The manuscript addresses an important topic with valuable data, but significant reorganization, clearer figures, and expanded discussion are needed to improve clarity and scientific impact. I therefore recommend major revisions. With the above changes, the paper would be much more accessible to a broader readership and would better highlight its contributions.

Major Comments

1. Figures and Visualization

• Figure 1: The current layout is difficult to interpret: the legends are too small, and colors are used simultaneously to represent ¹³⁷Cs deposition and river catchments. I recommend separating these elements:
• Create a first panel where colors represent only ¹³⁷Cs deposition, and identify river catchments with labels instead of colors.
• Zoom in slightly so that Fukushima Dai-ichi Nuclear Power Plant is clearly visible.
• Include an inset map showing the study area within Japan to orient readers unfamiliar with the region.
• Describe all acronyms in the figure caption.
• Contextual location: There is no figure linking the lagoon to its main ¹³⁷Cs source (Fukushima). A simple map indicating the relative position of the lagoon and nuclear plant would strengthen context for non-local readers.

1. Methods vs. Results

• Some parameters discussed in Results and Discussion could be introduced in Methods:
• K_d calculation: Please specify explicitly how K_d was obtained (e.g., Bq m⁻³ divided by Bq kg⁻¹).
• Voronoi partitioning: Mention and briefly describe this procedure in Methods (currently introduced only in line 199).
• In general, all analytical and data-processing steps used to represent or interpret results should be described in Methods before they appear in the Results section.

1. Representation of Results

• Figure 2: The statistical plot combines all river catchment data, but much of the underlying data are not shown elsewhere (only partially in Table S1). Consider a clearer, more illustrative figure:
• For example, a four-panel plot—one per river—showing monthly dissolved and particulate ¹³⁷Cs concentrations (Bq m⁻³) as column plots.
• This would highlight the key findings: (a) Ume and Nikkeshi contribute most ¹³⁷Cs to the estuary; (b) fluxes peak in summer.
• Remaining parameters (Bq kg⁻¹, g m⁻³, K_d) could either share a secondary axis or be kept in a table placed in the main text rather than in the Supplementary Information.
• Tables:
• Table 1 seems to contain background or supporting data rather than primary results. Consider moving it to the Supplementary Information and bringing Table S1 into the main text.

1. Organization of Results and Discussion

• Section 3.2 (Lagoon results):
• Subsection 3.2.1 is titled “Relationship between dissolved ¹³⁷Cs and salinity” but includes all parameters and does not first present the spatial distribution of ¹³⁷Cs itself. Begin with a clear presentation of particulate and dissolved ¹³⁷Cs distributions—perhaps using isosurface plots or panel figures similar to those for rivers (even a subset of months for clarity).
• The investigation of relationships with salinity fits better in the Discussion rather than Results. Splitting Results and Discussion would improve flow and clarity.
• Figure 3 (salinity relationships): The panels suggest little to no correlation. Consider exploring:
• Subsets of data (individual rivers, specific lagoon stations near river mouths, or specific seasons).
• If no relationship exists under any subset, move these plots to the Supplementary Information or remove them.
• Section 3.2.2 and Figure 4: Similar concerns apply here—repeated salinity analysis without a clear signal. A deeper targeted analysis or removal is advised.
• Figure 5: This clearly shows seasonal variability (higher in summer, lower in winter) and could be placed under a dedicated subsection focused on temporal variability rather than under temperature–salinity relationships.

1. Section 3.3 (Modeling)

• Add a schematic diagram summarizing the model setup and assumptions for desorption/dissolution of Cs. Values adopted from the literature could be noted on the schematic for transparency.
• Include final model results within the main text (e.g., from Figure S1). Panels a–c of Figure S1 are particularly informative, illustrating that most ¹³⁷Cs is discharged in August and that the particulate fraction dominates.

1. Broader Perspective and Implications

• Quantify, if possible, the flux of ¹³⁷Cs from the lagoon to the Pacific Ocean, addressing one of the paper’s stated aims.
• Place findings in a broader context:
• How significant are lagoon discharges compared with other ¹³⁷Cs sources to the Pacific?
• Are there similar studies (in Japan or elsewhere) that could provide comparison?
• What are the implications of these results for understanding radionuclide transport or for monitoring strategies?
• Suggest potential next steps or research directions based on the conclusions.

General Comments

This study focuses on Matsukawa-ura, a brackish lagoon, where the authors conducted observations of dissolved and particulate ^137Cs concentrations in lagoon and inflowing river waters during 2021–2023, approximately ten years after the Fukushima Daiichi Nuclear Power Plant accident. The study demonstrates the seasonal dependence of dissolved ^137Cs concentrations in the lagoon and estimates particulate and dissolved fluxes from rivers. The results suggest that riverine fluxes alone cannot explain the elevated dissolved ^137Cs concentrations in the lagoon, and the authors cite previous studies to point out the possible contribution from bottom sediments. The observational dataset itself is valuable and provides important insights for understanding the long-term dynamics of radiocesium in Matsukawa-ura.

However, the section on mass flux analysis contains fundamental problems, and in its current form, the conclusions cannot be supported. The flux estimation includes many questionable assumptions and methodological flaws. I believe that this part requires major revision or should be removed entirely.

While the observational data are highly valuable, the mass balance analysis is methodologically and conceptually inappropriate and does not support the conclusions. I strongly recommend removing or substantially revising this section. If the manuscript focuses instead on the observational results and their comparison with previous studies, it would represent a useful and significant contribution.

Specific Comments

Mass balance section

L244: The authors estimated the supply flux from sediments by using the tidal prism to calculate “the volume of seawater inflow per 12 hours,” and assuming that this inflowing water uniformly reaches the lagoon’s mean concentration within 12 hours. The difference was then attributed to sedimentary fluxes. However, this estimation has several critical flaws:

(1) The tidal prism does not represent a “pure inflow” of new seawater. In reality, inflow and outflow occur simultaneously, and the exchange efficiency is less than 100%.

(2) The assumption that inflowing water reaches the lagoon-wide average concentration uniformly within 12 hours is unrealistic. Observations clearly show higher concentrations in the inner lagoon, demonstrating strong spatial heterogeneity.

(3) The authors assume a constant tidal prism exchange volume throughout the year, yet still discuss seasonal (summer vs. winter) differences. This is logically inconsistent.

For these reasons, the flux estimation is not valid and should be considered inappropriate.

L260: The method used to estimate the decrease of ^137Cs in bottom sediments is unclear. Given that the mass flux calculation itself is subject to large uncertainties, presenting the results as quantified values—and further highlighting them in the abstract—risks giving readers an impression of unwarranted accuracy. Considering the likelihood of these numbers being cited in future work, the current presentation lacks scientific integrity. This section should be deleted or, at the very least, thoroughly rewritten.

L31：Sanial et al. (2017) is not a study of sedimentary fluxes and its citation here is inappropriate. The positioning of this reference should be reconsidered.

Introduction：Kambayashi et al. (2021) is a highly relevant prior study and should be cited in the Introduction.

Fig. 4：The spatial arrangement of sampling sites is difficult to understand. I recommend improving the map by adding scale bars, landmarks, or other visual cues to make the locations more intuitive.