Tracing Nitrogen Distribution and Biotic Responses in Spring-Fed Karst Rivers: A Pilot Study

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper examines the nitrogen flow pathway and its (indirect) impact on biodiversity and drinking water management. It discusses the impact of excess nitrogen on ecosystem function. The overall nitrogen cycle was analysed by quantifying water, sediments, aquatic mosses, and benthic macroinvertebrates, attempting to determine whether water with high nitrogen content can be stored in organisms and sediments within a given aquatic ecosystem. This research is interdisciplinary in nature. It fits within the journal's thematic scope. I present my comments below.:

1. The abstract is missing numerical data. Please provide it.

2. Nitrogen-related issues in both terrestrial and aquatic ecosystems are widely analysed. Therefore, I propose expanding the literature review to include introductions from other regions. Currently, only five references are cited.

3. Present a table with the basic hydrological characteristics of these rivers.
4. Section 2.1 requires citations; much of the information is not supported by scientific sources.

5. List the number of water, sediment, and moss species samples collected. This information should be included in the tables.

6. What is the mineralogical composition of the sediments studied? Is there any literature data on this topic? If so, it should be included in the article.

7.  How were the bryophyte samples prepared for analysis?

8. Figure 1 is incomplete, lacking a scale, and the name of the town should be given on the figure and against the background of Greece, so that the reader could get an idea of ​​the location.

9. Although the abstract mentions drinking water, the text provides no information on this topic.

10. Was the species composition of mosses examined, or was only a specific species considered?

 

Author Response

We would like to express our sincere gratitude for your valuable and constructive feedback. Below we provide detailed responses to your comments (in italic).

 

This paper examines the nitrogen flow pathway and its (indirect) impact on biodiversity and drinking water management. It discusses the impact of excess nitrogen on ecosystem function. The overall nitrogen cycle was analysed by quantifying water, sediments, aquatic mosses, and benthic macroinvertebrates, attempting to determine whether water with high nitrogen content can be stored in organisms and sediments within a given aquatic ecosystem. This research is interdisciplinary in nature. It fits within the journal's thematic scope. I present my comments below.:

1. The abstract is missing numerical data. Please provide it.

In response to the reviewer’s comment, numerical data have now been incorporated into the abstract to improve clarity and quantitative support.

2. Nitrogen-related issues in both terrestrial and aquatic ecosystems are widely analysed. Therefore, I propose expanding the literature review to include introductions from other regions. Currently, only five references are cited.

We thank the reviewer for this suggestion. The literature review has been expanded.

3. Present a table with the basic hydrological characteristics of these rivers.
   Thank you – a table (Table 1 in the revised version) was added.
4. Section 2.1 requires citations; much of the information is not supported by scientific sources.

We conducted a thorough search of available hydrological literature and databases for the listed streams (Stylos, Agios Georgios, Armenoi, Zourpos). While detailed quantitative characteristics such as channel length, catchment area, and mean annual discharge are not available in published sources for these specific headwater streams, we have included contextual information and references for the broader Koiliaris watershed, within which Stylos and Armenoi springs are significant karstic contributors (Lilli et al., 2020; Nerantzaki & Nikolaidis, 2020).

5. List the number of water, sediment, and moss species samples collected. This information should be included in the tables.

We thank the Reviewer for this suggestion. The number of samples collected has now been clearly indicated in Table 3 and specified in the Methods section (2.2). One composite sample per matrix was collected at each site and analyzed in triplicate in the laboratory.

6. What is the mineralogical composition of the sediments studied? Is there any literature data on this topic? If so, it should be included in the article.

We thank the Reviewer for this comment. Although no site-specific mineralogical analyses were performed in the present pilot study, published data from the Koiliaris River Basin CZO indicate that sediments are derived predominantly from carbonate sedimentary rocks (limestone and cherty limestone) and consist mainly of calcareous alluvial materials. This contextual information has now been added to the manuscript with the appropriate reference.

7. How were the bryophyte samples prepared for analysis?

We thank the reviewer for this comment. The preparation of bryophyte samples was described in Section 2.3; however, for clarity, the procedure has now been explicitly specified. Bryophyte samples were rinsed in the field, cleaned of mineral and organic particles in the laboratory, oven-dried, homogenized, and subsequently analyzed for total nitrogen.

8. Figure 1 is incomplete, lacking a scale, and the name of the town should be given on the figure and against the background of Greece, so that the reader could get an idea of the location.

Thank you – the figure was revised.

9. Although the abstract mentions drinking water, the text provides no information on this topic.

We thank the reviewer for this comment. We agree that the initial version of the manuscript placed disproportionate emphasis on drinking water aspects without providing sufficient supporting information. To avoid overextension and maintain a clear focus, references to drinking water have been reduced, and the manuscript has been revised to concentrate specifically on the distribution of nitrogen across abiotic and biotic matrices within the studied system.

10. Was the species composition of mosses examined, or was only a specific species considered?

We thank the reviewer for this question. The full species composition of bryophytes was examined at each site during the field survey. However, for the purpose of nitrogen analysis, only one species per site was selected, namely the dominant and most abundant species, in order to ensure representativeness and comparability among sites. Importantly, the same species was dominant at three of the four sites, which strengthens the reliability of inter-site comparisons of tissue nitrogen concentrations and reduces potential interspecific variability effects. This approach was adopted to minimize the influence of species-specific physiological differences in nitrogen uptake and to ensure that observed differences primarily reflect environmental variation rather than taxonomic composition.

The Cover Letter to Editors and Reviewers is attached. 

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

142: 4°C

159: 800°C

162: (NH₄)₂SO₄

162: και

178: The image of the island of Crete is good, but you can zoom in more on the sampling area.

191-192: Did you find only these two moss species?

203 Glossosomatidae

211 Gammarus

218-233: In the text, you report TN levels with 2 decimal places; in the table, only 1.

301: Lowest

Author Response

We sincerely thank the Reviewer 2 for the careful reading of the manuscript and for the helpful corrections, which have contributed to improving the clarity and quality of the text. All identified errors were corrected, which greatly helped improve the manuscript. Below we provide detailed responses to your comments (in italic).

 

178: The image of the island of Crete is good, but you can zoom in more on the sampling area.

Thank you – the figure was revised.

191-192: Did you find only these two moss species?

We thank the reviewer for this question. The full species composition of bryophytes was examined at each site during the field survey. However, for the purpose of nitrogen analysis, only one species per site was selected, namely the dominant and most abundant species, in order to ensure representativeness and comparability among sites. Importantly, the same species was dominant at three of the four sites, which strengthens the reliability of inter-site comparisons of tissue nitrogen concentrations and reduces potential interspecific variability effects. This approach was adopted to minimize the influence of species-specific physiological differences in nitrogen uptake and to ensure that observed differences primarily reflect environmental variation rather than taxonomic composition.

218-233: In the text, you report TN levels with 2 decimal places; in the table, only 1.

Thank you! TN levels have been rounded to one decimal place.

The Cover Letter to the Editors and Reviwers is attached. 

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

General comments:

The paper is well-written, easy to follow, and coherent. Excellent work.

I have very few comments, but I do have one biggie. I think you should re-consider how you’ve examined these data. While I haven’t seen the raw data (and your paper is missing some details about how many individual samples you obtained for each location and matrix type (e.g., how many sediment grabs?) something like a generalized linear mixed effects model might help you preserve the information rather than summarizing the data beforehand. That potentially resolves your sample size issue, but a more general way to deal with that is to use the Akaike Information Criterion (maybe the one corrected for small sample sizes). Using AIC might help you better understand how including various predictors for a given response variable improves the model fit. I’m suggesting major revisions, but only because if you agree the stats could be changed, then that’s a major revision. 

I also thought the methods section could use some more details about the work.

But overall the paper is very well-written and easy to understand.

Specific comments:

Lines 15-17: The subject of this sentence is a little buried. I can’t really tell what the major thrust of your paper is: drinking water protection, identifying indicators, effective environmental monitoring, and river basin management are all mentioned. You can probably just drop these lines and lose very little.

Line 21: I suggest inserting a statement here after the sentence ending with ‘drinking water quality’ about the general challenges to this sort of thing, like the programs are expensive, all programs must in some way be customized to the study area (e.g., general ideas work fine, but there will always be some area-specific adjustments), the common inapplicability of guidelines (and their uncertainty), mixture toxicity, and all the other stuff that makes understanding even ‘simple’ things like nitrogen in rivers very hard.

Lines 27-28: do you mean you quantified nitrogen in tissues of moss and benthic invertebrates? If yes, please be explicit. I wasn’t sure if you looked at moss and benthic communities since those are so common in this type of work.

Lines 32-35: information theoretic approaches, like evaluating improvement in model fit using things like AICc might be a better tool for you.

Lines 54-57: I suggest mentioning this briefly in your abstract

Lines 52-65: excellent and well-structured paragraph

Lines 66-74: some words here about what tissue nitrogen is capturing would be helpful for people like me. Obviously nitrogen is very common in tissues, so I’m slightly ignorant about what signals you are looking for in tissue nitrogen of benthic invertebrates. Do benthic invertebrates in nitrogen rich areas also have comparatively more nitrogen in their tissues? If so, what’s the story with that? I don’t really understand how that happens. I’m talking molecularly.

Lines 82-84: also where you (or whomever) doesn’t want to start an exact program that becomes embedded

Line 142: typo; ‘0’ instead of ‘°’

Lines 145-147: did you do any taxonomy on the moss? Looks like you did from the results, but didn’t see that information here.

Lines 160: typo; O₂

Lines 171-177: so what were the predictors? Did you have any non-detects? If yes, what did you do with them? Also, do you only care about the slopes? You may want to consider using a correlation coefficient as well. Presumably you used the means or medians per location as the input variables? I haven’t looked in a while, but there may be techniques you can use which allows you to use the raw data; maybe a generalized linear model with nesting? I’m not sure if Statistica does that, but R certainly has at least 2 popular packages which can fit GLMs. How did you deal with differences in benthic and moss assemblages across your sites? Did you standardize them somehow, like only pick a certain family of caddisflies?

But just for clarity, I also suggest being explicit about how many individual replicates you have for each compartment. For example, did you only take one water grab for each sampling location? You mention 5-10 sub samples for moss on line 146, but I didn’t see the same info for water, sediment, or benthos.

Figure 1: you can probably zoom in more in the main map showing the numbered sites.

Table 1: make sure to define any acronyms/abbreviations in the table in the caption.

Lines 200-215: I think it makes sense to show all of these individual measurements in some sort of figure; you could probably have benthic invertebrate genus or family on the y -axis and nitrogen concentration on the x with colored symbols for each study location. If that’s too messy, you could consider some other visualization to help the reader understand the results very quickly.

Lines 218-233: I also suggest making these results into some sort of coherent figure that tells this story easily.

Line 285: Use the full word instead of Fam./fam. You can start the sentence with ‘The Gammaridae and Hydrobiidae families….’

Line 286: typo

Lines 284-296: that you looked at TN within genera (or maybe families) would be clearer with a figure showing that raw. The results from Table 2 suggest your main interest was all of the benthos at a site. And maybe it was, but still. Presenting the measurements for each taxon in a figure will help the reader understand your work and make reading the results much easier.

Lines 349-353: but this is also with rolling up/summarizing your data at the level of the site before examining the data. Something like a generalized linear model with taxon as a random factor nested within site (maybe; random vs. fixed vs. nested can get confusing) might help you preserve the information in the individual measurements while still answering your question.

Lines 363-391: the conclusions are long; one paragraph is a good target. Just main messages should be the goal of the conclusion.

Comments on the Quality of English Language

english is mostly great - just a few typos here and there

Author Response

We sincerely thank Reviewer 3 for the valuable and constructive comments. All of them have been addressed, and we hope that our revisions adequately meet the Reviewer’s concerns. Please note that this revised version also incorporates the recommendations provided by Reviewer 1, 2 and 4. In response to the Reviewer’s comments, we have applied Principal Component Analysis to explore variation across matrices and sites. The Materials and Methods, Results, and Discussion sections have been updated accordingly.

Below we provide detailed responses to your comments (in italic).

 

General comments: The paper is well-written, easy to follow, and coherent. Excellent work. I have very few comments, but I do have one biggie. I think you should re-consider how you’ve examined these data. While I haven’t seen the raw data (and your paper is missing some details about how many individual samples you obtained for each location and matrix type (e.g., how many sediment grabs?) something like a generalized linear mixed effects model might help you preserve the information rather than summarizing the data beforehand. That potentially resolves your sample size issue, but a more general way to deal with that is to use the Akaike Information Criterion (maybe the one corrected for small sample sizes). Using AIC might help you better understand how including various predictors for a given response variable improves the model fit. I’m suggesting major revisions, but only because if you agree the stats could be changed, then that’s a major revision. I also thought the methods section could use some more details about the work. But overall the paper is very well-written and easy to understand.

We sincerely thank the reviewer for the positive evaluation of our manuscript and for the thoughtful and constructive suggestions regarding the statistical approach. We carefully considered the possibility of applying generalized linear mixed-effects models and model selection based on AIC. However, given the pilot-scale design of this study (four sites, one composite sample per matrix per site), the dataset does not provide sufficient degrees of freedom to support mixed-effects modeling or reliable model selection procedures. The limited number of independent observations would likely result in model overparameterization and unstable estimates. For this reason, we intentionally treated the statistical analysis as exploratory and focused on effect size (standardized coefficients) rather than formal model comparison. We have now clarified this rationale more explicitly in the Methods and Discussion sections to improve transparency. We agree that a larger dataset with replication across sites and seasons would allow the application of mixed-effects models and AIC-based model selection, and this will be considered in future work. Additionally, we have expanded the Methods section to provide further methodological details as suggested.

 

Specific comments:

Lines 15-17: The subject of this sentence is a little buried. I can’t really tell what the major thrust of your paper is: drinking water protection, identifying indicators, effective environmental monitoring, and river basin management are all mentioned. You can probably just drop these lines and lose very little.

We thank the Reviewer for this helpful comment. We agree that the original sentence was overly broad. It has now been revised to better reflect the primary focus of the study, namely tracing nitrogen pathways and evaluating ecosystem response indicators in spring-fed river systems.

Line 21: I suggest inserting a statement here after the sentence ending with ‘drinking water quality’ about the general challenges to this sort of thing, like the programs are expensive, all programs must in some way be customized to the study area (e.g., general ideas work fine, but there will always be some area-specific adjustments), the common inapplicability of guidelines (and their uncertainty), mixture toxicity, and all the other stuff that makes understanding even ‘simple’ things like nitrogen in rivers very hard.

We thank the Reviewer for this insightful suggestion. A paragraph has now been added to the Abstract discussing the broader challenges associated with nitrogen monitoring and management, including site-specific variability, uncertainty in guideline applicability, and the influence of interacting stressors in complex karst systems.

Lines 27-28: do you mean you quantified nitrogen in tissues of moss and benthic invertebrates? If yes, please be explicit. I wasn’t sure if you looked at moss and benthic communities since those are so common in this type of work.

We thank the reviewer for this helpful clarification. Yes, we quantified tissue nitrogen concentrations in aquatic mosses and benthic macroinvertebrates. The sentence has now been revised to explicitly state this.

Lines 32-35: information theoretic approaches, like evaluating improvement in model fit using things like AICc might be a better tool for you.

We thank the reviewer for this valuable suggestion. We carefully considered the use of information-theoretic approaches such as AICc-based model comparison. However, given the pilot-scale design of the study (n = 4 sites, one composite sample per matrix), the dataset does not provide sufficient degrees of freedom to support reliable model selection procedures without overparameterization. For this reason, we treated the analysis as exploratory and focused on the direction and magnitude of observed associations rather than formal model comparison. To clarify this point, we have revised the Abstract and Methods sections to explicitly state that the limited sample size precluded formal statistical inference.

 

Lines 54-57: I suggest mentioning this briefly in your abstract

Thank you for the suggestion; it has been implemented.

Lines 52-65: excellent and well-structured paragraph

Lines 66-74: some words here about what tissue nitrogen is capturing would be helpful for people like me. Obviously nitrogen is very common in tissues, so I’m slightly ignorant about what signals you are looking for in tissue nitrogen of benthic invertebrates. Do benthic invertebrates in nitrogen rich areas also have comparatively more nitrogen in their tissues? If so, what’s the story with that? I don’t really understand how that happens. I’m talking molecularly.

We thank the reviewer for this insightful comment. We agree that further clarification of what tissue nitrogen represents biologically is helpful. We have now expanded the Introduction to explain that tissue nitrogen reflects the incorporation of environmentally available nitrogen into proteins and other nitrogen-containing biomolecules, and that in bryophytes and macroinvertebrates this occurs via direct uptake and trophic transfer, respectively.

Lines 82-84: also where you (or whomever) doesn’t want to start an exact program that becomes embedded

Thank you for this helpful clarification. The text has been revised to emphasize that the pilot approach provides a flexible, exploratory framework, particularly useful in contexts where establishing a fixed, long-term monitoring program is not intended.

Line 142: typo; ‘0’ instead of ‘°’

Corrected.

Lines 145-147: did you do any taxonomy on the moss? Looks like you did from the results, but didn’t see that information here.

Thank you for the suggestion; the information on the identification was added.

Lines 160: typo; O₂

Corrected.

Lines 171-177: so what were the predictors? Did you have any non-detects? If yes, what did you do with them? Also, do you only care about the slopes? You may want to consider using a correlation coefficient as well. Presumably you used the means or medians per location as the input variables? I haven’t looked in a while, but there may be techniques you can use which allows you to use the raw data; maybe a generalized linear model with nesting? I’m not sure if Statistica does that, but R certainly has at least 2 popular packages which can fit GLMs. How did you deal with differences in benthic and moss assemblages across your sites? Did you standardize them somehow, like only pick a certain family of caddisflies?

But just for clarity, I also suggest being explicit about how many individual replicates you have for each compartment. For example, did you only take one water grab for each sampling location? You mention 5-10 sub samples for moss on line 146, but I didn’t see the same info for water, sediment, or benthos.

We thank the reviewer for these helpful questions and suggestions. The predictors explored were TN concentrations in abiotic matrices (water and sediments), while response variables were tissue TN concentrations in mosses and macroinvertebrates. Mean values per site were used as input variables, based on one composite sample per compartment at each site.

No non-detects were present (all TN concentrations were above analytical DLs). Laboratory triplicates represent technical replication only and were not treated as independent observations.

Given the pilot-scale design (n = 4 sites) and the lack of biological replication, analyses were treated as exploratory. Correlation coefficients and generalized linear or mixed-effects models were not applied, as these would not yield meaningful or stable results with the available dataset.

For biotic matrices, one dominant and representative moss species per site was selected for TN analysis to minimize interspecific variability; assemblage-level standardization was therefore not applied. We have revised the Methods section (2.4) to clarify these points and explicitly state the limitations of the statistical approach.

Figure 1: you can probably zoom in more in the main map showing the numbered sites.

Thank you – the figure was revised.

Table 1: make sure to define any acronyms/abbreviations in the table in the caption.

Thank you – a legend was provided.

Lines 200-215: I think it makes sense to show all of these individual measurements in some sort of figure; you could probably have benthic invertebrate genus or family on the y -axis and nitrogen concentration on the x with colored symbols for each study location. If that’s too messy, you could consider some other visualization to help the reader understand the results very quickly.

Lines 218-233: I also suggest making these results into some sort of coherent figure that tells this story easily.

We thank the reviewer for this helpful suggestion. To improve clarity and provide an integrated visualization of the results, we have added a new figure (PCA ordination; Figure 2) that summarizes TN patterns across abiotic and biotic matrices and highlights their relationships across sites. Given the pilot-scale design and limited sample size, this approach was preferred to avoid overinterpretation and excessive complexity.

Line 285: Use the full word instead of Fam./fam. You can start the sentence with ‘The Gammaridae and Hydrobiidae families….’

Thank you; it has been implemented.

Line 286: typo

Lines 284-296: that you looked at TN within genera (or maybe families) would be clearer with a figure showing that raw. The results from Table 2 suggest your main interest was all of the benthos at a site. And maybe it was, but still. Presenting the measurements for each taxon in a figure will help the reader understand your work and make reading the results much easier.

We thank the Reviewer for this comment and for pointing out the need for clarification. We would like to emphasize that total nitrogen was not analyzed separately for individual genera or families. Instead, all benthic macroinvertebrates collected at each site were pooled into a single composite sample to represent site-level benthic nitrogen content. This has been clarified in the Methods section to avoid misunderstanding.

Lines 349-353: but this is also with rolling up/summarizing your data at the level of the site before examining the data. Something like a generalized linear model with taxon as a random factor nested within site (maybe; random vs. fixed vs. nested can get confusing) might help you preserve the information in the individual measurements while still answering your question.

We thank the Reviewer for this suggestion. We agree that models incorporating taxon-level replication could better preserve variability in larger datasets. However, in the present pilot study, nitrogen measurements were based on site-level composite samples, and taxon-specific TN data were not available. Consequently, the analysis focused on site-level patterns and was treated as exploratory. This limitation has been clarified in the Methods and Discussion sections.

Lines 363-391: the conclusions are long; one paragraph is a good target. Just main messages should be the goal of the conclusion.

Comments on the Quality of English Language: English is mostly great - just a few typos here and there

Thank you for this suggestion. The Conclusions section has been substantially shortened and consolidated into a single paragraph focusing on the main findings of the study.

The Cover Letter to the Editors and Reviewers is attached. 

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

This so-called pilot exploratory study purports to trace nitrogen pathways and evaluate ecosystem response indicators in Mediterranean karst spring-fed rivers, yet it fails to deliver meaningful, robust findings due to severe methodological flaws, insufficient data support, ambiguous result presentation, and overreaching conclusions that are disconnected from the limited scope of the research—relying on a mere 4 sampling sites and lacking statistical rigor, it cannot substantiate its claims regarding biodiversity implications or drinking water management. The research fails to meet basic standards for scientific rigor, and as a result, the manuscript is not suitable for publication in any peer-reviewed journal.

1. The title is excessively broad and misleading: Claiming “Implications for Biodiversity and Drinking Water Management” is unjustified, as the study only reports total nitrogen (TN) concentrations across 4 matrices at 4 sites without providing direct evidence of how these data inform biodiversity conservation or drinking water protection strategies. The title also overstates the study’s contribution. Tracing nitrogen pathways requires comprehensive analysis of sources, transformations, and fluxes, which is not achieved with the limited sampling and simplistic correlation analysis conducted.
2. Abstract: The abstract dedicates disproportionate space to background information (e.g., nitrogen’s role in freshwater ecosystems, karst system complexity) rather than emphasizing key results—only a single sentence vaguely describes TN concentration differences across matrices, lacking specific data or meaningful trends. The abstract’s conclusion about “aquatic bryophytes and macroinvertebrates as integrative indicators” is unsupported, as no statistical validation or comparative analysis with established indicators is provided.
3. Methodology: 1) Only 4 sampling sites are included (n=4), which is statistically inadequate for detecting meaningful spatial patterns or correlations—multiple regression analysis with n=4 cannot yield reliable effect-size estimates or causal inferences. 2) The study claims to represent “contrasting hydrobiological source areas and degrees of anthropogenic influence”, but no quantitative data (e.g., land use intensity, nitrogen input loads) are provided to validate the gradient of anthropogenic pressure across sites. 3) Sampling was conducted only once in July 2025, failing to account for seasonal variability in nitrogen dynamics, hydrology, or biotic communities—critical for karst systems with variable groundwater-surface water interactions. 4) The protocol mentions collecting “5 to 10 subsamples” but does not specify the spatial distribution of subsamples or how “patchiness” was quantified—this lack of standardization introduces sampling bias. 5) Individuals were identified only to family or genus level, which is too coarse for assessing biodiversity responses to nitrogen exposure—species-level identification is necessary for detecting subtle changes in sensitive taxa. 6) The calibration curve for TN analysis used KNO₃, (NH₄)₂SO₄, and EDTA, but no information is provided on the concentration ranges of standards, R² values, or detection limits—essential for evaluating data accuracy. 7) Treating exploratory regression analysis with n=4 as meaningful is inappropriate; the study emphasizes “direction and magnitude of standardized coefficients” but ignores that such small sample sizes render effect-size estimates highly unstable.
4. Results: 1) All TN concentration data (Table 2) and physicochemical parameters (Table 1) lack measures of variability (SD/SE), making it impossible to assess data precision or compare differences between sites. 2) As required, lines 227–233 describe correlations between TN in different matrices (e.g., water-moss, water-macroinvertebrates) but provide no figure or table to illustrate these relationships—readers cannot verify the strength or pattern of associations. 3) The study reports dominant moss species (H. tenax, C. fontinaloides) but provides no data on moss biomass, cover, or density—critical for evaluating their utility as indicators. 4) Taxonomic richness is mentioned (e.g., “poorer community at Zourpos”) but no raw data on taxon counts, relative abundance, or diversity indices (Shannon-Wiener, Simpson) are provided—quantitative biodiversity metrics are necessary to link nitrogen exposure to community changes. 5) The study states moss TN is “15- to 85-fold higher than in water and sediments” but provides no raw data for these calculations (e.g., water TN in mg/L vs. moss TN in g/kg requires unit conversion verification)—the fold difference may be miscalculated. 6) The claim of “site-specific nitrogen pathways” is not supported by quantitative spatial analysis (e.g., ANOVA to test for site differences in TN concentrations)—no statistical tests are reported to validate spatial variability.
5. Discussion: 1) The discussion claims “aquatic mosses and macroinvertebrates proved valuable as time-integrated indicators” but provides no evidence of their performance relative to established indicators or how they integrate nitrogen exposure over time. 2) Seawater intrusion at Zourpos is identified as a stressor, but the study does not disentangle the effects of nitrogen enrichment from salinity on biotic communities. 3) The study cites comparisons to other river systems (e.g., Wuding River, Taipu River) but does not account for differences in climate, geology, or land use. 4) The discussion downplays the lack of statistical significance (p>0.05) by blaming “limited sample size” but does not acknowledge that the observed trends (e.g., b*=0.78 for water-moss TN) are statistically meaningless with n=4. 5) The study links nitrogen levels to "simplified communities of tolerant taxa" but provides no data on functional traits (e.g., feeding guilds, pollution tolerance values) of macroinvertebrates. 6) The discussion mentions “drinking-water vulnerability” but provides no specific guidance on how TN concentrations in biota or sediments relate to drinking water quality (e.g., no data on nitrate levels, a critical drinking water contaminant). 7) The study notes weak coupling between sediment TN and biotic TN but does not discuss potential reasons (e.g., sediment nitrogen bioavailability, mineralization rates)—a critical gap in understanding nitrogen pathways. 8) The discussion mentions “global climate change” but provides no data on how nitrogen dynamics might respond to climate variability. 9) The suggestion to "prioritize conservation of high-quality headwaters" is based on anecdotal observations (e.g., sensitive taxa at upstream sites) rather than quantitative assessment of habitat quality or nitrogen fluxes.

Author Response

We thank the Reviewer for the detailed and critical evaluation of our manuscript. We acknowledge that the study is based on a limited pilot-scale dataset and that several aspects required clearer framing and additional methodological transparency. In response to the review, we have substantially revised the manuscript to: clarify the exploratory nature of the study, temper conclusions and avoid overinterpretation, improve methodological detail and transparency, add measures of analytical precision, refine the title and abstract to better reflect the scope of the work. While the study does not aim to provide definitive causal inference, it offers preliminary insights intended to guide future, larger-scale investigations in Mediterranean karst systems. Below we provide detailed responses to your comments (in italic).

 

This so-called pilot exploratory study purports to trace nitrogen pathways and evaluate ecosystem response indicators in Mediterranean karst spring-fed rivers, yet it fails to deliver meaningful, robust findings due to severe methodological flaws, insufficient data support, ambiguous result presentation, and overreaching conclusions that are disconnected from the limited scope of the research—relying on a mere 4 sampling sites and lacking statistical rigor, it cannot substantiate its claims regarding biodiversity implications or drinking water management. The research fails to meet basic standards for scientific rigor, and as a result, the manuscript is not suitable for publication in any peer-reviewed journal.

1. The title is excessively broad and misleading: Claiming “Implications for Biodiversity and Drinking Water Management” is unjustified, as the study only reports total nitrogen (TN) concentrations across 4 matrices at 4 sites without providing direct evidence of how these data inform biodiversity conservation or drinking water protection strategies. The title also overstates the study’s contribution. Tracing nitrogen pathways requires comprehensive analysis of sources, transformations, and fluxes, which is not achieved with the limited sampling and simplistic correlation analysis conducted.

We thank the Reviewer for this thoughtful comment. We acknowledge that the original title may have conveyed a broader scope than supported by the pilot-scale dataset. In a response, we have revised the title to better reflect the exploratory nature of the study and to avoid overstatement regarding management implications. The revised title emphasizes nitrogen distribution and biotic responses within a pilot-scale framework.

2. Abstract: The abstract dedicates disproportionate space to background information (e.g., nitrogen’s role in freshwater ecosystems, karst system complexity) rather than emphasizing key results—only a single sentence vaguely describes TN concentration differences across matrices, lacking specific data or meaningful trends. The abstract’s conclusion about “aquatic bryophytes and macroinvertebrates as integrative indicators” is unsupported, as no statistical validation or comparative analysis with established indicators is provided.

We thank the Reviewer for this comment. We agree that the original abstract contained excessive background information and insufficient emphasis on the study results. The abstract has been substantially revised to (i) reduce contextual background, (ii) present quantitative findings more clearly, and (iii) temper conclusions to reflect the exploratory pilot-scale nature of the study.

3. Methodology: 1) Only 4 sampling sites are included (n=4), which is statistically inadequate for detecting meaningful spatial patterns or correlations—multiple regression analysis with n=4 cannot yield reliable effect-size estimates or causal inferences. 2) The study claims to represent “contrasting hydrobiological source areas and degrees of anthropogenic influence”, but no quantitative data (e.g., land use intensity, nitrogen input loads) are provided to validate the gradient of anthropogenic pressure across sites. 3) Sampling was conducted only once in July 2025, failing to account for seasonal variability in nitrogen dynamics, hydrology, or biotic communities—critical for karst systems with variable groundwater-surface water interactions. 4) The protocol mentions collecting “5 to 10 subsamples” but does not specify the spatial distribution of subsamples or how “patchiness” was quantified—this lack of standardization introduces sampling bias. 5) Individuals were identified only to family or genus level, which is too coarse for assessing biodiversity responses to nitrogen exposure—species-level identification is necessary for detecting subtle changes in sensitive taxa. 6) The calibration curve for TN analysis used KNO₃, (NH₄)₂SO₄, and EDTA, but no information is provided on the concentration ranges of standards, R²values, or detection limits—essential for evaluating data accuracy. 7) Treating exploratory regression analysis with n=4 as meaningful is inappropriate; the study emphasizes “direction and magnitude of standardized coefficients” but ignores that such small sample sizes render effect-size estimates highly unstable.

We acknowledge that the limited number of sites (n = 4) does not allow robust statistical inference or causal interpretation. The study was designed as a pilot-scale exploratory assessment, and regression analyses were used only to explore potential associations rather than to establish statistically supported relationships. To avoid overinterpretation, we have revised the Methods and Discussion sections to clearly emphasize the exploratory nature of the analysis and to temper the interpretation of standardized coefficients.

We would like to clarify that the objective of this study was not to quantify or compare a gradient of anthropogenic pressure across sites. The sites were selected based on previous studies documenting contrasting hydrobiological settings and land-use influences within the basin, which are cited to provide contextual background. The manuscript has been revised to avoid implying a quantified anthropogenic gradient and to clarify that these references were used solely to justify site selection.

We acknowledge that sampling was conducted during a single summer campaign and therefore does not capture seasonal variability. However, the objective of this pilot study was not to assess seasonal dynamics but to provide an exploratory snapshot of nitrogen distribution across abiotic and biotic matrices under comparable hydrological conditions.

We respectfully clarify that bryophyte subsampling followed Bowden et al. (2017), which was cited in the manuscript. The number of subsamples (5–10 per site) was adjusted according to stream width and spatial heterogeneity of bryophyte cover, as recommended in the reference.

The primary objective of this study was not to conduct a detailed taxonomic biodiversity assessment but to evaluate tissue nitrogen content as an exploratory indicator of nitrogen exposure. Identification to family or genus level followed standardized biomonitoring protocols and was sufficient for contextual ecological interpretation within the scope of this pilot study.

Additional information on calibration ranges and detection limits has been added to the Methods section.

We agree with the Reviewer and have revised the text to clarify that all regression results are exploratory only, with no inferential interpretation given the limited sample size.

4. Results: 1) All TN concentration data (Table 2) and physicochemical parameters (Table 1) lack measures of variability (SD/SE), making it impossible to assess data precision or compare differences between sites. 2) As required, lines 227–233 describe correlations between TN in different matrices (e.g., water-moss, water-macroinvertebrates) but provide no figure or table to illustrate these relationships—readers cannot verify the strength or pattern of associations. 3) The study reports dominant moss species (H. tenax, C. fontinaloides) but provides no data on moss biomass, cover, or density—critical for evaluating their utility as indicators. 4) Taxonomic richness is mentioned (e.g., “poorer community at Zourpos”) but no raw data on taxon counts, relative abundance, or diversity indices (Shannon-Wiener, Simpson) are provided—quantitative biodiversity metrics are necessary to link nitrogen exposure to community changes. 5) The study states moss TN is “15- to 85-fold higher than in water and sediments” but provides no raw data for these calculations (e.g., water TN in mg/L vs. moss TN in g/kg requires unit conversion verification)—the fold difference may be miscalculated. 6) The claim of “site-specific nitrogen pathways” is not supported by quantitative spatial analysis (e.g., ANOVA to test for site differences in TN concentrations)—no statistical tests are reported to validate spatial variability.

Physicochemical parameters, Table 1, (e.g., pH) are reported as single in situ measurements, consistent with standard practice.

Analytical variability for TN measurements has now been clarified based on laboratory triplicate analyses, with an estimated precision of ±5% (relative standard deviation). This information has been added to the Methods section and Table 2 caption.

A PCA ordination figure has been added to visually summarize the relationships among TN across matrices and sites.

Moss biomass or cover was not assessed, as the focus of this pilot study was on tissue nitrogen content rather than abundance-based indicator performance.

Community composition was described qualitatively for contextual purposes only, and no quantitative biodiversity assessment was intended within the scope of this pilot study.

We acknowledge that the original wording was imprecise and did not clearly account for unit differences between water (mg/L) and solid matrices (g/kg). The statement has been corrected following proper unit conversion, and the text now distinguishes between moss–water and moss–sediment differences to avoid misinterpretation.

We acknowledge that no formal statistical tests were applied to assess spatial differences among sites, as the limited sample size of this pilot study does not support robust spatial inference. References to site-specific patterns were therefore intended as descriptive and exploratory observations rather than statistically validated pathways. The manuscript has been revised to clarify this and to avoid overinterpretation of spatial variability.

5. Discussion: 1) The discussion claims “aquatic mosses and macroinvertebrates proved valuable as time-integrated indicators” but provides no evidence of their performance relative to established indicators or how they integrate nitrogen exposure over time. 2) Seawater intrusion at Zourpos is identified as a stressor, but the study does not disentangle the effects of nitrogen enrichment from salinity on biotic communities. 3) The study cites comparisons to other river systems (e.g., Wuding River, Taipu River) but does not account for differences in climate, geology, or land use. 4) The discussion downplays the lack of statistical significance (p>0.05) by blaming “limited sample size” but does not acknowledge that the observed trends (e.g., b*=0.78 for water-moss TN) are statistically meaningless with n=4. 5) The study links nitrogen levels to "simplified communities of tolerant taxa" but provides no data on functional traits (e.g., feeding guilds, pollution tolerance values) of macroinvertebrates. 6) The discussion mentions “drinking-water vulnerability” but provides no specific guidance on how TN concentrations in biota or sediments relate to drinking water quality (e.g., no data on nitrate levels, a critical drinking water contaminant). 7) The study notes weak coupling between sediment TN and biotic TN but does not discuss potential reasons (e.g., sediment nitrogen bioavailability, mineralization rates)—a critical gap in understanding nitrogen pathways. 8) The discussion mentions “global climate change” but provides no data on how nitrogen dynamics might respond to climate variability. 9) The suggestion to "prioritize conservation of high-quality headwaters" is based on anecdotal observations (e.g., sensitive taxa at upstream sites) rather than quantitative assessment of habitat quality or nitrogen fluxes.

The text has been revised to frame mosses and macroinvertebrates as potential rather than validated time-integrated indicators, consistent with the exploratory nature of the study.

The role of seawater intrusion is discussed only as contextual information, and no attempt is made to separate salinity and nitrogen effects within this pilot study.

The cited examples (e.g., Wuding River, Taipu River) are used for contextual comparison only, not as directly comparable systems.

We agree with the reviewer that trends derived from n = 4 observations are statistically unstable and should not be interpreted as meaningful effect sizes. The manuscript has been revised to explicitly acknowledge this limitation, and all such results are now framed strictly as exploratory observations.

We acknowledge that the study does not provide direct assessment of drinking water quality (e.g., nitrate concentrations or regulatory thresholds). References to drinking-water vulnerability were therefore not intended to imply direct applicability to drinking water standards but to provide broader environmental context. The manuscript has been revised to reduce this emphasis and to avoid overstating links between TN in biota or sediments and drinking water quality.

Potential mechanisms underlying the sediment–biotic TN coupling are now acknowledged in the Discussion.

We acknowledge that the present study does not include data on climate variability or its effects on nitrogen dynamics. References to global climate change were therefore intended only to provide broader environmental context. The manuscript has been revised to reduce this emphasis and to avoid implying climate-driven interpretations not supported by the data.

Conservation-related statements have been tempered to avoid management recommendations not supported by quantitative analyses.

The Cover Letter to the Editors and Reviewers is attached. 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

All my comments were analysed, and the authors answered. I am satisfied. 

Author Response

We thank Reviewer 1 for the careful evaluation of the manuscript and for confirming that all comments were adequately addressed. Attached is a cover letter with responses to the comments of the Academic Editor.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Thank you for the extensive revisions. No further comments.

Author Response

We thank Reviewer 3 for the constructive feedback and for acknowledging the extensive revisions made to the manuscript. Attached is a cover letter with responses to the comments of the Academic Editor.

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

 No Comments.

Author Response

We thank Reviewer 4 for the time invested in reviewing the manuscript and for confirming that no further comments remain. Attached is a cover letter with responses to the comments of the Academic Editor.

Author Response File: Author Response.pdf