the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 02 Jun 2026
A new method for updating snow fields in the NWP models using satellite snow extent based ’Snow Barrel’ pseudo-observations as applied to HARMONIE-AROME cycle 43
Abstract. This paper introduces the "snow barrel" method, a new approach to integrate satellite-derived snow observations into NWP models. Snow is a key component of the environment, helping to regulate surface temperature, atmospheric and soil conditions, and playing a key role in the water cycle. However, assimilating satellite snow data into NWP models remains challenging due to resolution mismatches and the complexity of handling snow extent observations. The snow barrel method addresses these challenges by aggregating satellite pixel observations from the EUMETSAT H-SAF H32 intermediate product into 10x10 pixel areas and creating pseudo-observations that align with NWP model scales. Implemented within the HARMONIE-AROME model in the MetCoOp operational domain covering northern Europe, this approach selectively applies snow barrel observations where they conflict with the model's background field, particularly in regions with thin or patchy snow cover. The results demonstrate improved representation of snow cover during the transitional seasons without disrupting areas of solid cover or imposing a significant computational burden. The method effectively combines the spatial coverage advantages of satellite data with the precision of in situ measurements, particularly benefiting areas with sparse ground observations. Although constrained by cloud cover and lighting conditions inherent to optical satellite products, the snow barrel methodology offers a flexible framework that could be expanded to other satellite platforms and potentially adapted for additional surface parameters beyond snow cover.
- Preprint
(7922 KB) - Metadata XML
- BibTeX
- EndNote
Status: final response (author comments only)
-
RC1: 'Comment on egusphere-2026-1408', Anonymous Referee #1, 07 Sep 2026
-
AC1: 'Reply on RC1', Mikael Hasu, 16 Oct 2026
RESPONSE TO REVIEWER #1
General comments
Reviewer comment:
This manuscript describes the use of a new “Snow Barrel” methodology to process satellite snow cover observations from the EUMETSAT H SAF H32 product for assimilation within the HARMONIE-AROME regional NWP system. The manuscript is well written and clearly presented, and the assimilation experiments demonstrate clear benefits to the analysed snow field relative to the use of in situ observations alone. Although it represents a potentially valuable contribution to the NWP and snow DA community, I remain unconvinced that the manuscript fully establishes (1) the novelty of the Snow Barrel concept relative to existing observation-processing approaches for satellite snow cover, (2) the added value of the Snow Barrel methodology itself relative to the underlying satellite observations, and (3) the operational benefit of the approach for NWP forecasting.
Response:
We thank Reviewer #1 for the valuable comments. We have taken them into account, which has helped us improve the clarity of the manuscript. Our detailed responses are provided below.Reviewer comment:
The manuscript would benefit from a clearer justification of the novelty of the Snow Barrel concept. As presented, the methodology appears largely to comprise established observation-processing steps: aggregation (superobbing) of high-resolution observations to the model scale, empirical conversion to a model variable, and quality-control and data-selection procedures. While the implementation is tailored to the target system, it is unclear what constitutes a substantive methodological advance beyond existing practice. The authors should more clearly distinguish the novel aspects of the approach from established operational processing frameworks.
Response:
The novelty of the Snow Barrel method is primarily technical. To construct the barrels, we use the intermediate H32 product rather than the final product. The intermediate product is provided on the satellite grid, whereas the final product is provided on a regular latitude-longitude grid. This allows us to avoid unnecessary interpolation from the satellite grid to the latitude-longitude grid and then to the atmospheric model grid. It also avoids overly aggressive thinning, because in polar regions the physical spacing of the latitude-longitude grid approaches zero. Another novel aspect is superobbing. Although superobbing is widely used in the assimilation of other data types, to the best of our knowledge it has not previously been applied to surface data assimilation, particularly snow data assimilation. We have revised the text to clarify the method and highlight these aspects.Reviewer comment:
While the manuscript demonstrates a positive impact from assimilating satellite snow-cover information, it does not isolate the contribution of the Snow Barrel methodology itself. Several aspects of the approach, including the observation-selection criteria, observation-to-SWE conversion, and prescribed error characteristics, appear to be based on heuristic choices, yet little evidence is provided that they have been optimized and demonstrate benefit compared to alternative approaches. A direct comparison between assimilation of Snow Barrels and assimilation of the original H32 product would be particularly valuable, as it would distinguish the impact of the satellite observations from the impact of the Snow Barrel processing. Without such evidence, it is difficult to determine how much added value the proposed approach provides.
Response:
We agree that several aspects of the data assimilation methodology are based on heuristic choices. These parameters were inherited from earlier snow-analysis development and were not systematically optimized as part of this study. We have clarified this more explicitly in the revised manuscript.We also agree that a comparison with the original H32 product would be valuable. However, such a comparison is not straightforward within the current operational framework. The Snow Barrel methodology uses the intermediate H32 satellite product in its native satellite geometry and converts it into SWE pseudo-observations suitable for assimilation. In contrast, the final H32 product is first reprojected onto a latitude-longitude grid and subsequently spatially processed. Assimilating the final H32 product would therefore require a substantially different observation-processing chain, including additional thinning and handling of spatially correlated observations.
The present study evaluates the complete Snow Barrel methodology, including the aggregation of satellite information into SWE pseudo-observations suitable for snow analysis. As a result, the experiments do not isolate the impact of the aggregation step itself. We have revised the manuscript to clarify this limitation and to state more clearly that several methodological choices should be regarded as practical configuration choices rather than systematically optimized parameters.
Reviewer comment:
Given that Snow Barrels are intended for use in a regional NWP system, I was surprised that the evaluation focuses almost exclusively on the snow analysis. The operational value of improved snow analyses ultimately depends on whether they translate into improved weather forecasts. I would therefore expect some assessment of forecast impacts on relevant variables such as 2 m temperature and humidity. Without such evidence, it is difficult to judge the practical benefit of the approach for operational NWP.Response:
We agree that the operational value of improved snow analyses should ultimately be assessed through their impact on weather forecasts. To address this, we conducted an additional one-month experiment covering March 2025 and added a new forecast-impact evaluation to the revised manuscript. The analysis includes verification of snow depth, 2 m temperature, and 2 m relative humidity forecasts, demonstrating the impact of the Snow Barrel observations on both the snow analysis and subsequent forecasts.We note, however, that the primary effect of the Snow Barrel observations is expected in the spatial distribution of snow cover, particularly near the snowline and under patchy snow conditions. Furthermore, all available SYNOP snow-depth observations are assimilated by the operational system and therefore strongly constrain the analysed snow state at station locations. As a result, improvements in the surrounding snow-cover distribution are not always fully reflected in conventional station-based forecast verification metrics.
We also expanded the discussion to highlight that improved gridded snow analyses may be valuable not only for NWP applications but also for other applications, such as reanalyses and environmental modelling.
---
Specific comments
Reviewer comment (Lines 59–60):
“However, current snow data assimilation techniques in NWP remain limited, restricted to using in situ measurements” – this statement is overly restrictive as current snow DA approaches in NWP are not limited to use of in situ observations. Several operational centres also assimilate satellite-derived snow cover products, including the Met Office, ECMWF, NOAA, and potentially others. Please revise this statement and provide a more balanced overview of current operational practice.
Response:
We agree and have added a paragraph with relevant references to the Introduction. Please also see our response to the next comment.Reviewer comment (Line 82):
The manuscript would benefit from a brief explanation of why ground-based observations of snow depth need to be integrated with satellite retrievals in satellite-based snow water equivalent (SWE) products. A short addition would be useful highlighting the limitations of passive microwave retrievals alone, with respect to snowpack sensitivity and retrieval uncertainty under certain conditions, such as wet snow and forest cover.
Response:
We have added an explanation of why many satellite-derived SWE products incorporate ground-based snow observations.Reviewer comment (Line 93):
The mention of H43 alongside H31 and H34 is a little unclear. H43 is based on FCI observations from MTG, whereas the current wording could be interpreted as implying that it originates from MSG. Please clarify. In addition, it would be useful to comment on whether the higher spatial resolution of H43 (compared to its MSG-based predecessors) is actually sufficient for many NWP applications in northern Europe.
Response:
We agree that the original wording was unclear regarding satellite platforms. The manuscript has been revised to explicitly distinguish between the MSG/SEVIRI-based H31 and H34 products and the MTG/FCI-based H43 product. We have also clarified that, although H43 provides substantially improved nominal resolution compared with the earlier MSG products, its effective spatial resolution over the Nordic region remains degraded because of the viewing geometry of geostationary satellites. This was one of the reasons why the H32 intermediate product was selected for the present study.Reviewer comment (Line 154):
“Additionally, snow cover enhances the surface emissivity.” Please specify the relevant wavelength range or part of the electromagnetic spectrum when discussing changes in emissivity. This statement is currently too general. For example, snow tends to increase emissivity in the thermal infrared, reduce emissivity in the visible spectrum, and can have varying effects in the microwave depending on frequency.
Response:
We replaced the sentence with: “Additionally, snow cover affects the ground heat flux, modifies latent and sensible heat fluxes, and alters surface emissivity in a wavelength-dependent manner.”Reviewer comment (Lines 179–180):
Please clarify the meaning of “if more than 20% of pixels are classified as either ‘snow’ or ‘no snow’”. Does this mean that the combined total of snow and no-snow classifications must exceed 20% of pixels, or that a single category must individually exceed the 20% threshold (i.e. >20% snow or >20% no-snow)?
Response:
The criterion refers to the combined fraction of pixels classified as either "snow" or "no snow" (with "partial snow" counted as snow). In other words, a barrel is accepted when more than 20% of its pixels provide useful surface information, while the remaining pixels may be classified as unclassified, water, or not processed. The manuscript has been revised to clarify this point.Reviewer comment (Lines 183–184):
The statement “They are only applied where they conflict with the model's background field...” does not appear to be fully consistent with the decision tree shown in Figure 3. If I have understood the flowchart correctly, observations indicating snow can be accepted when the background also contains snow, provided the background SWE is below 25 kg m⁻². Please clarify or rephrase.
Response:
Thank you for pointing this out. We agree that the original wording was too restrictive and did not fully reflect the logic shown in Figure 3. Snow Barrel observations indicating snow are primarily used when the background field and observation disagree. However, they may also be assimilated when the model already indicates snow cover, provided that the background SWE remains below the 25 kg m⁻² threshold. In this case, the observations help adjust regions of shallow or patchy snow cover and improve the position of the snowline. We have revised the text accordingly to better reflect the actual selection criteria.Reviewer comment (Line 195 and Table 2):
It is unclear whether these climatological snow-density values derived from a regional or global climatology? I also wonder whether use of the model background snow density might provide a more physically consistent approach, since it would be location-specific, meteorologically dependent, and evolve with the modelled snowpack state. I note that the manuscript discusses this as possible future work, but some additional justification of the current approach would be helpful.
Response:
The climatological snow-density values for southern Finland were inherited from the HIRLAM model and are not generally representative of the entire model domain. As explained in the revised manuscript, snow-depth observations will be assimilated in the next HARMONIE-AROME cycle (cycle 46), with conversion to SWE based on the model's prognostic snow density, as suggested. We note, however, that using prognostic snow density introduces an additional dependence of the initial state on modelling assumptions. We have clarified that the current density values are based on data from southern Finland and are not generally representative.Reviewer comment (Lines 199–216 and Figure 3):
The rationale for several of the observation selection criteria described in lines 199-216 (and illustrated in Figure 3) is not sufficiently convincing to me, and I would appreciate further justification for the chosen thresholds:
• First, the use of snowy observations when the background snow water equivalent (SWE) is up to 25 kg m⁻² seems to imply that snow-presence observations can provide useful quantitative information about SWE in regions where the model already contains a shallow snowpack. However, the observations are only sensitive to the presence or absence of snow, rather than the amount of snow. I am therefore unsure why these observations should be expected to improve the analysed SWE. Furthermore, it is not clear that this criterion necessarily avoids erroneous reductions in snow amount. For example, if a deep snowpack exists in reality but the background contains only a shallow snowpack, a partially snow-covered Snow Barrel could result in an observation corresponding to an even lower SWE estimate, further reducing an already underestimated snowpack. Such an error may have more serious consequences than reducing an overly deep snowpack, since it could more rapidly lead to the erroneous removal of snow cover and the associated radiative impacts.
• Second, I am not fully persuaded by the decision to reject snow-free observations when the background SWE exceeds 100 kg m⁻². The manuscript appears to assume that, in these circumstances, the model background is more trustworthy than the observation. However, one common model error is the misplacement of snow cover. If the overwhelming majority, or even all, of the pixels within a Snow Barrel are classified as snow-free, this would seem to provide strong evidence that the location is snow-free in reality and could therefore be particularly valuable for correcting the model state. I could more readily justify this criterion in cases where only a small fraction of the constituent pixels are classified as snow-free.
• More generally, are these thresholds and selection criteria empirically optimised? If so, the manuscript would be strengthened by presenting evidence that they improve analysis or forecast performance relative to alternative choices. At present, the criteria appear somewhat heuristic, and it is difficult to assess whether they represent the best overall trade-off between accepting and rejecting observations.Response:
We agree that the manuscript did not sufficiently explain the rationale behind the observation-selection thresholds.
First, we would like to clarify that the primary purpose of the Snow Barrel observations is not to provide a quantitative estimate of SWE. The satellite observations contain information only on the presence or absence of snow cover and therefore cannot directly constrain the true snow mass. Rather, the objective of the method is to improve the spatial representation of snow-covered and snow-free areas, particularly in regions of thin or patchy snow cover and along the snowline. Because SWE is the prognostic variable analyzed by the current operational snow-analysis system, corrections to the snow-cover state must necessarily be introduced through adjustments to SWE. The pseudo-observation formulation should therefore be regarded as a practical mechanism for modifying the analyzed snow cover rather than as a retrieval of SWE from satellite snow-extent observations.
The 25 kg m⁻² threshold was introduced to restrict the influence of snow-extent observations to regions where the model already indicates only a shallow snow cover. In such situations, information on snow presence or absence is expected to be most useful for correcting the position of the snowline and patchy snow-cover areas. In regions with larger snow amounts, snow-extent observations provide little information about the actual snow mass and are therefore excluded.
The 100 kg m⁻² threshold was introduced as a conservative quality-control measure to prevent large negative analysis increments from removing established snowpacks based on individual satellite observations. During development, such corrections were considered operationally more harmful than retaining some excess snow. We acknowledge that this criterion may also reject valid observations in cases where the model has misplaced the snow cover. But this is a general feature of any background quality control.
More generally, the threshold values used in the current implementation were selected during the development phase of the system based on preliminary experiments and operational experience. They should therefore be regarded as practical configuration parameters rather than optimized components of the Snow Barrel methodology itself. The primary aim of this paper is to introduce the Snow Barrel concept and demonstrate its feasibility within an operational NWP framework. A systematic sensitivity analysis and optimization of threshold values was beyond the scope of the present study and remains an important topic for future work. We stress this in the revised version of the manuscript.Reviewer comment (Line 228):
As noted above, I am not fully convinced that restricting the allowable background SWE range maximises the beneficial value of the observations in the melting zone. It seems possible that potentially informative observations are being excluded in situations where they could provide useful corrections to the model state.
Response:
This concern is closely related to the discussion above regarding the 25 kg m⁻² and 100 kg m⁻² SWE thresholds. As clarified in the revised manuscript, these thresholds were introduced during the development phase of the system to restrict the influence of snow-barrel observations to situations where snow-cover information was considered most reliable and informative. The thresholds should therefore be regarded as practical implementation choices rather than optimized characteristics of the Snow Barrel methodology itself. We acknowledge that some potentially informative observations may be excluded by these criteria, and we have added text to the manuscript noting that systematic optimization and sensitivity testing of the threshold values remain topics for future work.
Reviewer comment (Line 239):
The statement “The magnitude of the difference….is approximately 30 kg m⁻²” requires clarification. Does this refer to the cumulative SWE difference across the entire domain, including both positive and negative increments, or some other measure?
Response:
The statement does not refer to a cumulative SWE difference over the domain. It refers to the approximate magnitude of local analysis increments introduced by Snow Barrel observations. Since the maximum pseudo-observation generated from a fully snow-covered barrel corresponds to approximately 30 kg m⁻² (depending on the monthly climatological snow-density value), the resulting local SWE differences between analyses with and without Snow Barrels are of a similar magnitude. The manuscript has been revised to clarify this point.Reviewer comment (Lines 274–275):
“A key strength of the snow barrels is…” This appears to be a general advantage of satellite snow-cover observations rather than a feature unique to the Snow Barrel methodology. The same benefit would presumably be obtained from the underlying H32 product. The wording should be revised accordingly.
Response:
We agree that the ability to capture spatial variations in snow cover is a general advantage of satellite-derived snow-cover observations rather than a unique feature of the Snow Barrel methodology. We have therefore revised the wording accordingly.The revised text now distinguishes between the general benefit of satellite snow-cover observations and the specific contribution of the Snow Barrel approach. We emphasize that the novelty of the methodology lies in the use of the H32 intermediate product on its original satellite grid at approximately 1 km resolution. This preserves the native observation geometry and allows the original high-resolution snow-cover information to be used when constructing pseudo-observations for assimilation within the NWP snow-analysis system.
Reviewer comment (Lines 338–339):
The specified observation errors appear surprisingly small. Given that the observations provide information primarily on snow presence or absence rather than snow quantity, I would have expected the observation error to be substantially larger relative to the background error. Additional justification of the chosen error values, or evidence supporting their calibration, would be helpful.
Response:
The purpose of the statement in the Discussion was to illustrate how the relative weighting used in the current SWE-based system has been transferred to the upcoming snow-depth analysis. The observation-error values of 2.5 cm for the background and in-situ observations and 3 cm for the snow barrel observations were not derived from a formal calibration study but were chosen to preserve approximately the same relative weighting between observation types as in the current operational configuration.
It should also be noted that the influence of snow barrel observations is not controlled solely by the specified observation error. Prior to assimilation, the observations are subject to the filtering and selection criteria shown in Figure 3, and the resulting pseudo-observations are constrained to represent only relatively small corrections to the snow field. Thus, the effective influence of snow barrel observations is governed both by the observation error and by the preceding quality-control and selection procedures.
We have added clarification to the manuscript and note that a systematic tuning of observation-error statistics remains future work.---
Technical corrections
Reviewer comment (Line 129):
“developed in frames of ACCORD” should be revised to “developed in the framework of ACCORD”.
Response:
The text has been revised accordingly.Reviewer comment (Lines 241–243):
The feature being discussed is difficult to identify in Figure 4. Please consider providing a zoomed panel or enlarged view of the relevant region.
Response:
We have revised Figure 4 by adding a zoomed panel over southern Finland to better highlight the feature discussed in the text.Reviewer comment (Figure 7):
Please improve the distinction between RMSE and bias curves. Different colours or line styles, or annotation of the lines, would make the figure easier to interpret.
Response:
To improve readability, we have revised the presentation of the verification statistics. The original Figure 7 has been updated with improved visual separation between the RMSE and bias curves. In addition, we have introduced a new figure showing the differences between the Snow Barrel and Reference experiments (Snow Barrel minus Reference). This additional figure allows the impact of snow barrel observations to be assessed more clearly throughout the annual cycle and makes the seasonal improvements and degradations easier to identify.Citation: https://doi.org/10.5194/egusphere-2026-1408-AC1
-
AC1: 'Reply on RC1', Mikael Hasu, 16 Oct 2026
-
RC2: 'Comment on egusphere-2026-1408', Anonymous Referee #2, 10 Sep 2026
Thank you very much for the chance to review, “A new method for updating snow fields in the NWP models using satellite snow extent based ’Snow Barrel’ pseudo-observations as applied to HARMONIE-AROME cycle 43.” The paper presents a novel technique to update snow field in NWP models using a “snow barrel” approach, whereby the method identifies snow cover and non-covered pixels from optical satellites and compares those pixels to the model background fields and the H32 product. If there is a difference, the method proceeds to update the model.
I found the technique to be very interesting, and the figures well done. I believe there are a couple sections that could be better organized and framed for the reader, which I elaborate more on below. Additionally, I understand that the authors found most discrepancy between days of thin snow cover, but this seems inherent on the threshold for snow from the initial SWE prediction (which is currently unclear). More description and justification on this would be helpful.
Introduction
- I found the introduction could be overall a bit tighter to the question at hand and could better highlight the gap that researchers are accomplishing. That is to say, why snow is currently tough for NWP models, how/why current methods are insufficient, and why new methods are needed. I found the current structure drifted a bit and it was hard to follow the research gaps they were addressing.
- Line 30: I found the introduction to be lacking one critical paragraph that focused on what is a NWP, why accurate SWE is important for NWP, and why it is currently challenging. Paragraph 3 of the introduction does this somewhat, but I think this could be strengthen to justify why snow is important for NWP and why new methods are needed.
- Line 37-55: I am not really sure what paragraph 4 is adding. I think Paragraphs 4 and 5 could be better framed around how these observations are or are not currently used by NWP models. This is touched on in Line 61, but I think these sections could be tighter and more direct to say why data assimilation of these observations is insufficient, hence why a barrel method is needed. The current method made it hard to follow some of the acronyms for the reader, so I feel this could be better organized.
Data
- I found some of the information here redundant to the introduction (Line 66 and 109 are very similar) and a lot of the information in the data section also belonged into the introduction to strengthen the research gap (for example lines 79 – 89). I recommend the authors streamline the introduction to present the research gap and current challenges and use the data section to focus on the datasets and processing steps used for this analysis.
- I noticed that the study period later appeared to be April 2024 to April 2025. Was there a reason for this study period? And the same thing for this study region. Both sets of information would be stronger in the data section.
Model and Methods
- I appreciated figures 1 and 2 to guide the description of the analysis. I think some subsection headings would also benefit the reader as it was at times hard to follow the various steps to develop the model runs. I also found that some decisions appeared to be less justified, and it would benefit the reader to know the impacts of those decisions.
- Line 161: typo for “similar”
- Line 171: what does all members mean?
- Line 179: Where did the 20% for snow or no snow come from? Other papers have cited 40% as a threshold for snow or no snow (Hall et al. 1995; 10.1016/0034-4257(95)00137-P). This is an example of a decision that was challenging to know how it was made and what impacts it might have on the analysis.
- Line 184-188: I am bit confused. I understand that snow barrel values are only applied when the model prognostic variable disagrees with the snow barrel for snow or no snow, and the model variable is SWE, but this implies that there is a cut-off for SWE that the authors deem “snow.” Was this threshold anything above >0? Could really small SWE values (e.g., <1) also be considered no snow, in which case barrel values that indicate no snow for these respective values get assimilated into the model?
- Line 190: It would be helpful to understand the breakdown of how the authors got to 0-10 cm snow depth. Is there any prior literature to back up this claim?
- Line 214-215: Where do these standard deviations come from? Were they calculated or assigned or something else?
Results
- Overall comment: do forested areas impacted utility of the barrel method? Given that optical cannot penetrate through the canopy these values would indicate no snow when the model would produce SWE. Could incorporating these values actually make the snow fields worse?
- There are some “Cloud Gap Filled” optical products. Do the authors think these could provide utility?
- For Figure 7 (Lines 259-265): Can you provide the RMSE and bias differences between Barrels and Reference across all months and then separated by season? It seems like the barrel method is particularly useful in the spring when clouds are scarce and snow is thin and patchy, but I didn’t see these explicit numbers, and I found it very hard to figure out the differences between barrel and reference in the plot because the points were very small.
- I think some subsection headings for the Results would help here too.
Discussion
- Line 278: The authors bring up “computationally efficient” multiple times, but I couldn’t find this comparison. Can the authors provide a metric for this efficiency compared to other methods?
- Some of the topics in the Discussion belonged in the methods or the results. For example, Lines 296-300, for example. Lines 311-318 was also unclear how it related to interpretation of the study as it was mostly descriptive. I’d recommend the authors tighten the Discussion and work on reorganizing for the key takeaways, strengths, and limitations.
Citation: https://doi.org/10.5194/egusphere-2026-1408-RC2 -
AC2: 'Reply on RC2', Mikael Hasu, 16 Oct 2026
RESPONSE TO REVIEWER #2
General comments
Reviewer comment:
Thank you very much for the chance to review, “A new method for updating snow fields in the NWP models using satellite snow extent based ’Snow Barrel’ pseudo-observations as applied to HARMONIE-AROME cycle 43.” The paper presents a novel technique to update snow field in NWP models using a “snow barrel” approach, whereby the method identifies snow cover and non-covered pixels from optical satellites and compares those pixels to the model background fields and the H32 product. If there is a difference, the method proceeds to update the model.
I found the technique to be very interesting, and the figures well done. I believe there are a couple sections that could be better organized and framed for the reader, which I will elaborate more on below. Additionally, I understand that the authors found most discrepancy between days of thin snow cover, but this seems inherent on the threshold for snow from the initial SWE prediction (which is currently unclear). More description and justification on this would be helpful.Response:
We thank Reviewer #2 for the useful comments. We have taken them into account, which has improved the quality of the manuscript. Our detailed responses are provided below.---
Introduction
Reviewer comment:
I found the introduction could be overall a bit tighter to the question at hand and could better highlight the gap that researchers are accomplishing. That is to say, why snow is currently tough for NWP models, how/why current methods are insufficient, and why new methods are needed. I found the current structure drifted a bit and it was hard to follow the research gaps they were addressing.
Response:
We agree. The Introduction was reorganized as follows:
- why snow is important and why monitoring is needed;
- NWP as a source of snow information and how snow affects NWP forecasts;
- observations as a source of snow information; and
- the combination of NWP and snow observations through data assimilation, including why it is needed, its current status, and the remaining gaps.Reviewer comment (Line 30):
I found the introduction to be lacking one critical paragraph that focused on what is a NWP, why accurate SWE is important for NWP, and why it is currently challenging. Paragraph 3 of the introduction does this somewhat, but I think this could be strengthen to justify why snow is important for NWP and why new methods are needed.
Response:
Yes, we revised the Introduction section, as explained in our reply to your previous comment.Reviewer comment (Lines 37–55):
I am not really sure what paragraph 4 is adding. I think Paragraphs 4 and 5 could be better framed around how these observations are or are not currently used by NWP models. This is touched on in Line 61, but I think these sections could be tighter and more direct to say why data assimilation of these observations is insufficient, hence why a barrel method is needed. The current method made it hard to follow some of the acronyms for the reader, so I feel this could be better organized.
Response:
Yes, we revised the Introduction section. We clearly explained why assimilation of satellite snow observations is problematic for NWP and what the barrel method suggests.---
Data
Reviewer comment:
I found some of the information here redundant to the introduction (Line 66 and 109 are very similar) and a lot of the information in the data section also belonged into the introduction to strengthen the research gap (for example lines 79–89). I recommend the authors streamline the introduction to present the research gap and current challenges and use the data section to focus on the datasets and processing steps used for this analysis.Response:
We substantially revised both the Introduction and the Data section. As suggested, the Introduction now describes observations only briefly and focuses on presenting the research gap. The Data section contains two subsections:
- “Satellite data related to snow,” which provides a more detailed overview of satellite observations and their limitations; and
- “Snow Barrel super-observations,” which describes our proposed processing of H32 observations.Reviewer comment:
I noticed that the study period later appeared to be April 2024 to April 2025. Was there a reason for this study period? And the same thing for this study region. Both sets of information would be stronger in the data section.
Response:
The period from April 2024 to March 2025 was chosen because the Snow Barrel methodology had completed its development phase in March 2024 and was ready for operational testing. This period provides the first complete annual cycle available for evaluating the system under operational conditions, including autumn snow onset, winter snow-cover conditions, spring melt, and summer snow-free conditions. We have added this explanation to the manuscript. The study domain corresponds to the operational MetCoOp configuration, which was the intended target system for the development and evaluation of the Snow Barrel methodology.
---Model and Methods
Reviewer comment:
I appreciated figures 1 and 2 to guide the description of the analysis. I think some subsection headings would also benefit the reader as it was at times hard to follow the various steps to develop the model runs. I also found that some decisions appeared to be less justified, and it would benefit the reader to know the impacts of those decisions.
Response:
We agree that the original Model and Methods section contained several methodological steps that were difficult to follow within a single continuous narrative. To improve readability, we reorganized Section 3 by introducing additional subsection structure.Reviewer comment (Line 161):
Typo for “similar.”
Response:
The typo has been corrected in the revised manuscript.Reviewer comment (Line 171):
What does “all members” mean?
Response:
We clarified the text by replacing "all members" with "all ensemble members analyzed during the 07, 08 and 09 UTC assimilation window". We also added a short explanation when introducing MEPS that ensemble members are individual forecasts with slightly different initial conditions and perturbations that together represent forecast uncertainty.Reviewer comment (Line 179):
Where did the 20% for snow or no snow come from? Other papers have cited 40% as a threshold for snow or no snow (Hall et al. 1995; 10.1016/0034-4257(95)00137-P). This is an example of a decision that was challenging to know how it was made and what impacts it might have on the analysis.
Response:
This is not a threshold to decide about “snow” or “no snow”. This is a threshold to decide if the snow barrel in question contains any useful information. Snow barrels may contain many “unclassified” pixels (due to cloudiness or darkness) or many “sea” pixels. If the snow barrel contains at least 20% of pixels with snow-related information (either “snow” or “no snow”), it is not filtered out and used in the analysis. In the revised manuscript, we reformulated the statement to avoid further misunderstandings.Reviewer comment (Lines 184–188):
I am bit confused. I understand that snow barrel values are only applied when the model prognostic variable disagrees with the snow barrel for snow or no snow, and the model variable is SWE, but this implies that there is a cut-off for SWE that the authors deem “snow.” Was this threshold anything above >0? Could really small SWE values (e.g., <1) also be considered no snow, in which case barrel values that indicate no snow for these respective values get assimilated into the model?
Response:
The assimilation does not rely on a SWE > 0 criterion for the model background. Instead, the use of Snow Barrel pseudo-observations is controlled by thresholds applied to the background SWE. Snow-indicating Snow Barrel pseudo-observations (SWE > 0) are assimilated only when the background SWE is below 25 kg m⁻², while no-snow pseudo-observations (SWE = 0) are used only when the background SWE is below 100 kg m⁻². Therefore, small positive background SWE values are not treated differently from other values below these thresholds. We have clarified this in the manuscript.Reviewer comment (Line 190):
It would be helpful to understand the breakdown of how the authors got to 0–10 cm snow depth. Is there any prior literature to back up this claim?
Response:
The use of 10 cm when new snow is detected by the satellite but not by the model first guess originates from the ECMWF snow-analysis method proposed by de Rosnay et al. (2015). The idea of reducing this value according to the fraction of snow among the useful pixels in a barrel arose from preliminary studies conducted during development of the method. The resulting approximate SWE value is then used in the optimal-interpolation analysis.Reviewer comment (Lines 214–215):
Where do these standard deviations come from? Were they calculated, assigned, or obtained in some other way?
Response:
The observation-error standard deviations were assigned empirically during the development and testing phase of the snow barrel implementation. They were not derived from a dedicated statistical estimation or observation-error tuning study. The values were chosen to maintain the existing balance between the background field and conventional SYNOP snow observations while giving slightly less weight to snow barrel observations, since these are indirect pseudo-observations derived from satellite snow-extent classifications rather than direct measurements of SWE. We have clarified this in the revised manuscript.
---Results
Reviewer comment:
Do forested areas impact the utility of the barrel method? Given that optical observations cannot penetrate through the canopy, these values would indicate no snow when the model would produce SWE. Could incorporating these values actually make the snow fields worse?Response:
Forests do not affect the Snow Barrel method itself, but they do affect the quality of the underlying H SAF snow-cover product. This issue is discussed extensively by Siljamo et al. (2020), which we cite. The Snow Barrel method creates SWE super-observations from existing satellite snow-extent data and therefore inherits limitations in the underlying product. We have added a note about forested areas to the Discussion section of the revised manuscript.
Reviewer comment:
There are some “Cloud Gap Filled” optical products. Do the authors think these could provide utility?
Response:
We do not think these products would be useful for assimilation in NWP models. In such products, gaps caused by cloud cover are usually filled using statistical methods. NWP data assimilation also uses statistical methods, but it incorporates an additional source of information: the model background field. We therefore prefer products that remain as close as possible to the original observations, such as the H32 intermediate product. The H SAF classification identifies snow-covered and snow-free pixels while leaving other pixels, including water surfaces and clouds, undefined. This is well suited to NWP data assimilation. The issue is discussed by Siljamo et al. (2020) and is now also mentioned in the Introduction.Reviewer comment (Figure 7, Lines 259–265):
Can you provide the RMSE and bias differences between Barrels and Reference across all months and then separated by season? It seems like the barrel method is particularly useful in the spring when clouds are scarce and snow is thin and patchy, but I didn’t see these explicit numbers, and I found it very hard to figure out the differences between barrel and reference in the plot because the points were very small.
Response:
We agree that the differences between the experiments were difficult to quantify from the original version of Figure 7 alone. To address this, we have added a new figure showing the differences in RMSE and bias statistics between the snow-barrel experiment and the reference experiment. This allows the impact of the snow barrel assimilation to be assessed directly without visually comparing two close curves.Reviewer comment:
I think some subsection headings for the Results would help here too.
Response:
We agree that the Results section covers several different aspects of the evaluation and that additional subsection headings improve readability. We have therefore reorganized Section 4 by introducing subsection headings.
---Discussion
Reviewer comment (Line 278):
The authors bring up “computationally efficient” multiple times, but I couldn’t find this comparison. Can the authors provide a metric for this efficiency compared to other methods?
Response:
We agree that the term “computationally efficient” may imply a comparison with alternative assimilation approaches, which was not performed in this study. Our intention was to indicate that adding Snow Barrel observations does not impose a significant computational burden on the operational system. We have therefore revised the wording throughout the manuscript. In the operational MEPS configuration, processing and assimilating Snow Barrel observations as an additional observation type increases the runtime of the snow-analysis cycle by less than one minute compared with the standard configuration without Snow Barrels. This additional cost was considered negligible for operational forecasting purposes.Reviewer comment:
Some of the topics in the Discussion belonged in the methods or the results. For example, Lines 296–300. Lines 311–318 were also unclear in how they related to the interpretation of the study, as they were mostly descriptive. I’d recommend the authors tighten the Discussion and work on reorganizing the key takeaways, strengths, and limitations.
Response:
We have revisited and tightened the Discussion section in the revised manuscript to improve its focus and readability.Citation: https://doi.org/10.5194/egusphere-2026-1408-AC2
-
RC3: 'Comment on egusphere-2026-1408', Anonymous Referee #3, 17 Sep 2026
My overall assessment is that the paper presents a potentially useful operational approach, but the evidence currently provided is insufficient to establish robust improvements in snow analyses or demonstrate benefits for NWP forecasts. My main scientific concern is that the method is intended primarily to improve shallow and partial snow cover, yet several of its assumptions may be particularly uncertain under these conditions. First, several influential methodological choices (including the 10 × 10 pixel barrel size, the 20% minimum classified-pixel threshold, the maximum pseudo-depth of 10 cm, and the background SWE thresholds of 25 and 100 kg m⁻²) are prescribed without sufficient justification or sensitivity analysis. Although these choices may be operationally reasonable, they directly influence the analysis increments and resulting snow fields. Their robustness therefore needs to be demonstrated. Second, shallow and partial snow cover can be difficult to identify using optical observations. A satellite “no snow” classification may indicate genuinely snow-free ground, but it may also reflect snow obscured by vegetation or unresolved within mixed pixels. The approximately 1 km resolution, followed by further aggregation into barrels, also raises questions about how adequately the observations represent shallow snow heterogeneity. Third, shallow snow can change substantially between the satellite overpass and assimilation the following morning, particularly during active melt or snowfall in shoulder seasons. The main quantitative evaluation covers one annual cycle and does not separately assess these rapidly changing conditions, limiting the evidence for the method’s performance during such events. Fourth, monthly climatological density values may poorly represent shallow, newly fallen snow, particularly during warmer seasons when the prescribed density reflects denser seasonal snow. The resulting uncertainty in the conversion from pseudo-depth to SWE should be assessed. Finally, for shallow snow, changes in surface albedo may be more consequential for near-surface weather than changes in SWE. Assimilating SWE alone may therefore be insufficient to address snow-related shortcomings in operational NWP systems. My other comments are listed below:
- The Introduction does not clearly establish the novelty of the proposed approach relative to previous snow-assimilation efforts. The most relevant prior work is introduced only in the Methods section. Much of the Introduction explains basic snow variables, while insufficient attention is given to existing snow-cover assimilation approaches, their limitations, and the specific gap addressed by the proposed method.
- Method and data section is insufficiently reproducible in its present form. Please make sure all the defined variables and processing step are defined clearly. For example:
- Please define precisely how barrel snow fraction is calculated and how invalid pixels are handled. Now, the paper does not explicitly say whether the snow fraction is subsequently calculated relative to all 100 pixels or only the valid snow/no-snow pixels
- Also, clearly state how the assimilation workflow handles the impacts of snow removal on other prognostic variables
- The Results section would benefit from stronger quantitative support for the claims, particularly under thin or partial snow cover.
- Although Figure 7 presents RMSE and bias, several key conclusions rely primarily on visual interpretation. For example, the closer agreement with satellite-observed snow coverage in southern Finland is assessed visually in the discussion of Figure 5 (lines 235–245). Similarly, the concentration of accepted barrels along the snowline in Figure 4 is interpreted as indicating their added value in these regions (lines 225–230). However, because barrel acceptance explicitly depends on background SWE and satellite classification, this spatial distribution partly reflects the selection rules and does not, by itself, establish where assimilation improves accuracy. I suggest supplementing these examples with quantitative comparisons against independent snow-cover observations, reporting both missed snow and false snow detection with and without barrel assimilation. Where suitable observations are available, separating performance under shallow versus deeper snow, or partial versus complete snow coverage, would help substantiate the claimed benefits under transitional conditions.
- For Figure 7, please clearly identify the observations used to calculate RMSE and bias, explain how these statistics were calculated, and clarify whether these same observations were assimilated.
- Figure 7 is particularly difficult to interpret because the absolute differences between experiments are small relative to the large event- scale RMSE peaks. A difference plot or relative improvement plot would be much more informative.
- Please also report seasonal RMSE and bias values for both experiments, sample sizes, and uncertainty estimates for their differences to establish the magnitude and robustness of the improvements beyond the qualitative changes illustrated in the maps.
- The primary quantitative evaluation presented in the manuscript is limited to a single annual cycle, from April 2024 through March 2025. Although, the manuscript briefly describes preliminary experiments conducted over the CARRA-East domain during 2017, but no quantitative results or sufficiently detailed methodology are presented to support the analysis.
- Several advantages attributed to the snow-barrel approach in the Discussion are plausible but not directly demonstrated by the results presented. In particular, lines 282–285 state that aggregation reduces classification errors and ensures greater representativeness and reliability. However, the comparison between analyses with and without barrels evaluates the combined effect of adding satellite information through the proposed method; it does not isolate the benefit of aggregation itself. Please provide supporting evidence, such as an independent evaluation of aggregated versus unaggregated or alternatively aggregated H32 information, or qualify these statements as potential advantages. Similarly, lines 289–291 suggest that using barrels under partially cloudy conditions significantly enhances snow-data reliability. An evaluation of accuracy as a function of valid-pixel fraction would help support this claim. Otherwise, please distinguish improved availability under partial cloud from demonstrated improvements in reliability.
- The discussion should engage more deeply with the central compromise of the proposed method (meaning it gains operational compatibility by turning a categorical areal observation into a point-like quantitative SWE pseudo-observation, but that transformation introduces substantial representativeness and structural uncertainty.)
- The statement that current NWP snow assimilation is “restricted to using in situ measurements” is contradicted later by the discussion of prior satellite snow assimilation and should be corrected.
- Line 30: Please clarify that both snow depth and density vary spatially and jointly determine SWE. The current wording emphasizes density variability without acknowledging spatial variability in depth.
- Line 38: Please clarify which in situ snow observations are being discussed (for example, snow depth, SWE, or snow cover).
- Line 46: Please explicitly identify the principal limitations of microwave-derived SWE estimates rather than referring generally to “certain limitations.”
- Line 53- 55: seems out of place and break the flow
- The sentence beginning “Refine the assimilation scheme…” is grammatically incomplete.
- SWE units should be consistently formatted as kg m⁻²
- Snow depth observations” should not be called “satellite snow depth observations” when the satellite measures snow extent and depth is empirically assigned. Same about SWE. Reframe these derived values as empirical pseudo-observations, not satellite-derived SWE or snow depth.
- The text alternates between “snow field,” “snow cover field,” and “SWE field”; these are not interchangeable.
- The computational efficiency claim should report actual runtime or percentage overhead.
- “Significant” should be avoided unless statistical significance was tested.
Citation: https://doi.org/10.5194/egusphere-2026-1408-RC3 -
AC3: 'Reply on RC3', Mikael Hasu, 16 Oct 2026
RESPONSE TO REVIEWER #3
General comments
Reviewer comment:
My overall assessment is that the paper presents a potentially useful operational approach, but the evidence currently provided is insufficient to establish robust improvements in snow analyses or demonstrate benefits for NWP forecasts. My main scientific concern is that the method is intended primarily to improve shallow and partial snow cover, yet several of its assumptions may be particularly uncertain under these conditions. First, several influential methodological choices (including the 10 × 10 pixel barrel size, the 20% minimum classified-pixel threshold, the maximum pseudo-depth of 10 cm, and the background SWE thresholds of 25 and 100 kg m⁻²) are prescribed without sufficient justification or sensitivity analysis. Although these choices may be operationally reasonable, they directly influence the analysis increments and resulting snow fields. Their robustness therefore needs to be demonstrated. Second, shallow and partial snow cover can be difficult to identify using optical observations. A satellite “no snow” classification may indicate genuinely snow-free ground, but it may also reflect snow obscured by vegetation or unresolved within mixed pixels. The approximately 1 km resolution, followed by further aggregation into barrels, also raises questions about how adequately the observations represent shallow snow heterogeneity. Third, shallow snow can change substantially between the satellite overpass and assimilation the following morning, particularly during active melt or snowfall in shoulder seasons. The main quantitative evaluation covers one annual cycle and does not separately assess these rapidly changing conditions, limiting the evidence for the method’s performance during such events. Fourth, monthly climatological density values may poorly represent shallow, newly fallen snow, particularly during warmer seasons when the prescribed density reflects denser seasonal snow. The resulting uncertainty in the conversion from pseudo-depth to SWE should be assessed. Finally, for shallow snow, changes in surface albedo may be more consequential for near-surface weather than changes in SWE. Assimilating SWE alone may therefore be insufficient to address snow-related shortcomings in operational NWP systems.Response:
Thank you for these constructive comments. We agree that several methodological choices and assumptions influence the behaviour of the Snow Barrel methodology and that their implications should be discussed more explicitly. In the revised manuscript, we have expanded the discussion of the methodology, its assumptions, and associated uncertainties.We now state more clearly that parameters such as the barrel size, threshold values, pseudo-observation limits, and observation/background error settings should be regarded as practical configuration parameters selected during system development rather than optimised characteristics of the methodology itself. A systematic sensitivity analysis of these settings was beyond the scope of the present study and remains a topic for future work.
Regarding the detection of shallow and patchy snow cover, the present study focuses on the use of the H32 snow-extent product within the Snow Barrel framework rather than on the validation of the satellite product itself. To clarify this distinction, we now refer more explicitly to previous validation studies of the H32 product and discuss the implications of uncertainties in the underlying satellite observations.
We also expanded the discussion of temporal representativeness and note that referring all observations to the morning analysis time is a practical compromise in the current operational setup. Furthermore, we discuss limitations related to the use of climatological snow-density values in the conversion from snow extent to SWE pseudo-observations and describe ongoing developments toward the direct assimilation of snow-depth observations.
Finally, we agree that snow-related impacts on weather forecasts involve processes beyond SWE alone, including effects on surface albedo and the surface energy balance. To address the operational relevance of the method, we have added a new forecast-impact experiment and verification of near-surface forecast variables in the revised manuscript. While the detected forecast impacts are relatively modest, they provide an initial assessment of the influence of Snow Barrel observations on NWP forecasts.
---
Specific comments
Reviewer comment:
The Introduction does not clearly establish the novelty of the proposed approach relative to previous snow-assimilation efforts. The most relevant prior work is introduced only in the Methods section. Much of the Introduction explains basic snow variables, while insufficient attention is given to existing snow-cover assimilation approaches, their limitations, and the specific gap addressed by the proposed method.
Response:
We have revised the Introduction to clarify the novelty of the method, better describe previous snow-assimilation approaches and their limitations, and state the specific research gap addressed by this study.Reviewer comment:
The Method and Data section is insufficiently reproducible in its present form. Please make sure all defined variables and processing steps are described clearly. For example, please define precisely how the barrel snow fraction is calculated and how invalid pixels are handled. The paper does not explicitly state whether the snow fraction is calculated relative to all 100 pixels or only the valid snow/no-snow pixels.
Response:
We have clarified the description of the snow-fraction calculation in the manuscript. The snow fraction is calculated as the number of pixels classified as snow or partial snow divided by the total number of pixels in the barrel (100 pixels). Consequently, pixels classified as unclassified, not processed, or water are included in the calculation and may reduce the resulting snow fraction. We have also clarified that barrels are only accepted when at least 20% of the pixels provide valid snow information.Reviewer comment:
Also, clearly state how the assimilation workflow handles the impacts of snow removal on other prognostic variables.
Response:
The assimilation workflow does not directly handle impacts on other prognostic variables; the forecast model represents these interactions. The exception is the appearance of new snow, when the surface albedo is assigned the fresh-snow albedo value. This is now stated explicitly in the manuscript.Reviewer comment:
The Results section would benefit from stronger quantitative support for the claims, particularly under thin or partial snow cover.
Response:
We have strengthened the quantitative evaluation in the revised manuscript by adding additional verification metrics and forecast verification results. While the largest impact of the Snow Barrel observations is expected in regions with thin and patchy snow cover, the revised manuscript now includes a more comprehensive quantitative assessment of both the snow analysis and forecast fields.Reviewer comment (Figures 4 and 5):
Although Figure 7 presents RMSE and bias, several key conclusions rely primarily on visual interpretation. For example, the closer agreement with satellite-observed snow coverage in southern Finland is assessed visually in the discussion of Figure 5 (lines 235–245). Similarly, the concentration of accepted barrels along the snowline in Figure 4 is interpreted as indicating their added value in these regions (lines 225–230). However, because barrel acceptance explicitly depends on background SWE and satellite classification, this spatial distribution partly reflects the selection rules and does not, by itself, establish where assimilation improves accuracy. I suggest supplementing these examples with quantitative comparisons against independent snow-cover observations, reporting both missed snow and false snow detection with and without barrel assimilation. Where suitable observations are available, separating performance under shallow versus deeper snow, or partial versus complete snow coverage, would help substantiate the claimed benefits under transitional conditions.
Response:
We agree that quantitative verification against independent snow-cover observations would provide valuable additional information, particularly under shallow and patchy snow conditions. However, the primary impact of the Snow Barrel observations is expected in the spatial distribution of snow cover, while the current verification framework relies mainly on SYNOP observations. Since these observations are assimilated by the operational snow analysis, improvements in the surrounding snow-cover distribution may not be fully captured by station-based scores. We have therefore expanded the discussion of the validation approach and its limitations in the revised manuscript, and acknowledge that a more comprehensive spatial verification would be a valuable topic for future work.Reviewer comment (Figure 7):
For Figure 7, please clearly identify the observations used to calculate RMSE and bias, explain how these statistics were calculated, and clarify whether these same observations were assimilated.
Response:
We have clarified the verification methodology in the revised manuscript. The RMSE and bias statistics are calculated using snow-depth observations from 1078 SYNOP stations across the MetCoOp domain. Since SWE is the prognostic snow variable used by the operational data assimilation system, the observed snow depth is converted to SWE prior to verification. We have also explicitly stated that the SYNOP observations used for verification are routinely assimilated by the operational MEPS data-assimilation system and have added this information to both the main text and the figure caption.Reviewer comment (Figure 7):
Figure 7 is particularly difficult to interpret because the absolute differences between experiments are small relative to the large event-scale RMSE peaks. A difference plot or relative improvement plot would be much more informative.
Response:
We agree that the original figure made it difficult to distinguish the impact of the Snow Barrel observations from the large event variability. In the revised manuscript, we have redesigned the figure by separating the RMSE and bias statistics into individual subplots and added a new figure showing the differences between the Snow Barrel and reference experiments.Reviewer comment:
Please also report seasonal RMSE and bias values for both experiments, sample sizes, and uncertainty estimates for their differences to establish the magnitude and robustness of the improvements beyond the qualitative changes illustrated in the maps.
Response:
To strengthen the quantitative evaluation, we have added seasonal RMSE and bias statistics for both experiments together with the corresponding differences. We have also expanded the verification section and added forecast verification results to provide a broader quantitative assessment of the impact of the Snow Barrel observations. However, as discussed in the revised manuscript, the main benefit of the Snow Barrel observations is expected in the spatial distribution of snow cover and is therefore not always fully reflected in station-based verification metrics.Reviewer comment:
The primary quantitative evaluation presented in the manuscript is limited to a single annual cycle, from April 2024 through March 2025. Although the manuscript briefly describes preliminary experiments conducted over the CARRA-East domain during 2017, no quantitative results or sufficiently detailed methodology are presented to support the analysis.Response:
In NWP (in contrast to climate modelling), analysis of one annual cycle is believed to be enough to make conclusions, especially over large domains. This is primarily because of the computational cost of NWP experiments with data assimilation. During one annual cycle over a large domain, we can expect a variety of weather conditions, which already cover a large number of possible caveats.
The Snow Barrel approach was initially developed and tested within the CARRA-East framework, where it helped identify challenges related to the use of satellite snow observations in the CANARI surface analysis system and guided the development of the current methodology. However, these experiments were exploratory in nature, were not part of the final CARRA reanalysis, and are not directly comparable with the experiments presented here. To avoid confusion, we have revised the manuscript to clarify the role of the CARRA-East experiments and removed any wording that could be interpreted as a comparison or as supporting evidence for the results presented in this paper.Reviewer comment (Discussion, Lines 282–291):
Several advantages attributed to the Snow Barrel approach in the Discussion are plausible but not directly demonstrated by the results presented. In particular, lines 282–285 state that aggregation reduces classification errors and ensures greater representativeness and reliability. However, the comparison between analyses with and without barrels evaluates the combined effect of adding satellite information through the proposed method; it does not isolate the benefit of aggregation itself. Please provide supporting evidence, such as an independent evaluation of aggregated versus unaggregated or alternatively aggregated H32 information, or qualify these statements as potential advantages. Similarly, lines 289–291 suggest that using barrels under partially cloudy conditions significantly enhances snow-data reliability. An evaluation of accuracy as a function of valid-pixel fraction would help support this claim. Otherwise, please distinguish improved availability under partial cloud from demonstrated improvements in reliability.Response:
We agree that the experiments presented in this study evaluate the impact of the Snow Barrel methodology as a whole and do not isolate the specific contribution of the aggregation step. A direct comparison with unaggregated H32 pixels is not possible within the current operational framework, as the data assimilation system is designed for observations that represent spatial scales larger than an individual model grid cell and cannot make direct use of pixel-scale satellite classifications. The H32 information must therefore first be transformed into pseudo-observations suitable for assimilation. We have revised the manuscript to clarify that the reduced sensitivity to pixel-scale classification errors and improved representativeness are expected advantages of the aggregation approach rather than effects demonstrated separately by the experiments presented here.Reviewer comment:
The Discussion should engage more deeply with the central compromise of the proposed method: it gains operational compatibility by turning a categorical areal observation into a point-like quantitative SWE pseudo-observation, but that transformation introduces substantial representativeness and structural uncertainty.Response:
We agree. We added a paragraph to the Discussion explaining that, although the Snow Barrel method addresses problems related to resolution and coordinate systems, it still inherits major challenges associated with assimilating snow-extent observations. These include converting categorical observations into continuous SWE pseudo-observations and reconciling area-averaged satellite observations with data-assimilation methods developed primarily for point observations. Both transformations rely on practical rules and parameters that introduce representativeness and structural uncertainty.Reviewer comment:
The statement that current NWP snow assimilation is “restricted to using in situ measurements” is contradicted later by the discussion of prior satellite snow assimilation and should be corrected.
Response:
We agree that this statement was incorrect. It has been corrected in the Introduction, and relevant references on the use of satellite snow observations in NWP models have been added.Reviewer comment (Line 30):
Please clarify that both snow depth and density vary spatially and jointly determine SWE. The current wording emphasizes density variability without acknowledging spatial variability in depth.
Response:
The text has been revised to clarify that both snow depth and snow density vary spatially and jointly determine SWE.Reviewer comment (Line 38):
Please clarify which in situ snow observations are being discussed (for example, snow depth, SWE, or snow cover).
Response:
We have clarified that the in situ observations referred to here are snow-depth observations.Reviewer comment (Line 46):
Please explicitly identify the principal limitations of microwave-derived SWE estimates rather than referring generally to “certain limitations.”
Response:
We have revised the text to explicitly describe the main limitations of microwave-derived SWE estimates.Reviewer comment (Lines 53–55):
This text seems out of place and breaks the flow.
Response:
We have revised the lines to improve the flow and overall readability of the introduction.Reviewer comment:
The sentence beginning “Refine the assimilation scheme…” is grammatically incomplete.
Response:
The sentence has been corrected in the revised manuscript.Reviewer comment:
SWE units should be consistently formatted as kg m⁻².
Response:
The SWE units have been checked throughout the manuscript and are now formatted consistently as kg m⁻².Reviewer comment:
“Snow depth observations” should not be called “satellite snow depth observations” when the satellite measures snow extent and depth is empirically assigned. The same applies to SWE. Reframe these derived values as empirical pseudo-observations, not satellite-derived SWE or snow depth.
Response:
The terminology has been revised throughout the manuscript to consistently refer to snow-depth and SWE pseudo-observations rather than incorrectly implying direct satellite measurements of snow depth or SWE.Reviewer comment:
The text alternates between “snow field,” “snow cover field,” and “SWE field”; these are not interchangeable.
Response:
We have carefully reviewed the manuscript and corrected the terminology throughout to clearly distinguish between the snow field, snow-cover field, and SWE field where relevant.Reviewer comment:
The computational-efficiency claim should report actual runtime or percentage overhead.
Response:
We agree that the term “computationally efficient” may imply a runtime comparison, which was not performed in this study. Our intention was to indicate that adding Snow Barrel observations does not introduce a noticeable computational burden to the operational system. We have therefore revised the wording and now provide a quantitative estimate of the impact on runtime. In the operational MEPS configuration, processing and assimilating Snow Barrel observations increases the runtime of the snow-analysis cycle by less than one minute compared with the standard configuration without Snow Barrels. This additional computational cost is considered negligible for operational forecasting purposes.Reviewer comment:
“Significant” should be avoided unless statistical significance was tested.
Response:
We have reviewed the manuscript and replaced the term “significant” with more appropriate wording where statistical significance was not formally assessed.Citation: https://doi.org/10.5194/egusphere-2026-1408-AC3 -
AC4: 'Reply on AC3', Mikael Hasu, 05 Nov 2026
To complement our previous response, we noticed that our explanation regarding the definition of the snow fraction was incorrect. We apologize for the confusion this may have caused. The correct definition is:
The snow fraction is calculated as the number of pixels classified as snow divided by the total number of useful pixels (snow + partial snow + no snow). Pixels classified as partial snow are not counted as snow in the numerator.
The description in the revised manuscript has now also been corrected to ensure consistency with this definition.
Citation: https://doi.org/10.5194/egusphere-2026-1408-AC4
-
AC4: 'Reply on AC3', Mikael Hasu, 05 Nov 2026
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 314 | 186 | 53 | 553 | 35 | 40 |
- HTML: 314
- PDF: 186
- XML: 53
- Total: 553
- BibTeX: 35
- EndNote: 40
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1
General comments
This manuscript describes the use of a new “Snow Barrel” methodology to process satellite snow cover observations from the EUMETSAT H SAF H32 product for assimilation within the HARMONIE-AROME regional NWP system. The manuscript is well written and clearly presented, and the assimilation experiments demonstrate clear benefits to the analysed snow field relative to the use of in situ observations alone. Although it represents a potentially valuable contribution to the NWP and snow DA community, I remain unconvinced that the manuscript fully establishes (1) the novelty of the Snow Barrel concept relative to existing observation-processing approaches for satellite snow cover, (2) the added value of the Snow Barrel methodology itself relative to the underlying satellite observations, and (3) the operational benefit of the approach for NWP forecasting. These issues are discussed in the general comments below:
-----------------------------------------------
Specific comments
Line 59-60: “However, current snow data assimilation techniques in NWP remain limited, restricted to using in situ measurements” – this statement is overly restrictive as current snow DA approaches in NWP are not limited to use of in situ observations. Several operational centres also assimilate satellite-derived snow cover products, including the Met Office, ECMWF, NOAA, and potentially others. Please revise this statement and provide a more balanced overview of current operational practice.
Line 82: The manuscript would benefit from a brief explanation of why ground-based observations of snow depth need to be integrated with satellite retrievals in satellite-based snow water equivalent (SWE) products. A short addition would be useful highlighting the limitations of passive microwave retrievals alone, with respect to snowpack sensitivity and retrieval uncertainty under certain conditions, such as wet snow and forest cover.
Line 93: The mention of H43 alongside H31 and H34 is a little unclear. H43 is based on FCI observations from MTG, whereas the current wording could be interpreted as implying that it originates from MSG. Please clarify. In addition, it would be useful to comment on whether the higher spatial resolution of H43 (compared to its MSG-based predecessors) is actually sufficient for many NWP applications in northern Europe.
Line 154: “Additionally, snow cover enhances the surface emissivity.” Please specify the relevant wavelength range or part of the electromagnetic spectrum when discussing changes in emissivity. This statement is currently too general. For example, snow tends to increase emissivity in the thermal infrared, reduce emissivity in the visible spectrum, and can have varying effects in the microwave depending on frequency.
Line 179-180: Please clarify the meaning of “if more than 20% of pixels are classified as either ‘snow’ or ‘no snow’”. Does this mean that the combined total of snow and no-snow classifications must exceed 20% of pixels, or that a single category must individually exceed the 20% threshold (i.e. >20% snow or >20% no-snow)?
Line 183-184: The statement “They are only applied where they conflict with the model's background field...” does not appear to be fully consistent with the decision tree shown in Figure 3. If I have understood the flowchart correctly, observations indicating snow can be accepted when the background also contains snow, provided the background SWE is below 25 kg m⁻². Please clarify or rephrase.
Line 195 and Table 2: It is unclear whether these climatological snow-density values derived from a regional or global climatology? I also wonder whether use of the model background snow density might provide a more physically consistent approach, since it would be location-specific, meteorologically dependent, and evolve with the modelled snowpack state. I note that the manuscript discusses this as possible future work, but some additional justification of the current approach would be helpful.
Line 199-216 and Figure 3: The rationale for several of the observation selection criteria described in lines 199-216 (and illustrated in Figure 3) is not sufficiently convincing to me, and I would appreciate further justification for the chosen thresholds:
Line 228: As noted above, I am not fully convinced that restricting the allowable background SWE range maximises the beneficial value of the observations in the melting zone. It seems possible that potentially informative observations are being excluded in situations where they could provide useful corrections to the model state.
Line 239: The statement “The magnitude of the difference….is approximately 30 kg m⁻²” requires clarification. Does this refer to the cumulative SWE difference across the entire domain, including both positive and negative increments, or some other measure?
Line 274-275: “A key strength of the snow barrels is…” This appears to be a general advantage of satellite snow-cover observations rather than a feature unique to the Snow Barrel methodology. The same benefit would presumably be obtained from the underlying H32 product. The wording should be revised accordingly.
Line 338-339: The specified observation errors appear surprisingly small. Given that the observations provide information primarily on snow presence or absence rather than snow quantity, I would have expected the observation error to be substantially larger relative to the background error. Additional justification of the chosen error values, or evidence supporting their calibration, would be helpful.
-----------------------------------------------
Technical corrections
Line 129: “developed in frames of ACCORD” should be revised to “developed in the framework of ACCORD”.
Line 241-243: The feature being discussed is difficult to identify in Figure 4. Please consider providing a zoomed panel or enlarged view of the relevant region.
Figure 7: Please improve the distinction between RMSE and bias curves. Different colours or line styles, or annotation of the lines, would make the figure easier to interpret.