Tanhapour et al. (2026) Potential of Sentinel-3 snow cover fraction data for improving hydrological simulations at the regional scale

Identification

• Journal: Scientific Reports
• Year: 2026
• Date: 2026-03-30
• Authors: Mitra Tanhapour, Juraj Parajka, Gabriele Schwaizer, Mariëtte Vreugdenhil, Silvia Kohnová, Kamila Hlavčová, Roman Výleta, Jan Szolgay
• DOI: 10.1038/s41598-026-46403-2

Research Groups

• Department of Land and Water Resources Management, Slovak University of Technology, Bratislava, Slovakia
• Centre for Water Resource Systems, TU Wien, Vienna, Austria
• Institute of Hydraulic Engineering and Water Resources Management, TU Wien, Vienna, Austria
• ENVEO-Environmental Earth Observation IT GmbH, Innsbruck, Austria
• Department of Geodesy and Geoinformation, TU Wien, Vienna, Austria

Short Summary

This study evaluates the accuracy and potential of a new Sentinel-3 snow cover fraction (SCF) product for improving hydrological simulations at a regional scale in Austria, demonstrating that its use in multiple-objective model calibration significantly enhances snow and runoff simulations, particularly in lowland catchments.

Objective

• To evaluate the potential of a new Sentinel-3 snow cover fraction (SCF) product for improving hydrological simulations at the regional scale.
• To compare the snow cover mapping accuracy of the Sentinel-3 SCF product with daily snow depth observations at 631 climate stations.
• To assess and compare the runoff and snow model efficiencies obtained from multiple-objective calibration (runoff and SCF) versus calibration to runoff only.

Study Configuration

• Spatial Scale: Austria (approximately 84,000 km²), covering 188 lowland and alpine catchments (ranging from 13.7 km² to 6214 km²). Sentinel-3 SCF product at 200 m spatial resolution. Hydroclimatic inputs at 1 km resolution. Hydrological model applied in 200 m elevation zones.
• Temporal Scale:
  • Sentinel-3 SCF product: Daily, January 2017 to December 2023.
  • Snow depth observations: Daily, January 2017 to December 2023.
  • Hydrological modeling inputs and runoff observations: Daily, 1 September 2017 to 31 August 2022.
  • Model calibration period: 1 September 2017 to 31 August 2020.
  • Model validation period: 1 September 2020 to 31 August 2022.
  • Model warm-up period: 365 days prior to calibration/validation.

Methodology and Data

• Models used:
  • Conceptual hydrological TUW model (semi-distributed, 15 parameters).
  • Locally Adaptive Multi-Spectral Unmixing (LAMSU) algorithm for Sentinel-3 SCF product generation.
• Data sources:
  • Satellite: Sentinel-3 SLSTR and OLCI observations for Snow Cover Fraction (SCF).
  • Observation: Daily snow depth measurements from 631 climate stations in Austria (Hydrographic Service of Austria). Daily runoff observations for 188 catchments (Hydrographic Service of Austria).
  • Reanalysis/Gridded Data: SPARTACUS database (Geosphere Austria) for daily grid maps of precipitation and air temperature at 1 km resolution. Mean daily potential evaporation derived from gridded air temperature and potential sunshine duration index using a modified Blaney–Criddle approach.
  • Ancillary Data: Copernicus Global Surface Water dataset for water body masking in SCF. Adapted Simple Cloud Detection Algorithm for cloud masking.

Main Results

• The Sentinel-3 SCF product shows very high agreement with daily snow depth observations at climate stations, with a median overall accuracy exceeding 95.3% (interquartile range of 2.65%) for the period 2017–2023.
• Snow cover overestimation errors (SCF shows snow, ground observation is zero) are generally higher than underestimation errors, but are significantly lower in higher elevations (less than 3% from December to March) compared to previous assessments using other products.
• The mean cloud coverage over Austria in the SCF product is 57.8%, with lower cloud coverage in alpine regions during winter months compared to lowland areas.
• Multiple-objective calibration (CALM), incorporating Sentinel-3 SCF data, improves runoff simulations in 39% of the overall catchments during the validation period, with a notable improvement in 54% of lowland catchments.
• CALM significantly enhances snow simulations in 84% of the overall catchments during the validation period, improving 98% of alpine catchments and 68% of lowland catchments.
• The use of SCF data in calibration impacts key hydrological model parameters related to snow accumulation and melt (e.g., degree-day factor, threshold temperatures) and soil moisture (e.g., soil field capacity), particularly in alpine catchments for soil field capacity.

Contributions

• First comprehensive assessment of the accuracy and potential of a new Sentinel-3 SCF product, derived using a physically based spectral unmixing approach, for hydrological modeling at a regional scale.
• Demonstrates that the Sentinel-3 SCF product offers reduced cloud coverage and lower overestimation errors in complex alpine terrain compared to previous NDSI-based products.
• Provides evidence that integrating Sentinel-3 SCF data into multiple-objective hydrological model calibration improves both snow and runoff simulations in the validation period, especially enhancing runoff predictions in lowland catchments and snow dynamics in alpine catchments.
• Highlights the impact of Sentinel-3 SCF data on the calibration of sensitive hydrological model parameters, suggesting improved internal consistency and reduced uncertainty in hydrological simulations.

Funding

• Slovak Research and Development Agency (contract No. APVV-23-0332 and No. VV-MVP-24-0208)
• VEGA grant agency (No. VEGA 1/0657/25)
• EU NextGenerationEU through the recovery and resilience plan for Slovakia (project No. 09I03-03-V05-00005)
• ESA EXPRO+ AlpSnow project (Contract No. 4000132770/20/I-NB)
• FFG ASAP project Digital Twin for Austria – Alpine Hydrology and Future Hazards (DTA-Hydro, project No. FO999918403)

Citation

@article{Tanhapour2026Potential,
  author = {Tanhapour, Mitra and Parajka, Juraj and Schwaizer, Gabriele and Vreugdenhil, Mariëtte and Kohnová, Silvia and Hlavčová, Kamila and Výleta, Roman and Szolgay, Jan},
  title = {Potential of Sentinel-3 snow cover fraction data for improving hydrological simulations at the regional scale},
  journal = {Scientific Reports},
  year = {2026},
  doi = {10.1038/s41598-026-46403-2},
  url = {https://doi.org/10.1038/s41598-026-46403-2}
}