We would like to express our sincere appreciation to the reviewer for the thorough and careful evaluation of our manuscript, as well as for the constructive and insightful comments and suggestions. We have carefully considered all comments and have revised the manuscript accordingly. Detailed responses to each comment are provided below. In addition, a supplementary document highlighting the revisions made in the manuscript has been provided for ease of reference.

Introduction

・Your introduction is very general and elaborate. The literature review adequately covers relevant studies but could not better highlight the novelty of this work. The rationale for focusing on small floodplain water bodies is not enough.

Response1

In accordance with the reviewer’s comments, we have streamlined the literature review and reorganized the description of the floodplains and floodplain waters in the study area, together with relevant previous studies. The text has also been revised to clarify the positioning and novelty of this study (L30–L74).

Materials and methods

・Specify the number, depth, and spatial distribution of shallow groundwater wells relative to the backwaters. Clarify how these wells represent the aquifer’s lateral and vertical heterogeneity, as this is critical for validating groundwater-endmember calculations.

Response2

We agree with the reviewer’s comment and have revised Figure 2 to more clearly illustrate the spatial distribution of shallow groundwater and its relationship with topography. In addition, based on previous studies, we have incorporated information on the stratigraphic distribution in the vicinity of the study area, as well as the flow direction of shallow groundwater (Figure 2, Lines 88–89).

・Since CDOM was only measured in November 2022, justify why this single snapshot is sufficient to infer seasonal contributions of river water, groundwater, and sewage. Provide details on sensor calibration, analytical precision, and replicate measurements to support CDOM as a reliable mixing tracer.

Response3

We agree with the reviewer’s comment and, to supplement the results based solely on data from November, have added data from approximately 40 locations each for rivers and shallow groundwater, including the study site, collected during a different period (L284–286). The additional data show that river CDOM concentrations are higher during periods of low flow (November in this study) than during the high-flow summer months (approximately July to September). In contrast, seasonal variations in CDOM concentrations in the surrounding shallow groundwater are minimal, and a clear difference in concentration between river water and shallow groundwater is maintained throughout the year. These results suggest that CDOM is an effective indicator for distinguishing backwater contributions. In addition, information on the basic performance of the CDOM sensor has been added (L122–L124).

Discussion

・L327：The inference of denitrification in Backwater C relies on the 2:1 δ¹⁵N:δ¹⁸O enrichment ratio, but additional evidence (e.g., dissolved N₂ gas concentrations, sediment denitrification potential assays, or microbial marker genes) is needed to confirm this process. Explain why denitrification is prominent in Backwater C but not in other backwaters, considering factors like organic matter availability, hydrological residence time, and sediment properties.

Response4

We were unable to obtain the additional evidence regarding denitrification suggested by the reviewer within the scope of this study. Therefore, we have clarified in the relevant section (L327–L337) that denitrification is presented as a possible explanation. Furthermore, as this study does not fully resolve the differences in the extent of denitrification inferred from NO₃ concentrations and δ¹⁵N− NO₃ and δ¹⁸O−NO₃ values among individual backwaters, we have cited relevant literature and emphasized the need for further detailed investigation.

・Clarify the potential influence of the factory drainage (600 m upstream of Backwater A) on water chemistry. Provide data on the drainage’s nutrient and tracer concentrations (e.g., CDOM, ions) to rule out or quantify its contribution to Backwater A’s unique ion composition (high Na⁺ and SO₄²⁻).

Response5

We agree with the reviewer’s comment and, as a supplement, have added water quality data for industrial wastewater upstream of Backwater A, obtained by a different research organization, to the main text (L280–L283). Furthermore, water quality parameters not included in the main text have been compiled into a table in the supplementary materials.

・L389：Define "DIN" (Dissolved Inorganic Nitrogen) upon first use, as it includes NO₃⁻, NH₄⁺, and NO₂⁻—a distinction critical for readers unfamiliar with aquatic biogeochemistry.

Response6

We agree with the reviewer’s comment. As this study measures only NO₃⁻ and NH₄⁺, we have added a statement at their first occurrence defining NO₃⁻ + NH₄⁺ as DIN (L22 and L133).

・There is no Conclusion.

Response7

As suggested by the reviewer, we have added a corresponding statement to the conclusion.

We sincerely thank the reviewer for the careful assessment of our manuscript and for the valuable comments and suggestions. We have carefully addressed all points raised and revised the manuscript accordingly. Detailed point-by-point responses are provided below. For clarity and ease of reference, we have also included a supplementary document summarizing the revisions made in the manuscript.

General Comments:

This manuscript investigates the origins and pathways of water and associated nutrient dynamics (N, P, Si) in three small backwater systems (“Wando”) within the urban Tama River, Tokyo. The authors employ a multi-tracer approach, including ²²²Rn, stable water isotopes (δ²H, δ¹⁸O), stable nitrate isotopes (δ¹⁵N, δ¹⁸O–NO₃), ion balance, and CDOM, to distinguish between river water underflow and shallow groundwater contributions. The study finds that water sources and nutrient conditions vary both spatially among the three backwaters and seasonally, with implications for nutrient stoichiometry (N:P:Si ratios) and, by extension, potential phytoplankton community composition. In my opinion the paper may be published with minor revisions after addressing comments provided below:

Abstract

The abstract is generally well-structured and informative, providing clear ranges of nutrient concentrations. It is somewhat long and could be tightened.

Response1

Thank you for your helpful comment. We have revised the abstract by simplifying several redundant sentences and reducing the overall length (L 8–11 and 22–26).

L17: The claim that CDOM serves as “an effective indicator” might overstated given the singlecampaign measurement as you stated the measurement “only in November 2022” (L122).

Response2

Thank you for your valuable comment. To supplement the results based solely on data from November, we have added data from approximately 40 locations each for rivers and shallow groundwater—including the survey site—from other periods (L284–286). Since it was confirmed that river CDOM concentrations are higher during periods of low flow (November in this study) compared to the summer months (around July to September) when flow is high, we included a statement in the abstract regarding the universality of its effectiveness as an indicator.

L25: Subject–verb disagreement, “…the backwaters, which was…”

Response3

Thank you for your helpful comment. We have deleted this sentence as part of the revision of the Abstract section.

L27–28: The final sentence is vague and reads more like a future research direction than a conclusion.

Response4

Thank you for your helpful comment. We have revised the final sentence to more clearly convey the main findings of this study. (Lines 27–28).

Introduction

The introduction provides adequate context on floodplain ecology, the flood pulse concept, and the specific Japanese context of embankment-confined floodplains, which is really good. The transition from large floodplain studies to small backwater systems is clearly motivated.

L60–62: The hypothesis statement is somewhat generic. The specific testable predictions could be sharpened, for example, what nutrient dynamics differences were expected, and why? The introduction would also benefit from briefly mentioning previous multi-tracer studies in floodplain contexts to better position the novelty of this work.

Response5

Thank you for your valuable comment. Following the suggestion of Reviewer 1, we have revised and streamlined a concise overview of floodplain research. We have also added background information to introduce the hypothesis that environmental characteristics unique to Japanese rivers (i.e., short river lengths and steep gradients) influence small-scale floodplain systems, thereby clarifying the novelty of this study. Furthermore, we have incorporated, as you suggested, a description of the hypothesis for the study area as well as previous tracer-based studies in floodplains. (L30-L74).

Methods

The study site description is detailed and supported by clear figures. Analytical methods are generally well-described with appropriate references to standard protocols. However, it would be nice to provide analytical precision values for measurements, especially the isotope measurement.

L94: “Pond A” appears to refer to Backwater A?

Response 6

Thank you for pointing this out. The incorrect term has been corrected to “Backwater A” (Line 94).

L109: Are the number of replicate samples (n=3) at each site likely refers to field replicates?

Response 7

Thank you for your question. We collected water samples in three separate bottles during field sampling. To avoid misunderstanding, the sentence has been revised for clarity (Line 109).

L122–124: CDOM measurement is described only briefly, and calibration is not detailed sufficiently for reproducibility.

Response 8

Thank you for your helpful comment. We have added information on the basic specifications of the commercially available CDOM sensor and clarified that calibration was conducted only at the factory before shipment (Lines 112–124).

L138: “Standard materials USGS32, USGS34, and IAEA were used”. Which IAEA standard? IAEA-NO-3? Please specify the full standard details.

Response 9

Thank you for your comment. We have corrected the notation to IAEA-NO₃, as we used this standard reference material (Line 138).

Results

Results are presented systematically. The data are rich and generally well-organized.

L236: DSi concentrations for Backwater C are reported as “14 to 40 μmol”, the unit L−1 appears to be missing. Additionally, these values are dramatically lower than those for Backwater A and B (175–276 μmol L⁻¹), yet this large discrepancy is not discussed. Is this an artifact of diatom uptake, dilution by low-DSi rainwater, or a data entry error?

Response10

Thank you for pointing this out. The value “14–40 μmol” was the difference in concentration between river water and backwater. We mistakenly left this value, which had been included in the first draft. The correct concentration range for Backwater C is 189–252 µmol L^-1. We have revised the text accordingly to avoid any misunderstanding.

L250: “corresponding” should be “corresponding”?

Response11

Thank you for your helpful comment. The spelling has been corrected(L249-L250)

L268: “Ellins’ et al., 1990”, the apostrophe placement is unusual.

Response12

Thank you for your comment. We have reviewed the cited reference and corrected the notation accordingly (Line 268).

Discussion

The discussion is structured around water origin estimation and nutrient dynamics, which is logical. However, the discussion remains largely qualitative and descriptive, it would be nice to have quantitative mixing estimates, but it seems out of the scope of this research.

Response13

Thank you for your valuable comment. We attempted to estimate mixing ratios based on data from major ions, ²²²Rn, and δ²H & δ¹⁸O–H₂O to support a quantitative discussion. However, due to the variability in shallow groundwater and the lack of data on discharge rates, these estimates involved substantial uncertainty. Therefore, determining mixing ratios at multiple sites remains a subject for future work.

L322: The δ¹⁵N range for river water is cited as “15–20‰, Kumazawa, 1999” but the authors’ own data show a range of 12.7‰–18.1‰ (L240). Using the authors’ measured values rather than a literature value would be more appropriate here.

Response14

Thank you for your comment. The text has been revised so that the discussion is explicitly based on the δ¹⁵N-NO₃ measurements obtained in this study.

L364: “groudwater” should read “groundwater” and “compostion” should read “composition.”

Response15

Thank you for your helpful comment. The spelling has been corrected (L364).

Conclusion?

The manuscript ends abruptly after Section 4.4 without a Conclusions section. For Biogeosciences, a concise synthesis of the main findings, their broader significance, and key limitations is expected. The current Discussion does not adequately summarize the study's contribution. A dedicated Conclusions section (approximately 150–250 words) should be added.

Response16

Thank you for your comment. The conclusion has been revised to provide a concise summary of the study’s findings, significance, and limitations based on the discussion.

Data availability

L395: The statement “Available from the corresponding author upon reasonable request” does not meet Biogeosciences’ data policy. The data should be deposited in a public repository prior to acceptance.

Response17

Thank you for your comment. In response, the data used in this study have been made publicly available through the Zenodo repository under the title “Water chemistry and geochemical tracers dataset from the Tama River Watershed, Japan (2022)” (DOI: 10.5281/zenodo.20115076). The corresponding information has been added to the revised manuscript (L395).