Enhancing simulations of snowpack properties in land surface models with the Soil, Vegetation and Snow scheme v2.0 (SVS2)

Vionnet, Vincent; Leroux, Nicolas R.; Fortin, Vincent; Abrahamowicz, Maria; Woolley, Georgina; Mazzotti, Giulia; Gaillard, Manon; Lafaysse, Matthieu; Royer, Alain; Domine, Florent; Gauthier, Nathalie; Rutter, Nick; Derksen, Chris; Bélair, Stéphane

doi:10.5194/gmd-18-9119-2025

Articles | Volume 18, issue 22

https://doi.org/10.5194/gmd-18-9119-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-18-9119-2025

© Author(s) 2025. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 18, issue 22

Model description paper

|

27 Nov 2025

Model description paper |

| 27 Nov 2025

Enhancing simulations of snowpack properties in land surface models with the Soil, Vegetation and Snow scheme v2.0 (SVS2)

Vincent Vionnet, Nicolas R. Leroux, Vincent Fortin, Maria Abrahamowicz, Georgina Woolley, Giulia Mazzotti, Manon Gaillard, Matthieu Lafaysse, Alain Royer, Florent Domine, Nathalie Gauthier, Nick Rutter, Chris Derksen, and Stéphane Bélair

Download

Final revised paper (published on 27 Nov 2025)
Supplement to the final revised paper
Preprint (discussion started on 22 Jul 2025)
Supplement to the preprint

Interactive discussion

Status: closed

CEC1:
'Comment on egusphere-2025-3396 - No compliance with the policy of the journal', Juan Antonio Añel, 28 Jul 2025

Dear authors,
Unfortunately, after checking your manuscript, it has come to our attention that it does not comply with our "Code and Data Policy".

https://www.geoscientific-model-development.net/policies/code_and_data_policy.html
In your Code and Data Availability section you state that "The SVS2 code for distributed applications will be soon made available on https://github.com/ECCC-ASTD-MRD/sps". Firstly, we can not accept statements on future availability of assets used in a manuscript. Our policy clearly states that all the code and data necessary to replicate a manuscript must be published openly and freely to anyone before submission. Therefore, your manuscript should have never been accepted for Discussions given such lack of compliance with the policy. Secondly, GitHub, which you reference several times in your manuscript, is not a suitable repository for scientific publication. GitHub itself instructs authors to use other long-term archival and publishing alternatives for purposes of scientific research.
Additionally, for part of the data used in your manuscript you cite websites that are not suitable repositories, such as ulaval.ca or nesdis.noaa.gov, or point to generic main pages that do not contain the specific data used in your work, but are portals to access generic data. You must store the specific data used in your work in suitable repositories according to our policy.
Therefore, the current situation with your manuscript is irregular. Please, publish all the code and data necessary to replicate your work in one of the appropriate repositories and reply to this comment with the relevant information (link and a permanent identifier for it (e.g. DOI)) as soon as possible, as we can not accept manuscripts in Discussions that do not comply with our policy.
Also, you must include a modified 'Code and Data Availability' section in any potentially reviewed manuscript, containing the information of the new repositories.
I must note that if you do not fix this problem, we cannot continue with the peer-review process or accept your manuscript for publication in our journal.
Juan A. Añel

Geosci. Model Dev. Executive Editor

Citation: https://doi.org/10.5194/egusphere-2025-3396-CEC1
- AC1:
  'Reply on CEC1', Vincent Vionnet, 29 Jul 2025
  
  Dear Dr. Juan A. Añel,
  
  Thank you very much for your detailed comments regarding our manuscript.
  Following your comments, we have initiated actions at the Meteorological Research Division of Environment and Climate Change Canada to release on Github the SVS2 code for distributed applications. A corresponding permanent repository will be created on Zenodo and the corresponding DOI will be cited in the 'Code and Data Availability' section of our manuscript.
  In a Zenodo repository, all the observation datasets used in the results section will be made publicly available as well as simulation configurations, and the scripts and extracts of simulation outputs used to generate the results and figures of our manuscript. The corresponding DOI will be cited in the 'Code and Data Availability' section of our manuscript.
  We expect the Zenodo repositories to be available by August 8^th. We will then post the corresponding information in this discussion to make sure that the peer-review process can continue.
  Thank you very much for your time and consideration.
  Best regards,
  Vincent Vionnet
  
  Citation: https://doi.org/10.5194/egusphere-2025-3396-AC1
  - AC2: 'Reply on AC1', Vincent Vionnet, 07 Aug 2025
    
    Dear Dr. Juan A. Añel,
    
    Thank you very much for your detailed comments regarding our manuscript.
    The SVS2 code within the ECCC Surface Prediction System is now available in a permanent repository: https://doi.org/10.5281/zenodo.16740463 (Vionnet et al., 2025b) This code has been used for the distributed simulations presented in our manuscript.
    The scripts and data used to generate all the figures of the manuscript are also available on Zenodo: https://doi.org/10.5281/zenodo.16760831 (Vionnet et al., 2025c). In particular, this repository contains the observations used to evaluate the distributed simulations: (i) the snow pits data and (ii) the GMASI snow cover data. Both observation datasets are distributed in a NetCDF format. The configurations files for the point-scale and distributed simulations are also included in this repository.
    The modified 'Code and Data Availability' in any future version of the manuscript will read as:
    The SVS 2.0 code used for the point scale applications at Snow Crested Butte is available in a permanent repository: https://doi.org/10.5281/zenodo.14859640 (Vionnet et al., 2025a). The SVS 2.0 code used for the distributed simulations presented in this paper is available from https://doi.org/10.5281/zenodo.16740463 (Vionnet et al., 2025b). Scripts and data to produce the figures in this paper as well as SVS2 configurations files are available from https://doi.org/10.5281/zenodo.16760831 (Vionnet et al., 2025c). This repository also contains the observations used to evaluate the distributed snowpack simulations (snow pit data and GMASI snow cover data).
    The other freely-available datasets used in this work are:
    ERA5-Land data: https://doi.org/10.24381/cds.e2161bac
    ERA5/Crocus dataset: https://doi.org/10.5281/zenodo.10943718
    Global map of snow darkening coefficient for the snowpack model Crocus (Gaillard et al., 2024): https://doi.org/10.5281/zenodo.14194990
    Snow Crested Butte dataset: https://doi.org/10.5281/zenodo.6618553
    We hope these modifications make our manuscript compliant with the policy of the journal and guarantee the replicability of our results.
    Thank you very much for your time and consideration.
    
    Best regards,
    
    Vincent Vionnet, on behalf of all the authors.
    References used in this answer:
    Gaillard, M., Vionnet, V., Lafaysse, M., Dumont, M., and Ginoux, P.: Improved snow darkening coefficient for large-scale albedo modelling with Crocus [Dataset], Zenodo, https://doi.org/10.5281/zenodo.14194990, 2024.
    Vionnet, V., Leroux, N., Fortin, V., Abrahamowicz, M., Woolley, G., Mazzotti, G., Gaillard, M., Lafaysse, M., Royer, A., Domine, F., Gauthier, N., Rutter, N., Derksen, C., & Belair, S. (2025a). Code of the land surface scheme Soil Vegetation and Snow version 2 integrated in the MESH platform (v1.0.0). Zenodo. https://doi.org/10.5281/zenodo.14859640
    Vionnet, V., Leroux, N., Fortin, V., Abrahamowicz, M., Woolley, G., Mazzotti, G., Gaillard, M., Lafaysse, M., Royer, A., Domine, F., Gauthier, N., Rutter, N., Derksen, C., & Belair, S. (2025b). Code of the land surface scheme Soil Vegetation and Snow version 2 integrated in the ECCC Surface Prediction System. Zenodo. https://doi.org/10.5281/zenodo.16740463
    Vionnet, V., Leroux, N., Royer, A., Domine, F., Fortin, V., Abramowicz, M., Woolley, G., Mazzotti, G., Gaillard, M., Lafaysse, M., Gauthier, N., Rutter, N., Derksen, C., & Belair, S. (2025c). Scripts and data to produce figures for the SVS2 paper submitted to GMD (Version v1) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.16760831
    
    Citation: https://doi.org/10.5194/egusphere-2025-3396-AC2
    
    CEC2: 'Reply on AC2', Juan Antonio Añel, 07 Aug 2025
    
    Dear authors,
    Many thanks for your point-by-point reply. As a minor issue, it would be good if you could provide the exact ERA5 data used in your work in a repository, instead of simply linking the ERA5 webpage, which is not a reliable permanent repository and makes harder for readers to know exactly the files that you have used.
    Juan A. Añel
    Geosci. Model Dev. Executive Editor
    
    Citation: https://doi.org/10.5194/egusphere-2025-3396-CEC2
    
    AC3: 'Reply on CEC2', Vincent Vionnet, 08 Aug 2025
    
    Dear Dr. Juan A. Añel,
    
    Thank you very much for your quick reply.
    Information about the ERA5 forcing data have been added to the permanent repository https://doi.org/10.5281/zenodo.16782503 (see the folder distributed_simulations/forcing). Scripts to download the ERA5 data from their original archives are provided as well as an example of ERA5-forcing file for SVS2 within the ECCC Surface Prediction System.
    The modified 'Code and Data Availability' in any future version of the manuscript will read as:
    The SVS 2.0 code used for the point scale applications at Snow Crested Butte is available in a permanent repository: https://doi.org/10.5281/zenodo.14859640 (Vionnet et al., 2025a). The SVS 2.0 code used for the distributed simulations presented in this paper is available from https://doi.org/10.5281/zenodo.16740463 (Vionnet et al., 2025b). Scripts and data to produce the figures in this paper as well as SVS2 configurations files are available from https://doi.org/10.5281/zenodo.16760831 (Vionnet et al., 2025c). This repository also contains the observations used to evaluate the distributed snowpack simulations (snow pit data and GMASI snow cover data). ERA5 atmospheric forcing have been obtained from https://doi.org/10.24381/cds.adbb2d47 and https://doi.org/10.24381/cds.143582cf as detailed in Vionnet et al. (2025c).
    The other freely-available datasets used in this work are:
    ERA5-Land data: https://doi.org/10.24381/cds.e2161bac
    ERA5/Crocus dataset: https://doi.org/10.5281/zenodo.10943718
    Global map of snow darkening coefficient for the snowpack model Crocus (Gaillard et al., 2024): https://doi.org/10.5281/zenodo.14194990
    Snow Crested Butte dataset: https://doi.org/10.5281/zenodo.6618553
    We hope these modifications make our manuscript compliant with the policy of the journal and guarantee the replicability of our results.
    Thank you very much for your time and consideration.
    
    Best regards,
    
    Vincent Vionnet, on behalf of all the authors.
    References used in this answer:
    Vionnet, V., Leroux, N., Fortin, V., Abrahamowicz, M., Woolley, G., Mazzotti, G., Gaillard, M., Lafaysse, M., Royer, A., Domine, F., Gauthier, N., Rutter, N., Derksen, C., & Belair, S. (2025a). Code of the land surface scheme Soil Vegetation and Snow version 2 integrated in the MESH platform (v1.0.0). Zenodo. https://doi.org/10.5281/zenodo.14859640
    Vionnet, V., Leroux, N., Fortin, V., Abrahamowicz, M., Woolley, G., Mazzotti, G., Gaillard, M., Lafaysse, M., Royer, A., Domine, F., Gauthier, N., Rutter, N., Derksen, C., & Belair, S. (2025b). Code of the land surface scheme Soil Vegetation and Snow version 2 integrated in the ECCC Surface Prediction System. Zenodo. https://doi.org/10.5281/zenodo.16740463
    Vionnet, V., Leroux, N., Royer, A., Domine, F., Fortin, V., Abramowicz, M., Woolley, G., Mazzotti, G., Gaillard, M., Lafaysse, M., Gauthier, N., Rutter, N., Derksen, C., & Belair, S. (2025c). Scripts and data to produce figures for the SVS2 paper submitted to GMD (Version v2) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.16782503
    
    Citation: https://doi.org/10.5194/egusphere-2025-3396-AC3
    
    AC4: 'Additional reply on CEC2', Vincent Vionnet, 21 Oct 2025
    
    Dear Dr. Juan A. Añel,
    
    as previously discussed, a Zenodo repository has been created to gather the scripts and data used to generate all the figures of the revised manuscript. This repository has been updated to include the configuration files corresponding to the SVS simulations at Snow Crested Butte that have will be added to the revised manuscript (see our answers to Reviewers 1). The SVS output files and the scripts used to generate the updated figures will be also added to the repository.
    The modified 'Code and Data Availability' in the revised version of the manuscript will read as:
    The SVS 2 .0 code used for the point scale applications at Snow Crested Butte is available in a permanent repository: https://doi.org/10.5281/zenodo.14859640 (Vionnet et al., 2025a). The SVS 1.0 code is also available in this repository. The SVS 2.0 code used for the distributed simulations presented in this paper is available from https://doi.org/10.5281/zenodo.16740463 (Vionnet et al., 2025b). Scripts and data to produce the figures in this paper as well as SVS 1.0 and SVS 2.0 configurations files are available from https://doi.org/10.5281/zenodo.16760830 (Vionnet et al., 2025c). This repository also contains the observations used to evaluate the distributed snowpack simulations (snow pit data and GMASI snow cover data). ERA5 atmospheric forcing have been obtained from https://doi.org/10.24381/cds.adbb2d47 and https://doi.org/10.24381/cds.143582cf as detailed in Vionnet et al. (2025c).
    The other freely-available datasets used in this work are:
    ERA5-Land data: https://doi.org/10.24381/cds.e2161bac
    ERA5/Crocus dataset: https://doi.org/10.5281/zenodo.10943718
    Global map of snow darkening coefficient for the snowpack model Crocus (Gaillard et al., 2024): https://doi.org/10.5281/zenodo.14194990
    Snow Crested Butte dataset: https://doi.org/10.5281/zenodo.6618553
    We hope these modifications make our manuscript compliant with the policy of the journal and guarantee the replicability of our results.
    Thank you very much for your time and consideration.
    
    Best regards,
    
    Vincent Vionnet, on behalf of all the authors.
    
    References used in this answer:
    Vionnet, V., Leroux, N., Fortin, V., Abrahamowicz, M., Woolley, G., Mazzotti, G., Gaillard, M., Lafaysse, M., Royer, A., Domine, F., Gauthier, N., Rutter, N., Derksen, C., & Belair, S. (2025a). Code of the land surface scheme Soil Vegetation and Snow version 2 integrated in the MESH platform (v1.0.0). Zenodo. https://doi.org/10.5281/zenodo.14859640
    Vionnet, V., Leroux, N., Fortin, V., Abrahamowicz, M., Woolley, G., Mazzotti, G., Gaillard, M., Lafaysse, M., Royer, A., Domine, F., Gauthier, N., Rutter, N., Derksen, C., & Belair, S. (2025b). Code of the land surface scheme Soil Vegetation and Snow version 2 integrated in the ECCC Surface Prediction System. Zenodo. https://doi.org/10.5281/zenodo.16740463
    Vionnet, V., Leroux, N., Royer, A., Domine, F., Fortin, V., Abramowicz, M., Woolley, G., Mazzotti, G., Gaillard, M., Lafaysse, M., Gauthier, N., Rutter, N., Derksen, C., & Belair, S. (2025c). Scripts and data to produce figures for the SVS2 paper submitted to GMD (Version v3) [Data set]. Zenodo. https://doi.org/10.5281/zenodo.16760830
    
    Citation: https://doi.org/10.5194/egusphere-2025-3396-AC4
RC1:
'Comment on egusphere-2025-3396', Anonymous Referee #1, 19 Aug 2025
The authors provide a comprehensive description of SVS2, including significant developments such as a revised snow albedo scheme, novel physical parameterizations for Arctic snowpack properties, and a new forest canopy module. They then demonstrate its capability to simulate snowpack properties, including microstructure, across extensive geographical regions. In general, this manuscript is well-written and suitable for publication in GMD. However, my major concern is that the authors have not clearly explained the impacts of these improvements on enhancing model performance.
Special comments:
Line 535: Why did the authors apply normalization of total depth to each pair of simulated and observed profiles?

Line 467 states that a dew point temperature threshold of 0 °C was used to separate total precipitation into liquid or solid precipitation in point-scale simulations. Line 510 mentions that precipitation phase in distributed simulations was determined based on hydrometeor temperature derived from downscaled air temperature and humidity. What is the difference between these two methods, and why were different methods applied in this study?

In section 4.1 "Point-scale evaluation at Snow Crested Butte", the authors present an evaluation of the model in point-scale mode at Snow Crested Butte. They conducted simulations at both the open site and the forest site, with comparisons to on-site measurements. However, they mainly present the results simulated by SVS2/Crocus without explaining how the new forest canopy module enhanced simulation performance. A comparison between SVS1 and SVS2 is recommended to highlight the role of the new forest canopy module.

The model did not accurately simulate the grain type of each layer, which is reasonable considering uncertainties in the observed snow type and in the model's snow grain diagnostics. My question is: if the simulation of grain type was unsatisfactory, was the inclusion of complex snow microstructure necessary? To what extent can the complex microstructure configuration improve model performance?

Lines 591-593: Why can the Arctic configuration also improve simulations for sites below the treeline?

The authors note that SVS2 has been previously applied in several studies, including Woolley et al. (2024), and they make comparisons with Woolley et al. (2024) throughout almost all parts of the manuscript. What, then, are the scientific advances of this study compared to Woolley et al. (2024)?

The authors refined the snow albedo parametrization and indicated that they incorporated significant developments such as a revised snow albedo scheme. However, in the Results section, they did not demonstrate the impacts of such developments on model performance.
Citation: https://doi.org/10.5194/egusphere-2025-3396-RC1
- AC5: 'Reply on RC1', Vincent Vionnet, 21 Oct 2025
  
  Please find our response in the pdf supplement.
  Vincent Vionnet, on behalf of all the co-authors.
  
  Citation: https://doi.org/10.5194/egusphere-2025-3396-AC5
RC2:
'Comment on egusphere-2025-3396', Anonymous Referee #2, 11 Sep 2025

The authors present an upgraded version of the Soil Vegetation and Snow (SVS), a land surface model employed for land surface climate and hydrological predictions. The model version presented here (SVS2) includes the snow scheme Crocus to improve the representation of snow processes, including snow microphysics and light absorbing particle deposition, a new canopy model to represent the effects of vegetation, and some improvements tailored to applications of the model to snow in arctic regions. In my opinion the paper is very well written and the model development presented here seems relevant. I believe the paper should be considered for publication once the authors address or respond to the reviewer’s comments.

Comments

1) Since the abstract mentions explicitly the fact that one of the improvement is the addition of light absorbing particles effects to snow in SVS, some more detail on how this is implemented (e.g., how the spatial distribution of the snow aging time constant gamma was derived) would be very helpful to the reader. In section 2.2.2, it might also be worth reporting the albedo formulation used in SVS2, since this - and the value of its parameter gamma - are discussed in that paragraph. From the discussion it seems like using the radiative transfer model TARTES would improve on this parameterization - I wonder, is the 10x increased in computation cost mentioned in the paper for the entire land model, snow model, or only snow radiative transfer calculations?

2) It seems the authors perform yearly simulations for each year starting in September 1st, where “The initial soil state on 1 September of each winter was taken from a 1-year spin-up of the land surface conditions.” Is a 1-year model spin up enough to equilibrate the thermal regimes of soils or the ice content in the soil? Is it possible that this would affect the seasonal snow simulations given the high-latitude domain of this study?

3) A large section of the paper focuses on the energy and mass balance of canopy and intercepted water. However, I did not find in the paper any discussion about the numerical scheme used, or order of operations in a model time step. Are there any particular strategies used for solving the land surface energy balance which are relevant for the reader? Is it the same as in the previous version of the parent model (SVS / Crocus)? Since this model was developed for coupled runs for numerical weather predictions or earth system modeling, some detail on the solution of the energy and mass balance of the land surface / canopy / snowpack might be useful.

Citation: https://doi.org/10.5194/egusphere-2025-3396-RC2
- AC6: 'Reply on RC2', Vincent Vionnet, 21 Oct 2025
  
  Please find our response in the pdf supplement.
  Vincent Vionnet, on behalf of all the co-authors.
  
  Citation: https://doi.org/10.5194/egusphere-2025-3396-AC6

Peer review completion

AR – Author's response | RR – Referee report | ED – Editor decision | EF – Editorial file upload

AR by Vincent Vionnet on behalf of the Authors (22 Oct 2025) Author's response Author's tracked changes Manuscript

ED: Referee Nomination & Report Request started (22 Oct 2025) by Dalei Hao

RR by Anonymous Referee #1 (23 Oct 2025)

ED: Publish as is (31 Oct 2025) by Dalei Hao

AR by Vincent Vionnet on behalf of the Authors (05 Nov 2025) Author's response Manuscript

Short summary

Snow microstructure controls snowpack properties, but most land surface models overlook this factor. To support future satellite missions, we created a new land surface model based on the Crocus scheme that simulates snow microstructure. Key improvements include better snow albedo representation, enhanced Arctic snow modeling, and improved forest module to capture Canada's diverse snow conditions. Results demonstrate improved simulations of snow density and melt across large regions of Canada.

Enhancing simulations of snowpack properties in land surface models with the Soil, Vegetation and Snow scheme v2.0 (SVS2)

Download

Interactive discussion

Peer review completion

Suggestions for revision or reasons for rejection