All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
This revised version is suitable for publication in PeerJ.
[# PeerJ Staff Note - this decision was reviewed and approved by Jörg Oehlmann, a PeerJ Section Editor covering this Section #]
I thank the author for taking into consideration my comments and suggestions. The paper now looks good and ready for publication. Good effort!
I thank the author for taking into consideration my comments and suggestions. The paper now looks good and ready for publication. Good effort!
I thank the author for taking into consideration my comments and suggestions. The paper now looks good and ready for publication. Good effort!
I thank the author for taking into consideration my comments and suggestions. The paper now looks good and ready for publication. Good effort!
Some dense figures lack legends or uncertainties, making interpretation difficult. Integrating confidence intervals or error bars would improve reporting clarity. The literature assessment is thorough, however this work fails to address MODIS's validation issues, such as cloud contamination and forest biases. Add legends and uncertainty ranges to figures to enhance readability, and include a brief section acknowledging MODIS biases and their implications for H43 validation.
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
I find the article well-written, but the introduction should be revised to avoid turning into a literature review. It should more clearly highlight the motivations behind this study.
Introduction: The introduction is too long and feels more like a literature review. Instead of describing each sensor in a separate paragraph, it would be much more relevant and focused on the topic to compare the different methods for observing snow using optical and radar remote sensing. For example, the paper does not mention the Landsat and Sentinel-2 sensors in the optical satellite section, even though they are widely used and highly valuable, especially in mountainous regions.
It would be useful to include a table comparing the advantages and disadvantages of each sensor, while keeping in mind that the focus of this article is on Meteosat. Thus, a comparison with Meteosat should be made, and the rationale for choosing Meteosat over other sensors should be explained. I find that the motivations for using these sensors, as opposed to SPOT, for instance, are not clearly addressed. Additionally, a comparison with MODIS could be useful.
Validation with In-situ Data: I believe the validation with in-situ data is quite limited because in mountainous areas, conditions can change significantly just 2 kilometers away from the measurement point.
Regarding line 669: What threshold is used to classify data as snow at the station? Is it greater than 0cm?
In section 3.1.1, when discussing the Alps, differences between the two products are mentioned, but no numerical values are provided, whereas figures are given for other regions. It would be helpful to include numbers in parentheses for the Alps as well.
In section 3.1.3, does performing analyses with altitude bands at 500m really make sense in mountainous regions when working with a 2km resolution?
Once again, in the conclusion, there is a description of the article, but the motivations are not clearly addressed. Why is this study innovative? Why should these products be used over others?
The manuscript is generally well-written in professional English, featuring a clear structure and sufficient references. The introduction offers a thorough overview of the current state of satellite-based snow monitoring and highlights the significance of the H SAF program. The figures and tables are relevant, and the data is effectively shared.
However, there are a few areas needing attention:
Some figures are quite dense and lack clear legends or indications of uncertainty, making them challenging to interpret. Incorporating confidence intervals or error bars would greatly enhance the clarity of the reporting.
Although the literature review is robust, the limitations of MODIS as a validation reference, such as cloud contamination and forest biases, are not adequately addressed in relation to the findings of this study.
To enhance the manuscript, consider improving the readability of the figures by adding legends and uncertainty ranges, and including a brief subsection that explicitly acknowledges known biases of MODIS and their implications for the validation of H43.
The research question, evaluating the inaugural MTG-FCI-based H43 snow product, is clearly articulated and appropriately aligned with the journal’s scope. The study design, involving validation against MODIS data and in-situ measurements across three mountainous regions, is well-conceived and substantively relevant.
Concerns:
The cloud masking methodology relies on SEVIRI-derived products resampled to the FCI scale. This approach may introduce artifacts that are not examined in sufficient detail.
In certain regions, particularly Georgia, the coverage of in-situ stations is limited, which constrains the statistical robustness of the findings. While this limitation is acknowledged in the manuscript, the associated uncertainty is not comprehensively quantified.
The analysis period (December–February 2024–25) is relatively brief. Although suitable for an initial evaluation, assertions regarding operational readiness should be moderated.
Suggested improvement: Provide explicit quantification of uncertainty, including confidence intervals and sample sizes in tabulated results, and adjust conclusions to reflect the restricted validation period and limited in-situ data coverage.
The findings from this analysis are generally sound, illustrating that the H43 model exhibits improvements in Probability of Detection (POD) and a decrease in False Alarm Rate (FAR) when compared to the H34 model. These results are consistent with earlier expectations stemming from the advancements provided by the MTG-FCI sensor technology.
However, several important concerns warrant attention:
Firstly, although the reported enhancements in POD are evident, they are, in certain instances, marginal, showing increases of only 0.004 to 0.01. The strong conclusions drawn from these modest improvements could be perceived as somewhat overconfident. It would be prudent to adopt a more cautious tone in these interpretations to more accurately reflect the subtle nature of the enhancements.
Additionally, the validation process utilizing MODIS data may unintentionally introduce its own biases into the overall evaluation. This potential for bias is acknowledged within the findings, yet the discussion surrounding it is somewhat limited and would benefit from a more thorough examination.
Furthermore, the lack of uncertainty ranges in the results complicates the assessment of statistical robustness. Without a clear understanding of the potential errors or variations in the data, it becomes challenging to fully evaluate the reliability of the reported findings.
To enhance the rigor of this analysis, I recommend reframing the conclusions as indicative of an initial evaluation phase rather than positioning them as conclusive proof of operational validation. Additionally, incorporating thorough uncertainty quantification will bolster the credibility of the claims made and provide a clearer understanding of the confidence level associated with the results.
This evaluation serves as a timely and significant addition to the field, as it represents the first assessment of the MTG-FCI-derived H43 snow product. This research is of great importance to both the remote sensing and hydrology communities, providing valuable insights into the effectiveness and performance of snow products.
To enhance the manuscript's quality and increase its chances for publication, I recommend several substantial revisions. In particular, the authors should emphasize a more thorough analysis of the uncertainties surrounding the findings and adopt a more cautious approach when interpreting marginal improvements. Additionally, improving the clarity and visual presentation of the figures will greatly enhance the overall comprehensibility of the manuscript.
In summary, while the study is fundamentally sound, concerns regarding the limited scope of validation, the need for precise uncertainty quantification, and occasional exaggeration of results necessitate comprehensive revisions. I would welcome the opportunity to re-review the manuscript following the implementation of these improvements.
-
-
-
1. General Evaluation
This manuscript presents an initial evaluation of the EUMETSAT H SAF H43 snow cover product, the first operational snow mask derived from the Flexible Combined Imager (FCI) aboard Meteosat Third Generation (MTG). The paper provides a clear and systematic validation against MODIS MOD10A1 (Collection 6.1) and WMO in-situ snow depth (SD) data across three geographically distinct regions — the Alps, Turkey, and the Caucasus.
The topic is timely and scientifically significant, particularly given the transition from MSG-SEVIRI to MTG-FCI platforms and the need to assess their implications for operational snow monitoring. The study is methodologically sound, well-structured, and written with commendable clarity. The choice of datasets and evaluation metrics (POD, FAR, ACC) is appropriate and aligns with established validation standards for remote sensing snow products.
Overall, the manuscript makes a valuable contribution to the cryospheric remote sensing community and the EUMETSAT H SAF user base. Only minor editorial and structural refinements are suggested to improve clarity and readability.
2. General Comments
2.1. Abstract
The abstract is generally clear and well-structured, effectively summarizing the study’s objectives and main findings. However, the following minor improvements are recommended:
• The authors should consider specifying the validation period explicitly (e.g., December 2024 to February 2025) to provide temporal context for the evaluation.
• It would be helpful to briefly mention the evaluation metrics employed (e.g., POD, FAR, ACC) to clarify the basis for performance assessment.
• The phrase “Particular attention was paid to...” appears twice (lines 30 and 32). Merging these into a single, more concise statement would improve clarity.
• The final sentence (lines 43–45) is overly long and may benefit from being split into two sentences to enhance readability and highlight the operational value of the H43 product.
2.2. Introduction
The introduction is generally well-structured and provides sufficient background on the topic. A few minor enhancements are suggested to improve clarity and scientific framing:
• Lack of Explicit Research Question or Hypothesis: The manuscript would benefit from a clearly stated research question or hypothesis to guide the study (e.g., assessing the reliability of the H43 product in complex mountainous terrain).
• Motivation Could Be More Focused: While the motivation is clear, it could be more concise and more directly linked to limitations in previous SEVIRI-based products (e.g., spatial constraints in H10 and H34).
• Objectives Are Clear but Slightly Verbose: The listed objectives (lines 354–361) are relevant but could be grouped or reworded into 2–3 concise goals to enhance readability and focus.
2.3. Materials & Methods
• Visualization of the H43 Algorithm: Consider including a schematic diagram to illustrate the processing chain of the H43 snow detection algorithm. This would help clarify the sequence of thresholding, filtering, and classification steps, especially for readers less familiar with FCI-based processing.
• MODIS Cloud Masking and Gap-Filling: It would be helpful to clarify whether cloud-covered MODIS pixels were excluded or addressed using any gap-filling techniques during the validation process. Since cloud contamination can significantly affect snow detection, explaining the treatment of cloudy pixels would enhance the methodological transparency.
• Justification for MODIS as Reference Dataset: The manuscript should briefly justify the choice of MODIS MOD10A1 (500 m) as the primary reference for snow cover, especially when higher-resolution alternatives such as Landsat (30 m) or Sentinel-2 (10 m) are available. A short discussion on the trade-offs (e.g., spatial vs. temporal coverage, cloud handling, consistency with previous studies) would strengthen the rationale.
2.4. Results
• Clarification on In-situ Validation (Section 3.1.1, Figure 6): The authors should clarify how the H43 and H34 products were matched with the in-situ snow depth data. It is not explicitly stated whether the comparison was performed on a daily basis (i.e., pixel-to-station matchups) or through temporal aggregation (e.g., monthly or multi-day averages). Clarifying the temporal matching procedure would help readers better understand the reliability and temporal consistency of the validation results.
• Clarification on MODIS-Based Validation (Section 3.1.2, Figures 7–9): The box plots in Figures 7–9 appear to represent daily validation statistics, but this is not clearly explained in the text. The authors should explicitly state whether both H43 and H34 were validated against daily MODIS MOD10A1 snow-cover maps using daily composites of each product. It is recommended to mention this clearly in the Materials and Methods section.
• Additionally, including a concise methodological flowchart summarizing the overall validation process (datasets, temporal matching, and evaluation steps) would greatly improve transparency and readability.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.