numerous articles missing – for example, Line 46 should say “are an important part of…” instead of “are important part of…”
I note, also, that the quality of writing noticeably changes throughout the manuscript. For example, the writing is much clearer in some sections, like “2.5 Statical analyses” – there are few missing articles (e.g., “the”). But then the writing changes right afterward, in section “3.1. Environmental variables” – for example, the grammar is slightly off in Line 153 articles and should instead be written as “Water temperatures in the studied lakes ranged from 10.8 to 25.6 C, reaching the highest values in July and August.” So, I would recommend whoever helped write Section 2.5 to help edit the entire manuscript.
Second, it was difficult for me to understand the meaning of some sentences. For example, when describing the results (Line 185):
“Mean values of δ13C were found in calanoid copepods accounted for -31.8‰ with range between -41.5‰ to -23.7. Mean δ13C values of cyclopoid copepods was -30.6‰ and range: -40‰ to -22.2‰). δ13C measurements of Daphnia sp. showed mean value of -31.2‰ and range between -42.5‰ and –24.4‰)”
There are typos, like the extra right parentheses (this: “)”) that show up. I am also confused by “accounted for” in the first sentence. Are you just trying to say that the mean δ13C value of calanoid copepods was -31.8‰ with a range from -41.5 to -23.7‰?
Discussion
Some additional information on the particulars of the author’s study sites, compared to prior studies, would increase the quality of the manuscript. For example, the authors find that the only factor that positively affected the δ13C of cyclopoid copepods was water temperature (Line 237) but was not a significant predictor for Daphnia sp. δ13C. What’s the seasonal temperature range that these water bodies experience in a year? Do these water bodies freeze over in the wintertime? Would this at all lead you to predict the differing responses of temperature for cyclopoid copepods vs. Daphnia?
In addition, as these water bodies are quite close to Cracow, I would expect high amounts of anthropogenic influence – would this affect your interpretations? For example, the authors counterintuitively find that higher concentrations of ammonium and nitrate to be correlated with slower phytoplankton growth rates – as they note, this is “opposite to what would be expected… but seasonal dynamics and declining water temperatures effects might be involved.” However, the classic understanding of phytoplankton growth rates increasing with reactive nitrogen availability is based on the assumption that these nutrients are typically rare, like in the open oceans. For shallow, freshwater and even artificial lakes – like the ones studied here – would reactive nitrogen species just be high all the time? Would the authors be able to provide additional data on the chemistry of these lakes?
In Line 288, the authors discuss the positive effect of “Other bacteria” on δ13C values. I’m not sure what exactly the authors can do to make this part stronger, but I found it to be speculative as the measurement of community diversity the authors used – alpha vs. beta vs. gamma vs. delta protobacteria – is quite coarse and does not reflect the current genome-based taxonomy used to classify bacteria today. For example, using marker genes beyond rRNA and those targeted in FISH, it has been proposed that delta proteobacteria be reclassified into clades that better reflect the diverse range of chemistry that different subclades can carry out, and which are masked by the broad paintbrush of ‘delta proteobacteria’ ( https://doi.org/10.1099/ijsem.0.004213).
Therefore, inferring mechanism from these coarse groups of bacteria is inherently speculative because each species of bacteria within these groups can perform a wide range of chemistry. As the authors note themselves, methane oxidation can be accomplished by bacteria from the beta, alpha, and gamma clades. Unless the authors can determine this at a finer resolution – i.e., they detect a phylum like Cyanobacteriota – I would suggest they walk back some of their claims. (I also note that, even knowing there is Cyanobacteria present, most studies do not speculate on mechanism as Cyanobacteria themselves can carry out extremely diverse reactions…)
FISH
I can’t figure out the exact sequences of the different probes that the authors used. I would strongly recommend that they have a table in the supplemental with the sequence. Based on the citations, I’m currently concerned that they are using some more outdated FISH probes that may either miss legitimate members of their target group, or hybridize to organisms that are now classified elsewhere. The citations are also confusing because they cite the Lücker et al. 2007 paper, but that is a methods paper that tests a variety of probes – just citing it does not tell me which specific one they used. In addition, the authors used NON338 to evaluate nonspecific fluorescent binding – this is good, as it detects fluorescence caused by nonspecific binding, sticking, etc. But it does not indicate that the taxon-specific probes have complete coverage or perfect phylogenetic specificity. Thus, “less than 5% nonspecific staining” addresses the analytical problem of just using the probes, but not the broader taxonomic problem. This should be discussed as a limitation of their study in the Discussion.
Data Availability
I strongly disagree with the authors statement that “The dataset used within this study will be made available on a reasonable request,” as it is a common convention in the field to have data available in a supplement or publicly available dataset upon submission. As noted above, I couldn’t even figure out what sequences the authors used for their FISH. For example, here is a recent preprint in Biogeosciences where the authors give 18S tables in their supplement (I know this is slightly different, but just wanted to give a recent example):
https://egusphere.copernicus.org/preprints/2026/egusphere-2026-4192/
Also, the authors should give their raw, measured values for environmental parameters (like pH, temperature, ammonium, etc.) as well as carbon isotope data. Not just the averaged data (e.g., Table 1). Here is a recent example from Biogeosciences, where they have made this data available in their supplemental:
https://bg.copernicus.org/articles/19/1/2022/
I also think this may violate the journal’s own policies. Although the authors technically have a statement on how their data can be accessed, in line with the journals request, the journal also states: “If the data are not publicly accessible at the time of final publication, the data statement should describe where and when they will appear, and provide information on how readers can obtain the data until then. Nevertheless, authors should make such embargoed data available to reviewers during the review process in order to foster reproducibility. The Copernicus review system allows to define such assets as 'access limited to reviewers' and reviewers must then sign that they will use such data only for the purpose of reviewing without making copies, sharing, or reusing. In rare cases where the data cannot be deposited publicly (e.g., because of commercial constraints), a detailed explanation of why this is the case is required. The data needed to replicate figures in a paper should in any case be publicly available, either in a public database (strongly recommended), or in a supplement to the paper.”
https://www.biogeosciences.net/policies/data_policy.html#data_availability
Figures
The figures are difficult to interpret. As a starter, the text on the figures is far too small and difficult to read. But, more importantly, there are no figures showing the actual data measured in this study besides Figure 1. Figures 2-4 are essentially a graphical rendering of Table 3, with model-based uncertainty bands. It basically does not allow the reader, or reviewer, to judge for themselves if the model is indeed a good fit for the measured data. Also – For Figure 4, the phrase “stepwise-selected significant predictors” is confusing because the manuscript states that the full Daphnia model was retained, while Figure 4 apparently displays only predictors with P<0.05. In addition, only 12 Daphnia samples were measured and the fitted model for Daphnia contains 11 predictors (day of the year, proteobacteria, alpha proteobacteria, etc.), the authors should address potential model overfitting – this could also be addressed by just showing the actual measured data. Again, since the data are not made available, it is difficult to really independently evaluate this manuscript.
Minor comments
Line 19: You should briefly define what d13C is – for example, “densities of various groups of bacteria and the carbon isotope composition (d13C) of calanoid copepods”
Line 39: Do you mean ‘reviewed by’ instead of ‘revised by’?
Line 90: Typo, should be “depths” not “depts”
Supplemental: This looks like raw outputs from R. The table caption should define what the column headers are – e.g., “Df; Degrees of freedom.”
Citation: https://doi.org/10.5194/egusphere-2026-3444-RC2 -
AC2: 'Reply on RC2', Wojciech Krztoń, 14 Sep 2026
Reviewer 2
Reviewer comments for Krzton et al. 2026, submitted to Biogeosciences
Krzton et al. attempt to understand the parameters affecting the carbon isotope composition (δ13C) of planktonic crustaceans. If I understand their introduction correctly, they are motivated by the larger topic of climate change impacting food webs and carbon flow within freshwater ecosystems. By correlating a variety of environmental parameters (e.g., pH, temperature, ammonium and nitrate, abundance of bacterial groups, etc.) with the δ13C of planktonic crustaceans using generalized linear models, they attempt to derive a mechanistic understanding of planktonic δ13C. This, in turn, would allow their results to be applied to freshwater ecosystems more broadly and not just their study sites. This is a topic of interest to many groups, as the δ13C of planktonic crustaceans (as well as other freshwater macro- and microorganisms) have been measured to disentangle the complex food webs of marine and freshwater ecosystems, and assess potential impacts of climate change. Therefore, the manuscript is of interest in that regard.
However, I have significant concerns about the manuscript as it is currently presented. My first concern regards readability; at times, it was difficult for me to understand what exactly the authors were trying to say. My second concern regards data availability. I was not able to fully independently evaluate the manuscript as the raw data measured by the authors was never presented, either in the main text or as a supplementary table. Instead, the authors only present averages of their data, or model outputs. For example, Figures 2-4 are just graphical renderings of Table 3, which are their modeled outputs. They also do not give the sequences used for their FISH probes. For the probes, this is extremely important as some FISH probes are outdated and will incorrectly classify certain bacteria in certain clades, largely because our knowledge of bacterial taxonomy & classification is rapidly changing as more genome-based approaches are used, and as sequence data repositories steadily grow. Instead, the authors state that “The dataset used within this study will be made available on a reasonable request” – I do not think this is sufficient (is it reasonable to expect a reviewer to ask for your data, when you send it in for review?) and wonder if it even violates the journal’s own policies regarding data availability (see more detail below). I also note that this is to the detriment to the authors – fewer people will cite your paper if the data is not readily available and they have to contact you (e.g. go through the trouble of finding your current institution, finding updated contact information, waiting for your response, potentially negotiating with you for what is a ‘reasonable’ request…). More often than not, people may just find a similar study and cite that instead.
Given all that, I had to take the authors at their word regarding what they found. In this regard, it does not appear that they can draw strong conclusions. They do not find significant differences in δ13C among the three planktonic crustacean groups they examined (Supplemental Table S2). Perhaps not surprisingly, they can therefore only conclude that there is a “complex interplay of environmental conditions, microbial processes and functional trait-driven response of δ13C values in planktonic crustaceans” (Line 315). They also note, correctly, that one of the weaknesses of their paper is that they don’t measure the δ13C of the actual dietary sources – this would be a much stronger paper if they had instead waited to publish after getting that data. Instead, this paper relies on correlation of many parameters with few data points. For example, only 12 Daphnia samples were measured for δ13C and the fitted model for Daphnia contains 11 predictors – the potential for model overfitting, and potential over-interpretation, is extremely high. And, as I noted above, since they do not include any of their actual measured data… I cannot independently confirm whether their data is indeed overfitted or not.
Therefore, I recommend that the manuscript should be rejected.
Response: We sincerely thank Reviewer 2 for the careful and constructive assessment of our manuscript. We appreciate the reviewer’s concerns regarding the readability of the manuscript, data availability, interpretation of the statistical analyses, and the presentation of the underlying data. We would also like to clarify that the sample sizes reported in the previous version were inadvertently carried over from an earlier dataset used for different analyses; the correct sample sizes for the isotope analyses are 48 calanoid copepod, 49 cyclopoid copepod, and 27 Daphnia samples. In revised manuscript we will carefully consider all of these points. We believe that these revisions will substantially improve overall presentation of the manuscript.
Below we include our comments and responses to all of the reviewer’s concerns, which are listed above, and as major and minor comments.
Major comments
Issues with manuscript readability
- I would strongly recommend that the authors work with the editor or other English speaker to edit the writing of the final manuscript. Alternatively, it appears that Biogeosciences allows use of AI tools in manuscript writing, as long as the usage is either described in the Methods section or Acknowledgments: https://www.biogeosciences.net/submission.html
- My concern is twofold. First, there are numerous small errors – e.g. grammar, typos – throughout the manuscript that need to be fixed. For example, just in the abstract: i) in the first sentence of the abstract, the word “in” appears twice; ii) “Day” and “Other” are capitalized in the abstract in Line 21, but should not be; iii) The last sentence of the abstract is missing a period at the end. These small errors make it difficult to read the manuscript smoothly. There are also numerous articles missing – for example, Line 46 should say “are an important part of…” instead of “are important part of…”
- I note, also, that the quality of writing noticeably changes throughout the manuscript. For example, the writing is much clearer in some sections, like “2.5 Statical analyses” – there are few missing articles (e.g., “the”). But then the writing changes right afterward, in section “3.1. Environmental variables” – for example, the grammar is slightly off in Line 153 articles and should instead be written as “Water temperatures in the studied lakes ranged from 10.8 to 25.6 C, reaching the highest values in July and August.” So, I would recommend whoever helped write Section 2.5 to help edit the entire manuscript.
- Second, it was difficult for me to understand the meaning of some sentences. For example, when describing the results (Line 185):
- “Mean values of δ13C were found in calanoid copepods accounted for -31.8‰ with range between -41.5‰ to -23.7. Mean δ13C values of cyclopoid copepods was -30.6‰ and range: -40‰ to -22.2‰). δ13C measurements of Daphnia sp. showed mean value of -31.2‰ and range between -42.5‰ and –24.4‰)”
- There are typos, like the extra right parentheses (this: “)”) that show up. I am also confused by “accounted for” in the first sentence. Are you just trying to say that the mean δ13C value of calanoid copepods was -31.8‰ with a range from -41.5 to -23.7‰?
Response: We thank the reviewer for their assessment of the manuscript. The specific examples provided in the review are helpful for identifying similar issues throughout the text. While preparing the revised version, we will work on English grammar, syntax, punctuation, use of articles, and the overall readability and consistency of the writing style. The final version of the manuscript will undergo professional proofreading by a native English speaker and/or the use of AI tools.
Discussion
- Some additional information on the particulars of the author’s study sites, compared to prior studies, would increase the quality of the manuscript. For example, the authors find that the only factor that positively affected the δ13C of cyclopoid copepods was water temperature (Line 237) but was not a significant predictor for Daphnia sp. δ13C. What’s the seasonal temperature range that these water bodies experience in a year? Do these water bodies freeze over in the wintertime? Would this at all lead you to predict the differing responses of temperature for cyclopoid copepods vs. Daphnia?
Response: Water temperatures in the studied waterbodies during the sampling period (April–October) ranged from 10.8 to 25.6 ℃. Temperature ranges for each waterbody are given in Table 1. The waterbodies freeze over in winter; however, the ice-on period has been observed to decline over the past few decades, which can trigger changes in the functioning of aquatic ecosystems. Milder winters and longer springs can lead to shifts in the species dominating during summer (e.g. a switch from cyanobacteria to dinoflagellates), as recently demonstrated by Rabow et al. (2026). Although this is particularly interesting in the context of climate change, the influence of the dominant algal species on crustacean δ13C was not considered in this manuscript.
In order to address the effects of temperature on δ13C we will provide following clarification:
“Changes in water temperature strongly influence the functioning of aquatic ecosystems, with even slight warming increasing phytoplankton growth rates and zooplankton metabolic demands, and therefore carbon transfer within the food web (Kratina et al. 2012). Rising temperature may affect consumers’ δ13C; however, differences in the responses of daphnia and cyclopoid copepods can be explained by their feeding biology. Daphnia are non-selective filter feeders, consuming mainly small-sized algae and bacteria. Cyclopoid copepods are selective feeders, able to switch between preferred food sources depending on their availability (e.g. De Meester et al. 2023). A recent study by Huỳnh et al. (2026) indicates that at higher temperatures, copepods can switch towards increased predation on rotifers to meet their higher metabolic demands. A similar mechanism may operate in the food webs studied in our manuscript, as we found water temperature to be a significant predictor of cyclopoid copepods’ δ13C. “
- In addition, as these water bodies are quite close to Cracow, I would expect high amounts of anthropogenic influence – would this affect your interpretations? For example, the authors counterintuitively find that higher concentrations of ammonium and nitrate to be correlated with slower phytoplankton growth rates – as they note, this is “opposite to what would be expected… but seasonal dynamics and declining water temperatures effects might be involved.” However, the classic understanding of phytoplankton growth rates increasing with reactive nitrogen availability is based on the assumption that these nutrients are typically rare, like in the open oceans. For shallow, freshwater and even artificial lakes – like the ones studied here – would reactive nitrogen species just be high all the time? Would the authors be able to provide additional data on the chemistry of these lakes?
Response: We thank the reviewer for pointing this out. We agree that proximity to Cracow exposes the studied sites to relatively high anthropogenic pressure; however, at this point we do not have any data from sites free from human activities. Therefore, although interesting, interpretation of these results in the light of the presence or absence of anthropogenic pressure is rather limited. In the revised version of the manuscript we will indicate that these results were obtained from water bodies experiencing high anthropogenic pressure, and that the interpretations require caution.
We would also like to explain that the studied water bodies are small and shallow, and although all are of artificial origin, they are not artificial reservoirs in the sense of dam reservoirs.
We did not analyse reactive nitrogen species (i.e. nitric oxide, nitrogen dioxide radical, peroxynitrite), and it is therefore impossible to consider their influence in this manuscript. In our study systems, nitrate and ammonium were detected during most samplings (summarised in Table 1), suggesting that these shallow water bodies were not consistently depleted in dissolved inorganic nitrogen. All measurements of nitrate and ammonia concentrations will be included in the dataset that we will make available alongside the revised manuscript. (https://github.com/wmkrzt/Environ.Bacteria.d13C).
- In Line 288, the authors discuss the positive effect of “Other bacteria” on δ13C values. I’m not sure what exactly the authors can do to make this part stronger, but I found it to be speculative as the measurement of community diversity the authors used – alpha vs. beta vs. gamma vs. delta protobacteria – is quite coarse and does not reflect the current genome-based taxonomy used to classify bacteria today. For example, using marker genes beyond rRNA and those targeted in FISH, it has been proposed that delta proteobacteria be reclassified into clades that better reflect the diverse range of chemistry that different subclades can carry out, and which are masked by the broad paintbrush of ‘delta proteobacteria’ ( https://doi.org/10.1099/ijsem.0.004213).
- Therefore, inferring mechanism from these coarse groups of bacteria is inherently speculative because each species of bacteria within these groups can perform a wide range of chemistry. As the authors note themselves, methane oxidation can be accomplished by bacteria from the beta, alpha, and gamma clades. Unless the authors can determine this at a finer resolution – i.e., they detect a phylum like Cyanobacteriota – I would suggest they walk back some of their claims. (I also note that, even knowing there is Cyanobacteria present, most studies do not speculate on mechanism as Cyanobacteria themselves can carry out extremely diverse reactions…)
Response: We thank the Reviewer for this valuable comment. We agree that the interpretation of the positive correlation between the proportion of “other bacteria” and δ¹³C values should be presented with greater caution, particularly in light of the limited taxonomic resolution of the FISH method used and the changes that have occurred in bacterial classification based on genomic data. We will add the following explanation in the discussion about FISH: “In this study, FISH probes were used to quantitatively assess broad phylogenetic groups, including Alpha-, Beta-, Gamma- and Delta-proteobacteria. This approach allowed us to determine the relative abundance of major bacterial groups in the samples studied and was particularly useful for comparing community structure across research periods. However, the taxonomic resolution of this approach is limited and does not allow the unambiguous identification of specific genera or species responsible for the observed relationships. FISH is primarily a method that enables the direct identification and quantitative assessment of specific groups of microorganisms, and the scope of the information obtained depends on the specificity of the probes used. However, our results should be interpreted with caution, because using FISH we obtained different microbial groups that do not constitute a coherent monophyletic unit. An example is the group “Deltaproteobacteria”, which may encompass microorganisms with significantly different metabolic potentials and ecological functions. For example, representatives of lineages currently classified within Desulfobacterota include numerous bacteria associated with sulphate reduction and sulphur metabolism, whereas representatives of Myxococcota are characterised by different metabolic and ecological strategies, including predatory behaviour. In turn, Bdellovibrionota includes bacteria with a specialised mode of predation on other bacteria (Waite et al. 2020).”
Consequently, in our revised manuscript, we propose to modify the relevant section and narrow the scope of our interpretation of the results. We will no longer interpret the positive correlation between the proportion of “other bacteria” and δ¹³C values as evidence of a specific metabolic mechanism but rather as an indication that an increase in the proportion of microorganisms belonging to broadly defined groups, which were not detected as the main groups analysed in the study, coincided with changes in δ¹³C values. However, based on the method used, it is not possible to identify which specific taxa or which metabolic processes are responsible for this relationship. We believe that this approach preserves the value of the statistical result obtained while avoiding overinterpretation.
As suggested by the Reviewer, we will also expand the discussion to include the need for methods with higher taxonomic resolution in future studies: “Sequencing of marker genes, such as 16S rRNA, would allow a more accurate characterisation of the taxonomic composition of communities, while a shotgun metagenomic approach would additionally enable the determination of the microbiome’s genetic and functional potential. Modern genomic taxonomy databases, such as the Genome Taxonomy Database (GTDB), base bacterial classification primarily on phylogenomics using sets of conserved protein markers, which further illustrates how significantly modern classification differs from classical schemes based mainly on rRNA markers.”
As a result, we will conclude that the observed relationship between “other bacteria” and δ¹³C should be treated as an interesting ecological signal and a starting point for further research, rather than as an unambiguously identified mechanism. We thank the Reviewer for drawing attention to this aspect, as it has allowed us to clarify the scope of our interpretation of the results and better relate them to modern bacterial taxonomy.
- FISH
- I can’t figure out the exact sequences of the different probes that the authors used. I would strongly recommend that they have a table in the supplemental with the sequence. Based on the citations, I’m currently concerned that they are using some more outdated FISH probes that may either miss legitimate members of their target group, or hybridize to organisms that are now classified elsewhere. The citations are also confusing because they cite the Lücker et al. 2007 paper, but that is a methods paper that tests a variety of probes – just citing it does not tell me which specific one they used. In addition, the authors used NON338 to evaluate nonspecific fluorescent binding – this is good, as it detects fluorescence caused by nonspecific binding, sticking, etc. But it does not indicate that the taxon-specific probes have complete coverage or perfect phylogenetic specificity. Thus, “less than 5% nonspecific staining” addresses the analytical problem of just using the probes, but not the broader taxonomic problem. This should be discussed as a limitation of their study in the Discussion.
Response: We thank the reviewer for this comment. We agree, that detailed information regarding the FISH probes used should be presented in a manner that allows for the unambiguous identification of each oligonucleotide and an assessment of its current specificity and taxonomic coverage.
Following the reviewer's suggestion, we will add the supplementary materials with a table containing, for each probe used: the probe name, the full nucleotide sequence in the 5′–3′ direction, the target phylogenetic group, the target rRNA region, the hybridization conditions used, and the reference where the probe was described or validated. This addition should enable full identification of the probe set used and facilitate an independent assessment of its usefulness.
We also agree, that the mere reference to Lücker et al. (2007) was insufficient and could be misleading. This publication concerned the development and evaluation of a set of FISH probes targeting bacteria related to Deltaproteobacteria, including updated information on the specificity of previously published probes and the development of new probes. Therefore, it does not constitute a definitive indication of the specific probe used in this study. In the revised version of the manuscript, we’ll have assigned each probe to its appropriate source and provided its full sequence in the Supplementary Materials table.
We also acknowledge the reviewer's comment regarding the need to distinguish between two aspects of FISH assay quality: (i) nonspecific fluorescent signal and (ii) taxonomic coverage and specificity of the probes used. The inclusion of the NON338 probe, as a negative control, allowed us to assess the level of signal unrelated to specific hybridization of the probe with the target rRNA. Therefore, a nonspecific staining level of less than 5% indicates a low analytical background and serves as a quality control for the detection procedure itself. However, it cannot be interpreted as evidence of full taxonomic coverage or full specificity of the target probes. As per the reviewer's comment, we’ll therefore clarify the interpretation of this result in the manuscript. Consistent with previous work on FISH, broad phylogenetic probes may not detect some organisms belonging to the nominal target group if they possess appropriate differences in the probe binding site sequence, and at the same time, they may exhibit cross-reactivity with organisms outside the target group. Consequently, the absence of a FISH signal does not necessarily indicate the absence of a given taxon, and the obtained signal does not always indicate a complete representation of all organisms belonging to the declared group. This is particularly important for the broad phylogenetic groups analyzed in this study. With the rapid increase in the number of available rRNA sequences and changes in bacterial taxonomic classification, the specificity and coverage of older probes may change. The introduction of the probeBase database already highlighted the need to re-validate the specificity of previously published probes as new sequence data becomes available. Newer versions of probeBase additionally enable in silico assessment of probe specificity and coverage against updated rRNA sequence databases. Therefore, in the extended Discussion section, we’ll intend to emphasize that the probe set used should be considered a tool for quantitatively assessing selected broad phylogenetic groups, not as a complete characterization of the entire bacterial community. Therefore, FISH results may be subject to both underestimation of the abundance of certain groups due to incomplete probe coverage and potential detection of organisms outside the nominal target group due to imperfect specificity. However, this does not mean that the approach used is without value. FISH remains a well-established method enabling the direct visualization and quantification of microbial cells in environmental samples. In this study, this method was primarily used to compare the relative proportions of broad bacterial groups between the studied water samples, and the obtained results should be interpreted within the resolution provided by the probe set used. We also purpose to introduce appropriate limitations in the interpretation of statistical results. The statement of "less than 5% nonspecific staining" will not be longer present as evidence of the full taxonomic validity of the probes used, but solely as a result of monitoring the level of nonspecific fluorescent signal. At the same time, in the Discussion, we’ll like to clearly indicate, that an up-to-date assessment of the coverage and specificity of the probes used, based on modern sequence databases, as well as the use of high-throughput sequencing methods, could in the future provide a more comprehensive characterization of the analyzed bacterial communities.
We will include following description in the discussion: ‘An important limitation of the FISH approach is that the specificity and taxonomic coverage of the probes may not fully reflect the current diversity and classification of the targeted bacterial groups. Although the use of the NON338 probe indicated a low level of nonspecific fluorescent signal, this control does not assess the phylogenetic coverage or specificity of the taxon-specific probes. Consequently, some members of the targeted groups may have remained undetected, while potential cross-hybridization with organisms outside the nominal target groups cannot be completely excluded. Therefore, the FISH results should be interpreted as estimates of the abundance of the targeted broad phylogenetic groups rather than as a comprehensive characterization of the bacterial community.’
- Data Availability
- I strongly disagree with the authors statement that “The dataset used within this study will be made available on a reasonable request,” as it is a common convention in the field to have data available in a supplement or publicly available dataset upon submission. As noted above, I couldn’t even figure out what sequences the authors used for their FISH. For example, here is a recent preprint in Biogeosciences where the authors give 18S tables in their supplement (I know this is slightly different, but just wanted to give a recent example): https://egusphere.copernicus.org/preprints/2026/egusphere-2026-4192/
- Also, the authors should give their raw, measured values for environmental parameters (like pH, temperature, ammonium, etc.) as well as carbon isotope data. Not just the averaged data (e.g., Table 1). Here is a recent example from Biogeosciences, where they have made this data available in their supplemental: https://bg.copernicus.org/articles/19/1/2022/
- I also think this may violate the journal’s own policies. Although the authors technically have a statement on how their data can be accessed, in line with the journals request, the journal also states: “If the data are not publicly accessible at the time of final publication, the data statement should describe where and when they will appear, and provide information on how readers can obtain the data until then. Nevertheless, authors should make such embargoed data available to reviewers during the review process in order to foster reproducibility. The Copernicus review system allows to define such assets as 'access limited to reviewers' and reviewers must then sign that they will use such data only for the purpose of reviewing without making copies, sharing, or reusing. In rare cases where the data cannot be deposited publicly (e.g., because of commercial constraints), a detailed explanation of why this is the case is required. The data needed to replicate figures in a paper should in any case be publicly available, either in a public database (strongly recommended), or in a supplement to the paper.”
- https://www.biogeosciences.net/policies/data_policy.html#data_availability
Response: Thank you for raising this issue. We agree that the original Data Availability statement was insufficient and that the underlying data should be made publicly accessible to facilitate transparency and reproducibility. Together with the revised version of the manuscript, we will make the dataset publicly available. The dataset will include all of the variables considered in our study (δ¹³C values, environmental variables, bacteria abundances). Additionally, as stated above, we will provide details on FISH probes. Accordingly, we will revise the Data Availability statement in the manuscript to provide the location and access information for the dataset. For the review purposes, both files can be accessed here: https://github.com/wmkrzt/Environ.Bacteria.d13C
- Figures
- The figures are difficult to interpret. As a starter, the text on the figures is far too small and difficult to read. But, more importantly, there are no figures showing the actual data measured in this study besides Figure 1. Figures 2-4 are essentially a graphical rendering of Table 3, with model-based uncertainty bands. It basically does not allow the reader, or reviewer, to judge for themselves if the model is indeed a good fit for the measured data. Also – For Figure 4, the phrase “stepwise-selected significant predictors” is confusing because the manuscript states that the full Daphnia model was retained, while Figure 4 apparently displays only predictors with P<0.05. In addition, only 12 Daphnia samples were measured and the fitted model for Daphnia contains 11 predictors (day of the year, proteobacteria, alpha proteobacteria, etc.), the authors should address potential model overfitting – this could also be addressed by just showing the actual measured data. Again, since the data are not made available, it is difficult to really independently evaluate this manuscript.
Response: If the Editor decision is further proceeding of our manuscript and we will be given the chance of making improvements, we will work on figures readability in revised version of the manuscript. Axes labels will be made more visible, and we will add data points to each predictor plot. Finally, we will make results description within the manuscript concise with figures captions.
We would also like to clarify an error in the sample sizes reported in the previous version of the manuscript. The numbers of samples reported previously (36 calanoid copepod, 29 cyclopoid copepod, and 12 Daphnia samples) were inadvertently retained from an earlier version of the dataset used for a different set of analyses. The isotope dataset used for the analyses presented in the current manuscript actually contains 48 calanoid copepod, 49 cyclopoid copepod, and 27 Daphnia samples. Out of this dataset, stepwise-selected models included slightly lower number of fitted observations (45 calanoid copepod samples, 45 cyclopoid copepod samples and 26 Daphnia samples). We apologize for this error and will carefully check and correct the sample sizes throughout the manuscript. Regarding the potential for overfitting of the Daphnia model, we acknowledge that the relatively small sample size (n = 26) in relation to the number of predictors included in the model warrants caution. The model was developed as an exploratory analysis of potential environmental and biological drivers of Daphnia stable isotope variability, rather than as a predictive model. We will add an explicit statement in the Discussion, acknowledging the possibility of model overfitting: “Considering the sample size fitted into a model for Daphnia (N = 26) and relatively large number of stepwise-selected environmental predictors (11), presented results for relationships of δ¹³C in this group should be interpreted with appropriate caution”.
- Minor comments
- Line 19: You should briefly define what d13C is – for example, “densities of various groups of bacteria and the carbon isotope composition (d13C) of calanoid copepods”
- Line 39: Do you mean ‘reviewed by’ instead of ‘revised by’?
- Line 90: Typo, should be “depths” not “depts”
- Supplemental: This looks like raw outputs from R. The table caption should define what the column headers are – e.g., “Df; Degrees of freedom.
Response: All of these comments will be used for corrections of the revised version of the manuscript
References:
De Meester, L., Declerck, S. A., & Ger, K. A. (2023). Beyond Daphnia: a plea for a more inclusive and unifying approach to freshwater zooplankton ecology. Hydrobiologia, 850(21), 4693-4703.
Huỳnh, T. H., Horváth, Z., Pálffy, K., Kardos, V., Szabó, B., Dobosy, P., & Vad, C. F. (2024). Heatwave‐induced functional shifts in zooplankton communities result in weaker top‐down control on phytoplankton. Ecology and Evolution, 14(8), e70096.
Kratina, P., Greig, H. S., Thompson, P. L., Carvalho-Pereira, T. S. A., & Shurin, J. B. (2012). Warming modifies trophic cascades and eutrophication in experimental freshwater communities. Ecology, 93(6), 1421–1430. https://doi.org/10.1890/11-1595.1
Kubera, Ł., Kalwasińska, A., & Perliński, P. (2025). Assessment of the Structure and Activity of Bacterial Communities as a Tandem Tool for Estimating the Ecological Condition of Lakes in Poland: Assessment of the Structure and Activity of Bacterial Communities. Ecosystems, 28(3), 32.
Kunzmann, A. J., Ehret, H., Yohannes, E., Straile, D., & Rothhaupt, K. O. (2019). Calanoid copepod grazing affects plankton size structure and composition in a deep, large lake. Journal of Plankton Research, 41(6), 955-966.
Landa, M., Blain, S., Christaki, U., Monchy, S., & Obernosterer, I. (2016). Shifts in bacterial community composition associated with increased carbon cycling in a mosaic of phytoplankton blooms. The ISME journal, 10(1), 39-50.
Rabow, S., Cai, S., Carlsson, P., & Rengefors, K. (2026). Earlier spring may explain dominance of the slow‐growing dinoflagellate Naiadinium polonicum over the cyanobacterium Microcystis during a heatwave. Journal of Phycology.
Citation: https://doi.org/10.5194/egusphere-2026-3444-AC2
-
AC2: 'Reply on RC2', Wojciech Krztoń, 14 Sep 2026
Status: closed
-
RC1: 'Comment on egusphere-2026-3444', Anonymous Referee #1, 17 Jul 2026
General comments:
The manuscript presents a valuable and timely contribution to research on carbon cycling in shallow freshwater ecosystems. The authors address the important issue of identifying the environmental and microbial factors shaping carbon isotope (δ¹³C) variability in the major groups of crustacean zooplankton. A particular strength of the study is the integration of stable isotope analyses with bacterial community characterization and environmental parameters, providing insights into pelagic ecosystem functioning from both ecological and biogeochemical perspectives. The study is based on an extensive field dataset collected seasonally from several shallow eutrophic water bodies. The analytical methods and statistical models are appropriate for addressing the stated research objectives, and the results are presented in a clear and logical manner. The Discussion deserves particular recognition for placing the findings within the context of current knowledge and interpreting them in light of the functional differences among the investigated zooplankton groups. I also appreciate that the authors openly acknowledge the main limitation of the study, namely the lack of δ¹³C measurements of potential food sources. Recognizing this limitation and identifying directions for future research strengthens the credibility of the conclusions and highlights promising avenues for further investigation.
Overall, I believe that the manuscript provides novel insights into the functioning of freshwater food webs and is suitable for publication after minor revisions.
Specific comments:
- In the Introduction, the authors hypothesize that the relationship between δ¹³C values and bacterial abundance would be observed only in Daphnia, as copepods are not expected to feed directly on bacteria. However, the results indicate that bacterial-related variables ("Other bacteria" and, marginally, β-proteobacteria) were also retained in the final model for calanoid copepods. It would be worthwhile to discuss this observation in greater detail and propose possible explanations, for example by considering indirect effects of bacteria mediated through the food web or through changes in the quality of phytoplankton food resources. Such an addition would help explain why the obtained results partially differ from the original hypothesis.
- The study focuses on the abundance of individual bacterial groups; however, it is worth noting that abundance does not necessarily reflect metabolic activity. Carbon cycling processes may be driven by the activity of only a small fraction of the bacterial community. A brief discussion of this limitation would provide a more balanced interpretation of the results.
- The analysis of bacterial communities represents one of the major strengths of the study and provides valuable information for interpreting the observed patterns. Nevertheless, it would be useful to discuss in greater detail the limitations of the FISH method, which enables the quantitative assessment of broad phylogenetic groups but does not provide information on species composition or the metabolic potential of individual taxa. This is particularly relevant for the "Other bacteria" group, which emerged as an important predictor in the statistical models while encompassing a highly diverse assemblage of microorganisms with different ecological functions. A brief discussion of this limitation would strengthen the manuscript. Furthermore, it could be noted that only high-throughput sequencing approaches, such as metagenomics or functional metagenomics, would provide a more comprehensive characterization of bacterial communities and enable the identification of specific taxa and metabolic pathways responsible for the observed relationships with δ¹³C values. Such a statement would not diminish the value of the present study but rather highlight an interesting direction for future research.
- The explanation of the negative effects of NH₄⁺ and NO₃⁻ on δ¹³C values is convincing and well supported by the literature. However, it would be beneficial to emphasize more explicitly that these relationships are correlative and do not necessarily imply direct cause-and-effect relationships. Such clarification would make the interpretation more balanced.
- The number of isotope samples available for Daphnia is considerably smaller than for the two copepod groups. This is understandable given the field conditions; however, it would be useful to briefly acknowledge in the Discussion that the smaller sample size may have reduced the statistical power of the analyses for this taxon.
Technical comments:
- Line 17: "important part of carbon cycling in in shallow..." – remove the duplicated word "in".
- Line 28: "Thanks to integration..." → "Thanks to the integration..."
- Throughout the manuscript, please ensure consistency in the formatting of units (e.g., "10 mL" instead of "10mL").
Citation: https://doi.org/10.5194/egusphere-2026-3444-RC1 -
AC1: 'Reply on RC1', Wojciech Krztoń, 14 Sep 2026
Reviewer 1:
General comments:
The manuscript presents a valuable and timely contribution to research on carbon cycling in shallow freshwater ecosystems. The authors address the important issue of identifying the environmental and microbial factors shaping carbon isotope (δ¹³C) variability in the major groups of crustacean zooplankton. A particular strength of the study is the integration of stable isotope analyses with bacterial community characterization and environmental parameters, providing insights into pelagic ecosystem functioning from both ecological and biogeochemical perspectives. The study is based on an extensive field dataset collected seasonally from several shallow eutrophic water bodies. The analytical methods and statistical models are appropriate for addressing the stated research objectives, and the results are presented in a clear and logical manner. The Discussion deserves particular recognition for placing the findings within the context of current knowledge and interpreting them in light of the functional differences among the investigated zooplankton groups. I also appreciate that the authors openly acknowledge the main limitation of the study, namely the lack of δ¹³C measurements of potential food sources. Recognizing this limitation and identifying directions for future research strengthens the credibility of the conclusions and highlights promising avenues for further investigation.
Overall, I believe that the manuscript provides novel insights into the functioning of freshwater food webs and is suitable for publication after minor revisions.
Response: We sincerely thank Reviewer 1 for the positive and constructive evaluation of our manuscript. We greatly appreciate the reviewer’s recognition of the value of integrating stable isotope analyses with environmental and microbial data, as well as the strengths of our field dataset and ecological interpretation. Please find responses to all of the comments below. We believe, that revisions that we will make will improve the presentation and interpretation of our results.
Specific comments:
- In the Introduction, the authors hypothesize that the relationship between δ¹³C values and bacterial abundance would be observed only in Daphnia, as copepods are not expected to feed directly on bacteria. However, the results indicate that bacterial-related variables ("Other bacteria" and, marginally, β-proteobacteria) were also retained in the final model for calanoid copepods. It would be worthwhile to discuss this observation in greater detail and propose possible explanations, for example by considering indirect effects of bacteria mediated through the food web or through changes in the quality of phytoplankton food resources. Such an addition would help explain why the obtained results partially differ from the original hypothesis.
Response: Thank you for this comment. We propose following explanation in the “Discussion”: “Calanoid copepods are able to feed predatory on ciliates, which was demonstrated e.g. by Kunzmann et al. (2019). Additionally, Kunzman et al. (2019) observed increases in bacteria abundance, which was most likely a cascading effect of ciliates removal by the predatory copepod. Although, calanoid copepods cannot feed directly on the bacteria, therefore positive effect of ‘Other bacteria’ on δ¹³C may indicate indirect effects. On the other hand, given findings of Kunzmann et al. (2019) this may be just a circumstantial correlation, suggesting increases in ‘Other bacteria’ abundance as an effect of grazing by the calanoid copepods”.
- The study focuses on the abundance of individual bacterial groups; however, it is worth noting that abundance does not necessarily reflect metabolic activity. Carbon cycling processes may be driven by the activity of only a small fraction of the bacterial community. A brief discussion of this limitation would provide a more balanced interpretation of the results.
Response: Thank you for pointing this out. We will provide a better explanation in the discussion: “Biogeochemical cycle of carbon are regulated by microbes (Landa et al. 2016). However, the efficiency of carbon transfer may depend on small fraction of the bacterial community that is metabolically active. It is worth to point out that this fraction can be different from total abundance of bacteria. Moreover, the activity may vary throughout the vegetation season (Kubera et al. 2025)”
- The analysis of bacterial communities represents one of the major strengths of the study and provides valuable information for interpreting the observed patterns. Nevertheless, it would be useful to discuss in greater detail the limitations of the FISH method, which enables the quantitative assessment of broad phylogenetic groups but does not provide information on species composition or the metabolic potential of individual taxa. This is particularly relevant for the "Other bacteria" group, which emerged as an important predictor in the statistical models while encompassing a highly diverse assemblage of microorganisms with different ecological functions. A brief discussion of this limitation would strengthen the manuscript. Furthermore, it could be noted that only high-throughput sequencing approaches, such as metagenomics or functional metagenomics, would provide a more comprehensive characterization of bacterial communities and enable the identification of specific taxa and metabolic pathways responsible for the observed relationships with δ¹³C values. Such a statement would not diminish the value of the present study but rather highlight an interesting direction for future research.
Response: We thank the reviewer for this valuable comment. We agree that a more detailed discussion of the methodological limitations of the FISH technique will improve the quality of the manuscript and provide a more adequate interpretation of the data regarding the bacterial community. We propose following clarification, in the revised version of the manuscript: “FISH is an effective method that allows for the direct visualization, counting, and spatial localization of microorganisms in their natural environment. Importantly, FISH probes can be designed to target phylogenetic groups, ranging from broad taxonomic levels down to the genus or species level under appropriate conditions. However, the taxonomic resolution of this method is inherently dependent on the specificity and range of the applied probes. In this study, the FISH method was primarily used to quantitatively determine broad phylogenetic groups; therefore, the obtained data should not be interpreted as a detailed characterization of the bacterial community at the species level.
FISH method provides information based on the detection of ribosomal RNA and thus does not directly reveal the genomic or metabolic potential of the individual detected taxa. Quantitative signals may also be influenced by differences in the range and specificity of probes, cell permeability, ribosomal RNA content, and the physiological state of the microorganisms. Consequently, differences in abundance determined by the FISH method between phylogenetic groups should be interpreted as differences in abundance/detectability of the studied groups, rather than as direct evidence of differences in their metabolic activity or ecological functions. These remarks are particularly important regarding the category "Other bacteria," which represents a heterogeneous residual group encompassing microorganisms that may differ significantly in terms of taxonomy, physiology, and ecological role. Therefore, although the "Other bacteria" group proved to be an important predictor in statistical models, this result should be interpreted at the level of the aggregated bacterial category detected using the applied FISH method. Unfortunately, the current data do not allow us to determine which specific taxa within this group contributed to the observed relationship with δ¹³C values, nor to establish what metabolic processes may underlie this relationship.
The application of high-throughput sequencing methods, such as 16SrRNA gene amplicon sequencing or shotgun metagenomics, could provide a much more comprehensive characterization of the structure and functional potential of the studied bacterial communities. High-throughput sequencing and integrated multi-omic approaches as an important direction for future research. Such approaches could help establish the taxonomic composition of the "Other bacteria" category and identify microbial genes and metabolic pathways potentially related to the observed variability in δ¹³C values.”.
We believe that the explanation will moderate the original formulations regarding functional interpretation in accordance with the obtained research results. The above explanation will strengthen, rather than diminish, the value of this study by defining the scope of conclusions that can be drawn from an assessment based on the FISH technique.
- The explanation of the negative effects of NH₄⁺ and NO₃⁻ on δ¹³C values is convincing and well supported by the literature. However, it would be beneficial to emphasize more explicitly that these relationships are correlative and do not necessarily imply direct cause-and-effect relationships. Such clarification would make the interpretation more balanced.
Response: We will include the following explanation in the Discussion: “The study was carried out as the fieldwork-based monitoring, therefore we can discuss only correlative effects”. Thank you for pointing this out.
- The number of isotope samples available for Daphnia is considerably smaller than for the two copepod groups. This is understandable given the field conditions; however, it would be useful to briefly acknowledge in the Discussion that the smaller sample size may have reduced the statistical power of the analyses for this taxon.
Response: We thank the reviewer for pointing this out. We apologize for the confusion regarding the sample sizes reported in the manuscript. The sample sizes stated in the previous version of the manuscript (36 calanoid copepod, 29 cyclopoid copepod, and 12 Daphnia samples) were inadvertently retained from an earlier version of the dataset used for a different set of analyses. In fact, the isotope dataset used for the analyses presented in the manuscript contains 45 calanoid copepod, 45 cyclopoid copepod, and 26 Daphnia samples. From this dataset, stepwise-selected models included slightly lower number of fitted observations (45 calanoid copepod samples, 45 cyclopoid copepod samples and 26 Daphnia samples). Thus, the sample sizes reported in the previous version of the manuscript were incorrect. We will correct this information in the revised manuscript.
We have carefully checked the dataset and revised the manuscript to report the correct sample sizes throughout. We also agree with the reviewer that the number of Daphnia samples is lower than that of the two copepod groups and therefore may provide lower statistical power for analyses involving Daphnia. However, the difference is considerably smaller than suggested by the sample sizes reported in the previous version (26 Daphnia samples compared with 45 samples for each copepod group). We will nevertheless add a following statement to the Discussion: “Considering the sample size for Daphnia (N = 26) and relatively large number of stepwise-selected environmental predictors (11), presented results for relationships of δ¹³C in this group should be interpreted with appropriate caution”.
Technical comments:
- Line 17: "important part of carbon cycling in in shallow..." – remove the duplicated word "in".
- Line 28: "Thanks to integration..." → "Thanks to the integration..."
- Throughout the manuscript, please ensure consistency in the formatting of units (e.g., "10 mL" instead of "10mL").
Response: Thank you. We will take these comments into account while revising the manuscript.
Citation: https://doi.org/10.5194/egusphere-2026-3444-AC1
-
RC2: 'Comment on egusphere-2026-3444', Anonymous Referee #2, 21 Jul 2026
Reviewer comments for Krzton et al. 2026, submitted to Biogeosciences
Krzton et al. attempt to understand the parameters affecting the carbon isotope composition (δ13C) of planktonic crustaceans. If I understand their introduction correctly, they are motivated by the larger topic of climate change impacting food webs and carbon flow within freshwater ecosystems. By correlating a variety of environmental parameters (e.g., pH, temperature, ammonium and nitrate, abundance of bacterial groups, etc.) with the δ13C of planktonic crustaceans using generalized linear models, they attempt to derive a mechanistic understanding of planktonic δ13C. This, in turn, would allow their results to be applied to freshwater ecosystems more broadly and not just their study sites. This is a topic of interest to many groups, as the δ13C of planktonic crustaceans (as well as other freshwater macro- and microorganisms) have been measured to disentangle the complex food webs of marine and freshwater ecosystems, and assess potential impacts of climate change. Therefore, the manuscript is of interest in that regard.
However, I have significant concerns about the manuscript as it is currently presented. My first concern regards readability; at times, it was difficult for me to understand what exactly the authors were trying to say. My second concern regards data availability. I was not able to fully independently evaluate the manuscript as the raw data measured by the authors was never presented, either in the main text or as a supplementary table. Instead, the authors only present averages of their data, or model outputs. For example, Figures 2-4 are just graphical renderings of Table 3, which are their modeled outputs. They also do not give the sequences used for their FISH probes. For the probes, this is extremely important as some FISH probes are outdated and will incorrectly classify certain bacteria in certain clades, largely because our knowledge of bacterial taxonomy & classification is rapidly changing as more genome-based approaches are used, and as sequence data repositories steadily grow. Instead, the authors state that “The dataset used within this study will be made available on a reasonable request” – I do not think this is sufficient (is it reasonable to expect a reviewer to ask for your data, when you send it in for review?) and wonder if it even violates the journal’s own policies regarding data availability (see more detail below). I also note that this is to the detriment to the authors – fewer people will cite your paper if the data is not readily available and they have to contact you (e.g. go through the trouble of finding your current institution, finding updated contact information, waiting for your response, potentially negotiating with you for what is a ‘reasonable’ request…). More often than not, people may just find a similar study and cite that instead.
Given all that, I had to take the authors at their word regarding what they found. In this regard, it does not appear that they can draw strong conclusions. They do not find significant differences in δ13C among the three planktonic crustacean groups they examined (Supplemental Table S2). Perhaps not surprisingly, they can therefore only conclude that there is a “complex interplay of environmental conditions, microbial processes and functional trait-driven response of δ13C values in planktonic crustaceans” (Line 315). They also note, correctly, that one of the weaknesses of their paper is that they don’t measure the δ13C of the actual dietary sources – this would be a much stronger paper if they had instead waited to publish after getting that data. Instead, this paper relies on correlation of many parameters with few data points. For example, only 12 Daphnia samples were measured for δ13C and the fitted model for Daphnia contains 11 predictors – the potential for model overfitting, and potential over-interpretation, is extremely high. And, as I noted above, since they do not include any of their actual measured data… I cannot independently confirm whether their data is indeed overfitted or not.
Therefore, I recommend that the manuscript should be rejected.
Major comments
Issues with manuscript readability
I would strongly recommend that the authors work with the editor or other English speaker to edit the writing of the final manuscript. Alternatively, it appears that Biogeosciences allows use of AI tools in manuscript writing, as long as the usage is either described in the Methods section or Acknowledgments: https://www.biogeosciences.net/submission.html
My concern is twofold. First, there are numerous small errors – e.g. grammar, typos – throughout the manuscript that need to be fixed. For example, just in the abstract: i) in the first sentence of the abstract, the word “in” appears twice; ii) “Day” and “Other” are capitalized in the abstract in Line 21, but should not be; iii) The last sentence of the abstract is missing a period at the end. These small errors make it difficult to read the manuscript smoothly. There are also numerous articles missing – for example, Line 46 should say “are an important part of…” instead of “are important part of…”
I note, also, that the quality of writing noticeably changes throughout the manuscript. For example, the writing is much clearer in some sections, like “2.5 Statical analyses” – there are few missing articles (e.g., “the”). But then the writing changes right afterward, in section “3.1. Environmental variables” – for example, the grammar is slightly off in Line 153 articles and should instead be written as “Water temperatures in the studied lakes ranged from 10.8 to 25.6 C, reaching the highest values in July and August.” So, I would recommend whoever helped write Section 2.5 to help edit the entire manuscript.
Second, it was difficult for me to understand the meaning of some sentences. For example, when describing the results (Line 185):
“Mean values of δ13C were found in calanoid copepods accounted for -31.8‰ with range between -41.5‰ to -23.7. Mean δ13C values of cyclopoid copepods was -30.6‰ and range: -40‰ to -22.2‰). δ13C measurements of Daphnia sp. showed mean value of -31.2‰ and range between -42.5‰ and –24.4‰)”
There are typos, like the extra right parentheses (this: “)”) that show up. I am also confused by “accounted for” in the first sentence. Are you just trying to say that the mean δ13C value of calanoid copepods was -31.8‰ with a range from -41.5 to -23.7‰?
Discussion
Some additional information on the particulars of the author’s study sites, compared to prior studies, would increase the quality of the manuscript. For example, the authors find that the only factor that positively affected the δ13C of cyclopoid copepods was water temperature (Line 237) but was not a significant predictor for Daphnia sp. δ13C. What’s the seasonal temperature range that these water bodies experience in a year? Do these water bodies freeze over in the wintertime? Would this at all lead you to predict the differing responses of temperature for cyclopoid copepods vs. Daphnia?
In addition, as these water bodies are quite close to Cracow, I would expect high amounts of anthropogenic influence – would this affect your interpretations? For example, the authors counterintuitively find that higher concentrations of ammonium and nitrate to be correlated with slower phytoplankton growth rates – as they note, this is “opposite to what would be expected… but seasonal dynamics and declining water temperatures effects might be involved.” However, the classic understanding of phytoplankton growth rates increasing with reactive nitrogen availability is based on the assumption that these nutrients are typically rare, like in the open oceans. For shallow, freshwater and even artificial lakes – like the ones studied here – would reactive nitrogen species just be high all the time? Would the authors be able to provide additional data on the chemistry of these lakes?
In Line 288, the authors discuss the positive effect of “Other bacteria” on δ13C values. I’m not sure what exactly the authors can do to make this part stronger, but I found it to be speculative as the measurement of community diversity the authors used – alpha vs. beta vs. gamma vs. delta protobacteria – is quite coarse and does not reflect the current genome-based taxonomy used to classify bacteria today. For example, using marker genes beyond rRNA and those targeted in FISH, it has been proposed that delta proteobacteria be reclassified into clades that better reflect the diverse range of chemistry that different subclades can carry out, and which are masked by the broad paintbrush of ‘delta proteobacteria’ ( https://doi.org/10.1099/ijsem.0.004213).
Therefore, inferring mechanism from these coarse groups of bacteria is inherently speculative because each species of bacteria within these groups can perform a wide range of chemistry. As the authors note themselves, methane oxidation can be accomplished by bacteria from the beta, alpha, and gamma clades. Unless the authors can determine this at a finer resolution – i.e., they detect a phylum like Cyanobacteriota – I would suggest they walk back some of their claims. (I also note that, even knowing there is Cyanobacteria present, most studies do not speculate on mechanism as Cyanobacteria themselves can carry out extremely diverse reactions…)
FISH
I can’t figure out the exact sequences of the different probes that the authors used. I would strongly recommend that they have a table in the supplemental with the sequence. Based on the citations, I’m currently concerned that they are using some more outdated FISH probes that may either miss legitimate members of their target group, or hybridize to organisms that are now classified elsewhere. The citations are also confusing because they cite the Lücker et al. 2007 paper, but that is a methods paper that tests a variety of probes – just citing it does not tell me which specific one they used. In addition, the authors used NON338 to evaluate nonspecific fluorescent binding – this is good, as it detects fluorescence caused by nonspecific binding, sticking, etc. But it does not indicate that the taxon-specific probes have complete coverage or perfect phylogenetic specificity. Thus, “less than 5% nonspecific staining” addresses the analytical problem of just using the probes, but not the broader taxonomic problem. This should be discussed as a limitation of their study in the Discussion.
Data Availability
I strongly disagree with the authors statement that “The dataset used within this study will be made available on a reasonable request,” as it is a common convention in the field to have data available in a supplement or publicly available dataset upon submission. As noted above, I couldn’t even figure out what sequences the authors used for their FISH. For example, here is a recent preprint in Biogeosciences where the authors give 18S tables in their supplement (I know this is slightly different, but just wanted to give a recent example):
https://egusphere.copernicus.org/preprints/2026/egusphere-2026-4192/
Also, the authors should give their raw, measured values for environmental parameters (like pH, temperature, ammonium, etc.) as well as carbon isotope data. Not just the averaged data (e.g., Table 1). Here is a recent example from Biogeosciences, where they have made this data available in their supplemental:
https://bg.copernicus.org/articles/19/1/2022/
I also think this may violate the journal’s own policies. Although the authors technically have a statement on how their data can be accessed, in line with the journals request, the journal also states: “If the data are not publicly accessible at the time of final publication, the data statement should describe where and when they will appear, and provide information on how readers can obtain the data until then. Nevertheless, authors should make such embargoed data available to reviewers during the review process in order to foster reproducibility. The Copernicus review system allows to define such assets as 'access limited to reviewers' and reviewers must then sign that they will use such data only for the purpose of reviewing without making copies, sharing, or reusing. In rare cases where the data cannot be deposited publicly (e.g., because of commercial constraints), a detailed explanation of why this is the case is required. The data needed to replicate figures in a paper should in any case be publicly available, either in a public database (strongly recommended), or in a supplement to the paper.”
https://www.biogeosciences.net/policies/data_policy.html#data_availability
Figures
The figures are difficult to interpret. As a starter, the text on the figures is far too small and difficult to read. But, more importantly, there are no figures showing the actual data measured in this study besides Figure 1. Figures 2-4 are essentially a graphical rendering of Table 3, with model-based uncertainty bands. It basically does not allow the reader, or reviewer, to judge for themselves if the model is indeed a good fit for the measured data. Also – For Figure 4, the phrase “stepwise-selected significant predictors” is confusing because the manuscript states that the full Daphnia model was retained, while Figure 4 apparently displays only predictors with P<0.05. In addition, only 12 Daphnia samples were measured and the fitted model for Daphnia contains 11 predictors (day of the year, proteobacteria, alpha proteobacteria, etc.), the authors should address potential model overfitting – this could also be addressed by just showing the actual measured data. Again, since the data are not made available, it is difficult to really independently evaluate this manuscript.
Minor comments
Line 19: You should briefly define what d13C is – for example, “densities of various groups of bacteria and the carbon isotope composition (d13C) of calanoid copepods”
Line 39: Do you mean ‘reviewed by’ instead of ‘revised by’?
Line 90: Typo, should be “depths” not “depts”
Supplemental: This looks like raw outputs from R. The table caption should define what the column headers are – e.g., “Df; Degrees of freedom.”
Citation: https://doi.org/10.5194/egusphere-2026-3444-RC2 -
AC2: 'Reply on RC2', Wojciech Krztoń, 14 Sep 2026
Reviewer 2
Reviewer comments for Krzton et al. 2026, submitted to Biogeosciences
Krzton et al. attempt to understand the parameters affecting the carbon isotope composition (δ13C) of planktonic crustaceans. If I understand their introduction correctly, they are motivated by the larger topic of climate change impacting food webs and carbon flow within freshwater ecosystems. By correlating a variety of environmental parameters (e.g., pH, temperature, ammonium and nitrate, abundance of bacterial groups, etc.) with the δ13C of planktonic crustaceans using generalized linear models, they attempt to derive a mechanistic understanding of planktonic δ13C. This, in turn, would allow their results to be applied to freshwater ecosystems more broadly and not just their study sites. This is a topic of interest to many groups, as the δ13C of planktonic crustaceans (as well as other freshwater macro- and microorganisms) have been measured to disentangle the complex food webs of marine and freshwater ecosystems, and assess potential impacts of climate change. Therefore, the manuscript is of interest in that regard.
However, I have significant concerns about the manuscript as it is currently presented. My first concern regards readability; at times, it was difficult for me to understand what exactly the authors were trying to say. My second concern regards data availability. I was not able to fully independently evaluate the manuscript as the raw data measured by the authors was never presented, either in the main text or as a supplementary table. Instead, the authors only present averages of their data, or model outputs. For example, Figures 2-4 are just graphical renderings of Table 3, which are their modeled outputs. They also do not give the sequences used for their FISH probes. For the probes, this is extremely important as some FISH probes are outdated and will incorrectly classify certain bacteria in certain clades, largely because our knowledge of bacterial taxonomy & classification is rapidly changing as more genome-based approaches are used, and as sequence data repositories steadily grow. Instead, the authors state that “The dataset used within this study will be made available on a reasonable request” – I do not think this is sufficient (is it reasonable to expect a reviewer to ask for your data, when you send it in for review?) and wonder if it even violates the journal’s own policies regarding data availability (see more detail below). I also note that this is to the detriment to the authors – fewer people will cite your paper if the data is not readily available and they have to contact you (e.g. go through the trouble of finding your current institution, finding updated contact information, waiting for your response, potentially negotiating with you for what is a ‘reasonable’ request…). More often than not, people may just find a similar study and cite that instead.
Given all that, I had to take the authors at their word regarding what they found. In this regard, it does not appear that they can draw strong conclusions. They do not find significant differences in δ13C among the three planktonic crustacean groups they examined (Supplemental Table S2). Perhaps not surprisingly, they can therefore only conclude that there is a “complex interplay of environmental conditions, microbial processes and functional trait-driven response of δ13C values in planktonic crustaceans” (Line 315). They also note, correctly, that one of the weaknesses of their paper is that they don’t measure the δ13C of the actual dietary sources – this would be a much stronger paper if they had instead waited to publish after getting that data. Instead, this paper relies on correlation of many parameters with few data points. For example, only 12 Daphnia samples were measured for δ13C and the fitted model for Daphnia contains 11 predictors – the potential for model overfitting, and potential over-interpretation, is extremely high. And, as I noted above, since they do not include any of their actual measured data… I cannot independently confirm whether their data is indeed overfitted or not.
Therefore, I recommend that the manuscript should be rejected.
Response: We sincerely thank Reviewer 2 for the careful and constructive assessment of our manuscript. We appreciate the reviewer’s concerns regarding the readability of the manuscript, data availability, interpretation of the statistical analyses, and the presentation of the underlying data. We would also like to clarify that the sample sizes reported in the previous version were inadvertently carried over from an earlier dataset used for different analyses; the correct sample sizes for the isotope analyses are 48 calanoid copepod, 49 cyclopoid copepod, and 27 Daphnia samples. In revised manuscript we will carefully consider all of these points. We believe that these revisions will substantially improve overall presentation of the manuscript.
Below we include our comments and responses to all of the reviewer’s concerns, which are listed above, and as major and minor comments.
Major comments
Issues with manuscript readability
- I would strongly recommend that the authors work with the editor or other English speaker to edit the writing of the final manuscript. Alternatively, it appears that Biogeosciences allows use of AI tools in manuscript writing, as long as the usage is either described in the Methods section or Acknowledgments: https://www.biogeosciences.net/submission.html
- My concern is twofold. First, there are numerous small errors – e.g. grammar, typos – throughout the manuscript that need to be fixed. For example, just in the abstract: i) in the first sentence of the abstract, the word “in” appears twice; ii) “Day” and “Other” are capitalized in the abstract in Line 21, but should not be; iii) The last sentence of the abstract is missing a period at the end. These small errors make it difficult to read the manuscript smoothly. There are also numerous articles missing – for example, Line 46 should say “are an important part of…” instead of “are important part of…”
- I note, also, that the quality of writing noticeably changes throughout the manuscript. For example, the writing is much clearer in some sections, like “2.5 Statical analyses” – there are few missing articles (e.g., “the”). But then the writing changes right afterward, in section “3.1. Environmental variables” – for example, the grammar is slightly off in Line 153 articles and should instead be written as “Water temperatures in the studied lakes ranged from 10.8 to 25.6 C, reaching the highest values in July and August.” So, I would recommend whoever helped write Section 2.5 to help edit the entire manuscript.
- Second, it was difficult for me to understand the meaning of some sentences. For example, when describing the results (Line 185):
- “Mean values of δ13C were found in calanoid copepods accounted for -31.8‰ with range between -41.5‰ to -23.7. Mean δ13C values of cyclopoid copepods was -30.6‰ and range: -40‰ to -22.2‰). δ13C measurements of Daphnia sp. showed mean value of -31.2‰ and range between -42.5‰ and –24.4‰)”
- There are typos, like the extra right parentheses (this: “)”) that show up. I am also confused by “accounted for” in the first sentence. Are you just trying to say that the mean δ13C value of calanoid copepods was -31.8‰ with a range from -41.5 to -23.7‰?
Response: We thank the reviewer for their assessment of the manuscript. The specific examples provided in the review are helpful for identifying similar issues throughout the text. While preparing the revised version, we will work on English grammar, syntax, punctuation, use of articles, and the overall readability and consistency of the writing style. The final version of the manuscript will undergo professional proofreading by a native English speaker and/or the use of AI tools.
Discussion
- Some additional information on the particulars of the author’s study sites, compared to prior studies, would increase the quality of the manuscript. For example, the authors find that the only factor that positively affected the δ13C of cyclopoid copepods was water temperature (Line 237) but was not a significant predictor for Daphnia sp. δ13C. What’s the seasonal temperature range that these water bodies experience in a year? Do these water bodies freeze over in the wintertime? Would this at all lead you to predict the differing responses of temperature for cyclopoid copepods vs. Daphnia?
Response: Water temperatures in the studied waterbodies during the sampling period (April–October) ranged from 10.8 to 25.6 ℃. Temperature ranges for each waterbody are given in Table 1. The waterbodies freeze over in winter; however, the ice-on period has been observed to decline over the past few decades, which can trigger changes in the functioning of aquatic ecosystems. Milder winters and longer springs can lead to shifts in the species dominating during summer (e.g. a switch from cyanobacteria to dinoflagellates), as recently demonstrated by Rabow et al. (2026). Although this is particularly interesting in the context of climate change, the influence of the dominant algal species on crustacean δ13C was not considered in this manuscript.
In order to address the effects of temperature on δ13C we will provide following clarification:
“Changes in water temperature strongly influence the functioning of aquatic ecosystems, with even slight warming increasing phytoplankton growth rates and zooplankton metabolic demands, and therefore carbon transfer within the food web (Kratina et al. 2012). Rising temperature may affect consumers’ δ13C; however, differences in the responses of daphnia and cyclopoid copepods can be explained by their feeding biology. Daphnia are non-selective filter feeders, consuming mainly small-sized algae and bacteria. Cyclopoid copepods are selective feeders, able to switch between preferred food sources depending on their availability (e.g. De Meester et al. 2023). A recent study by Huỳnh et al. (2026) indicates that at higher temperatures, copepods can switch towards increased predation on rotifers to meet their higher metabolic demands. A similar mechanism may operate in the food webs studied in our manuscript, as we found water temperature to be a significant predictor of cyclopoid copepods’ δ13C. “
- In addition, as these water bodies are quite close to Cracow, I would expect high amounts of anthropogenic influence – would this affect your interpretations? For example, the authors counterintuitively find that higher concentrations of ammonium and nitrate to be correlated with slower phytoplankton growth rates – as they note, this is “opposite to what would be expected… but seasonal dynamics and declining water temperatures effects might be involved.” However, the classic understanding of phytoplankton growth rates increasing with reactive nitrogen availability is based on the assumption that these nutrients are typically rare, like in the open oceans. For shallow, freshwater and even artificial lakes – like the ones studied here – would reactive nitrogen species just be high all the time? Would the authors be able to provide additional data on the chemistry of these lakes?
Response: We thank the reviewer for pointing this out. We agree that proximity to Cracow exposes the studied sites to relatively high anthropogenic pressure; however, at this point we do not have any data from sites free from human activities. Therefore, although interesting, interpretation of these results in the light of the presence or absence of anthropogenic pressure is rather limited. In the revised version of the manuscript we will indicate that these results were obtained from water bodies experiencing high anthropogenic pressure, and that the interpretations require caution.
We would also like to explain that the studied water bodies are small and shallow, and although all are of artificial origin, they are not artificial reservoirs in the sense of dam reservoirs.
We did not analyse reactive nitrogen species (i.e. nitric oxide, nitrogen dioxide radical, peroxynitrite), and it is therefore impossible to consider their influence in this manuscript. In our study systems, nitrate and ammonium were detected during most samplings (summarised in Table 1), suggesting that these shallow water bodies were not consistently depleted in dissolved inorganic nitrogen. All measurements of nitrate and ammonia concentrations will be included in the dataset that we will make available alongside the revised manuscript. (https://github.com/wmkrzt/Environ.Bacteria.d13C).
- In Line 288, the authors discuss the positive effect of “Other bacteria” on δ13C values. I’m not sure what exactly the authors can do to make this part stronger, but I found it to be speculative as the measurement of community diversity the authors used – alpha vs. beta vs. gamma vs. delta protobacteria – is quite coarse and does not reflect the current genome-based taxonomy used to classify bacteria today. For example, using marker genes beyond rRNA and those targeted in FISH, it has been proposed that delta proteobacteria be reclassified into clades that better reflect the diverse range of chemistry that different subclades can carry out, and which are masked by the broad paintbrush of ‘delta proteobacteria’ ( https://doi.org/10.1099/ijsem.0.004213).
- Therefore, inferring mechanism from these coarse groups of bacteria is inherently speculative because each species of bacteria within these groups can perform a wide range of chemistry. As the authors note themselves, methane oxidation can be accomplished by bacteria from the beta, alpha, and gamma clades. Unless the authors can determine this at a finer resolution – i.e., they detect a phylum like Cyanobacteriota – I would suggest they walk back some of their claims. (I also note that, even knowing there is Cyanobacteria present, most studies do not speculate on mechanism as Cyanobacteria themselves can carry out extremely diverse reactions…)
Response: We thank the Reviewer for this valuable comment. We agree that the interpretation of the positive correlation between the proportion of “other bacteria” and δ¹³C values should be presented with greater caution, particularly in light of the limited taxonomic resolution of the FISH method used and the changes that have occurred in bacterial classification based on genomic data. We will add the following explanation in the discussion about FISH: “In this study, FISH probes were used to quantitatively assess broad phylogenetic groups, including Alpha-, Beta-, Gamma- and Delta-proteobacteria. This approach allowed us to determine the relative abundance of major bacterial groups in the samples studied and was particularly useful for comparing community structure across research periods. However, the taxonomic resolution of this approach is limited and does not allow the unambiguous identification of specific genera or species responsible for the observed relationships. FISH is primarily a method that enables the direct identification and quantitative assessment of specific groups of microorganisms, and the scope of the information obtained depends on the specificity of the probes used. However, our results should be interpreted with caution, because using FISH we obtained different microbial groups that do not constitute a coherent monophyletic unit. An example is the group “Deltaproteobacteria”, which may encompass microorganisms with significantly different metabolic potentials and ecological functions. For example, representatives of lineages currently classified within Desulfobacterota include numerous bacteria associated with sulphate reduction and sulphur metabolism, whereas representatives of Myxococcota are characterised by different metabolic and ecological strategies, including predatory behaviour. In turn, Bdellovibrionota includes bacteria with a specialised mode of predation on other bacteria (Waite et al. 2020).”
Consequently, in our revised manuscript, we propose to modify the relevant section and narrow the scope of our interpretation of the results. We will no longer interpret the positive correlation between the proportion of “other bacteria” and δ¹³C values as evidence of a specific metabolic mechanism but rather as an indication that an increase in the proportion of microorganisms belonging to broadly defined groups, which were not detected as the main groups analysed in the study, coincided with changes in δ¹³C values. However, based on the method used, it is not possible to identify which specific taxa or which metabolic processes are responsible for this relationship. We believe that this approach preserves the value of the statistical result obtained while avoiding overinterpretation.
As suggested by the Reviewer, we will also expand the discussion to include the need for methods with higher taxonomic resolution in future studies: “Sequencing of marker genes, such as 16S rRNA, would allow a more accurate characterisation of the taxonomic composition of communities, while a shotgun metagenomic approach would additionally enable the determination of the microbiome’s genetic and functional potential. Modern genomic taxonomy databases, such as the Genome Taxonomy Database (GTDB), base bacterial classification primarily on phylogenomics using sets of conserved protein markers, which further illustrates how significantly modern classification differs from classical schemes based mainly on rRNA markers.”
As a result, we will conclude that the observed relationship between “other bacteria” and δ¹³C should be treated as an interesting ecological signal and a starting point for further research, rather than as an unambiguously identified mechanism. We thank the Reviewer for drawing attention to this aspect, as it has allowed us to clarify the scope of our interpretation of the results and better relate them to modern bacterial taxonomy.
- FISH
- I can’t figure out the exact sequences of the different probes that the authors used. I would strongly recommend that they have a table in the supplemental with the sequence. Based on the citations, I’m currently concerned that they are using some more outdated FISH probes that may either miss legitimate members of their target group, or hybridize to organisms that are now classified elsewhere. The citations are also confusing because they cite the Lücker et al. 2007 paper, but that is a methods paper that tests a variety of probes – just citing it does not tell me which specific one they used. In addition, the authors used NON338 to evaluate nonspecific fluorescent binding – this is good, as it detects fluorescence caused by nonspecific binding, sticking, etc. But it does not indicate that the taxon-specific probes have complete coverage or perfect phylogenetic specificity. Thus, “less than 5% nonspecific staining” addresses the analytical problem of just using the probes, but not the broader taxonomic problem. This should be discussed as a limitation of their study in the Discussion.
Response: We thank the reviewer for this comment. We agree, that detailed information regarding the FISH probes used should be presented in a manner that allows for the unambiguous identification of each oligonucleotide and an assessment of its current specificity and taxonomic coverage.
Following the reviewer's suggestion, we will add the supplementary materials with a table containing, for each probe used: the probe name, the full nucleotide sequence in the 5′–3′ direction, the target phylogenetic group, the target rRNA region, the hybridization conditions used, and the reference where the probe was described or validated. This addition should enable full identification of the probe set used and facilitate an independent assessment of its usefulness.
We also agree, that the mere reference to Lücker et al. (2007) was insufficient and could be misleading. This publication concerned the development and evaluation of a set of FISH probes targeting bacteria related to Deltaproteobacteria, including updated information on the specificity of previously published probes and the development of new probes. Therefore, it does not constitute a definitive indication of the specific probe used in this study. In the revised version of the manuscript, we’ll have assigned each probe to its appropriate source and provided its full sequence in the Supplementary Materials table.
We also acknowledge the reviewer's comment regarding the need to distinguish between two aspects of FISH assay quality: (i) nonspecific fluorescent signal and (ii) taxonomic coverage and specificity of the probes used. The inclusion of the NON338 probe, as a negative control, allowed us to assess the level of signal unrelated to specific hybridization of the probe with the target rRNA. Therefore, a nonspecific staining level of less than 5% indicates a low analytical background and serves as a quality control for the detection procedure itself. However, it cannot be interpreted as evidence of full taxonomic coverage or full specificity of the target probes. As per the reviewer's comment, we’ll therefore clarify the interpretation of this result in the manuscript. Consistent with previous work on FISH, broad phylogenetic probes may not detect some organisms belonging to the nominal target group if they possess appropriate differences in the probe binding site sequence, and at the same time, they may exhibit cross-reactivity with organisms outside the target group. Consequently, the absence of a FISH signal does not necessarily indicate the absence of a given taxon, and the obtained signal does not always indicate a complete representation of all organisms belonging to the declared group. This is particularly important for the broad phylogenetic groups analyzed in this study. With the rapid increase in the number of available rRNA sequences and changes in bacterial taxonomic classification, the specificity and coverage of older probes may change. The introduction of the probeBase database already highlighted the need to re-validate the specificity of previously published probes as new sequence data becomes available. Newer versions of probeBase additionally enable in silico assessment of probe specificity and coverage against updated rRNA sequence databases. Therefore, in the extended Discussion section, we’ll intend to emphasize that the probe set used should be considered a tool for quantitatively assessing selected broad phylogenetic groups, not as a complete characterization of the entire bacterial community. Therefore, FISH results may be subject to both underestimation of the abundance of certain groups due to incomplete probe coverage and potential detection of organisms outside the nominal target group due to imperfect specificity. However, this does not mean that the approach used is without value. FISH remains a well-established method enabling the direct visualization and quantification of microbial cells in environmental samples. In this study, this method was primarily used to compare the relative proportions of broad bacterial groups between the studied water samples, and the obtained results should be interpreted within the resolution provided by the probe set used. We also purpose to introduce appropriate limitations in the interpretation of statistical results. The statement of "less than 5% nonspecific staining" will not be longer present as evidence of the full taxonomic validity of the probes used, but solely as a result of monitoring the level of nonspecific fluorescent signal. At the same time, in the Discussion, we’ll like to clearly indicate, that an up-to-date assessment of the coverage and specificity of the probes used, based on modern sequence databases, as well as the use of high-throughput sequencing methods, could in the future provide a more comprehensive characterization of the analyzed bacterial communities.
We will include following description in the discussion: ‘An important limitation of the FISH approach is that the specificity and taxonomic coverage of the probes may not fully reflect the current diversity and classification of the targeted bacterial groups. Although the use of the NON338 probe indicated a low level of nonspecific fluorescent signal, this control does not assess the phylogenetic coverage or specificity of the taxon-specific probes. Consequently, some members of the targeted groups may have remained undetected, while potential cross-hybridization with organisms outside the nominal target groups cannot be completely excluded. Therefore, the FISH results should be interpreted as estimates of the abundance of the targeted broad phylogenetic groups rather than as a comprehensive characterization of the bacterial community.’
- Data Availability
- I strongly disagree with the authors statement that “The dataset used within this study will be made available on a reasonable request,” as it is a common convention in the field to have data available in a supplement or publicly available dataset upon submission. As noted above, I couldn’t even figure out what sequences the authors used for their FISH. For example, here is a recent preprint in Biogeosciences where the authors give 18S tables in their supplement (I know this is slightly different, but just wanted to give a recent example): https://egusphere.copernicus.org/preprints/2026/egusphere-2026-4192/
- Also, the authors should give their raw, measured values for environmental parameters (like pH, temperature, ammonium, etc.) as well as carbon isotope data. Not just the averaged data (e.g., Table 1). Here is a recent example from Biogeosciences, where they have made this data available in their supplemental: https://bg.copernicus.org/articles/19/1/2022/
- I also think this may violate the journal’s own policies. Although the authors technically have a statement on how their data can be accessed, in line with the journals request, the journal also states: “If the data are not publicly accessible at the time of final publication, the data statement should describe where and when they will appear, and provide information on how readers can obtain the data until then. Nevertheless, authors should make such embargoed data available to reviewers during the review process in order to foster reproducibility. The Copernicus review system allows to define such assets as 'access limited to reviewers' and reviewers must then sign that they will use such data only for the purpose of reviewing without making copies, sharing, or reusing. In rare cases where the data cannot be deposited publicly (e.g., because of commercial constraints), a detailed explanation of why this is the case is required. The data needed to replicate figures in a paper should in any case be publicly available, either in a public database (strongly recommended), or in a supplement to the paper.”
- https://www.biogeosciences.net/policies/data_policy.html#data_availability
Response: Thank you for raising this issue. We agree that the original Data Availability statement was insufficient and that the underlying data should be made publicly accessible to facilitate transparency and reproducibility. Together with the revised version of the manuscript, we will make the dataset publicly available. The dataset will include all of the variables considered in our study (δ¹³C values, environmental variables, bacteria abundances). Additionally, as stated above, we will provide details on FISH probes. Accordingly, we will revise the Data Availability statement in the manuscript to provide the location and access information for the dataset. For the review purposes, both files can be accessed here: https://github.com/wmkrzt/Environ.Bacteria.d13C
- Figures
- The figures are difficult to interpret. As a starter, the text on the figures is far too small and difficult to read. But, more importantly, there are no figures showing the actual data measured in this study besides Figure 1. Figures 2-4 are essentially a graphical rendering of Table 3, with model-based uncertainty bands. It basically does not allow the reader, or reviewer, to judge for themselves if the model is indeed a good fit for the measured data. Also – For Figure 4, the phrase “stepwise-selected significant predictors” is confusing because the manuscript states that the full Daphnia model was retained, while Figure 4 apparently displays only predictors with P<0.05. In addition, only 12 Daphnia samples were measured and the fitted model for Daphnia contains 11 predictors (day of the year, proteobacteria, alpha proteobacteria, etc.), the authors should address potential model overfitting – this could also be addressed by just showing the actual measured data. Again, since the data are not made available, it is difficult to really independently evaluate this manuscript.
Response: If the Editor decision is further proceeding of our manuscript and we will be given the chance of making improvements, we will work on figures readability in revised version of the manuscript. Axes labels will be made more visible, and we will add data points to each predictor plot. Finally, we will make results description within the manuscript concise with figures captions.
We would also like to clarify an error in the sample sizes reported in the previous version of the manuscript. The numbers of samples reported previously (36 calanoid copepod, 29 cyclopoid copepod, and 12 Daphnia samples) were inadvertently retained from an earlier version of the dataset used for a different set of analyses. The isotope dataset used for the analyses presented in the current manuscript actually contains 48 calanoid copepod, 49 cyclopoid copepod, and 27 Daphnia samples. Out of this dataset, stepwise-selected models included slightly lower number of fitted observations (45 calanoid copepod samples, 45 cyclopoid copepod samples and 26 Daphnia samples). We apologize for this error and will carefully check and correct the sample sizes throughout the manuscript. Regarding the potential for overfitting of the Daphnia model, we acknowledge that the relatively small sample size (n = 26) in relation to the number of predictors included in the model warrants caution. The model was developed as an exploratory analysis of potential environmental and biological drivers of Daphnia stable isotope variability, rather than as a predictive model. We will add an explicit statement in the Discussion, acknowledging the possibility of model overfitting: “Considering the sample size fitted into a model for Daphnia (N = 26) and relatively large number of stepwise-selected environmental predictors (11), presented results for relationships of δ¹³C in this group should be interpreted with appropriate caution”.
- Minor comments
- Line 19: You should briefly define what d13C is – for example, “densities of various groups of bacteria and the carbon isotope composition (d13C) of calanoid copepods”
- Line 39: Do you mean ‘reviewed by’ instead of ‘revised by’?
- Line 90: Typo, should be “depths” not “depts”
- Supplemental: This looks like raw outputs from R. The table caption should define what the column headers are – e.g., “Df; Degrees of freedom.
Response: All of these comments will be used for corrections of the revised version of the manuscript
References:
De Meester, L., Declerck, S. A., & Ger, K. A. (2023). Beyond Daphnia: a plea for a more inclusive and unifying approach to freshwater zooplankton ecology. Hydrobiologia, 850(21), 4693-4703.
Huỳnh, T. H., Horváth, Z., Pálffy, K., Kardos, V., Szabó, B., Dobosy, P., & Vad, C. F. (2024). Heatwave‐induced functional shifts in zooplankton communities result in weaker top‐down control on phytoplankton. Ecology and Evolution, 14(8), e70096.
Kratina, P., Greig, H. S., Thompson, P. L., Carvalho-Pereira, T. S. A., & Shurin, J. B. (2012). Warming modifies trophic cascades and eutrophication in experimental freshwater communities. Ecology, 93(6), 1421–1430. https://doi.org/10.1890/11-1595.1
Kubera, Ł., Kalwasińska, A., & Perliński, P. (2025). Assessment of the Structure and Activity of Bacterial Communities as a Tandem Tool for Estimating the Ecological Condition of Lakes in Poland: Assessment of the Structure and Activity of Bacterial Communities. Ecosystems, 28(3), 32.
Kunzmann, A. J., Ehret, H., Yohannes, E., Straile, D., & Rothhaupt, K. O. (2019). Calanoid copepod grazing affects plankton size structure and composition in a deep, large lake. Journal of Plankton Research, 41(6), 955-966.
Landa, M., Blain, S., Christaki, U., Monchy, S., & Obernosterer, I. (2016). Shifts in bacterial community composition associated with increased carbon cycling in a mosaic of phytoplankton blooms. The ISME journal, 10(1), 39-50.
Rabow, S., Cai, S., Carlsson, P., & Rengefors, K. (2026). Earlier spring may explain dominance of the slow‐growing dinoflagellate Naiadinium polonicum over the cyanobacterium Microcystis during a heatwave. Journal of Phycology.
Citation: https://doi.org/10.5194/egusphere-2026-3444-AC2
-
AC2: 'Reply on RC2', Wojciech Krztoń, 14 Sep 2026
Viewed
| HTML | XML | Total | Supplement | BibTeX | EndNote | |
|---|---|---|---|---|---|---|
| 155 | 51 | 32 | 238 | 36 | 26 | 26 |
- HTML: 155
- PDF: 51
- XML: 32
- Total: 238
- Supplement: 36
- BibTeX: 26
- EndNote: 26
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments:
The manuscript presents a valuable and timely contribution to research on carbon cycling in shallow freshwater ecosystems. The authors address the important issue of identifying the environmental and microbial factors shaping carbon isotope (δ¹³C) variability in the major groups of crustacean zooplankton. A particular strength of the study is the integration of stable isotope analyses with bacterial community characterization and environmental parameters, providing insights into pelagic ecosystem functioning from both ecological and biogeochemical perspectives. The study is based on an extensive field dataset collected seasonally from several shallow eutrophic water bodies. The analytical methods and statistical models are appropriate for addressing the stated research objectives, and the results are presented in a clear and logical manner. The Discussion deserves particular recognition for placing the findings within the context of current knowledge and interpreting them in light of the functional differences among the investigated zooplankton groups. I also appreciate that the authors openly acknowledge the main limitation of the study, namely the lack of δ¹³C measurements of potential food sources. Recognizing this limitation and identifying directions for future research strengthens the credibility of the conclusions and highlights promising avenues for further investigation.
Overall, I believe that the manuscript provides novel insights into the functioning of freshwater food webs and is suitable for publication after minor revisions.
Specific comments:
Technical comments: