Effects of snow redistribution parameterization on simulated snow thickness validated by MOSAiC observations

Gu, Fengguan; Liu, Changwei; Han, Bo; Yang, Qinghua; Liu, Jiping

doi:10.5194/egusphere-2026-2179

Preprints

https://doi.org/10.5194/egusphere-2026-2179

Preprints

02 Jun 2026

| 02 Jun 2026

Effects of snow redistribution parameterization on simulated snow thickness validated by MOSAiC observations

Fengguan Gu, Changwei Liu, Bo Han, Qinghua Yang, and Jiping Liu

Abstract. Snow plays a critical role in the mass and energy balance of sea ice through its insulating properties and high albedo. Based on observations from the Multidisciplinary Drifting Observatory for the Study of Arctic Climate (MOSAiC) campaign, we assess the influence of snow redistribution on snow thickness simulations using the Icepack column model. Our results show that, without snow redistribution, snow thickness is overestimated in winter and spring. The bulk redistribution scheme slightly reduces snow accumulation, while the blowing snow scheme (snwITDrdg) further increases agreement with observations but still shows biases during snowfall events. Sensitivity experiments indicate that setting the ratio of snow mass on ridges to that on level ice to 4 in the bulk scheme yields the best agreement with snow observations (MAE = 6.2 mm). In the snwITDrdg scheme, the snow erosion coefficient is treated as an effective tuning parameter. When observed sea ice concentration is prescribed, setting the snow erosion coefficient to 2.4×10^-5 produces simulated accumulated snow loss to leads consistent with observations and improves the simulated snow thickness (MAE = 6.4 mm). This study provides new insights into snow thickness simulation and the parameterization of snow redistribution, offering valuable guidance for improving Arctic snow thickness modeling.

Received: 16 Apr 2026 – Discussion started: 02 Jun 2026

Competing interests: At least one of the (co-)authors is a member of the editorial board of The Cryosphere.

Publisher's note: Copernicus Publications remains neutral with regard to jurisdictional claims made in the text, published maps, institutional affiliations, or any other geographical representation in this paper. While Copernicus Publications makes every effort to include appropriate place names, the final responsibility lies with the authors. Views expressed in the text are those of the authors and do not necessarily reflect the views of the publisher.

Download & links

Fengguan Gu, Changwei Liu, Bo Han, Qinghua Yang, and Jiping Liu

Status: final response (author comments only)

RC1:
'Comment on egusphere-2026-2179', Anonymous Referee #1, 22 Jun 2026
This manuscript assesses the influence of snow redistribution on snow-thickness simulations using the Icepack model and conducts sensitivity analyses to evaluate the parameterisation of snow redistribution. The topic is important for the snow and sea-ice modelling community. However, the manuscript requires further improvement before it can be recommended for publication.
Major points:
The authors use MOSAiC observations to compare with the model simulations. However, it is not clear which MOSAiC transects or loops were used in the analysis. The MOSAiC snow observations include the Southern Loop, Northern Loop, runway, ridges, and summer transects, which represent different sea-ice types and surface conditions. It is unclear whether the authors combined measurements from all ice types or selected a specific subset. More information should be provided in the revised manuscript. The underlying sea-ice type is highly relevant to snow depth and wind-driven snow redistribution. In Figures 1 and 2, sea-ice thickness is provided, but ice age or ice type, for example first-year ice vs second-year ice, is not clearly described. These different ice types may behave very differently in terms of snow accumulation and redistribution (https://doi.org/10.1525/elementa.2021.00115). In particular, the parameter P, defined as the ratio of snow mass on ridges to that on level ice, is likely to depend strongly on ice age, ice type, and surface roughness. This should be discussed and, where possible, supported by observational information.

Based on the observation dates in Fig 1, it appears that observations until late July were included. Therefore I assume the in-situ snow data from the summer transects were also used. However, the summer transects were not always located in the same areas as the Northern Loop and Southern Loop, which were sampled from November to May. Note that the summer transects has a small spatial overlap with the Nloop site (https://doi.org/10.1525/elementa.2021.000072). This change in sampling region may introduce bias into the time-series analysis shown in Figures 1, 2, and 4, because the ice conditions may differ substantially between these regions. The authors should clarify whether the same ice floe/region was followed throughout the analysis or whether different transects were combined. If different regions were combined, the potential effect on the interpretation of the time series should be discussed.

The study mainly analyses differences in the model simulations under different snow-redistribution schemes. However, more results and discussion based on the observations should be included when comparing the model results. For example, in lines 310–328, the authors interpret that the snwITDrdg scheme leads to increased snow-ice formation and increased snow thickness over thinner ice. Then my question is whether enhanced snow-ice formation was actually observed in the field? This means, more generally, for the results shown in Figures 1–4, the manuscript should not only describe how the simulations differ between schemes, but also integrate the available observations to evaluate which scheme is correct or more appropriate and under what conditions. This would help the authors generalise the results and identify the climatic, dynamic, or ice-type conditions under which each modelling scheme is most suitable.

Specific points:
Line 36: The statement that snow accumulation can lead to flooding should be clarified. Please explain under what conditions flooding occurs and what is meant by flooding in this context. This would make the sentence clearer for readers who are less familiar with sea-ice processes.
Section 2.2: Please add more information about the MOSAiC expedition data used in this study, including ice types, loop positions, transect types, and the temporal coverage of each dataset.
Section 2.3: Could it be combined with Section 2.1, which introduces the Icepack model?
Section 2.3: The equations are not numbered.
Lines 173–184: This paragraph seems somewhat disconnected from the surrounding text. It may fit better in the Introduction or Discussion?
line 179: it mentions that Icepack assumes a uniform snow thickness. However, observations show strong spatial variability in snow thickness at sub-grid scales (https://doi.org/10.1017/jog.2022.18 and https://doi.org/10.5194/tc-20-2825-2026). Would this uniform-snow assumption introduce bias or uncertainty into the results presented in the manuscript? This issue should be discussed.
Line 220: The authors state that the snwITDrdg scheme performs better than the none and bulk schemes for SWE. However, the simulated SWE remains substantially higher than the mean observations. Does this indicate that none of the three schemes fully captures the observed SWE evolution?
Lines 221–223: The authors state that the snwITDrdg scheme significantly reduces SWE, particularly during high-wind periods. However, I am not sure this is clearly supported by Figures 1b and 1d. Please provide more explanation or evidence to support this statement. In addition, the statement that “snow particles are suspended and redistributed or deposited in the leads” requires more supporting evidence or references too.
Line 248: Please explain how the range of P values from 0.3 to 4 was selected. Are these values based on observations? Since the minimum MAE appears to occur at the maximum tested value of P = 4, it would be useful to know whether MAE increases again for P values greater than 4.
Figure 2 caption: Explain the meaning of P and gamma in the caption so that the figure can be understood independently.
Citation: https://doi.org/10.5194/egusphere-2026-2179-RC1
- AC1: 'Reply on RC1', Fengguan Gu, 06 Aug 2026
  
  The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-2179/egusphere-2026-2179-AC1-supplement.pdf
  
  Citation: https://doi.org/10.5194/egusphere-2026-2179-AC1
RC2: 'Comment on egusphere-2026-2179', Anonymous Referee #2, 22 Jul 2026

The comment was uploaded in the form of a supplement: https://egusphere.copernicus.org/preprints/2026/egusphere-2026-2179/egusphere-2026-2179-RC2-supplement.pdf

Citation: https://doi.org/10.5194/egusphere-2026-2179-RC2

Fengguan Gu, Changwei Liu, Bo Han, Qinghua Yang, and Jiping Liu

Viewed

Total article views: 172 (including HTML, PDF, and XML)

HTML	PDF	XML	Total	BibTeX	EndNote
117	37	18	172	13	15

HTML: 117
PDF: 37
XML: 18
Total: 172
BibTeX: 13
EndNote: 15

Views and downloads (calculated since 02 Jun 2026)

Month	HTML	PDF	XML	Total
Jun 2026	78	14	11	103
Jul 2026	34	21	6	61
Aug 2026	5	2	1	8

Cumulative views and downloads (calculated since 02 Jun 2026)

Month	HTML	PDF	XML	Total
Jun 2026	78	14	11	103
Jul 2026	34	21	6	61
Aug 2026	5	2	1	8

Viewed (geographical distribution)

Total article views: 147 (including HTML, PDF, and XML) Thereof 147 with geography defined and 0 with unknown origin.

Country	#	Views	%

Latest update: 08 Aug 2026

Short summary

Snow strongly affects how sea ice grows and melts by insulating the ice and reflecting sunlight. Using observations from a year-long Arctic expedition, we evaluated how models represent snow movement by wind. We found that ignoring snow redistribution leads to too much snow, while improved methods better match observations. Adjusting key parameters further improves results. This work helps make sea ice models more accurate and reliable for climate studies.


Total:	0
HTML:	0
PDF:	0
XML:	0