Venäläinen et al. (2025) Updated monthly and new daily bias correction for assimilation-based passive microwave SWE retrieval

Identification

• Journal: ˜The œcryosphere
• Year: 2025
• Date: 2025-11-27
• Authors: Pinja Venäläinen, Colleen Mortimer, Kari Luojus, Lawrence Mudryk, Matias Takala, Jouni Pulliainen
• DOI: 10.5194/tc-19-6301-2025

Research Groups

• Finnish Meteorological Institute, Helsinki, Finland
• Climate Research Division, Environment Climate Change Canada, Toronto, Canada

Short Summary

This study updates and expands the bias correction for assimilation-based passive microwave Snow Water Equivalent (SWE) retrieval, introducing a new daily bias-corrected Snow CCI v3.1 product. The updated methodology, utilizing expanded snow course data, significantly improves Northern Hemisphere climatological snow mass estimates, bringing them into consistency with reanalysis products, especially for high SWE values and previously problematic months.

Objective

• To update and expand the monthly bias correction fields for the GlobSnow/Snow CCI passive microwave Snow Water Equivalent (SWE) retrieval methodology using newly available snow course data.
• To extend this bias correction to cover the months of December through May and to provide daily bias-corrected SWE estimates for the Snow CCI v3.1 product.

Study Configuration

• Spatial Scale: Northern Hemisphere (north of 15° N), focusing on snow-covered land areas. Regional analysis for Eurasia and North America.
• Temporal Scale: 1980–2022 for the daily bias-corrected Snow CCI v3.1 product. Monthly bias correction applied for December through May. Snow course data used for bias calculation covers the full observational period.

Methodology and Data

• Models used:
  • HUT (Helsinki University of Technology) snow emission model
  • Kriging interpolation (for snow depth fields and bias fields)
  • Modified Hall et al. (2002) algorithm for dry snow detection
• Data sources:
  • Satellite Passive Microwave Brightness Temperatures (Tb): Special Sensor Microwave/Imager (SSM/I), Special Sensor Microwave Imager/Sounder (SSMIS), NASA MEaSUREs Calibrated Enhanced-Resolution Passive Microwave Daily EASE-Grid 2.0 Brightness Temperature ESDR.
  • In-situ Snow Depth (SD) observations: European Centre for Medium-Range Weather Forecasts (ECMWF), All-Russia Research Institute of Hydrometeorological Information – World Data Center (RIHMI-WDC), Global Historical Climatology Network daily (GHCNd) by NOAA.
  • Snow Water Equivalent (SWE) reference data (for bias correction and validation): Snow course observations from North America, Finland, and Russia. Airborne gamma SWE estimates over the US and parts of southern Canada (for independent validation).
  • Snow Extent Data: CryoClim snow cover extent, passive microwave thresholding approach.
  • Reanalysis Products (for intercomparison/validation): SnowPEx Intercomparison Project product suites.

Main Results

• The updated monthly and new daily bias corrections for Snow CCI v3.1 (SCv3.1) significantly improve Northern Hemisphere climatological snow mass estimates, making them consistent with reanalysis products.
• The bias correction adds approximately 500 Gt to the Northern Hemisphere climatological snow mass, shifting the peak snow mass from the lower end to the middle of the reanalysis product spread.
• The updated bias correction extends coverage to December and January and provides daily SWE estimates, addressing user needs.
• Comparison with airborne gamma SWE validation shows large improvements in bias (e.g., from -18.8 mm to -3.3 mm for SWE < 500 mm) and marginal improvements in Mean Absolute Error (MAE) for both SWE < 200 mm and SWE < 500 mm, particularly for large SWE values.
• Changes in reference data had a greater impact on bias in North America (increased negative bias in Quebec/Alaska, reduced positive bias in central North America), adding approximately 100 Gt to March snow mass.
• Changes in the retrieval algorithm (e.g., spatially and temporally varying snow densities) reduced bias magnitude in SCv3.1 compared to GlobSnow v3.0, especially in Eurasia.
• While hemispheric SWE is clearly improved, local accuracy can be compromised, with bias correction sometimes overcorrecting small SWE values (e.g., RMSE for SWE < 200 mm slightly degraded from 41.8 mm to 43.1 mm in airborne gamma validation).

Contributions

• Development and application of an updated monthly and novel daily bias correction methodology for passive microwave assimilation-based SWE retrieval (Snow CCI v3.1).
• Expansion of the bias correction to cover the full December-May snow season, providing a more comprehensive product.
• Significant improvement in the accuracy and consistency of Northern Hemisphere climatological snow mass estimates, particularly for high SWE values and previously problematic months (April, May), aligning them with reanalysis products.
• Utilization of newly available and expanded snow course data, especially in North America, to refine bias fields.
• Addressing a key limitation of passive microwave SWE retrievals (systematic underestimation of large SWE values) through a robust bias correction.

Funding

• ESA SnowCCI+ project (grant no. 4000124098/18/I-NB)

Citation

@article{Venäläinen2025Updated,
  author = {Venäläinen, Pinja and Mortimer, Colleen and Luojus, Kari and Mudryk, Lawrence and Takala, Matias and Pulliainen, Jouni},
  title = {Updated monthly and new daily bias correction for assimilation-based passive microwave SWE retrieval},
  journal = {˜The œcryosphere},
  year = {2025},
  doi = {10.5194/tc-19-6301-2025},
  url = {https://doi.org/10.5194/tc-19-6301-2025}
}