The effect of beaver ponds on water physico-chemical composition in the Carpathians (Poland and Slovakia)

Wąs, Joanna; Siwek, Joanna P.; Kijowska-Strugała, Małgorzata; Gorczyca, Elżbieta

doi:10.5194/egusphere-2025-1184

Preprints

https://doi.org/10.5194/egusphere-2025-1184

Preprints

04 Apr 2025

| 04 Apr 2025

The effect of beaver ponds on water physico-chemical composition in the Carpathians (Poland and Slovakia)

Joanna Wąs, Joanna P. Siwek, Małgorzata Kijowska-Strugała, and Elżbieta Gorczyca

Abstract. In recent decades, the population of the Eurasian beaver (Castor fiber) has undergone a rapid recovery from near extinction to abundance across vast areas of Europe. The ability of this species to build dams makes its reintroduction an important environmental factor in the recolonised areas. This study investigated nine beaver-inhabited streams distributed across the Western Carpathians to assess the effects of geomorphic type, age of beaver pond sequence and seasonality on the physico-chemical changes to water in and below beaver ponds.

In general, greater reductions in NO₃^-and SO₄^2-were observed with increasing temperatures during the warm period (spring–summer). A comparison of two distinct types of beaver ponds revealed that there was a greater decrease in NO₃^- and Ca²⁺ in overflowing ponds and a greater decrease in pH downstream to these ponds compared to in-channel reservoirs. Beaver pond sequence age was positively related to decrease in dissolved oxygen, SO₄^2- and pH. Biogeochemical processes involving organic matter accumulated in beaver ponds, that include decomposition, aerobic/anaerobic oxidation and CaCO₃ precipitation, are responsible for changes of these physico-chemical parameters in stream water. The natural development of extensive beaver ponds and their persistence may be crucial for sustaining water purification processes. Further research based on a more frequent sampling strategy should aim to identify the biogeochemical processes that occur in beaver ponds under specific hydro-meteorological conditions: during low flow periods, snowmelt and rainfall events.

Received: 12 Mar 2025 – Discussion started: 04 Apr 2025

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Preprint (PDF, 3079 KB)

Notice on discussion status
The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.
Preprint (3079 KB)

Download & links

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Journal article(s) based on this preprint

07 Aug 2026

The effect of beaver dams on the physico-chemical properties of stream water in the Carpathians (Poland and Slovakia)

Joanna Wąs, Joanna P. Siwek, Małgorzata Kijowska-Strugała, and Elżbieta Gorczyca

Biogeosciences, 23, 5437–5459, https://doi.org/10.5194/bg-23-5437-2026,https://doi.org/10.5194/bg-23-5437-2026, 2026

Short summary

Joanna Wąs, Joanna P. Siwek, Małgorzata Kijowska-Strugała, and Elżbieta Gorczyca

Interactive discussion

Status: closed

CC1:
'Comment on egusphere-2025-1184', Mateusz Stolarczyk, 11 Jun 2025

The article “The effect of beaver ponds on water physico-chemical composition in the Carpathians (Poland and Slovakia)” examines how Eurasian beaver dams influence stream water chemistry in the Western Carpathians. Based on seasonal sampling of streams with varying geomorphology, pond types, and dam ages, the study found that beaver ponds significantly alter water parameters by reducing dissolved oxygen and nitrate levels, promoting sulfate reduction, increasing ammonium concentration, lowering pH, and in some cases facilitating calcium removal through precipitation. These effects are most pronounced during warm seasons and in older or overflowing ponds, where microbial and chemical processes intensify. The results underscore the important role of beaver ponds in modifying water quality. These valuable findings enhance our understanding of the complex influence of beaver activity on various environmental components, emphasizing the ecological significance of beaver-engineered habitats.
However, after reading the above study, several questions and comments arise regarding the conducted research, which may serve as useful guidance for the implementation of other research topics related to this subject:
1. Following an analysis of the results, conclusions, and the overall content of the paper, I would suggest considering a slight modification of the title (if it possible) to more accurately reflect the impact of beaver dams on the physico-chemical properties of stream water, rather than on the ponds as a whole. In my opinion, the ponds are a secondary consequence of dam construction, with the dam itself serving as the primary factor driving environmental modifications.
2. Sampling was conducted seasonally and did not account for hydrologically dynamic periods such as snowmelt or heavy rainfall events, which are known to cause substantial shifts in water flow and sediment transport. These events could temporarily alter or intensify the biogeochemical processes within beaver ponds, potentially affecting nutrient fluxes, oxygen dynamics, and pH levels. Incorporating high-resolution or event-based sampling in future research would provide a more comprehensive understanding of the short-term yet ecologically significant changes associated with such hydrological disturbances.
3. While the study offers valuable interpretations regarding biogeochemical processes in beaver ponds, it relies largely on indirect inference of organic matter dynamics and microbial activity. Direct measurements of parameters such as microbial community composition and microbial activity were not included, which limits the ability to fully validate the proposed mechanisms driving changes in water chemistry. Incorporating such data - through methods like enzyme activity assays or DNA sequencing - would strengthen the conclusions by providing process-level evidence and a clearer understanding of the microbial and organic matter contributions to the observed physico-chemical transformations.
4. The manuscript appropriately highlights the role of aquatic and riparian vegetation in nutrient cycling, particularly in relation to nitrate (NO₃⁻) reduction during the spring–summer period; however, this aspect was not quantitatively addressed. I agree that plant uptake likely contributes to observed nutrient dynamics, but this raises further questions about how pond age influences species diversity and vegetation cover? Specifically, is there a noticeable difference in plant growth or species composition between younger and older ponds, especially during the summer months? Addressing these questions with quantitative vegetation data would enhance understanding of biotic factors driving nutrient transformations in beaver ponds.

Citation: https://doi.org/10.5194/egusphere-2025-1184-CC1
- AC1: 'Reply on CC1', Joanna Wąs, 03 Nov 2025
  
  Thank you very much for voluntarily reviewing the paper. All the suggestions were carefully considered and we found them very helpful when revising the discussion section of the manuscript. You will find our response and the proposed changes to the manuscript below:
  1. Following an analysis of the results, conclusions, and the overall content of the paper, I would suggest considering a slight modification of the title (if it possible) to more accurately reflect the impact of beaver dams on the physico-chemical properties of stream water, rather than on the ponds as a whole. In my opinion, the ponds are a secondary consequence of dam construction, with the dam itself serving as the primary factor driving environmental modifications.
  Answer: Thank you for sharing your observation on the differences between beaver dams and ponds as environmental factors. Although a dam is essential for creating a pond, we believe that both ‘dams’ and ‘ponds’ are equally appropriate in this context, given that the processes underlying observed changes are related to stagnant water. However, we realise that using the term 'ponds' rather than 'dams' could suggest that the study focused mainly on pond properties and not on the rest of the stream. To avoid such confusion we will consider changing the title slightly, e.g. to ‘The effect of beaver dams on water physico-chemical composition in the Carpathians (Poland and Slovakia)’
  2. Sampling was conducted seasonally and did not account for hydrologically dynamic periods such as snowmelt or heavy rainfall events, which are known to cause substantial shifts in water flow and sediment transport. These events could temporarily alter or intensify the biogeochemical processes within beaver ponds, potentially affecting nutrient fluxes, oxygen dynamics, and pH levels. Incorporating high-resolution or event-based sampling in future research would provide a more comprehensive understanding of the short-term yet ecologically significant changes associated with such hydrological disturbances.
  Answer: Thank you for this valuable comment. We recognise that seasonal sampling did not capture short-term hydrological extremes, such as snowmelt or heavy rainfall, which can cause temporary but marked fluctuations in flow, sediment transport and nutrient dynamics. However, our sampling focused on capturing representative seasonal conditions under baseflow rather than event-driven variability, as the primary objective of the study was to assess the overall, sustained impact of beaver ponds on stream water chemistry under typical flow conditions. Nevertheless, we agree that high-resolution event-based sampling will provide valuable information on short-term biogeochemical responses and should be included in future studies.
  3. While the study offers valuable interpretations regarding biogeochemical processes in beaver ponds, it relies largely on indirect inference of organic matter dynamics and microbial activity. Direct measurements of parameters such as microbial community composition and microbial activity were not included, which limits the ability to fully validate the proposed mechanisms driving changes in water chemistry. Incorporating such data - through methods like enzyme activity assays or DNA sequencing - would strengthen the conclusions by providing process-level evidence and a clearer understanding of the microbial and organic matter contributions to the observed physico-chemical transformations
  Answer: Thank you for this important suggestion regarding the expansion and refinement of our future research. In our research, we did not measure the composition of microbial communities or the activity of microorganisms. The interpretations presented in this study are based on indicators derived from physicochemical parameters such as EC, pH, oxygen, and nutrients. While these variables reflect integrated biogeochemical responses at the ecosystem level, we acknowledge that the absence of direct measurements of microbial community composition or enzymatic activity limits our ability to conclusively verify the proposed mechanisms. Using molecular and biochemical methods, such as enzymatic activity tests or DNA sequencing, would provide evidence at the process level and offer a more complete understanding of the impact of microorganisms and organic matter on the observed changes in beaver ponds.
  4. The manuscript appropriately highlights the role of aquatic and riparian vegetation in nutrient cycling, particularly in relation to nitrate (NO₃⁻) reduction during the spring–summer period; however, this aspect was not quantitatively addressed. I agree that plant uptake likely contributes to observed nutrient dynamics, but this raises further questions about how pond age influences species diversity and vegetation cover? Specifically, is there a noticeable difference in plant growth or species composition between younger and older ponds, especially during the summer months? Addressing these questions with quantitative vegetation data would enhance understanding of biotic factors driving nutrient transformations in beaver ponds.
  Answer: We appreciate this insightful comment. The study did indeed highlight the potential role of aquatic and riparian vegetation in nutrient cycling, particularly with regard to nitrate reduction in spring and summer. However, as the main focus was on the physicochemical properties of the water, quantitative data on vegetation (e.g. biomass, species composition or percentage cover) were not collected. Future studies combining vegetation surveys, biomass estimates, and species-level analyses will help clarify the influence of biotic factors on nutrient cycling in beaver ponds.
  
  Citation: https://doi.org/10.5194/egusphere-2025-1184-AC1
RC1:
'Comment on egusphere-2025-1184', Gabriel Singer, 30 Sep 2025

General comments:
The abstract currently cannot stand on its own, i.e., it is not understandable on its own without having read more text of the manuscript. Needs to be rewritten. In the introduction please be much clearer about the actual research questions/hypotheses of your study. For instance, it is unclear what is “an environmental response to beaver dams”, meaning a change in upstream-downstream gradients in physicochemistry? Or do you mean a change after-before (in a temporal sense!) at larger spatial scale? Take care to robustly derive your study´s questions from past work AND explain them in an understandable way. The three questions are currently a slightly loose list without much connection and with poor derivation.
A weakness of the study is its focus on upstream-downstream comparisons without any controls. Water chemical conditions may change a lot between two sites along a stream continuum, independent of beaver dams. The study investigates only stream sites with beaver dams and compares data from above and below. There was no investigation before beaver dams were constructed, nor are there any control sites where no dams exist. I acknowledge that a full before-after-control-impact (BACI) study design was deemed infeasible, but such shortcomings of the study design must at least be discussed. Typically, in studies lacking controls, a higher level auf causality is needed for hypothesized processes, here in this case all processes that may affect physicochemical conditions. For ions without much explanation for change in beaver ponds, the current design does not allow to strongly pin down beavers as being the origin for any change, especially if the direction of change was not consistent. This should be reflected in the discussion.
The paper also suffers from a relatively inefficient presentation of results. There are too many figures. Consider moving some figures to the Supplement and look for ways to more efficiently present data (e.g. do not use separate tables for PCA variable loadings, but rather include these as a separate panel in the figure). Consider removing, shortening the more detailed results on dam cascades or vertical profiles or move to supplement to give the paper more focus. Some plots are simply superfluous and don´t present much exciting (e.g. SEC in Fig 10).
The discussion of the various anaerobic processes should be organized along the redox ladder.
The lower differences from upstream to downstream compared to upstream-pond may to some extent be related to the fact that downstream samples are from a well-mixed turbulent stream but pond sampling was likely not done in a spatially “integrative” way. Thus, pond samples may not represent average pond conditions. This needs discussion, also in light of anyway existing vertical profile data.
I strongly recommend authors to discuss the observed beaver pond effects in the light of residence time of water in the ponds. Doing this may offer a chance to integrate pond/stream size effects and seasons (differences in stream discharge). It may even be able to integrate geomorphic pond type (for which I find any observed differences inadequately derived as well as discussed anyway). Residence time in concert with temperature may be able to explain a lot.
Overall, I urge authors to shorten the paper and find more efficient wording in particular in the discussion. This will make the paper more accessible to a larger audience.
More specific comments:
55: “attenuation of downstream …. concentrations” – please express more clearly, do you mean compared to upstream? Then is it measured relative to a control? Or does this imply some before-after comparison as well? The same problem applies to the text in the abstract.
58: maybe two sentences instead of semicolon?
60-65: Could be worthwhile to dive deeper into whether these studies aiming at investigating “reductions” and thus working with upstream-downstream differences actually follow BACI-study designs to statistically robust conclusions.
66: maybe explain a bit better why and how geomorphic factors could actually have such an influence.
115: explain pond-to-channel ratio
155: were both the original data and the differences (percent change) used in the same PCA? Please provide justification why this procedure would make sense.
161: There is no N in this formula.
171: Variables are all concentrations or differences thereoef. I suggest to not mix untransformed with log-transformed variables in the same PCA. Concentrations may just all be log-transformed, as they are often log-normal distributed anyway. I cannot see any step in PCA where significance tests would become relevant. And to use significance for normality tests is a questionable approach anyway, as normally one “hopes” for high P to identify a variable as “not identifiable as not normal”. Try to work without those debated and criticized tests. Same applies to Shapiro-Wilk used prior to ANOVA.
178-180: Before doing hard-core hypothesis tests, please cleanly derive why these hypotheses would anyway make sense to be tested in the intro. The derivation of your questions and subsequent hypotheses needs to be clearer. Currently it seems that the factor “pond type” was defined after the study was done.
199: Reword “controls the values”, not appropriate wording to describe PCA results.
Fig 3: What are the various regions indicating? Information for A,A/B, etc. missing from the legend. I suggest you also present factor loadings as arrows, could be done in a separate plot (panel).
Fig 4: This figure does not allow to recognize which points belong to the same pond and date, i.e. the paired nature of the data is not shown (or “used” because in fact its consideration could make data analysis stronger). Consider working with differences or connections between points.
Table 4: This table presents A LOT OF P-values. Since the same dataset was used multiple times, some P-value adjustment (e.g. Bonferroni) should be done. Certainly, P-values >0.05 (<0.1) should not be reported. I fear that not too many P-values will remain significant then, making clear why this strategy of extensive dataset testing may not be the best. Also, stronger analysis may take into account multiple factors (pond type, age, seasons) at once.
269: Was this PCA entirely done with “differences”? A graph of variable scores as arrows as a panel in Fig 6 may be more efficient than Table 5. I suggest to not speak of “water samples” in the legend of Fig 6 as two samples were needed to create one point in these graphs, as I understand. The site codes in fig 6 are hard to distinguish.
285: Similar comments apply to this PCA as to the previously presented one.
Chapter 3.4. needs to be incorporated in the Discussion.
354: pointing out high TOC only makes sense in comparison to upstream.
365: any idea about CH4 in these ponds? It would make sense to organize the discussion of the various anaerobic processes along the redox ladder.
366: ammonification is a fairly specific process and not a synonym for OM oxidation.
370: Prefer “aerobic/anaerobic” for microbial processes, but “oxic/anoxic” or low/high redox potential for environmental conditions.
Fig 12: Very appreciated conceptual figure. Consider using coloring the pond sequence in a color gradient instead of arrows?

Citation: https://doi.org/10.5194/egusphere-2025-1184-RC1
- AC2:
  'Reply on RC1', Joanna Wąs, 03 Nov 2025
  Thank you for taking your time to review the paper. We are grateful for all the comments on our work - they helped us improve both the structure and content of the manuscript. Below, we address all comments point-by-point discussing the subsequent modifications. All of the comments and suggestions were carefully considered, and the manuscript has been revised accordingly.
  GENERAL COMMENTS
  1. The abstract currently cannot stand on its own, i.e., it is not understandable on its own without having read more text of the manuscript. Needs to be rewritten.
  Answer: Some parts of the abstract were previously omitted to make it more concise, however, you are right that this reduced the overall clarity of the study and led to the omission of some important information. We have now integrated additional details, such as clearer description of the methods, and partially rewritten the abstract into more structured paragraphs. The current version of the abstract is as follows:
  ‘Abstract. In recent decades, the population of the Eurasian beaver (Castor fiber) has undergone a rapid recovery from near extinction to abundance across vast areas of Europe. The ability of this species to build dams and the resulting impact on ecological processes makes the reintroduction of beavers an important environmental factor in the recolonised areas. Although the effects of beaver dams on stream water chemistry have been studied extensively, general conclusions are often limited by site specific conditions. Therefore, the aim of this study was to assess the impact of a geomorphic type and an age of beaver pond sequences on physico-chemical water properties in and below beaver ponds compared to upstream channel sections throughout the seasons.
  This study was conducted on nine beaver-inhabited streams distributed across the Western Carpathians (Poland and Slovakia). The beaver ponds were classified into two geomorphic categories: overflowing river banks or confined to the river channel. The age of the ponds varied from <2 years to >10 years. Water samples were collected above, within, and below beaver pond sequences during four seasons in 2022-2023.
  In general, a greater decrease in NO₃^- and SO₄^2-concentrations and increase in water temperatures within and below the ponds were observed during the warm period (spring–summer) than the rest of the year (autumn-winter). A comparison between two geomorphic types of beaver ponds revealed that there was a greater decrease in NO₃^- and Ca²⁺ concentrations and a greater decline in pH downstream of overflowing ponds compared to in-channel reservoirs. The age of beaver pond sequences was found to be a more significant factor than the pond type. It was positively related to decrease in dissolved oxygen, SO₄^2- and pH in the ponds and decrease in SO₄^2- in downstream channel sections.
  The role of the age of beaver ponds suggest that their long-term persistence may be important for sustaining water purification processes. However, further research based on a larger dataset and more frequent sampling strategy is needed to identify the biogeochemical processes occurring in beaver ponds under specific hydro-meteorological conditions: during low flow periods, snowmelt and rainfall events.’
  
  2. In the introduction please be much clearer about the actual research questions/hypotheses of your study. For instance, it is unclear what is “an environmental response to beaver dams”, meaning a change in upstream-downstream gradients in physicochemistry? Or do you mean a change after-before (in a temporal sense!) at larger spatial scale?
  Answer: We removed this general sentence and replaced it with a paragraph describing the idea in more detail. This includes a description of the role of specific beaver pond characteristics (age and geomorphic type) on the physico-chemical parameters, from which we further derived our study questions. The questions were also reformulated with more straightforward wording. It is now:
  ‘The objective of this study was to determine the effects of beaver dams on physico-chemical water properties in mountain streams of the Western Carpathians (Poland and Slovakia), with regard to seasonality and pond characteristics. The paper focuses on answering the following questions:
  How do physico-chemical water parameters change between river sections located upstream, within, and downstream of the beaver ponds, and what biogeochemical processes underlie these differences?
  
  How do these differences vary seasonally?
  
  How are these differences affected by the geomorphic type and age of the beaver ponds?’
  
  3. Take care to robustly derive your study´s questions from past work AND explain them in an understandable way. The three questions are currently a slightly loose list without much connection and with poor derivation.
  Answer: Thank you for pointing out this aspect that needed improvement. Some information was missing, while other (e.g. brief reasoning behind the pond classification) were scattered in the manuscript. We reorganized part of the introduction into three paragraphs that correspond more clearly with the research questions. We took special care to explain potential and already documented impacts of the age and geomorphic character of beaver ponds.
  
  4. A weakness of the study is its focus on upstream-downstream comparisons without any controls. Water chemical conditions may change a lot between two sites along a stream continuum, independent of beaver dams. The study investigates only stream sites with beaver dams and compares data from above and below. There was no investigation before beaver dams were constructed, nor are there any control sites where no dams exist. I acknowledge that a full before-after-control-impact (BACI) study design was deemed infeasible, but such shortcomings of the study design must at least be discussed. Typically, in studies lacking controls, a higher level of causality is needed for hypothesized processes, here in this case all processes that may affect physicochemical conditions. For ions without much explanation for change in beaver ponds, the current design does not allow to strongly pin down beavers as being the origin for any change, especially if the direction of change was not consistent. This should be reflected in the discussion.
  Answer: We agree that the lack of data from before the construction of the beaver dams limits our ability to clearly determine the cause of changes in the chemical composition of the water. However, this study used a comparative ‘upper river – lower river’ method, where the upper river sections were used as reference sections. This research approach is commonly used in field studies where it is impossible to obtain data 'before' the construction of beaver dams for logistical or environmental reasons (e.g. Green and Westbrook, 2009; Puttock et al., 2017). The revised manuscript has been supplemented with a discussion of these limitations, indicating that the presented results are correlational in nature and should be interpreted as indicating potential links between the presence of beaver dams and changes in water physicochemical properties. Additionally, the ‘Study limitations’ section highlights that a complete Before–After–Control–Impact (BACI) design would be the most appropriate approach to testing the causal effects of beaver activity (Puttock et al., 2020; Dittbrenner et al., 2022), but its application to natural river ecosystems is often impractical. It also highlights directions for future research that combine a spatial approach with a temporal component, which would allow for a more definitive assessment of the processes occurring in beaver-modified systems.
  4.1. Study limitations
  ‘... Although a Before–After–Control–Impact (BACI) model would be the most appropriate approach for studying the impact of beavers on the chemical composition of water, it was not possible to use this method in this area due to the lack of investigations prior to dam construction. Instead, the study focused on spatial and temporal contrasts throughout the year and used a comparative ‘upper-lower river’ method, where upstream sections were used as reference sections. This research approach is commonly used in field studies to assess hydrochemical gradients generated by beaver dams when it is logistically or environmentally impossible to obtain ‘before’ data on dam construction (Green and Westbrook, 2009; Klotz, 2010; Fuller and Peckarsky, 2011; Puttock et al., 2017; Stevenson et al., 2022). It enables the identification of local hydrological and chemical effects caused by beaver damming while minimising the confounding influence of catchment heterogeneity. Although this approach lacks independent control sections, as in a full BACI model, it provides robust comparative evidence for evaluating the magnitude and spatial extent of changes induced by beaver activity under natural field conditions. Future research should be more detailed and extensive to enable the application of the BACI model.…’
  Dittbrenner, B. J., Schilling, J. W., Torgersen, C. E., & Lawler, J. J. (2022). Relocated beaver can increase water storage and decrease stream temperature in headwater streams. Ecosphere, 13(7), e4168.
  Green, K. C., & Westbrook, C. J. (2009). Changes in riparian area structure, channel hydraulics, and sediment yield following loss of beaver dams. Journal of Ecosystems and Management, 10 (1):68-79.
  Puttock A,. Graham H.A., Cunliffe A.M., Elliott M., Brazier R.E. (2017). Eurasian beaver activity increases water storage, attenuates flow and mitigates diffuse pollution from intensively-managed grasslands. Science of the Total Environment, 576: 430–443.
  Puttock, A., Graham, H. A., Ashe, J., Luscombe, D. J., & Brazier, R. E. (2020). Beaver dams attenuate flow: A multi‐site study. Hydrological processes, 35(2), e14017.
  
  5. The paper also suffers from a relatively inefficient presentation of results. There are too many figures. Consider moving some figures to the Supplement and look for ways to more efficiently present data (e.g. do not use separate tables for PCA variable loadings, but rather include these as a separate panel in the figure). Consider removing, shortening the more detailed results on dam cascades or vertical profiles or move to supplement to give the paper more focus. Some plots are simply superfluous and don´t present much exciting (e.g. SEC in Fig 10).
  Answer: We took your advice on incorporating information from the tables resulting from PCA analysis into the figures. We also managed to improve efficiency by combining the two sites and multiple variables on the graphs depicting longitudinal and vertical profiles. Despite this, some of them can be moved into an Appendix. We will also consider removing part of the section ‘3.1.’ concerning PCA analyses of the physico-chemical background to shorten and focus the text. This would include removing one figure.
  
  6. The discussion of the various anaerobic processes should be organized along the redox ladder.
  Answer: We agree with the reviewer. Despite the lack of redox potential measurements, we attempted to organise the discussion of the aerobic/anaerobic processes along the redox ladder based on changes in the physico-chemical properties of the water, which indicate the occurrence of these processes. The example is below:
  ‘... As a result, there are intensive physico-chemical transformations in the stream water flowing into the ponds in the warm seasons. The increase in NH₄⁺ concentrations and the decrease in NO₃^- concentrations in the studied beaver ponds were particularly prominent in the warm period of the year due to the intensive ammonification (NH₄⁺ production) and denitrification processes (NO₃^- depletion). This is consistent with the findings of Margolis et al. (2001) and Law et al. (2016): the greatest NO₃^- decrease downstream to beaver ponds occurs in the summer. NO₃^- ion is one of the most effective electron acceptors (oxidants) in the process of anaerobic oxidation of organic matter; therefore, NO₃ is the preferred oxidising agent for anaerobic microbes. The process of NO₃^- reduction (denitrification) under anaerobic oxidation of organic matter occurs at a relatively high redox potential level on the redox ladder (Anderson and Fidel, 2025). However, it should be noted that the decrease in NO₃^- concentrations in the beaver ponds and streams below the ponds in the summer is partly related to the nitrogen fixation process, which is associated with increased microbial activity in sediments mediated by anaerobic conditions, as well as uptake by aquatic vegetation (Naiman and Melillo, 1984; Maret et al., 1987; Songster-Alpin and Klotz, 1995). Furthermore, the decrease in SO₄^2- concentrations in the studied ponds and downstream of the beaver dams was also observed during the warm period. This is likely due to the high rate of anaerobic oxidation of organic matter during that time. According to Smith et al. (1991), Cirmo and Driscoll (1993), and Margolis et al. (2001), the intensive anaerobic oxidation of organic matter can occur despite the relatively low redox potential of the water during the low-flow conditions of summer. During that time anaerobic microbes used even relatively inefficient oxidants, such as SO₄^2-, in the anaerobic oxidation of organic matter due to the low concentration of NO₃^- in the studied waters; the average concentration of NO₃^- in ponds and streams below them was very low during the summer, the average was up to 1.5 mg/L.’
  
  7. The lower differences from upstream to downstream compared to upstream-pond may to some extent be related to the fact that downstream samples are from a well-mixed turbulent stream but pond sampling was likely not done in a spatially “integrative” way. Thus, pond samples may not represent average pond conditions. This needs discussion, also in light of anyway existing vertical profile data.
  Answer: We agree that due to the spatial complexity of beaver ponds it’s difficult to derive general conclusions concerning the whole pond without comprehensive sampling across the system. The regional scope of this research didn’t allow for more detailed sampling at every pond each season. However, we applied consistent criteria for the location of the ‘pond’ sampling point across the whole study area and time, to increase their comparability. The detailed description that would help interpret the data was missing in the methodology section, therefore we clarified the sampling locations in the revised manuscript:
  ‘...The ‘pond’ sample was taken from the main channel section of the pond, i.e. excluding channel margins and backwaters. It was collected approximately 1m above the dam, from the middle of a water column….’
  We have complemented the discussion with a description of the identified constraints:
  ‘...Although the sample locations were standardised to a certain extent, they may not reflect typical stream or pond conditions. Stream samples are more likely to be well-mixed and representative due to turbulent flow conditions. This is also evidenced by the homogeneity of most longitudinal profiles upstream and downstream of beaver ponds. However, the spatial distribution of the pond samples may be more biased as it did not cover the full spectrum of hydrological units observed within the ponds (Majerova et al., 2020). Additionally, as the samples were taken from the middle of the water column, they do not represent the surface and bottom water layers, which according to the vertical sampling scheme may substantially affect observed physico-chemical parameters...’
  
  8. I strongly recommend authors to discuss the observed beaver pond effects in the light of residence time of water in the ponds. Doing this may offer a chance to integrate pond/stream size effects and seasons (differences in stream discharge). It may even be able to integrate geomorphic pond type (for which I find any observed differences inadequately derived as well as discussed anyway). Residence time in concert with temperature may be able to explain a lot.
  Answer: Thank you for this valuable suggestion. Although water residence time plays an important role in shaping the observed processes in beaver ponds, it was only briefly mentioned in the text. While modifying the introduction, we included it as one of the pond characteristics that may differ between the two geomorphic types, instead of the description previously made in the methodology section. The ‘age’ and ‘geomorphic type’ are not the direct causes of the difference, but this classification is likely connected to different hydrological and thermal conditions that influence biogeochemical processes. Much of the discussion on the impact of beaver pond age and geomorphological type was reorganised and supplemented with information regarding effects of water residence time and temperature, to clarify processes that may be responsible for the observed differences between the ponds.
  
  9. Overall, I urge authors to shorten the paper and find more efficient wording in particular in the discussion. This will make the paper more accessible to a larger audience.
  Answer: After reading the manuscript with a fresh perspective, we agree that it was somewhat overloaded with information. We revised the text and figures carefully. We removed part of section 3.1 from the results completely, because it didn’t carry information that would be necessary for understanding the impact of beavers, which is the main focus of the paper. We also removed or rephrased some sentences that were redundant or purely structural, especially in the discussion and conclusions. Moreover, the combination of several variables and study sites on the plots decreased the size of the figures. An incorporation of the PCA loadings into the figures decreased an overall number of figures and tables. All the changes resulted in the shortening of the main body of the manuscript by a few pages.
  
  SPECIFIC COMMENTS:
  55: “attenuation of downstream …. concentrations” – please express more clearly, do you mean compared to upstream? Then is it measured relative to a control? Or does this imply some before-after comparison as well? The same problem applies to the text in the abstract.
  Answer: We clarified the upstream-downstream relation in this fragment of the text, as well as in the description of our research objectives in the abstract.
  
  58: maybe two sentences instead of semicolon?
  Answer: Thank you, changed.
  
  60-65: Could be worthwhile to dive deeper into whether these studies aiming at investigating “reductions” and thus working with upstream-downstream differences actually follow BACI-study designs to statistically robust conclusions.
  Answer: We used the word 'reduction' incorrectly. It should have been 'decrease' — we meant a decrease in concentration. This error appeared in several other places in the text and has now been corrected.
  In this study, it was not possible to use the BACI model due to, among other reasons, the lack of pre-dam surveys. The study focused on spatial contrasts and seasonal changes within a single beaver-affected streamway. In our study, we used a stream-scale gradient approach, with sampling points located upstream of the beaver pond, within the pond, and downstream of the dam. This spatial model, commonly referred to as an upstream-downstream or longitudinal comparison (e.g., Klotz 2010; Fuller and Peckarsky 2011; Stevenson et al. 2022), has been widely adopted to assess in situ hydrochemical gradients generated by beaver dams. It allows for the identification of local effects of dam-induced hydrological and chemical transformations while minimizing the confounding effects of catchment heterogeneity. Although this approach does not include independent control sections, as in the full BACI project, it provides robust comparative evidence for assessing the direction and magnitude of changes induced by beaver activity in natural field conditions. Therefore, our sampling framework is consistent with established methods for assessing changes in stream water chemistry and sediment dynamics induced by beavers.
  Fuller, M. R., & Peckarsky, B. L. (2011). Does the morphology of beaver ponds alter downstream ecosystems?. Hydrobiologia, 668(1), 35-48.
  Klotz, R. L. (2010). Reduction of high nitrate concentrations in a Central New York State stream impounded by beaver. Northeastern Naturalist, 17(3), 349-356.
  Stevenson, J. R., Dunham, J. B., Wondzell, S. M., & Taylor, J. (2022). Dammed water quality—Longitudinal stream responses below beaver ponds in the Umpqua River Basin, Oregon. Ecohydrology, 15(4), e2430.
  
  66: maybe explain a bit better why and how geomorphic factors could actually have such an influence.
  Answer: Thank you for this suggestion. We rewrote a part of the Introduction and included a more detailed explanation of the expected influence of beaver pond characteristics in this section:
  ‘…The identification of factors reflecting the diversity of beaver ponds and their effects on water chemistry, would improve the understanding of the effects of beaver presence in river systems. The age of beaver ponds was commonly proposed as a such feature in previous studies on upstream-downstream gradients of physico-chemical parameters in beaver-occupied streams. For example Bason et al. (2017) found a greater reduction of suspended solids in older dam sequences compared to younger ones. Metal sequestration rates and a sediment organic matter content inside beaver ponds also rise with the pond age according to the study of Murray et al. (2021). Hydrological and thermal differences were observed between various geomorphic units within beaver ponds, representing e.g. main channel, its margins and backwater areas (Majerova et al., 2020). The share of such units, especially related to the floodplain inundation, was then suggested as a potential geomorphic factor that could explain differences between study sites (Murray et al., 2021). Spatially expansive beaver ponds are expected to facilitate a higher proportion of low velocity areas, increase in water residence time, overbank sedimentation and growth of aquatic vegetation, compared to the ponds constrained to the river channel. Both beaver pond age and its geomorphic type may correspond with the prolonged sediment-water interactions resulting in the enhanced biogeochemical processes…’
  
  115: explain pond-to-channel ratio
  Answer: The term 'pond-to-channel ratio' was originally intended to describe the distinction between 'overflowing' and 'in-channel' beaver ponds. To improve clarity and ensure consistency with the terminology used throughout the manuscript, we have replaced it with 'type of beaver pond'.
  
  155: were both the original data and the differences (percent change) used in the same PCA? Please provide justification why this procedure would make sense.
  Answer: We used the original (raw) data of solute concentrations for PCA to determine the physico-chemical background of the studied streams. For this purpose, we used the raw concentrations of solutes in the stream water above the beaver ponds that were unaffected by beaver dams. These concentrations were standardised prior to analysis. However, to determine the factors influencing changes in solute concentrations in ponds and streams below due to beaver dams, we performed PCA using the percentage difference in concentrations between the stream water above the beaver dams and the pond water/stream water below the dams. In this case, we used concentration differences rather than raw concentrations because we focused not on the factors determining solute concentrations in pond and stream waters below the dams, but rather on the factors shaping the differences in these concentrations due to the presence of beaver dams. Ultimately, in light of suggestions to shorten the manuscript, we decided to remove the results of the factor analysis based on the raw data from Section 3.1. The physico-chemical background of the studied streams.
  
  161:There is no N in this formula.
  Answer: It was our mistake — N was included in the second formula, but not the first. We have corrected this.
  
  171: Variables are all concentrations or differences thereoef. I suggest to not mix untransformed with log-transformed variables in the same PCA. Concentrations may just all be log-transformed, as they are often log-normal distributed anyway. I cannot see any step in PCA where significance tests would become relevant. And to use significance for normality tests is a questionable approach anyway, as normally one “hopes” for high P to identify a variable as “not identifiable as not normal”. Try to work without those debated and criticized tests. Same applies to Shapiro-Wilk used prior to ANOVA.
  Answer: We have removed the results of the PCA analysis in which both non-logarithmic and logarithmic variables (Ca²⁺) were used (see section 3.1. The physico-chemical background of the studied streams). Therefore, we removed the information about PCA analysis on raw data from 2.3. Statistical analysis section. The information about significance threshold applies to all the other analyses, so we moved it to a different part of this section. To our knowledge, the use of the normality test remains a standard procedure, as it allows to select parametric or nonparametric tests for further analyses based on the data distribution. Both Shapiro-Wilk and Kolmogorov-Smirnov tests have been used prior to ANOVA in previous study on beaver impacts on water chemistry (Błędzki et al., 2011; Bylak et al, 2024).
  
  178-180: Before doing hard-core hypothesis tests, please cleanly derive why these hypotheses would anyway make sense to be tested in the intro. The derivation of your questions and subsequent hypotheses needs to be clearer. Currently it seems that the factor “pond type” was defined after the study was done.
  Answer: The indicated fragment has been removed from this part of the text. We have clarified our underlying assumptions in the introduction instead. Please see point ‘66:’ above.
  
  199: Reword “controls the values”, not appropriate wording to describe PCA results.
  Answer: We agree with this comment. We have corrected this throughout the text.
  
  Fig. 3: What are the various regions indicating? Information for A,A/B, etc. missing from the legend. I suggest you also present factor loadings as arrows, could be done in a separate plot (panel).
  Answer: The regions A/B represented geological locations (corresponding with dominance of shale or sandstone in the catchments), while C/D represented sampling seasons. However, we removed Figure 3 in order to shorten the article and focus the result section on the impact of beavers.
  
  Fig. 4: This figure does not allow to recognize which points belong to the same pond and date, i.e. the paired nature of the data is not shown (or “used” because in fact its consideration could make data analysis stronger). Consider working with differences or connections between points.
  Answer: The purpose of this figure was not to distinguish solute concentrations in detail at each location and sampling time, but rather to illustrate a general pattern. The figure allows to distinguish between the solute concentrations in ponds and streams below the beaver dams (represented by squares and crosses, respectively) in relation to the stream above the pond, and categorises them according to the age of the beaver dam: less than three years old, four to nine years old, and more than ten years old (represented by small red signatures, larger yellow signatures, the largest green signatures, respectively). Inclusion of the season as another variable or links between the points would make the figure illegible. However, in most cases the relations between paired data (pond and downstream differences at particular location and season) can be distinguished on the figure. The points representing the same sampling site at the same time are located at the same position on the horizontal axis (which results from the same inlet value), therefore vertical distance between both points illustrates the difference.
  
  Table 4:This table presents A LOT OF P-values. Since the same dataset was used multiple times, some P-value adjustment (e.g. Bonferroni) should be done. Certainly, P-values >0.05 (<0.1) should not be reported. I fear that not too many P-values will remain significant then, making clear why this strategy of extensive dataset testing may not be the best. Also, stronger analysis may take into account multiple factors (pond type, age, seasons) at once.
  Answer: We didn’t consider results with p<0.1 statistically significant. They served as additional information and were presented in the table in regard to a scientific discussion on the arbitrariness of significance thresholds. However, we agree that their inclusion was unnecessary and could distract from the interpretation of truly significant results. Therefore, only results meeting the conventional significance threshold (p < 0.05) were retained in the revised table. Moreover, in response to the very appreciated comment concerning multiple analyses on the same dataset we applied recommended correction to improve the interpretation of the results and updated methodology section:
  ‘...The significance level of p<0.05 was used for all the calculations. Additionally, to reduce the risk of family-wise error for multiple comparisons based on the same dataset, we applied a Bonferroni correction to the significance threshold (p<0.017)...’
  
  269, 285: Was this PCA entirely done with “differences”? A graph of variable scores as arrows as a panel in Fig 6 may be more efficient than Table 5. I suggest to not speak of “water samples” in the legend of Fig 6 as two samples were needed to create one point in these graphs, as I understand. The site codes in fig 6 are hard to distinguish.
  Answer: The results of the PCA analysis that were based on raw data in Section 3.1 were removed to shorten the text. The remaining PCA analyses are both based only on concentration differences (%), so subsequent update of the methodology description made it less confusing. Exact information on the type of data used can be found in Section 2.3. Statistical analyses. As suggested, Table 5 has been incorporated into Figure 6 in graphical form as both contain the results of the same factor analysis (factor loadings). That applies to Table 6 and Figure 7 as well. Tables 5 and 6 have been moved to the supplements because, in addition to the factor loadings of F1 and F2, they also contain information about the factor loadings of F3. In the legend for Figure 6, we wrote 'water samples' because each point represents a single water sample, and the factor scores for factors 1 and 2 are marked for each sample. We have corrected the site codes in the figures relating to PCA.
  
  Chapter 3.4: needs to be incorporated in the Discussion.
  Answer: You are right. It became Chapter 4.1.
  
  354: pointing out high TOC only makes sense in comparison to upstream.
  Answer: We agree with this comment. We have corrected the text as follows:
  ‘...Detailed investigations carried out in July 2024 at sites 3 and 6 showed that beaver pond water contains higher concentrations of total organic carbon (TOC) compared to the stream water upstream. For example, the TOC concentration in the beaver pond at site 3 was found to be more than 13 mg/L, which was almost three times higher than in the stream flowing into the pond...’
  
  365: any idea about CH4 in these ponds? It would make sense to organize the discussion of the various anaerobic processes along the redox ladder.
  Answer: Unfortunately, we did not examine the concentrations of CO₂ and CH₄ in the stream water and pond water, so we cannot address this issue in the discussion. However, we realise that it is essential to take this into account when studying the impact of beaver ponds on the chemistry of the stream below in future research.
  
  366: ammonification is a fairly specific process and not a synonym for OM oxidation.
  Answer: We agree with this comment. The sentence has been revised as follows:
  ‘...The decomposition and mineralization of organic matter in the process of ammonification, under both aerobic and anaerobic conditions, leads to the production of NH₄⁺ ions in water….’
  
  370: Prefer “aerobic/anaerobic” for microbial processes, but “oxic/anoxic” or low/high redox potential for environmental conditions.
  Answer: We agree with this suggestion. We have corrected the terminology throughout the text.
  
  Fig. 12: Very appreciated conceptual figure. Consider using coloring the pond sequence in a color gradient instead of arrows?
  Answer: Thank you. As the original figure was found to be difficult to understand, we considered various ways of transforming it to improve its clarity. The use of arrows has been modified to indicate direction and the average magnitude of change observed in each period, type and age of beaver pond, instead of the simple indication of one difference resulting from a comparison. In this way, even though there is more information presented in the figure, the content of the image should be more clear. Please find the revised figure (Figure 11.pdf) in the attached document.
  
  Overall, we greatly appreciate this insightful review and we believe that the revised manuscript is substantially improved thanks to the consideration of all the suggestions.
  
  Citation: https://doi.org/10.5194/egusphere-2025-1184-AC2
RC2:
'Comment on egusphere-2025-1184', Miloš Rusnák, 21 Oct 2025
The study examines the impact of beaver ponds on biogeochemical processes and stream water chemistry. For testing, 9 mountain streams in Poland and Slovakia were selected. The study focuses on the physico-chemical parameters with a detailed sampling design and focuses on pond control effects, seasonality and typology. Based on the presented result, it is necessary to state that the manuscript is well written and structured. However, in several places (methodology and description of results), it contains a significant amount of detail that distracts from the main idea of the article. Conversely, some details and method explanations are missing. I also recommend revising several diagrams and charts, which would significantly improve the readability of the article. The manuscript requires small reconfiguration and improvement in readability. Hopefully, my comments will be helpful in that process.

For the paper, I have several general comments:
Study area (85-120): Here, I recommend the description of 2 factors, which are used for the interpretation of results in the discussion, and these are: geology and agricultural activities. Geology is completely missing and is described in the discussion that different structures affected stream chemistry. From human activities, it is important to add more information on how intensive agricultural activities are preserved in upstream conditions. Pastures, fertilisation or direct soil cultivation can affect chemical composition. The second important parameter is human presence in the upstream catchment area (population, number of buildings) that can affect a lot of selected chemical parameters. At the same time, it should be noted that the selected design of sampling strategy minimises ambiguities, but a broader description of these limits should be part of the discussion section.

Methods (145 - 150): For increasing the readability of the paper, I recommend adding an exact complex statement of all physico-chemical parameters with a description and interpretation of what the source is and how it is theoretically interpreted when its values change. Also, the exact morphological description of pond type, age, and other typologies that are used in interpretation should be stated here. For example, in the methods section, the type is defined as overflowing and in-channel, but in the Results section, it is interpreted as channel and overbank flow. Is it considered the same type of beaver pond?

Statistical analyses (170-180): Please add more information on how the clustering of the PCA analysis was performed, as is presented in Fig.3.

Discussion section (Fig. 12): I am not sure if I can understand the scheme in Fig. 12. Complex and comprehensive information output, what image (period, age, geomorphology type) represents and how this typology can affect chemical composition is missing. From the image, it is not clear how different periods, pond types and age are related to chemical water properties (which typology affects the increase or decrease of chemical composition) and what changes are related to which presented type.

Discussion: Even though there is a chapter in the Results section devoted to the limitations of the study, I would recommend moving it and expanding it in the Discussion section as a separate chapter.
Citation: https://doi.org/10.5194/egusphere-2025-1184-RC2
- AC3: 'Reply on RC2', Joanna Wąs, 03 Nov 2025
  
  Thank you for this review. We appreciate the chance to improve the manuscript following all the helpful comments and suggestions. We revised the paper accordingly. Please find our responses to each comment below:
  The study examines the impact of beaver ponds on biogeochemical processes and stream water chemistry. For testing, 9 mountain streams in Poland and Slovakia were selected. The study focuses on the physico-chemical parameters with a detailed sampling design and focuses on pond control effects, seasonality and typology. Based on the presented result, it is necessary to state that the manuscript is well written and structured. However, in several places (methodology and description of results), it contains a significant amount of detail that distracts from the main idea of the article. Conversely, some details and method explanations are missing. I also recommend revising several diagrams and charts, which would significantly improve the readability of the article. The manuscript requires small reconfiguration and improvement in readability. Hopefully, my comments will be helpful in that process.
  Answer: Thank you for the positive feedback and all the constructive remarks. We agree that the original manuscript was not very clear due to the amount of content and inefficient presentation of the results. However, we managed to shorten the text and reduce the amount of distracting or redundant information. The reduction included, e.g., removal of the part of Section 3.1. from the Results as it was found to be not relevant enough for the understanding of the effects of beaver activity. Section 2 Methods was slightly modified to eliminate repeated and include missing information, and to direct the reader to the information already present in the table. The overall readability was also enhanced by modifying the graphic materials, i.e., by combining several variables and the two study sites into the profiles, and by incorporating the PCA from the tables into the figures. The size and number of the figures was therefore reduced, while retaining the same amount of information and enhancing comparability and visual appeal.
  Study area (85-120): Here, I recommend the description of 2 factors, which are used for the interpretation of results in the discussion, and these are: geology and agricultural activities. Geology is completely missing and is described in the discussion that different structures affected stream chemistry. From human activities, it is important to add more information on how intensive agricultural activities are preserved in upstream conditions. Pastures, fertilisation or direct soil cultivation can affect chemical composition. The second important parameter is human presence in the upstream catchment area (population, number of buildings) that can affect a lot of selected chemical parameters. At the same time, it should be noted that the selected design of sampling strategy minimises ambiguities, but a broader description of these limits should be part of the discussion section.
  Answer: The geology of the study area was presented briefly in the first paragraph of section 2.1, including an explanation of its regional diversity. As part of the discussion, we refer to the geology at a similar regional level of detail. Unfortunately, based on the available data we could not provide a more detailed description of each river channel. However, we found that the reference made in the discussion could be confusing, as the phrase ‘correspond to various geological features’ implies that we are referring to more detailed components instead of the differences in general geological structure. We have modified this fragment to: ‘corresponded to the geological structure.’ Additional information on the water chemistry was also included:
  ‘...Streams draining areas underlain by shales and mudstones are typically characterised by higher concentrations of bicarbonates, calcium and magnesium ions due to increased weathering of carbonates and clay minerals, while streams in areas composed mainly of sandstone are usually more diluted and characterised by lower specific conductivity (e.g. Siwek et al., 2011; Solár et al., 2023)...’
  As suggested, we added a paragraph in the Section 2.1 Study area describing land use and some of the anthropogenic impacts:
  ‘...All of the catchments consisted of a mixture of forest and agricultural areas, i.e. arable lands and meadows. Despite the widespread abandonment of agricultural land and the generally low-to-moderate intensity of agricultural practices in the mountainous catchments (Pazúr et al., 2014; Kolecka et al 2017), the impact of these activities may be significant locally, as evidenced by the increase in nitrate concentrations observed in the Ondava catchment due to the use of fertilisers (Pekarova et al., 2006; Balejčíková et al., 2020). Overall, agricultural practices in the Polish and Slovakian Carpathians are mostly extensive, characterised by low–input management and mowing and grazing on a small-scale (Pazúr et al., 2014; Kolecka et al., 2017; Affek et al., 2023). According to national agricultural statistics (Central Statistical Office of Poland, 2025), the average consumption of mineral fertilisers (NPK) in the Małopolska and Podkarpackie provinces remains one of the lowest in the country (80–95 kg NPK ha⁻¹ year⁻¹ compared to the national average of ~125 kg ha⁻¹) and has shown little change over the last two decades. Similar situation is observed in the northern part of Slovakia where the average use of industrial fertilisers is less than 60 kg NPK ha⁻¹ year⁻¹ (Siman and Velísková, 2018). Consequently, the direct inflow of nutrients from fertilisers into spring watercourses is minimal, although local enrichment may occur near farms or areas of intensive grazing. In contrast, domestic wastewater was identified as the main source of stream water pollution in the rural areas of the Polish Carpathians (Siwek, 2020). The intensity of housing development across the study area ranged from negligible levels to artificial surfaces covering up to 4% of the area and containing a few hundred buildings…’
  Table 1 was also supplemented with information on the number of buildings in the studied catchments.
  The potential impact of catchment diversity on the obtained results was described in the discussion:
  ‘...As there was no identified relationship between the initial values and the percentage changes in the physico-chemical parameters, it was assumed that the observed changes would not differ significantly based on upstream conditions. This allowed various catchments with different land uses and levels of anthropogenic influence to be included in the study. However, it should be noted that the inclusion of the streams characterised by particularly high concentrations of certain components, e.g. highly anthropogenically modified, may have resulted in different observations...’
  
  Methods (145 - 150): For increasing the readability of the paper, I recommend adding an exact complex statement of all physico-chemical parameters with a description and interpretation of what the source is and how it is theoretically interpreted when its values change.
  Answer: The complex list of the measured parameters was previously presented on the bottom of Figure 2. However, you are right that listing those parameters in the text makes them more clearly visible. Therefore, we included the list at the beginning of the Methods section, before the detailed description of the sampling:
  ‘..The research followed a three-dimensional sampling scheme (Fig. 2). The following water parameters were measured: temperature (T), dissolved oxygen (DO), specific electrolytic conductivity (SEC), pH and the ionic composition (Ca²⁺, Mg²⁺, Na⁺, K⁺, NO₃^-, NH₄⁺, SO₄^2-, Cl^-)...’
  Although we agree that a description of each water parameter would be valuable, we are concerned that it would take up too much space and make the paper even longer.
  We did not make any assumptions about the source of each component in advance. Instead, we tried to identify patterns relating inlet values to land use and other catchment characteristics, such as size and slope. The PCA analyses indicated that the inlet values differed based on the location (geological setting). Additionally, we observed correlation between Cl^-, Na⁺, K⁺ and land use. Those informations were collected in the revised Results section ‘3.1 Physico-chemical background of the studied streams’ as follows:
  ‘...Significantly higher SEC and concentrations of major ions were found in streams that drained areas with a high proportion of shale in the bedrock (sites 1, 2, 4, and 6). Lower SEC and concentrations of major ions were found in streams draining catchments dominated by sandstone (sites 3, 7–9). The stream at site 5 has physico-chemical characteristics that fall between those of the two groups of streams. Concentrations of Cl^-, Na⁺ and K⁺ showed high correlations with the share of artificial surfaces in the studied catchments (r=0.85, 0.78 and 0.61 respectively). Additionally, concentration of Cl^- correlated with the number of buildings in the catchments (r=0.61)...’
  We modified the first paragraph of the discussion by adding a brief interpretation of those results:
  ‘...Higher SEC and higher concentrations of most of the major ions were found in streams that drain catchments dominated by shales in their bedrock, and they were lower in streams that drain catchments dominated by sandstones. This pattern is characteristic of the whole of the Outer Carpathians and it is likely associated with a different resistance to weathering (Siwek, 2021). The relation between Cl^-, Na⁺, K⁺ and anthropogenic modification of the catchment suggests human activity as an important factor for concentration of these ions in the study area...’
  Also, the exact morphological description of pond type, age, and other typologies that are used in interpretation should be stated here. For example, in the methods section, the type is defined as overflowing and in-channel, but in the Results section, it is interpreted as channel and overbank flow. Is it considered the same type of beaver pond?
  Answer: Following this suggestion, we have supplemented the Methods section with this clarification:
  ‘...For further analyses, the study sites were classified based on the beaver pond characteristics. Two geomorphic types were distinguished considering the extent of a beaver dam: in-channel (dams within river banks) or overflowing (dams extending over river banks). The age of beaver ponds was assessed based on orthophotos and classified into three categories (Table 1)...’
  We have unified the references to the pond types throughout the manuscript.
  
  Statistical analyses (170-180): Please add more information on how the clustering of the PCA analysis was performed, as is presented in Fig.3.
  Answer: We have added some information on PCA analysis, which can be found in the text below. The section of text referring to Figure 3 has been removed, as has the Figure 3 itself.
  ‘...Principal component analysis (PCA) was performed on the percentage change (Dc [%]) in the values of individual physico-chemical parameters of water in ponds and streams below the beaver dams relative to their values above the ponds, which was calculated using the following formula:
  Dc[%]=(Bx-Ax)/Ax‧100, (1)
  where:
  
  Bx is the value of the physico-chemical parameter x in the pond or stream below the dam, while Ax is the value of the physico-chemical parameter x in the stream above the dam during individual sampling sessions.
  Physico-chemical water parameters were included in the PCA if their mean absolute difference in Dc [%] (for all sampling sites combined) exceeded 5%. This calculation used the following formula (N is the total number of sampling sessions):
  ǀDcǀ[%]=∑(|(Bx-Ax)/Ax|‧100/N), (2)
  The variables that fulfilled the aforementioned assumption were: DO, H⁺, Ca²⁺, K⁺, HCO₃^-, SO₄^2-, Cl^-, NO₃^-, and NH₄⁺ (Table A1). The H⁺ concentration was used instead of the pH value to avoid the analysis involving logarithmic variables. The aim of using DC [%] in the PCA analysis was to identify the factors that determine changes in the physico-chemical parameters of water caused by the presence of beaver dams. The PCA method reduces the complexity of a large data set to a smaller set of easily interpretable factors that can be associated with specific processes (Drever, 1997). A matrix of factor scores, one of the most important parts of the PCA output, was also analysed. Factor scores provide a measure of the relation between each observation (i.e., each sample) and the identified factors (Shaw, Wheeler, 1997). The Kolmogorov–Smirnov and Lilliefors tests were used to determine the normality of the distributions of the variables. The Kaiser criterion was used to separate factors (the eigenvalues >1)...’
  
  Discussion section (Fig. 12): I am not sure if I can understand the scheme in Fig. 12. Complex and comprehensive information output, what image (period, age, geomorphology type) represents and how this typology can affect chemical composition is missing. From the image, it is not clear how different periods, pond types and age are related to chemical water properties (which typology affects the increase or decrease of chemical composition) and what changes are related to which presented type.
  Answer: We modified this figure (now Fig. 11) completely to make its content clearer and more comprehensive. Previously it was intended to emphasize the difference between periods, types or age of the beaver ponds by describing the direction of the difference alone. The revised figure presents the average magnitude and direction of change observed in each period, type, and age of beaver pond. We believe that enriched content of the figure makes it more informative, and at the same time more understandable. Please find the revised figure in the attached document (Figure 11.pdf).
  
  Discussion: Even though there is a chapter in the Results section devoted to the limitations of the study, I would recommend moving it and expanding it in the Discussion section as a separate chapter.
  Answer: As suggested, we moved this chapter to the Discussion section. This chapter has been substantially revised and complemented with more content:
  ‘...The wide spatial scope of the study and the high diversity of beaver-influenced streams introduced certain limitations. The spread of the study sites across a vast area restricted the number of samples collected to one sample per site each season, due to the time and cost constraints. In addition, dam failures resulting from floods and their targeted elimination during the research period decreased the planned sample size by half. The remaining study sites may have been affected by continuous dam maintenance by beavers. In the most extreme cases, temporal breaches of the dam may have been overlooked after its reconstruction. Moreover, since the study focused on the base flow conditions, early spring high flows have been omitted, which likely explains the similarities between the results from the late spring and the summer.
  Although a Before–After–Control–Impact (BACI) model would be the most appropriate approach for studying the impact of beavers on the chemical composition of water, it was not possible to use this method in this area due to the lack of investigations prior to dam construction. Instead, the study focused on spatial and temporal contrasts throughout the year and used a comparative ‘upper-lower river’ method, where upstream sections were used as reference sections. This research approach is commonly used in field studies to assess hydrochemical gradients generated by beaver dams when it is logistically or environmentally impossible to obtain ‘before’ data on dam construction (Green and Westbrook, 2009; Klotz, 2010; Fuller and Peckarsky, 2011; Puttock et al., 2017; Stevenson et al., 2022). It enables the identification of local hydrological and chemical effects caused by beaver damming while minimising the confounding influence of catchment heterogeneity. Although this approach lacks independent control sections, as in a full BACI model, it provides robust comparative evidence for evaluating the magnitude and spatial extent of changes induced by beaver activity under natural field conditions. Future research should be more detailed and extensive to enable the application of the BACI model.
  As there was no identified relationship between the initial values and the percentage change in the physicochemical parameters, it was assumed that the observed changes would not differ significantly based on upstream conditions. This allowed various catchments with different land uses and levels of anthropogenic influence to be included in the study. However, it should be noted that the inclusion of the streams characterised by particularly high concentrations of certain components, e.g. highly anthropogenically modified, may have resulted in different observations.
  Although the sample locations were standardised to a certain extent, they may not reflect typical stream or pond conditions. Stream samples are more likely to be well-mixed and representative due to turbulent flow conditions. This is also evidenced by the homogeneity of most longitudinal profiles upstream and downstream of beaver ponds. However, the spatial distribution of the pond samples may be more biased as it did not cover the full spectrum of hydrological units observed within the ponds (Majerova et al., 2020). Additionally, as the samples were taken from the middle of the water column, they do not represent the surface and bottom water layers, which according to the vertical sampling scheme could substantially affect observed physico-chemical parameters.
  The assessment of the impact of individual characteristics of beaver dam sequences may be constrained by the interconnections between some variables. Beaver ponds tend to be more persistent in smaller streams, and the length of the beaver occupancy seems to be related, to some extent, to the longitudinal and lateral development of the pond sequences. However, the estimation of mutual correlations was limited by the use of categorical values. Continuous values for the factors representing beaver pond age could not be obtained from the available sources, and the categorisation of the number of ponds was carried out because of the dynamics of this variable during the study period…’
  
  Thank you for taking the time and effort to review this paper. The review definitely helped us improve both the content and readability of the manuscript.
  
  Citation: https://doi.org/10.5194/egusphere-2025-1184-AC3

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

ED: Reconsider after major revisions (11 Nov 2025) by Gabriel Singer

Dear authors,
Thank you very much for preparing your answers to the reviewer critique. I have in the meantime acquired a third reviewer opinion that arrived after the reviewing window was closed yet is rich in very constructive comments that I urge you to address. I believe addressing these comments will significantly improve your paper. I would like to specifically comment the following points, which were raised in this third review and also bear some resemblance with other points of critique by other reviewers:

1) Reviewers share a general feeling that geomorphological factors and hydrology were incompletely treated. This starts with the lack of a control stream. Please prepare more background information on the various ponds in this regard. Residence time is a variable (and concept) that must be discussed, ideally you actually find a way to explicitly include it in your analysis. The last reviewer suggests using proxies for residence time and the manuscript actually included a "pond-to-channel" ratio that may be exactly that.

2) Analytical precision and raw data: This point raised by reviewer 3 seems very important to me. Please answer this point of critique with great care and provide information that lets readers trust your physico-chemical data. The important issue is that some differences among sites may be within the uncertainty range of the used methods. With regard to data, I urge you to provide the original data, likely in a supplement, besides the relative differences caused by beaver ponds. Unreliable differences may be flagged as such in graphs and tables (note that this quality of a measurement gets lost in the "percentage change"). Also, the issue of data quality is connected to how sampling was actually done, in particular the accompanying flow conditions and day time of sampling. Note that the fact that water remains in reservoirs behind beaver dams for quite a while translates to an averaging effect, thus upstream vs. downstream comparisons must be assessed under consideration of eventually dynamic situations upstream or in the pond (probable less so downstream).

3) Hyporheic exchange should be discussed also in the light of potentially being promoted by higher water pressures in the beaver ponds. The downstream water chemistry may thus be different not because of beaver pond-specific processes but because of forced increased hyporheic exchange flow (which of course in itself is a consequence of beaver activity).

Please move forward with preparing a revised manuscript. Given the many reviewer comments and their recommendations I consider this a request for major revision. Your revised manuscript will very likely be assessed by the same and/or additional reviewers once again.

Regards, Gabriel Singer

Comments to the Author: PDF

Hide

AR by Joanna Wąs on behalf of the Authors (15 Dec 2025) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (22 Dec 2025) by Gabriel Singer

RR by Miloš Rusnák (28 Dec 2025)

RR by Anonymous Referee #3 (26 Jan 2026)

Suggestions for revision or reasons for rejection

First, I would very much like to see this manuscript published, as it represents an important contribution and a clear stepping stone for further work on the controls of water chemistry in beaver-influenced reaches. At the same time, a degree of caution is warranted. We have seen cases in which NGOs/governments have taken marginal or context-dependent positive evidence of beaver impacts, combined with the occasional overstatement often typical of scientific papers, and used it out of context to inform policy. I therefore wish to emphasize a note of caution, and ask that my review be read in this light.
I believe the main argument of this manuscript is that reach geomorphology and pond age controls water chemistry, but the abstract is confusing.
Firstly, regarding geomorphology, the current distinctions appear largely limited to overbank flooding versus in-channel conditions, which I would not characterize as geomorphology sensu stricto. You further suggest that water chemistry is related to substrate, yet there is no table or figure that clearly situates the study sites with respect to geological units. If geomorphology, or perhaps more precisely landform, is a key explanatory factor, I would expect at least a basic characterization of the valley bottom, for example valley width, channel and hillslope configuration, longitudinal river profile, and substrate. What you appear to be arguing for is a hydro-geomorphic framework, which would also need to include parameters such as estimated discharge, water residence time, and catchment size. If this line of argument is pursued consistently, it would imply that the primary driver of change, namely the hydro-geomorphic context, has not been adequately investigated in the present study.
As a practical way forward, you could consider quantifying the flooded area and framing the observed differences between upstream and downstream primarily in terms of flooding extent and age, which may account for the variation you observe? If it makes sense, you might consider a simplified classification of beaver-influenced reaches into “large and old” versus “small and young.” This would allow you to streamline the argument by discussing area and age jointly as key explanatory variables. The main exception appears to be site 5, which is described as old but is very small and apparently in-channel; this case would require specific discussion.
Further, I am concerned that the analysis may primarily be capturing noise rather than signal. I find it frustrating that there is no straightforward comparison of upstream and downstream values. Figure 3 appears intended to address this, but it doesn’t give me the absolute changes between up- and downstream – did I miss this somewhere? I tried to download the original dataset from Zenodo, but didn’t find it under the cited DOI. In Figure 3, most values cluster around the 1:1 line, suggesting that they may largely reflect natural variability, with the exception of DO, ammonium, and pH. Even for these variables, however, a substantial number of data points exhibit trends opposite to those being argued for. This concern is fueled by the way the ‘less-than values’ (data below detection limits) are treated. The cited reference (Palarea-Albaladejo and Martín-Fernández, 2013) in fact argues against the approach adopted here for handling this data. In my view, values below the detection limit should be excluded from the analysis, but not sure they are? The authors state that NO₃⁻ and NH₄⁺ concentrations were below detection in 17 and 20 out of 102 samples, respectively, corresponding to nearly one fifth of the dataset, which is substantial. Figure 3 legend has a typo (stream). Here, I also don’t understand the different symbols – each value is plotting upstream versus downstream?
My concern about this manuscript interpreting mostly noise is also caused by the fact that measurement uncertainty is not propagated in the percent-change calculations, but as far as I know, ratios amplify and propagate measurement error, resulting in increased variability and poorer statistical behaviour compared to analyses based on absolute values or log-transformed changes, especially when original concentrations are low. I still don’t understand why the authors have not chosen a multiple regression but pca, it seems an odd choice, as in my opinion is a clustering exercise and very helpful in developing an argument, but should not be the final result?
I like Figure 6 (downstream changes) which has absolute numbers, and I assume contrasts an old and a young site? If this is true, then please annotate and describe in the captions, the same is true for Figure 7. Please write in the captions why those are relevant.
The discussion starts with variability of water properties depending on catchment geology, but I don’t see a figure showing this - there is also no reference to one in the text. The next point of the discussion is about DO, and states that there is a coherent change at all sites, but Figure 3 clearly shows values on both sites of the 1:1 line. This section does not cite a table or figure. I believe you don’t refer to TOC, but rather DOC or POC? Why is this data not shown? After that, you report a marked difference in NO3 concentrations, but again, I cannot really see this in Figure 3. Overall, the argument makes sense, but I specifically don’t see the decrease in NO3 in the dataset (Figure 3).
The discussion would benefit from substantial streamlining and a clearer focus. As I understand your central argument, you propose that decomposition of organic matter is the primary driver of the observed changes in water chemistry, with higher organic matter availability in older and larger ponds leading to the reported chemical patterns. If your evidence supports this claim, then this should be stated clearly and directly, without complicating the argument with secondary explanations, I am still doubtful if the PCA is the best method, as the axes cannot be clearly attributed? Why does your summary figure (Figure 8) has numbers without standard errors?
Please check on the English as well, there are a few unclear sentences, e.g. “The catchment characteristics refer to the catchments closed at the sampling sites below the beaver pond sequences”.

Referee Report: PDF

Hide

ED: Reconsider after major revisions (25 Feb 2026) by Gabriel Singer

Dear authors,

I have meanwhile received two reviews for your revised manuscript. One reviewer recommends acceptance, the other reviewer still sees some major shortcomings that need to be addressed and suggests a major revision. I side with the more critical reviewer and urge you to in particular take care of:

*) Streamlining the paper, clarify focus, leave away side aspects. This may include a change in the presentation framework such as replacement of the inadequately treated hydro-geomorphic scheme by a simpler comparison of large/old vs. small/young ponds. Data presentation must be improved, this includes minor things such as using more clearly distinguishable color gradients and symbols as well as homogeneity in data presentation (e.g. sites are numbered, at other places they are named). The current results presentation suggests confounding issues among type, age and landuse - these may be addressed up-front with a changed framework, or through improved results/discussion.

*) Adequate treatment of data quality issues. Noise has already been a concern in a previous round. Are there calibration standards below the detection limit? How do measurement errors propagate when computing ratios or percent change? When a careful assessment of noise shows its importance, this may be reflected in the wording of the discussion.

There are a lot of minor issues you may still want to take care of. The absolute differences used in the PCA are not clearly defined (is equ (2) showing a mean or the measure used in the PCA?). What happens if differences were negative as well as positive, but only their absolute value was used in the PCA? Detail how Bonferroni correction was used, Table 3 currently resembles a "fishing expedition" for P-values. Also, in Table 3 you report usage of data in an aggregated version - clarify what this means (a random factor? pooling? averaging?). Fig 7 is really hard to read, Fig 8 is a potentially nice conceptual figure but currently hard to read - why are variables ordered along a pond sequence and seemingly associated with specific ponds in the sequence?

Consider this a request for major revision. I support publication of this study, but the manuscript still needs substantial attention and major changes. Given the ecopolitical context of beavers, their removal or support, the study must be presented in a scientifically sound AND accessible way, thus making sure it actually achieves impact while preventing abuse.

Best regards, Gabriel Singer

Hide

AR by Joanna Wąs on behalf of the Authors (01 Apr 2026) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (review by editor) (04 Apr 2026) by Gabriel Singer

Dear authors,

Thank you for this revised version of your manuscript. At this point, I can see that many of the previous reviewer comments have been taken care of. There is a number of issues that remain, however, or that arose in this new version of your manuscript. Please, in particular take care of clarifying usage of statistics (this may require adaptation of your data analysis). I consider this a request for minor revision, but have to make acceptance of your work dependent on seeing the statistical issues solved.

More specifically, you describe usage of U-tests or Kruskal-Wallis Anovas. Both tests accept only one factor as structuring your dataset. You either use pond age (3 levels), pond type (2) or season (4? or 2?) as factors in your analyses. You do this by pooling data across all levels of factors not targeted in any single test. This strategy is in partly just statistically weak, in partly, however, also plainly wrong. Your study design is, in fact, a two-way factorial design (with two treatment or fixed factors pond age and pond type) and one additional within-subject factor that codes for seasons. The latter is because the SAME ponds were repeatedly sampled, in earlier times this would have been called a “repeated measures ANOVA”, nowadays such designs are handled by random factors. To account for the repeated measuring in the same ponds is necessary because otherwise by pooling across the 4 seasons you pseudoreplicate your dataset. The sample size for your two treatment factors must be 9 and not 9*4. When you ignore the repeated measures nature of your dataset, you run a U-test assuming a sample size of 9*4, i.e. you actually work with strongly inflated degrees of freedom. So, while ignoring pond type in a 1-way analysis targeting pond age may just mean a statistically weak strategy (the variance caused by pond type will increase apparent measurement noise), the ignorance of seasons being a within-subjects factor is statistically incorrect. Proper treatment of seasons will also make an analysis more powerful as the seasonal effect can be identified given the paired nature of your dataset. To illustrate, the problem I describe here is the same issue as you try to solve by using Dc to get rid of catchment-caused differences in upstream-pond-downstream data. You could also think of a factor coding for location along any river (with 3 levels up/pond/down). This factor is a within-subject factor requiring a random factor river in a proper statistical design. I am fine with your solution of using Dc, however, these thoughts just for illustration of the problem.

Please adapt your statistics. Describe clearly why you decided to run a 1-way design instead of a two-factorial design and that this means pooling of data across levels of factors not targeted in any particular test. Much cleaner would be to actually run a 2-way analysis. This is not possible using ordinal scaled data (i.e. using a test like Kruskal-Wallis), but will require more sophisticated GLM-type models. Going this way would also allow you to include the seasons factor in a correct way. From my experience, STATISTICA should be able to handle this.

I also urge you to invest a bit into briefer text expressions. The paper still has some potential for shortening without compromising content.

Minor line-by-line comments:

13: better “pond types differing in flooding extent”, so you already establish the word TYPE here.
17: types instead of categories
17/25/27: please make sure to always use the same words when addressing the two levels of the factor pond type to avoid confusion.
18: make clear that age has 3 CATEGORIES
21: processes instead of activity. It makes no sense to phrase “more statistically significant”. In a statistical sense the word “significant” cannot be intensified. You may use other qualifieers like “marked”, “pronounced”, etc.
26: when addressing a process like denitrification (or any other) just name the process, don´t call it “the denitrification process”. So here, this has to be “…which suggests more intense denitrification”. This applies to many instances throughout the MS, also for other processes.
31: Delete this last sentence, lame and meaningless. Close with an implication of your findings or some statement on the importance of your results even if limited.
34: Castor needs italics
45: the population statement is valid for where? Global?
48-49: Cannot see relevance of this sentence at this position, maybe move to (end of) next paragraph.
76-77: source and sink for what? Changes and transitions = redundant, reword. Cool to bring up self-purification, but you may want to also point to gases CO2, CH4, N2O, which build up as mineralization products in beaver ponds.
82: delete “complex”.
83: bother to give an example for these contradictory patterns?
91: find briefer expressions, here is a good example. The whole phrase “especially….could” can be replaced by “may”.
160ff and Fig 1b): I cannot see the added benefit of this figure. No chance to see how the valley actually drives a particular pond type. If you want to make this useful, then show cross-sections at a much smaller scale, so that readers can see in-channel vs. overflowing ponds.
164: replace “the archival” by “archived”.
Fig 2: Decide whether you want to use the word sequence or cascade in concert with ponds, but not both. In general use this less often, also in abstract, just confusing.
191+196: make the order in which these ponds pop up in the text agree with the figure.
210: sulphate and sulfide.
215-225: The detection limits are fairly high in comparison to the used standards. In essence, of your 6 standards, only the highest one was relevant, all others were below detection. This meanse 1-point calibrations for all parameters. Please clarify. Use L-1 instead of dm-3.
234-235 and eq2: The MEAN absolute difference should be denoted with a horizontal bar on top.
238-244: This is not understandable to me.
250: There was no research question for this particular analysis. At least provide a motivation for doing this here at this location, not necessarily in the intro.
253-254: Not understandable to me. If a result is significant, then it is significant and it does not matter whether P is 0.04 or 0.0001 – you cannot make an extra choice step. Also, you here bring up a potential confounding between the factors age and type. This confounding should be made clear in the design section. It may qualify as reasoning for 1-way instead of 2-way designs.
254-259: This section is hard to understand. Make clear how many comparisons you assumed for your Bonferroni correction. The logic behind what the last sentence here describes is really obscure to me.
262: what are “baseline conditions”.
266-267: P-values for these correlations?
Fig 3: use a color gradient that reflects the rank order of pond age levels, e.g. red/orange/yellow.
286: Does this statement refer to absolute Dc values or MEAN absolute Dc values?
Fig 4: The asterisks in the panel headers should sshow “seasonal difference” – what is this exactly referring to?
Table 3: Note that Kruskal-Wallis and U-tests cannot deal with dependence in the data set (repeated measures). Make clear how many repeated data usages you actually accounted for in the Bonferroni correction. As far as I can see, the same data was used many times, especially upstream data of spring and summer. It´s not trivial to here decide about an appropriate Bonferroni correction.
309: do not intensify “significant”.
319-321: clearly describe confounding (“overlap”) of factors much earlier in the MS (methods). I cannot follow the reasoning of this sentence: How can identifying a pH effect be driven by confounding?

Fig 7: This plot (also in Appendix) must be done differently, it´s confusing, hard to read and also wrong. Why not computing averages from the two profiles? There is no justification to show single data points. Also the plot implies pairing of measurements 1-2-3 with 9-10-11 but that is simply not correct. 1-2-3 may also be opposed to 12-13-14, right?

354: Please clean up the “deposition” argument. Certainly beaver ponds will be depositional habitats where PARTICULATE organic matter will accumulate by sinking to the bottom or getting filtered by dam and hyporheic interfacts. However you measure TOC from a water sample, likely without having disturbed the bottom, so this can only include small suspended particles.
359: downwelling “under” (?) the dam?
365: delete “a specific”
366: delete “a relatively”
373-375: Please revisit nitrogen-transforming processes. Standard denitrification does not produce NH4+, this is a different process called DNRA = dissimilatory nitrate reduction to ammonia. The current text ignores textbook knowledge.
370-378: When coming up with denitrification, it would be good to also bring up N2 and N2O as gaseous products. This creates a link to gaseous mineralization products to be explained in the intro.
386-387: I don´t understand the Margolis statement. A reduction of SO4 will increase pH but hardly ANC itself.
380-395: You may want to actually include increase CO2 (mineralization product) as a driver of pH changes. Outgassing in the pond will be slower than up- or downstream.
404-406: “over” the top of the dams?
420: consider CO2 emissions.
429: statement only makes sense for ice-cover – were the ponds covered by ice?
438: delete “one of”.
442: Meaning of “can be partly related to” is unclear to me.
447: logic of “can occur despite” unclear to me.

Fig 8 Use “season” instead of period in headline, get rid of “geomorphic”.

470: “significance… lower” is not a statistically meaningful statement.
497-505: Upstream is your reference, so I would call that the “control”. So, what you have is a control-impact design.
508: Any evidence shown for this statement?
515: why “most of”? there were only 2, right?
524-526: Unclear statement. Do you refer to the confounding issue?
528. Last sentence of this paragraph no clear to me.

Table A1: abs(Dc) was described as being a MEAN, so how can there be a SD for i? What is shown in this table seems to not agree with the description in Methods.

Hide

AR by Joanna Wąs on behalf of the Authors (23 Apr 2026) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (review by editor) (05 Jul 2026) by Gabriel Singer

AR by Joanna Wąs on behalf of the Authors (08 Jul 2026) Author's response Author's tracked changes Manuscript

ED: Publish as is (10 Jul 2026) by Gabriel Singer

AR by Joanna Wąs on behalf of the Authors (13 Jul 2026) Manuscript

Interactive discussion

Status: closed

CC1:
'Comment on egusphere-2025-1184', Mateusz Stolarczyk, 11 Jun 2025

The article “The effect of beaver ponds on water physico-chemical composition in the Carpathians (Poland and Slovakia)” examines how Eurasian beaver dams influence stream water chemistry in the Western Carpathians. Based on seasonal sampling of streams with varying geomorphology, pond types, and dam ages, the study found that beaver ponds significantly alter water parameters by reducing dissolved oxygen and nitrate levels, promoting sulfate reduction, increasing ammonium concentration, lowering pH, and in some cases facilitating calcium removal through precipitation. These effects are most pronounced during warm seasons and in older or overflowing ponds, where microbial and chemical processes intensify. The results underscore the important role of beaver ponds in modifying water quality. These valuable findings enhance our understanding of the complex influence of beaver activity on various environmental components, emphasizing the ecological significance of beaver-engineered habitats.
However, after reading the above study, several questions and comments arise regarding the conducted research, which may serve as useful guidance for the implementation of other research topics related to this subject:
1. Following an analysis of the results, conclusions, and the overall content of the paper, I would suggest considering a slight modification of the title (if it possible) to more accurately reflect the impact of beaver dams on the physico-chemical properties of stream water, rather than on the ponds as a whole. In my opinion, the ponds are a secondary consequence of dam construction, with the dam itself serving as the primary factor driving environmental modifications.
2. Sampling was conducted seasonally and did not account for hydrologically dynamic periods such as snowmelt or heavy rainfall events, which are known to cause substantial shifts in water flow and sediment transport. These events could temporarily alter or intensify the biogeochemical processes within beaver ponds, potentially affecting nutrient fluxes, oxygen dynamics, and pH levels. Incorporating high-resolution or event-based sampling in future research would provide a more comprehensive understanding of the short-term yet ecologically significant changes associated with such hydrological disturbances.
3. While the study offers valuable interpretations regarding biogeochemical processes in beaver ponds, it relies largely on indirect inference of organic matter dynamics and microbial activity. Direct measurements of parameters such as microbial community composition and microbial activity were not included, which limits the ability to fully validate the proposed mechanisms driving changes in water chemistry. Incorporating such data - through methods like enzyme activity assays or DNA sequencing - would strengthen the conclusions by providing process-level evidence and a clearer understanding of the microbial and organic matter contributions to the observed physico-chemical transformations.
4. The manuscript appropriately highlights the role of aquatic and riparian vegetation in nutrient cycling, particularly in relation to nitrate (NO₃⁻) reduction during the spring–summer period; however, this aspect was not quantitatively addressed. I agree that plant uptake likely contributes to observed nutrient dynamics, but this raises further questions about how pond age influences species diversity and vegetation cover? Specifically, is there a noticeable difference in plant growth or species composition between younger and older ponds, especially during the summer months? Addressing these questions with quantitative vegetation data would enhance understanding of biotic factors driving nutrient transformations in beaver ponds.

Citation: https://doi.org/10.5194/egusphere-2025-1184-CC1
- AC1: 'Reply on CC1', Joanna Wąs, 03 Nov 2025
  
  Thank you very much for voluntarily reviewing the paper. All the suggestions were carefully considered and we found them very helpful when revising the discussion section of the manuscript. You will find our response and the proposed changes to the manuscript below:
  1. Following an analysis of the results, conclusions, and the overall content of the paper, I would suggest considering a slight modification of the title (if it possible) to more accurately reflect the impact of beaver dams on the physico-chemical properties of stream water, rather than on the ponds as a whole. In my opinion, the ponds are a secondary consequence of dam construction, with the dam itself serving as the primary factor driving environmental modifications.
  Answer: Thank you for sharing your observation on the differences between beaver dams and ponds as environmental factors. Although a dam is essential for creating a pond, we believe that both ‘dams’ and ‘ponds’ are equally appropriate in this context, given that the processes underlying observed changes are related to stagnant water. However, we realise that using the term 'ponds' rather than 'dams' could suggest that the study focused mainly on pond properties and not on the rest of the stream. To avoid such confusion we will consider changing the title slightly, e.g. to ‘The effect of beaver dams on water physico-chemical composition in the Carpathians (Poland and Slovakia)’
  2. Sampling was conducted seasonally and did not account for hydrologically dynamic periods such as snowmelt or heavy rainfall events, which are known to cause substantial shifts in water flow and sediment transport. These events could temporarily alter or intensify the biogeochemical processes within beaver ponds, potentially affecting nutrient fluxes, oxygen dynamics, and pH levels. Incorporating high-resolution or event-based sampling in future research would provide a more comprehensive understanding of the short-term yet ecologically significant changes associated with such hydrological disturbances.
  Answer: Thank you for this valuable comment. We recognise that seasonal sampling did not capture short-term hydrological extremes, such as snowmelt or heavy rainfall, which can cause temporary but marked fluctuations in flow, sediment transport and nutrient dynamics. However, our sampling focused on capturing representative seasonal conditions under baseflow rather than event-driven variability, as the primary objective of the study was to assess the overall, sustained impact of beaver ponds on stream water chemistry under typical flow conditions. Nevertheless, we agree that high-resolution event-based sampling will provide valuable information on short-term biogeochemical responses and should be included in future studies.
  3. While the study offers valuable interpretations regarding biogeochemical processes in beaver ponds, it relies largely on indirect inference of organic matter dynamics and microbial activity. Direct measurements of parameters such as microbial community composition and microbial activity were not included, which limits the ability to fully validate the proposed mechanisms driving changes in water chemistry. Incorporating such data - through methods like enzyme activity assays or DNA sequencing - would strengthen the conclusions by providing process-level evidence and a clearer understanding of the microbial and organic matter contributions to the observed physico-chemical transformations
  Answer: Thank you for this important suggestion regarding the expansion and refinement of our future research. In our research, we did not measure the composition of microbial communities or the activity of microorganisms. The interpretations presented in this study are based on indicators derived from physicochemical parameters such as EC, pH, oxygen, and nutrients. While these variables reflect integrated biogeochemical responses at the ecosystem level, we acknowledge that the absence of direct measurements of microbial community composition or enzymatic activity limits our ability to conclusively verify the proposed mechanisms. Using molecular and biochemical methods, such as enzymatic activity tests or DNA sequencing, would provide evidence at the process level and offer a more complete understanding of the impact of microorganisms and organic matter on the observed changes in beaver ponds.
  4. The manuscript appropriately highlights the role of aquatic and riparian vegetation in nutrient cycling, particularly in relation to nitrate (NO₃⁻) reduction during the spring–summer period; however, this aspect was not quantitatively addressed. I agree that plant uptake likely contributes to observed nutrient dynamics, but this raises further questions about how pond age influences species diversity and vegetation cover? Specifically, is there a noticeable difference in plant growth or species composition between younger and older ponds, especially during the summer months? Addressing these questions with quantitative vegetation data would enhance understanding of biotic factors driving nutrient transformations in beaver ponds.
  Answer: We appreciate this insightful comment. The study did indeed highlight the potential role of aquatic and riparian vegetation in nutrient cycling, particularly with regard to nitrate reduction in spring and summer. However, as the main focus was on the physicochemical properties of the water, quantitative data on vegetation (e.g. biomass, species composition or percentage cover) were not collected. Future studies combining vegetation surveys, biomass estimates, and species-level analyses will help clarify the influence of biotic factors on nutrient cycling in beaver ponds.
  
  Citation: https://doi.org/10.5194/egusphere-2025-1184-AC1
RC1:
'Comment on egusphere-2025-1184', Gabriel Singer, 30 Sep 2025

General comments:
The abstract currently cannot stand on its own, i.e., it is not understandable on its own without having read more text of the manuscript. Needs to be rewritten. In the introduction please be much clearer about the actual research questions/hypotheses of your study. For instance, it is unclear what is “an environmental response to beaver dams”, meaning a change in upstream-downstream gradients in physicochemistry? Or do you mean a change after-before (in a temporal sense!) at larger spatial scale? Take care to robustly derive your study´s questions from past work AND explain them in an understandable way. The three questions are currently a slightly loose list without much connection and with poor derivation.
A weakness of the study is its focus on upstream-downstream comparisons without any controls. Water chemical conditions may change a lot between two sites along a stream continuum, independent of beaver dams. The study investigates only stream sites with beaver dams and compares data from above and below. There was no investigation before beaver dams were constructed, nor are there any control sites where no dams exist. I acknowledge that a full before-after-control-impact (BACI) study design was deemed infeasible, but such shortcomings of the study design must at least be discussed. Typically, in studies lacking controls, a higher level auf causality is needed for hypothesized processes, here in this case all processes that may affect physicochemical conditions. For ions without much explanation for change in beaver ponds, the current design does not allow to strongly pin down beavers as being the origin for any change, especially if the direction of change was not consistent. This should be reflected in the discussion.
The paper also suffers from a relatively inefficient presentation of results. There are too many figures. Consider moving some figures to the Supplement and look for ways to more efficiently present data (e.g. do not use separate tables for PCA variable loadings, but rather include these as a separate panel in the figure). Consider removing, shortening the more detailed results on dam cascades or vertical profiles or move to supplement to give the paper more focus. Some plots are simply superfluous and don´t present much exciting (e.g. SEC in Fig 10).
The discussion of the various anaerobic processes should be organized along the redox ladder.
The lower differences from upstream to downstream compared to upstream-pond may to some extent be related to the fact that downstream samples are from a well-mixed turbulent stream but pond sampling was likely not done in a spatially “integrative” way. Thus, pond samples may not represent average pond conditions. This needs discussion, also in light of anyway existing vertical profile data.
I strongly recommend authors to discuss the observed beaver pond effects in the light of residence time of water in the ponds. Doing this may offer a chance to integrate pond/stream size effects and seasons (differences in stream discharge). It may even be able to integrate geomorphic pond type (for which I find any observed differences inadequately derived as well as discussed anyway). Residence time in concert with temperature may be able to explain a lot.
Overall, I urge authors to shorten the paper and find more efficient wording in particular in the discussion. This will make the paper more accessible to a larger audience.
More specific comments:
55: “attenuation of downstream …. concentrations” – please express more clearly, do you mean compared to upstream? Then is it measured relative to a control? Or does this imply some before-after comparison as well? The same problem applies to the text in the abstract.
58: maybe two sentences instead of semicolon?
60-65: Could be worthwhile to dive deeper into whether these studies aiming at investigating “reductions” and thus working with upstream-downstream differences actually follow BACI-study designs to statistically robust conclusions.
66: maybe explain a bit better why and how geomorphic factors could actually have such an influence.
115: explain pond-to-channel ratio
155: were both the original data and the differences (percent change) used in the same PCA? Please provide justification why this procedure would make sense.
161: There is no N in this formula.
171: Variables are all concentrations or differences thereoef. I suggest to not mix untransformed with log-transformed variables in the same PCA. Concentrations may just all be log-transformed, as they are often log-normal distributed anyway. I cannot see any step in PCA where significance tests would become relevant. And to use significance for normality tests is a questionable approach anyway, as normally one “hopes” for high P to identify a variable as “not identifiable as not normal”. Try to work without those debated and criticized tests. Same applies to Shapiro-Wilk used prior to ANOVA.
178-180: Before doing hard-core hypothesis tests, please cleanly derive why these hypotheses would anyway make sense to be tested in the intro. The derivation of your questions and subsequent hypotheses needs to be clearer. Currently it seems that the factor “pond type” was defined after the study was done.
199: Reword “controls the values”, not appropriate wording to describe PCA results.
Fig 3: What are the various regions indicating? Information for A,A/B, etc. missing from the legend. I suggest you also present factor loadings as arrows, could be done in a separate plot (panel).
Fig 4: This figure does not allow to recognize which points belong to the same pond and date, i.e. the paired nature of the data is not shown (or “used” because in fact its consideration could make data analysis stronger). Consider working with differences or connections between points.
Table 4: This table presents A LOT OF P-values. Since the same dataset was used multiple times, some P-value adjustment (e.g. Bonferroni) should be done. Certainly, P-values >0.05 (<0.1) should not be reported. I fear that not too many P-values will remain significant then, making clear why this strategy of extensive dataset testing may not be the best. Also, stronger analysis may take into account multiple factors (pond type, age, seasons) at once.
269: Was this PCA entirely done with “differences”? A graph of variable scores as arrows as a panel in Fig 6 may be more efficient than Table 5. I suggest to not speak of “water samples” in the legend of Fig 6 as two samples were needed to create one point in these graphs, as I understand. The site codes in fig 6 are hard to distinguish.
285: Similar comments apply to this PCA as to the previously presented one.
Chapter 3.4. needs to be incorporated in the Discussion.
354: pointing out high TOC only makes sense in comparison to upstream.
365: any idea about CH4 in these ponds? It would make sense to organize the discussion of the various anaerobic processes along the redox ladder.
366: ammonification is a fairly specific process and not a synonym for OM oxidation.
370: Prefer “aerobic/anaerobic” for microbial processes, but “oxic/anoxic” or low/high redox potential for environmental conditions.
Fig 12: Very appreciated conceptual figure. Consider using coloring the pond sequence in a color gradient instead of arrows?

Citation: https://doi.org/10.5194/egusphere-2025-1184-RC1
- AC2:
  'Reply on RC1', Joanna Wąs, 03 Nov 2025
  Thank you for taking your time to review the paper. We are grateful for all the comments on our work - they helped us improve both the structure and content of the manuscript. Below, we address all comments point-by-point discussing the subsequent modifications. All of the comments and suggestions were carefully considered, and the manuscript has been revised accordingly.
  GENERAL COMMENTS
  1. The abstract currently cannot stand on its own, i.e., it is not understandable on its own without having read more text of the manuscript. Needs to be rewritten.
  Answer: Some parts of the abstract were previously omitted to make it more concise, however, you are right that this reduced the overall clarity of the study and led to the omission of some important information. We have now integrated additional details, such as clearer description of the methods, and partially rewritten the abstract into more structured paragraphs. The current version of the abstract is as follows:
  ‘Abstract. In recent decades, the population of the Eurasian beaver (Castor fiber) has undergone a rapid recovery from near extinction to abundance across vast areas of Europe. The ability of this species to build dams and the resulting impact on ecological processes makes the reintroduction of beavers an important environmental factor in the recolonised areas. Although the effects of beaver dams on stream water chemistry have been studied extensively, general conclusions are often limited by site specific conditions. Therefore, the aim of this study was to assess the impact of a geomorphic type and an age of beaver pond sequences on physico-chemical water properties in and below beaver ponds compared to upstream channel sections throughout the seasons.
  This study was conducted on nine beaver-inhabited streams distributed across the Western Carpathians (Poland and Slovakia). The beaver ponds were classified into two geomorphic categories: overflowing river banks or confined to the river channel. The age of the ponds varied from <2 years to >10 years. Water samples were collected above, within, and below beaver pond sequences during four seasons in 2022-2023.
  In general, a greater decrease in NO₃^- and SO₄^2-concentrations and increase in water temperatures within and below the ponds were observed during the warm period (spring–summer) than the rest of the year (autumn-winter). A comparison between two geomorphic types of beaver ponds revealed that there was a greater decrease in NO₃^- and Ca²⁺ concentrations and a greater decline in pH downstream of overflowing ponds compared to in-channel reservoirs. The age of beaver pond sequences was found to be a more significant factor than the pond type. It was positively related to decrease in dissolved oxygen, SO₄^2- and pH in the ponds and decrease in SO₄^2- in downstream channel sections.
  The role of the age of beaver ponds suggest that their long-term persistence may be important for sustaining water purification processes. However, further research based on a larger dataset and more frequent sampling strategy is needed to identify the biogeochemical processes occurring in beaver ponds under specific hydro-meteorological conditions: during low flow periods, snowmelt and rainfall events.’
  
  2. In the introduction please be much clearer about the actual research questions/hypotheses of your study. For instance, it is unclear what is “an environmental response to beaver dams”, meaning a change in upstream-downstream gradients in physicochemistry? Or do you mean a change after-before (in a temporal sense!) at larger spatial scale?
  Answer: We removed this general sentence and replaced it with a paragraph describing the idea in more detail. This includes a description of the role of specific beaver pond characteristics (age and geomorphic type) on the physico-chemical parameters, from which we further derived our study questions. The questions were also reformulated with more straightforward wording. It is now:
  ‘The objective of this study was to determine the effects of beaver dams on physico-chemical water properties in mountain streams of the Western Carpathians (Poland and Slovakia), with regard to seasonality and pond characteristics. The paper focuses on answering the following questions:
  How do physico-chemical water parameters change between river sections located upstream, within, and downstream of the beaver ponds, and what biogeochemical processes underlie these differences?
  
  How do these differences vary seasonally?
  
  How are these differences affected by the geomorphic type and age of the beaver ponds?’
  
  3. Take care to robustly derive your study´s questions from past work AND explain them in an understandable way. The three questions are currently a slightly loose list without much connection and with poor derivation.
  Answer: Thank you for pointing out this aspect that needed improvement. Some information was missing, while other (e.g. brief reasoning behind the pond classification) were scattered in the manuscript. We reorganized part of the introduction into three paragraphs that correspond more clearly with the research questions. We took special care to explain potential and already documented impacts of the age and geomorphic character of beaver ponds.
  
  4. A weakness of the study is its focus on upstream-downstream comparisons without any controls. Water chemical conditions may change a lot between two sites along a stream continuum, independent of beaver dams. The study investigates only stream sites with beaver dams and compares data from above and below. There was no investigation before beaver dams were constructed, nor are there any control sites where no dams exist. I acknowledge that a full before-after-control-impact (BACI) study design was deemed infeasible, but such shortcomings of the study design must at least be discussed. Typically, in studies lacking controls, a higher level of causality is needed for hypothesized processes, here in this case all processes that may affect physicochemical conditions. For ions without much explanation for change in beaver ponds, the current design does not allow to strongly pin down beavers as being the origin for any change, especially if the direction of change was not consistent. This should be reflected in the discussion.
  Answer: We agree that the lack of data from before the construction of the beaver dams limits our ability to clearly determine the cause of changes in the chemical composition of the water. However, this study used a comparative ‘upper river – lower river’ method, where the upper river sections were used as reference sections. This research approach is commonly used in field studies where it is impossible to obtain data 'before' the construction of beaver dams for logistical or environmental reasons (e.g. Green and Westbrook, 2009; Puttock et al., 2017). The revised manuscript has been supplemented with a discussion of these limitations, indicating that the presented results are correlational in nature and should be interpreted as indicating potential links between the presence of beaver dams and changes in water physicochemical properties. Additionally, the ‘Study limitations’ section highlights that a complete Before–After–Control–Impact (BACI) design would be the most appropriate approach to testing the causal effects of beaver activity (Puttock et al., 2020; Dittbrenner et al., 2022), but its application to natural river ecosystems is often impractical. It also highlights directions for future research that combine a spatial approach with a temporal component, which would allow for a more definitive assessment of the processes occurring in beaver-modified systems.
  4.1. Study limitations
  ‘... Although a Before–After–Control–Impact (BACI) model would be the most appropriate approach for studying the impact of beavers on the chemical composition of water, it was not possible to use this method in this area due to the lack of investigations prior to dam construction. Instead, the study focused on spatial and temporal contrasts throughout the year and used a comparative ‘upper-lower river’ method, where upstream sections were used as reference sections. This research approach is commonly used in field studies to assess hydrochemical gradients generated by beaver dams when it is logistically or environmentally impossible to obtain ‘before’ data on dam construction (Green and Westbrook, 2009; Klotz, 2010; Fuller and Peckarsky, 2011; Puttock et al., 2017; Stevenson et al., 2022). It enables the identification of local hydrological and chemical effects caused by beaver damming while minimising the confounding influence of catchment heterogeneity. Although this approach lacks independent control sections, as in a full BACI model, it provides robust comparative evidence for evaluating the magnitude and spatial extent of changes induced by beaver activity under natural field conditions. Future research should be more detailed and extensive to enable the application of the BACI model.…’
  Dittbrenner, B. J., Schilling, J. W., Torgersen, C. E., & Lawler, J. J. (2022). Relocated beaver can increase water storage and decrease stream temperature in headwater streams. Ecosphere, 13(7), e4168.
  Green, K. C., & Westbrook, C. J. (2009). Changes in riparian area structure, channel hydraulics, and sediment yield following loss of beaver dams. Journal of Ecosystems and Management, 10 (1):68-79.
  Puttock A,. Graham H.A., Cunliffe A.M., Elliott M., Brazier R.E. (2017). Eurasian beaver activity increases water storage, attenuates flow and mitigates diffuse pollution from intensively-managed grasslands. Science of the Total Environment, 576: 430–443.
  Puttock, A., Graham, H. A., Ashe, J., Luscombe, D. J., & Brazier, R. E. (2020). Beaver dams attenuate flow: A multi‐site study. Hydrological processes, 35(2), e14017.
  
  5. The paper also suffers from a relatively inefficient presentation of results. There are too many figures. Consider moving some figures to the Supplement and look for ways to more efficiently present data (e.g. do not use separate tables for PCA variable loadings, but rather include these as a separate panel in the figure). Consider removing, shortening the more detailed results on dam cascades or vertical profiles or move to supplement to give the paper more focus. Some plots are simply superfluous and don´t present much exciting (e.g. SEC in Fig 10).
  Answer: We took your advice on incorporating information from the tables resulting from PCA analysis into the figures. We also managed to improve efficiency by combining the two sites and multiple variables on the graphs depicting longitudinal and vertical profiles. Despite this, some of them can be moved into an Appendix. We will also consider removing part of the section ‘3.1.’ concerning PCA analyses of the physico-chemical background to shorten and focus the text. This would include removing one figure.
  
  6. The discussion of the various anaerobic processes should be organized along the redox ladder.
  Answer: We agree with the reviewer. Despite the lack of redox potential measurements, we attempted to organise the discussion of the aerobic/anaerobic processes along the redox ladder based on changes in the physico-chemical properties of the water, which indicate the occurrence of these processes. The example is below:
  ‘... As a result, there are intensive physico-chemical transformations in the stream water flowing into the ponds in the warm seasons. The increase in NH₄⁺ concentrations and the decrease in NO₃^- concentrations in the studied beaver ponds were particularly prominent in the warm period of the year due to the intensive ammonification (NH₄⁺ production) and denitrification processes (NO₃^- depletion). This is consistent with the findings of Margolis et al. (2001) and Law et al. (2016): the greatest NO₃^- decrease downstream to beaver ponds occurs in the summer. NO₃^- ion is one of the most effective electron acceptors (oxidants) in the process of anaerobic oxidation of organic matter; therefore, NO₃ is the preferred oxidising agent for anaerobic microbes. The process of NO₃^- reduction (denitrification) under anaerobic oxidation of organic matter occurs at a relatively high redox potential level on the redox ladder (Anderson and Fidel, 2025). However, it should be noted that the decrease in NO₃^- concentrations in the beaver ponds and streams below the ponds in the summer is partly related to the nitrogen fixation process, which is associated with increased microbial activity in sediments mediated by anaerobic conditions, as well as uptake by aquatic vegetation (Naiman and Melillo, 1984; Maret et al., 1987; Songster-Alpin and Klotz, 1995). Furthermore, the decrease in SO₄^2- concentrations in the studied ponds and downstream of the beaver dams was also observed during the warm period. This is likely due to the high rate of anaerobic oxidation of organic matter during that time. According to Smith et al. (1991), Cirmo and Driscoll (1993), and Margolis et al. (2001), the intensive anaerobic oxidation of organic matter can occur despite the relatively low redox potential of the water during the low-flow conditions of summer. During that time anaerobic microbes used even relatively inefficient oxidants, such as SO₄^2-, in the anaerobic oxidation of organic matter due to the low concentration of NO₃^- in the studied waters; the average concentration of NO₃^- in ponds and streams below them was very low during the summer, the average was up to 1.5 mg/L.’
  
  7. The lower differences from upstream to downstream compared to upstream-pond may to some extent be related to the fact that downstream samples are from a well-mixed turbulent stream but pond sampling was likely not done in a spatially “integrative” way. Thus, pond samples may not represent average pond conditions. This needs discussion, also in light of anyway existing vertical profile data.
  Answer: We agree that due to the spatial complexity of beaver ponds it’s difficult to derive general conclusions concerning the whole pond without comprehensive sampling across the system. The regional scope of this research didn’t allow for more detailed sampling at every pond each season. However, we applied consistent criteria for the location of the ‘pond’ sampling point across the whole study area and time, to increase their comparability. The detailed description that would help interpret the data was missing in the methodology section, therefore we clarified the sampling locations in the revised manuscript:
  ‘...The ‘pond’ sample was taken from the main channel section of the pond, i.e. excluding channel margins and backwaters. It was collected approximately 1m above the dam, from the middle of a water column….’
  We have complemented the discussion with a description of the identified constraints:
  ‘...Although the sample locations were standardised to a certain extent, they may not reflect typical stream or pond conditions. Stream samples are more likely to be well-mixed and representative due to turbulent flow conditions. This is also evidenced by the homogeneity of most longitudinal profiles upstream and downstream of beaver ponds. However, the spatial distribution of the pond samples may be more biased as it did not cover the full spectrum of hydrological units observed within the ponds (Majerova et al., 2020). Additionally, as the samples were taken from the middle of the water column, they do not represent the surface and bottom water layers, which according to the vertical sampling scheme may substantially affect observed physico-chemical parameters...’
  
  8. I strongly recommend authors to discuss the observed beaver pond effects in the light of residence time of water in the ponds. Doing this may offer a chance to integrate pond/stream size effects and seasons (differences in stream discharge). It may even be able to integrate geomorphic pond type (for which I find any observed differences inadequately derived as well as discussed anyway). Residence time in concert with temperature may be able to explain a lot.
  Answer: Thank you for this valuable suggestion. Although water residence time plays an important role in shaping the observed processes in beaver ponds, it was only briefly mentioned in the text. While modifying the introduction, we included it as one of the pond characteristics that may differ between the two geomorphic types, instead of the description previously made in the methodology section. The ‘age’ and ‘geomorphic type’ are not the direct causes of the difference, but this classification is likely connected to different hydrological and thermal conditions that influence biogeochemical processes. Much of the discussion on the impact of beaver pond age and geomorphological type was reorganised and supplemented with information regarding effects of water residence time and temperature, to clarify processes that may be responsible for the observed differences between the ponds.
  
  9. Overall, I urge authors to shorten the paper and find more efficient wording in particular in the discussion. This will make the paper more accessible to a larger audience.
  Answer: After reading the manuscript with a fresh perspective, we agree that it was somewhat overloaded with information. We revised the text and figures carefully. We removed part of section 3.1 from the results completely, because it didn’t carry information that would be necessary for understanding the impact of beavers, which is the main focus of the paper. We also removed or rephrased some sentences that were redundant or purely structural, especially in the discussion and conclusions. Moreover, the combination of several variables and study sites on the plots decreased the size of the figures. An incorporation of the PCA loadings into the figures decreased an overall number of figures and tables. All the changes resulted in the shortening of the main body of the manuscript by a few pages.
  
  SPECIFIC COMMENTS:
  55: “attenuation of downstream …. concentrations” – please express more clearly, do you mean compared to upstream? Then is it measured relative to a control? Or does this imply some before-after comparison as well? The same problem applies to the text in the abstract.
  Answer: We clarified the upstream-downstream relation in this fragment of the text, as well as in the description of our research objectives in the abstract.
  
  58: maybe two sentences instead of semicolon?
  Answer: Thank you, changed.
  
  60-65: Could be worthwhile to dive deeper into whether these studies aiming at investigating “reductions” and thus working with upstream-downstream differences actually follow BACI-study designs to statistically robust conclusions.
  Answer: We used the word 'reduction' incorrectly. It should have been 'decrease' — we meant a decrease in concentration. This error appeared in several other places in the text and has now been corrected.
  In this study, it was not possible to use the BACI model due to, among other reasons, the lack of pre-dam surveys. The study focused on spatial contrasts and seasonal changes within a single beaver-affected streamway. In our study, we used a stream-scale gradient approach, with sampling points located upstream of the beaver pond, within the pond, and downstream of the dam. This spatial model, commonly referred to as an upstream-downstream or longitudinal comparison (e.g., Klotz 2010; Fuller and Peckarsky 2011; Stevenson et al. 2022), has been widely adopted to assess in situ hydrochemical gradients generated by beaver dams. It allows for the identification of local effects of dam-induced hydrological and chemical transformations while minimizing the confounding effects of catchment heterogeneity. Although this approach does not include independent control sections, as in the full BACI project, it provides robust comparative evidence for assessing the direction and magnitude of changes induced by beaver activity in natural field conditions. Therefore, our sampling framework is consistent with established methods for assessing changes in stream water chemistry and sediment dynamics induced by beavers.
  Fuller, M. R., & Peckarsky, B. L. (2011). Does the morphology of beaver ponds alter downstream ecosystems?. Hydrobiologia, 668(1), 35-48.
  Klotz, R. L. (2010). Reduction of high nitrate concentrations in a Central New York State stream impounded by beaver. Northeastern Naturalist, 17(3), 349-356.
  Stevenson, J. R., Dunham, J. B., Wondzell, S. M., & Taylor, J. (2022). Dammed water quality—Longitudinal stream responses below beaver ponds in the Umpqua River Basin, Oregon. Ecohydrology, 15(4), e2430.
  
  66: maybe explain a bit better why and how geomorphic factors could actually have such an influence.
  Answer: Thank you for this suggestion. We rewrote a part of the Introduction and included a more detailed explanation of the expected influence of beaver pond characteristics in this section:
  ‘…The identification of factors reflecting the diversity of beaver ponds and their effects on water chemistry, would improve the understanding of the effects of beaver presence in river systems. The age of beaver ponds was commonly proposed as a such feature in previous studies on upstream-downstream gradients of physico-chemical parameters in beaver-occupied streams. For example Bason et al. (2017) found a greater reduction of suspended solids in older dam sequences compared to younger ones. Metal sequestration rates and a sediment organic matter content inside beaver ponds also rise with the pond age according to the study of Murray et al. (2021). Hydrological and thermal differences were observed between various geomorphic units within beaver ponds, representing e.g. main channel, its margins and backwater areas (Majerova et al., 2020). The share of such units, especially related to the floodplain inundation, was then suggested as a potential geomorphic factor that could explain differences between study sites (Murray et al., 2021). Spatially expansive beaver ponds are expected to facilitate a higher proportion of low velocity areas, increase in water residence time, overbank sedimentation and growth of aquatic vegetation, compared to the ponds constrained to the river channel. Both beaver pond age and its geomorphic type may correspond with the prolonged sediment-water interactions resulting in the enhanced biogeochemical processes…’
  
  115: explain pond-to-channel ratio
  Answer: The term 'pond-to-channel ratio' was originally intended to describe the distinction between 'overflowing' and 'in-channel' beaver ponds. To improve clarity and ensure consistency with the terminology used throughout the manuscript, we have replaced it with 'type of beaver pond'.
  
  155: were both the original data and the differences (percent change) used in the same PCA? Please provide justification why this procedure would make sense.
  Answer: We used the original (raw) data of solute concentrations for PCA to determine the physico-chemical background of the studied streams. For this purpose, we used the raw concentrations of solutes in the stream water above the beaver ponds that were unaffected by beaver dams. These concentrations were standardised prior to analysis. However, to determine the factors influencing changes in solute concentrations in ponds and streams below due to beaver dams, we performed PCA using the percentage difference in concentrations between the stream water above the beaver dams and the pond water/stream water below the dams. In this case, we used concentration differences rather than raw concentrations because we focused not on the factors determining solute concentrations in pond and stream waters below the dams, but rather on the factors shaping the differences in these concentrations due to the presence of beaver dams. Ultimately, in light of suggestions to shorten the manuscript, we decided to remove the results of the factor analysis based on the raw data from Section 3.1. The physico-chemical background of the studied streams.
  
  161:There is no N in this formula.
  Answer: It was our mistake — N was included in the second formula, but not the first. We have corrected this.
  
  171: Variables are all concentrations or differences thereoef. I suggest to not mix untransformed with log-transformed variables in the same PCA. Concentrations may just all be log-transformed, as they are often log-normal distributed anyway. I cannot see any step in PCA where significance tests would become relevant. And to use significance for normality tests is a questionable approach anyway, as normally one “hopes” for high P to identify a variable as “not identifiable as not normal”. Try to work without those debated and criticized tests. Same applies to Shapiro-Wilk used prior to ANOVA.
  Answer: We have removed the results of the PCA analysis in which both non-logarithmic and logarithmic variables (Ca²⁺) were used (see section 3.1. The physico-chemical background of the studied streams). Therefore, we removed the information about PCA analysis on raw data from 2.3. Statistical analysis section. The information about significance threshold applies to all the other analyses, so we moved it to a different part of this section. To our knowledge, the use of the normality test remains a standard procedure, as it allows to select parametric or nonparametric tests for further analyses based on the data distribution. Both Shapiro-Wilk and Kolmogorov-Smirnov tests have been used prior to ANOVA in previous study on beaver impacts on water chemistry (Błędzki et al., 2011; Bylak et al, 2024).
  
  178-180: Before doing hard-core hypothesis tests, please cleanly derive why these hypotheses would anyway make sense to be tested in the intro. The derivation of your questions and subsequent hypotheses needs to be clearer. Currently it seems that the factor “pond type” was defined after the study was done.
  Answer: The indicated fragment has been removed from this part of the text. We have clarified our underlying assumptions in the introduction instead. Please see point ‘66:’ above.
  
  199: Reword “controls the values”, not appropriate wording to describe PCA results.
  Answer: We agree with this comment. We have corrected this throughout the text.
  
  Fig. 3: What are the various regions indicating? Information for A,A/B, etc. missing from the legend. I suggest you also present factor loadings as arrows, could be done in a separate plot (panel).
  Answer: The regions A/B represented geological locations (corresponding with dominance of shale or sandstone in the catchments), while C/D represented sampling seasons. However, we removed Figure 3 in order to shorten the article and focus the result section on the impact of beavers.
  
  Fig. 4: This figure does not allow to recognize which points belong to the same pond and date, i.e. the paired nature of the data is not shown (or “used” because in fact its consideration could make data analysis stronger). Consider working with differences or connections between points.
  Answer: The purpose of this figure was not to distinguish solute concentrations in detail at each location and sampling time, but rather to illustrate a general pattern. The figure allows to distinguish between the solute concentrations in ponds and streams below the beaver dams (represented by squares and crosses, respectively) in relation to the stream above the pond, and categorises them according to the age of the beaver dam: less than three years old, four to nine years old, and more than ten years old (represented by small red signatures, larger yellow signatures, the largest green signatures, respectively). Inclusion of the season as another variable or links between the points would make the figure illegible. However, in most cases the relations between paired data (pond and downstream differences at particular location and season) can be distinguished on the figure. The points representing the same sampling site at the same time are located at the same position on the horizontal axis (which results from the same inlet value), therefore vertical distance between both points illustrates the difference.
  
  Table 4:This table presents A LOT OF P-values. Since the same dataset was used multiple times, some P-value adjustment (e.g. Bonferroni) should be done. Certainly, P-values >0.05 (<0.1) should not be reported. I fear that not too many P-values will remain significant then, making clear why this strategy of extensive dataset testing may not be the best. Also, stronger analysis may take into account multiple factors (pond type, age, seasons) at once.
  Answer: We didn’t consider results with p<0.1 statistically significant. They served as additional information and were presented in the table in regard to a scientific discussion on the arbitrariness of significance thresholds. However, we agree that their inclusion was unnecessary and could distract from the interpretation of truly significant results. Therefore, only results meeting the conventional significance threshold (p < 0.05) were retained in the revised table. Moreover, in response to the very appreciated comment concerning multiple analyses on the same dataset we applied recommended correction to improve the interpretation of the results and updated methodology section:
  ‘...The significance level of p<0.05 was used for all the calculations. Additionally, to reduce the risk of family-wise error for multiple comparisons based on the same dataset, we applied a Bonferroni correction to the significance threshold (p<0.017)...’
  
  269, 285: Was this PCA entirely done with “differences”? A graph of variable scores as arrows as a panel in Fig 6 may be more efficient than Table 5. I suggest to not speak of “water samples” in the legend of Fig 6 as two samples were needed to create one point in these graphs, as I understand. The site codes in fig 6 are hard to distinguish.
  Answer: The results of the PCA analysis that were based on raw data in Section 3.1 were removed to shorten the text. The remaining PCA analyses are both based only on concentration differences (%), so subsequent update of the methodology description made it less confusing. Exact information on the type of data used can be found in Section 2.3. Statistical analyses. As suggested, Table 5 has been incorporated into Figure 6 in graphical form as both contain the results of the same factor analysis (factor loadings). That applies to Table 6 and Figure 7 as well. Tables 5 and 6 have been moved to the supplements because, in addition to the factor loadings of F1 and F2, they also contain information about the factor loadings of F3. In the legend for Figure 6, we wrote 'water samples' because each point represents a single water sample, and the factor scores for factors 1 and 2 are marked for each sample. We have corrected the site codes in the figures relating to PCA.
  
  Chapter 3.4: needs to be incorporated in the Discussion.
  Answer: You are right. It became Chapter 4.1.
  
  354: pointing out high TOC only makes sense in comparison to upstream.
  Answer: We agree with this comment. We have corrected the text as follows:
  ‘...Detailed investigations carried out in July 2024 at sites 3 and 6 showed that beaver pond water contains higher concentrations of total organic carbon (TOC) compared to the stream water upstream. For example, the TOC concentration in the beaver pond at site 3 was found to be more than 13 mg/L, which was almost three times higher than in the stream flowing into the pond...’
  
  365: any idea about CH4 in these ponds? It would make sense to organize the discussion of the various anaerobic processes along the redox ladder.
  Answer: Unfortunately, we did not examine the concentrations of CO₂ and CH₄ in the stream water and pond water, so we cannot address this issue in the discussion. However, we realise that it is essential to take this into account when studying the impact of beaver ponds on the chemistry of the stream below in future research.
  
  366: ammonification is a fairly specific process and not a synonym for OM oxidation.
  Answer: We agree with this comment. The sentence has been revised as follows:
  ‘...The decomposition and mineralization of organic matter in the process of ammonification, under both aerobic and anaerobic conditions, leads to the production of NH₄⁺ ions in water….’
  
  370: Prefer “aerobic/anaerobic” for microbial processes, but “oxic/anoxic” or low/high redox potential for environmental conditions.
  Answer: We agree with this suggestion. We have corrected the terminology throughout the text.
  
  Fig. 12: Very appreciated conceptual figure. Consider using coloring the pond sequence in a color gradient instead of arrows?
  Answer: Thank you. As the original figure was found to be difficult to understand, we considered various ways of transforming it to improve its clarity. The use of arrows has been modified to indicate direction and the average magnitude of change observed in each period, type and age of beaver pond, instead of the simple indication of one difference resulting from a comparison. In this way, even though there is more information presented in the figure, the content of the image should be more clear. Please find the revised figure (Figure 11.pdf) in the attached document.
  
  Overall, we greatly appreciate this insightful review and we believe that the revised manuscript is substantially improved thanks to the consideration of all the suggestions.
  
  Citation: https://doi.org/10.5194/egusphere-2025-1184-AC2
RC2:
'Comment on egusphere-2025-1184', Miloš Rusnák, 21 Oct 2025
The study examines the impact of beaver ponds on biogeochemical processes and stream water chemistry. For testing, 9 mountain streams in Poland and Slovakia were selected. The study focuses on the physico-chemical parameters with a detailed sampling design and focuses on pond control effects, seasonality and typology. Based on the presented result, it is necessary to state that the manuscript is well written and structured. However, in several places (methodology and description of results), it contains a significant amount of detail that distracts from the main idea of the article. Conversely, some details and method explanations are missing. I also recommend revising several diagrams and charts, which would significantly improve the readability of the article. The manuscript requires small reconfiguration and improvement in readability. Hopefully, my comments will be helpful in that process.

For the paper, I have several general comments:
Study area (85-120): Here, I recommend the description of 2 factors, which are used for the interpretation of results in the discussion, and these are: geology and agricultural activities. Geology is completely missing and is described in the discussion that different structures affected stream chemistry. From human activities, it is important to add more information on how intensive agricultural activities are preserved in upstream conditions. Pastures, fertilisation or direct soil cultivation can affect chemical composition. The second important parameter is human presence in the upstream catchment area (population, number of buildings) that can affect a lot of selected chemical parameters. At the same time, it should be noted that the selected design of sampling strategy minimises ambiguities, but a broader description of these limits should be part of the discussion section.

Methods (145 - 150): For increasing the readability of the paper, I recommend adding an exact complex statement of all physico-chemical parameters with a description and interpretation of what the source is and how it is theoretically interpreted when its values change. Also, the exact morphological description of pond type, age, and other typologies that are used in interpretation should be stated here. For example, in the methods section, the type is defined as overflowing and in-channel, but in the Results section, it is interpreted as channel and overbank flow. Is it considered the same type of beaver pond?

Statistical analyses (170-180): Please add more information on how the clustering of the PCA analysis was performed, as is presented in Fig.3.

Discussion section (Fig. 12): I am not sure if I can understand the scheme in Fig. 12. Complex and comprehensive information output, what image (period, age, geomorphology type) represents and how this typology can affect chemical composition is missing. From the image, it is not clear how different periods, pond types and age are related to chemical water properties (which typology affects the increase or decrease of chemical composition) and what changes are related to which presented type.

Discussion: Even though there is a chapter in the Results section devoted to the limitations of the study, I would recommend moving it and expanding it in the Discussion section as a separate chapter.
Citation: https://doi.org/10.5194/egusphere-2025-1184-RC2
- AC3: 'Reply on RC2', Joanna Wąs, 03 Nov 2025
  
  Thank you for this review. We appreciate the chance to improve the manuscript following all the helpful comments and suggestions. We revised the paper accordingly. Please find our responses to each comment below:
  The study examines the impact of beaver ponds on biogeochemical processes and stream water chemistry. For testing, 9 mountain streams in Poland and Slovakia were selected. The study focuses on the physico-chemical parameters with a detailed sampling design and focuses on pond control effects, seasonality and typology. Based on the presented result, it is necessary to state that the manuscript is well written and structured. However, in several places (methodology and description of results), it contains a significant amount of detail that distracts from the main idea of the article. Conversely, some details and method explanations are missing. I also recommend revising several diagrams and charts, which would significantly improve the readability of the article. The manuscript requires small reconfiguration and improvement in readability. Hopefully, my comments will be helpful in that process.
  Answer: Thank you for the positive feedback and all the constructive remarks. We agree that the original manuscript was not very clear due to the amount of content and inefficient presentation of the results. However, we managed to shorten the text and reduce the amount of distracting or redundant information. The reduction included, e.g., removal of the part of Section 3.1. from the Results as it was found to be not relevant enough for the understanding of the effects of beaver activity. Section 2 Methods was slightly modified to eliminate repeated and include missing information, and to direct the reader to the information already present in the table. The overall readability was also enhanced by modifying the graphic materials, i.e., by combining several variables and the two study sites into the profiles, and by incorporating the PCA from the tables into the figures. The size and number of the figures was therefore reduced, while retaining the same amount of information and enhancing comparability and visual appeal.
  Study area (85-120): Here, I recommend the description of 2 factors, which are used for the interpretation of results in the discussion, and these are: geology and agricultural activities. Geology is completely missing and is described in the discussion that different structures affected stream chemistry. From human activities, it is important to add more information on how intensive agricultural activities are preserved in upstream conditions. Pastures, fertilisation or direct soil cultivation can affect chemical composition. The second important parameter is human presence in the upstream catchment area (population, number of buildings) that can affect a lot of selected chemical parameters. At the same time, it should be noted that the selected design of sampling strategy minimises ambiguities, but a broader description of these limits should be part of the discussion section.
  Answer: The geology of the study area was presented briefly in the first paragraph of section 2.1, including an explanation of its regional diversity. As part of the discussion, we refer to the geology at a similar regional level of detail. Unfortunately, based on the available data we could not provide a more detailed description of each river channel. However, we found that the reference made in the discussion could be confusing, as the phrase ‘correspond to various geological features’ implies that we are referring to more detailed components instead of the differences in general geological structure. We have modified this fragment to: ‘corresponded to the geological structure.’ Additional information on the water chemistry was also included:
  ‘...Streams draining areas underlain by shales and mudstones are typically characterised by higher concentrations of bicarbonates, calcium and magnesium ions due to increased weathering of carbonates and clay minerals, while streams in areas composed mainly of sandstone are usually more diluted and characterised by lower specific conductivity (e.g. Siwek et al., 2011; Solár et al., 2023)...’
  As suggested, we added a paragraph in the Section 2.1 Study area describing land use and some of the anthropogenic impacts:
  ‘...All of the catchments consisted of a mixture of forest and agricultural areas, i.e. arable lands and meadows. Despite the widespread abandonment of agricultural land and the generally low-to-moderate intensity of agricultural practices in the mountainous catchments (Pazúr et al., 2014; Kolecka et al 2017), the impact of these activities may be significant locally, as evidenced by the increase in nitrate concentrations observed in the Ondava catchment due to the use of fertilisers (Pekarova et al., 2006; Balejčíková et al., 2020). Overall, agricultural practices in the Polish and Slovakian Carpathians are mostly extensive, characterised by low–input management and mowing and grazing on a small-scale (Pazúr et al., 2014; Kolecka et al., 2017; Affek et al., 2023). According to national agricultural statistics (Central Statistical Office of Poland, 2025), the average consumption of mineral fertilisers (NPK) in the Małopolska and Podkarpackie provinces remains one of the lowest in the country (80–95 kg NPK ha⁻¹ year⁻¹ compared to the national average of ~125 kg ha⁻¹) and has shown little change over the last two decades. Similar situation is observed in the northern part of Slovakia where the average use of industrial fertilisers is less than 60 kg NPK ha⁻¹ year⁻¹ (Siman and Velísková, 2018). Consequently, the direct inflow of nutrients from fertilisers into spring watercourses is minimal, although local enrichment may occur near farms or areas of intensive grazing. In contrast, domestic wastewater was identified as the main source of stream water pollution in the rural areas of the Polish Carpathians (Siwek, 2020). The intensity of housing development across the study area ranged from negligible levels to artificial surfaces covering up to 4% of the area and containing a few hundred buildings…’
  Table 1 was also supplemented with information on the number of buildings in the studied catchments.
  The potential impact of catchment diversity on the obtained results was described in the discussion:
  ‘...As there was no identified relationship between the initial values and the percentage changes in the physico-chemical parameters, it was assumed that the observed changes would not differ significantly based on upstream conditions. This allowed various catchments with different land uses and levels of anthropogenic influence to be included in the study. However, it should be noted that the inclusion of the streams characterised by particularly high concentrations of certain components, e.g. highly anthropogenically modified, may have resulted in different observations...’
  
  Methods (145 - 150): For increasing the readability of the paper, I recommend adding an exact complex statement of all physico-chemical parameters with a description and interpretation of what the source is and how it is theoretically interpreted when its values change.
  Answer: The complex list of the measured parameters was previously presented on the bottom of Figure 2. However, you are right that listing those parameters in the text makes them more clearly visible. Therefore, we included the list at the beginning of the Methods section, before the detailed description of the sampling:
  ‘..The research followed a three-dimensional sampling scheme (Fig. 2). The following water parameters were measured: temperature (T), dissolved oxygen (DO), specific electrolytic conductivity (SEC), pH and the ionic composition (Ca²⁺, Mg²⁺, Na⁺, K⁺, NO₃^-, NH₄⁺, SO₄^2-, Cl^-)...’
  Although we agree that a description of each water parameter would be valuable, we are concerned that it would take up too much space and make the paper even longer.
  We did not make any assumptions about the source of each component in advance. Instead, we tried to identify patterns relating inlet values to land use and other catchment characteristics, such as size and slope. The PCA analyses indicated that the inlet values differed based on the location (geological setting). Additionally, we observed correlation between Cl^-, Na⁺, K⁺ and land use. Those informations were collected in the revised Results section ‘3.1 Physico-chemical background of the studied streams’ as follows:
  ‘...Significantly higher SEC and concentrations of major ions were found in streams that drained areas with a high proportion of shale in the bedrock (sites 1, 2, 4, and 6). Lower SEC and concentrations of major ions were found in streams draining catchments dominated by sandstone (sites 3, 7–9). The stream at site 5 has physico-chemical characteristics that fall between those of the two groups of streams. Concentrations of Cl^-, Na⁺ and K⁺ showed high correlations with the share of artificial surfaces in the studied catchments (r=0.85, 0.78 and 0.61 respectively). Additionally, concentration of Cl^- correlated with the number of buildings in the catchments (r=0.61)...’
  We modified the first paragraph of the discussion by adding a brief interpretation of those results:
  ‘...Higher SEC and higher concentrations of most of the major ions were found in streams that drain catchments dominated by shales in their bedrock, and they were lower in streams that drain catchments dominated by sandstones. This pattern is characteristic of the whole of the Outer Carpathians and it is likely associated with a different resistance to weathering (Siwek, 2021). The relation between Cl^-, Na⁺, K⁺ and anthropogenic modification of the catchment suggests human activity as an important factor for concentration of these ions in the study area...’
  Also, the exact morphological description of pond type, age, and other typologies that are used in interpretation should be stated here. For example, in the methods section, the type is defined as overflowing and in-channel, but in the Results section, it is interpreted as channel and overbank flow. Is it considered the same type of beaver pond?
  Answer: Following this suggestion, we have supplemented the Methods section with this clarification:
  ‘...For further analyses, the study sites were classified based on the beaver pond characteristics. Two geomorphic types were distinguished considering the extent of a beaver dam: in-channel (dams within river banks) or overflowing (dams extending over river banks). The age of beaver ponds was assessed based on orthophotos and classified into three categories (Table 1)...’
  We have unified the references to the pond types throughout the manuscript.
  
  Statistical analyses (170-180): Please add more information on how the clustering of the PCA analysis was performed, as is presented in Fig.3.
  Answer: We have added some information on PCA analysis, which can be found in the text below. The section of text referring to Figure 3 has been removed, as has the Figure 3 itself.
  ‘...Principal component analysis (PCA) was performed on the percentage change (Dc [%]) in the values of individual physico-chemical parameters of water in ponds and streams below the beaver dams relative to their values above the ponds, which was calculated using the following formula:
  Dc[%]=(Bx-Ax)/Ax‧100, (1)
  where:
  
  Bx is the value of the physico-chemical parameter x in the pond or stream below the dam, while Ax is the value of the physico-chemical parameter x in the stream above the dam during individual sampling sessions.
  Physico-chemical water parameters were included in the PCA if their mean absolute difference in Dc [%] (for all sampling sites combined) exceeded 5%. This calculation used the following formula (N is the total number of sampling sessions):
  ǀDcǀ[%]=∑(|(Bx-Ax)/Ax|‧100/N), (2)
  The variables that fulfilled the aforementioned assumption were: DO, H⁺, Ca²⁺, K⁺, HCO₃^-, SO₄^2-, Cl^-, NO₃^-, and NH₄⁺ (Table A1). The H⁺ concentration was used instead of the pH value to avoid the analysis involving logarithmic variables. The aim of using DC [%] in the PCA analysis was to identify the factors that determine changes in the physico-chemical parameters of water caused by the presence of beaver dams. The PCA method reduces the complexity of a large data set to a smaller set of easily interpretable factors that can be associated with specific processes (Drever, 1997). A matrix of factor scores, one of the most important parts of the PCA output, was also analysed. Factor scores provide a measure of the relation between each observation (i.e., each sample) and the identified factors (Shaw, Wheeler, 1997). The Kolmogorov–Smirnov and Lilliefors tests were used to determine the normality of the distributions of the variables. The Kaiser criterion was used to separate factors (the eigenvalues >1)...’
  
  Discussion section (Fig. 12): I am not sure if I can understand the scheme in Fig. 12. Complex and comprehensive information output, what image (period, age, geomorphology type) represents and how this typology can affect chemical composition is missing. From the image, it is not clear how different periods, pond types and age are related to chemical water properties (which typology affects the increase or decrease of chemical composition) and what changes are related to which presented type.
  Answer: We modified this figure (now Fig. 11) completely to make its content clearer and more comprehensive. Previously it was intended to emphasize the difference between periods, types or age of the beaver ponds by describing the direction of the difference alone. The revised figure presents the average magnitude and direction of change observed in each period, type, and age of beaver pond. We believe that enriched content of the figure makes it more informative, and at the same time more understandable. Please find the revised figure in the attached document (Figure 11.pdf).
  
  Discussion: Even though there is a chapter in the Results section devoted to the limitations of the study, I would recommend moving it and expanding it in the Discussion section as a separate chapter.
  Answer: As suggested, we moved this chapter to the Discussion section. This chapter has been substantially revised and complemented with more content:
  ‘...The wide spatial scope of the study and the high diversity of beaver-influenced streams introduced certain limitations. The spread of the study sites across a vast area restricted the number of samples collected to one sample per site each season, due to the time and cost constraints. In addition, dam failures resulting from floods and their targeted elimination during the research period decreased the planned sample size by half. The remaining study sites may have been affected by continuous dam maintenance by beavers. In the most extreme cases, temporal breaches of the dam may have been overlooked after its reconstruction. Moreover, since the study focused on the base flow conditions, early spring high flows have been omitted, which likely explains the similarities between the results from the late spring and the summer.
  Although a Before–After–Control–Impact (BACI) model would be the most appropriate approach for studying the impact of beavers on the chemical composition of water, it was not possible to use this method in this area due to the lack of investigations prior to dam construction. Instead, the study focused on spatial and temporal contrasts throughout the year and used a comparative ‘upper-lower river’ method, where upstream sections were used as reference sections. This research approach is commonly used in field studies to assess hydrochemical gradients generated by beaver dams when it is logistically or environmentally impossible to obtain ‘before’ data on dam construction (Green and Westbrook, 2009; Klotz, 2010; Fuller and Peckarsky, 2011; Puttock et al., 2017; Stevenson et al., 2022). It enables the identification of local hydrological and chemical effects caused by beaver damming while minimising the confounding influence of catchment heterogeneity. Although this approach lacks independent control sections, as in a full BACI model, it provides robust comparative evidence for evaluating the magnitude and spatial extent of changes induced by beaver activity under natural field conditions. Future research should be more detailed and extensive to enable the application of the BACI model.
  As there was no identified relationship between the initial values and the percentage change in the physicochemical parameters, it was assumed that the observed changes would not differ significantly based on upstream conditions. This allowed various catchments with different land uses and levels of anthropogenic influence to be included in the study. However, it should be noted that the inclusion of the streams characterised by particularly high concentrations of certain components, e.g. highly anthropogenically modified, may have resulted in different observations.
  Although the sample locations were standardised to a certain extent, they may not reflect typical stream or pond conditions. Stream samples are more likely to be well-mixed and representative due to turbulent flow conditions. This is also evidenced by the homogeneity of most longitudinal profiles upstream and downstream of beaver ponds. However, the spatial distribution of the pond samples may be more biased as it did not cover the full spectrum of hydrological units observed within the ponds (Majerova et al., 2020). Additionally, as the samples were taken from the middle of the water column, they do not represent the surface and bottom water layers, which according to the vertical sampling scheme could substantially affect observed physico-chemical parameters.
  The assessment of the impact of individual characteristics of beaver dam sequences may be constrained by the interconnections between some variables. Beaver ponds tend to be more persistent in smaller streams, and the length of the beaver occupancy seems to be related, to some extent, to the longitudinal and lateral development of the pond sequences. However, the estimation of mutual correlations was limited by the use of categorical values. Continuous values for the factors representing beaver pond age could not be obtained from the available sources, and the categorisation of the number of ponds was carried out because of the dynamics of this variable during the study period…’
  
  Thank you for taking the time and effort to review this paper. The review definitely helped us improve both the content and readability of the manuscript.
  
  Citation: https://doi.org/10.5194/egusphere-2025-1184-AC3

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

ED: Reconsider after major revisions (11 Nov 2025) by Gabriel Singer

Comments to the Author: PDF

Hide

AR by Joanna Wąs on behalf of the Authors (15 Dec 2025) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (22 Dec 2025) by Gabriel Singer

RR by Miloš Rusnák (28 Dec 2025)

RR by Anonymous Referee #3 (26 Jan 2026)

Suggestions for revision or reasons for rejection

Referee Report: PDF

Hide

ED: Reconsider after major revisions (25 Feb 2026) by Gabriel Singer

Hide

AR by Joanna Wąs on behalf of the Authors (01 Apr 2026) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (review by editor) (04 Apr 2026) by Gabriel Singer

Hide

AR by Joanna Wąs on behalf of the Authors (23 Apr 2026) Author's response Author's tracked changes Manuscript

ED: Publish subject to minor revisions (review by editor) (05 Jul 2026) by Gabriel Singer

AR by Joanna Wąs on behalf of the Authors (08 Jul 2026) Author's response Author's tracked changes Manuscript

ED: Publish as is (10 Jul 2026) by Gabriel Singer

AR by Joanna Wąs on behalf of the Authors (13 Jul 2026) Manuscript

Journal article(s) based on this preprint

07 Aug 2026

The effect of beaver dams on the physico-chemical properties of stream water in the Carpathians (Poland and Slovakia)

Joanna Wąs, Joanna P. Siwek, Małgorzata Kijowska-Strugała, and Elżbieta Gorczyca

Biogeosciences, 23, 5437–5459, https://doi.org/10.5194/bg-23-5437-2026,https://doi.org/10.5194/bg-23-5437-2026, 2026

Short summary

Joanna Wąs, Joanna P. Siwek, Małgorzata Kijowska-Strugała, and Elżbieta Gorczyca

Viewed

Total article views: 8,046 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
6,175	1,650	221	8,046	146	279

HTML: 6,175
PDF: 1,650
XML: 221
Total: 8,046
BibTeX: 146
EndNote: 279

Views and downloads (calculated since 04 Apr 2025)

Month	HTML	PDF	XML	Total
Apr 2025	477	105	25	607
May 2025	225	114	10	349
Jun 2025	233	90	20	343
Jul 2025	170	45	0	215
Aug 2025	521	69	0	590
Sep 2025	2,537	59	5	2,601
Oct 2025	610	115	20	745
Nov 2025	200	240	60	500
Dec 2025	239	209	15	463
Jan 2026	249	156	25	430
Feb 2026	192	147	13	352
Mar 2026	273	159	15	447
Apr 2026	60	40	0	100
May 2026	37	26	0	63
Jun 2026	7	5	0	12
Jul 2026	13	10	2	25
Aug 2026	23	13	2	38
Sep 2026	89	35	5	129
Oct 2026	20	13	4	37

Cumulative views and downloads (calculated since 04 Apr 2025)

Month	HTML	PDF	XML	Total
Apr 2025	477	105	25	607
May 2025	225	114	10	349
Jun 2025	233	90	20	343
Jul 2025	170	45	0	215
Aug 2025	521	69	0	590
Sep 2025	2,537	59	5	2,601
Oct 2025	610	115	20	745
Nov 2025	200	240	60	500
Dec 2025	239	209	15	463
Jan 2026	249	156	25	430
Feb 2026	192	147	13	352
Mar 2026	273	159	15	447
Apr 2026	60	40	0	100
May 2026	37	26	0	63
Jun 2026	7	5	0	12
Jul 2026	13	10	2	25
Aug 2026	23	13	2	38
Sep 2026	89	35	5	129
Oct 2026	20	13	4	37

Viewed (geographical distribution)

Total article views: 7,997 (including HTML, PDF, and XML) Thereof 7,997 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 12 Oct 2026

Download

The requested preprint has a corresponding peer-reviewed final revised paper. You are encouraged to refer to the final revised version.

Preprint (3079 KB)
Metadata XML

Short summary

The recent return of beavers to many regions has drawn scientific attention to the effects of their dam-building activity. This study identified beaver pond characteristics that significantly influence biogeochemical processes occurring in beaver-inhabited rivers. The study shows that more expansive and older ponds are more effective in altering physico-chemical water parameters. Such results support the need for the conservation of beaver ponds as a nature-based water management solution.


Total:	0
HTML:	0
PDF:	0
XML:	0