Review of “Ensemble-based data assimilation improves hyperresolution snowpack simulations in forests” by Alonso-Gonzalez et al. (2025)

General comments

This study explores whether assimilation of remotely sensed snow depth observations available for forest clearings improve snowpack simulations below forest canopies where these measurements are missing. The authors performed six different data assimilation experiments using various configurations that affect prior correlations, and thereby the ability of the data assimilation schemes to propagate information from observed (open) to unobserved (forested areas) locations. The results show that four out of the six experiments improved the simulations in forested areas compared to the reference simulation. In the discussion, the authors provide an informative judgment of the results and specify future research possibilities for improving their methods further. Overall, the research presented in this study is highly relevant, as snow in forests can be important for a large range of scientific (e.g., ecological studies) and practical (e.g., water resources management) applications. The methods presented here are at the forefront of snow data assimilation research and demonstrate promising results. The study is well-written and provides valuable insights, and is therefore an excellent study that only requires a few minor adjustments before eventual publication in my opinion.

Specific comments

L 71-83: I think this paragraph can be shortened and should focus on why snowpack monitoring in forests, in particular below forest canopies, is challenging.

L 84-134: I recommend to shorten these two paragraphs too since the introduction is rather verbose. Just state the main problems concisely with references to the extensive literature, such as on challenges with forcing data as one example.

L 42-201: Overall, I find the introduction a bit long and verbose. Please shorten where possible.

L 209-214: I don’t understand this sentence. Please clarify.

L 251-253: Please provide a scientific valid justification why this model was selected. Technical simplicity is not enough in my opinion. Why is the model appropriate for the experiments performed in this study? Which previous studies supports the choice of this model for the particular region, snow conditions and canopy properties?

L 504-506 and Table 1: I assume this table is for the snow depths of the canopy-covered cells? This is what L 231-233 states. Nevertheless, please specify this in the table caption and the text to avoid misunderstanding, or start the result section with repeating the information on L 231-233.

L 602-606: Please add some more text here to guide the reader what the figure shows. It seems that low precipitation multipliers are associated with negative temperature adjustments, for instance. Is there an elevation trend in the values for precipitation multipliers and temperature additions?

L 623-636: It would be interesting to know how much time was spent running the model and how much time was needed for the DA algorithm separately. If I understand correctly, you run 100 ensembles with 40401 cells, and repeat this simulation 4 times in the iterative framework. Correct?

L 669-747: Maybe split these two long paragraphs into shorter ones.

L 788-792: Please simplify this sentence since it is hard to read and understand.

Technical comments

L 137: Capital letter after comma.

L 186-188: “Available” twice. Remove one.

L 209: “snow-off”?

L 275: What is “2m air fields”?

L 288: Missing period.

L 308: I guess “using” is missing.

L 364: On period too much.

L 475: Abbreviation PCA has not been introduced.

Please refer to the PDF for comments.