General comments:

The manuscript entitled "Terrestrial browning from colored dissolved organic matter (CDOM) changes the seasonal phenology of the coastal Arctic carbon cycle" explores the influence of CDOM on both biogeochemical and physical processes in the Mackenzie River region. The subject is both relevant and timely, addressing key concerns about the role of land-ocean interactions in a rapidly changing Arctic environment. The study provides promising insights that could significantly advance our understanding of how CDOM modulates seasonal carbon fluxes and ecosystem functioning in coastal regions.

However, a few issues require clarification and refinement before the manuscript can be considered for publication.

Major comments:

1. Clarification needed on the consistency of the validation datasets and years used

The authors state that they focus their analysis on the year 2012, and most of the figures throughout the manuscript appear to reflect results from that year. However, it remains unclear whether the model-data comparisons presented are also based solely on 2012 data. For instance, in Appendix B (Figure B1), the terrestrial CDOM ratio is compared to observations from 2009. The manuscript would benefit from a clear statement on whether these validation figures are meant to support the 2012 model run specifically, or whether they are included to assess model performance more generally.

In addition, a major part of the validation appears to rely on the study by Lewis and Arrigo (2020), yet the exact years used in that comparison are not clearly stated in the present manuscript. Upon reviewing Lewis and Arrigo (2020), it is also not obvious which datasets were used by the authors of the current manuscript to reproduce the validation figures, particularly Figure D1.

Greater clarity is needed on:

• which years of observational data were used in each validation figure,
• whether these years correspond to the 2012 simulation,
• and which specific datasets were extracted from Lewis and Arrigo (2020) or other sources.

2. Terminology in the manuscript may overstate the scope of the analysis

In a few places, the manuscript uses terminology that suggests a broader ecosystem analysis than is actually conducted. For example, in the introduction, the authors state that they explore the effect of CDOM on the “seasonal cycle of plankton biomass, productivity, and carbon cycling,” which implies a study of both autotrophic and heterotrophic plankton communities. Similarly, in Section 3.4, the title “CDOM effect on marine productivity” gives the impression that both primary and secondary production are addressed, whereas the analysis focuses solely on phytoplankton primary production.

Another example appears in the sentence: “We explore the specific effects of CDOM light attenuation and ocean heating on the coastal ecosystems and the carbon cycle.” Again, the wording implies a broader analysis than what is presented in the paper.

To improve clarity and avoid overstatement, I recommend that the authors revise general terms such as plankton productivity, ecosystem dynamics, and carbon cycle to more specific phrases such as phytoplankton production, chlorophyll concentration, or primary production. This is especially important given the study’s focus on a single trophic level (phytoplankton only), while terms such as “ecosystem” may imply a broader biological or biogeochemical system.

3. Potential confounding effects due to differing nutrient concentrations in simulations

In Figure 7a, it appears that the initial nutrient concentrations are higher in the CDOM ON simulation than in the control simulation (CDOM OFF). However, the manuscript does not mention any differences in nutrient forcing or initial conditions between the two simulations.

Given that nutrient availability is a key limiting factor for phytoplankton growth, such differences could strongly influence the interpretation of CDOM-related effects.

If nutrient fields or boundary fluxes differ between the simulations, the observed variations in phytoplankton biomass and productivity may not be attributable solely to the effects of CDOM on light attenuation or temperature. I recommend that the authors clarify whether nutrient initial conditions and boundary forcings were kept identical between simulations. If they differ, a justification should be provided, and the interpretation of the results should be revisited accordingly.

4. Clarification needed on bloom timing and comparison to observations

In the discussion (line 370), the authors state that the model "successfully simulates the timing of the bloom peak compared to Lewis et al. (2020),” while also noting that improvements to the sea ice model are needed to simulate bloom initiation more accurately. However, it is not clearly stated which simulation is being referenced (CDOM ON or CDOM OFF).

Based on the figures comparing model results to observations (e.g., Figure D1), it appears that the CDOM OFF simulation initiates the bloom closer to the observed timing, while the CDOM ON simulation results in a delayed bloom. This seems to contradict the statement that the model accurately reproduces bloom timing — unless the authors are referring specifically to the amplitude of the bloom rather than its onset.

I recommend that the authors clarify:

• which simulation is being evaluated when claiming agreement with observations,
• whether CDOM improves or delays bloom timing relative to the data,
• and how this affects the interpretation of CDOM's role in modulating bloom phenology.

A more nuanced discussion would be helpful here, especially if CDOM appears to delay the bloom in a way that is less consistent with observed phenology.

5. Apparent inconsistency between NPP and chlorophyll concentrations

In line 365, the authors state that the simulated phytoplankton amplitude is 85% higher in the CDOM ON simulation, which is consistent with the values shown in Figure 7a (NPP ≈ 4.6 Gg C d⁻¹ without CDOM vs. ≈ 8.1 Gg C d⁻¹ with CDOM). However, in Figure D1 — which compares surface chlorophyll concentrations from both simulations to observations — the amplitude of the two simulations appears quite similar.

Since chlorophyll is commonly used as a proxy for phytoplankton biomass, one would expect a notable difference in amplitude if NPP nearly doubles. The manuscript does not provide an explanation for this apparent discrepancy. Could the authors clarify whether this discrepancy reflects a decoupling between biomass and productivity in the model (e.g., due to C:Chl ratio dynamics or other processes such as loss terms or export)? Additional clarification would help interpret how CDOM influences not only the rate of primary production but also the standing stock and its comparison to observations.

Conclusion:

Overall, the manuscript addresses an important and timely topic with a solid modeling approach. With the clarifications and revisions suggested above, the study will present a clearer and more robust contribution to our understanding of CDOM effects in Arctic coastal systems.

Please find my review of the paper in the attachment.