Muñoz et al. (2026) Assessing the effect of land cover on ISBA snow water equivalent and land surface temperature simulations over Europe

Identification

• Journal: ˜The œcryosphere
• Year: 2026
• Date: 2026-07-24
• Authors: Oscar Javier Rojas Muñoz, Constantin Ardilouze, Bertrand Bonan, Diane Tzanos, Darren J. Ghent, Céline Lamarche, Thomas Nägler, Jean‐Christophe Calvet
• DOI: 10.5194/tc-20-4099-2026

Research Groups

• Météo-France, CNRS, Univ. Toulouse, CNRM, Toulouse, France
• Department of Physics and Astronomy, National Centre for Earth Observation, Univ. Leicester, Leicester, UK
• Earth and Life Institute, Environmental Sciences, Univ. catholique de Louvain, Louvain-la-Neuve, Belgium
• ENVEO IT GmbH, Innsbruck, Austria

Short Summary

This study evaluates the impact of updated land cover (LC) representation on snow water equivalent (SWE) and land surface temperature (LST) simulations in the ISBA land surface model over Europe. It finds that integrating the latest ESA CCI LC data reduces ISBA SWE bias by approximately 23%, primarily driven by specific vegetation transitions, while having a limited impact on LST.

Objective

• To evaluate the impact of land cover (LC) representation on snow simulations (Snow Water Equivalent - SWE and Land Surface Temperature - LST) in the Interactions Between Soil, Biosphere and Atmosphere (ISBA) land surface model over Europe between 2010 and 2022.
• To assess the influence of an updated LC dataset (ECOCLIMAP-SG, based on ESA CCI LC) on the simulation of SWE and LST compared to an older LC dataset (ECOCLIMAP-II).

Study Configuration

• Spatial Scale: Europe (28.125° N–71.875° N, 25.875° W–63.875° E) at a horizontal resolution of 0.25° × 0.25°.
• Temporal Scale: January 2010 to September 2022 (13 years), with hourly atmospheric forcing and model outputs updated every three hours.

Methodology and Data

• Models used:
  • ISBA (Interactions Between Soil, Biosphere and Atmosphere) land surface model, integrated within the SURFEX v9 framework, operating in offline mode.
  • ISBA-ES (Explicit Snow) scheme for snow physics.
  • ISBA-A-gs configuration for the soil-plant system, dynamically computing leaf area index (LAI).
  • ERA5 reanalysis (ECMWF Integrated Forecasting System with a multi-layer snow scheme) for comparison.
• Data sources:
  • Land Cover (LC):
    • ECOCLIMAP-II (referred to as "Old LC"), based on early 2000s inventories (e.g., Corine Land Cover 2000, GLC2000).
    • ECOCLIMAP-SG (referred to as "New LC"), integrating ESA Climate Change Initiative (CCI) LC v2.0.7 product for 2010 at 300 m resolution.
  • Atmospheric Forcing: Hourly ERA5 atmospheric reanalysis data, interpolated to a 0.25° × 0.25° grid.
  • Snow Water Equivalent (SWE) for benchmarking:
    • ESA CCI Snow project (CCI SWE, version 3.1), daily coverage at 0.10° resolution (regridded to 0.25°), derived from passive microwave radiometer observations (1979–2022).
    • ERA5 reanalysis SWE, hourly at approximately 31 km resolution (regridded to 0.25°), assimilating IMS product.
  • Land Surface Temperature (LST) for benchmarking:
    • ESA CCI LST project (CCI LST, pre-release version 4 Aqua MODIS), daily at 0.05° resolution (regridded to 0.25°), corresponding to satellite overpass times (approximately 13:30 LT and 01:30 LT).
    • ERA5 reanalysis LST, hourly at approximately 31 km resolution (regridded to 0.25°).

Main Results

• ISBA simulations (both Old and New LC) consistently overestimate peak SWE values compared to the ESA CCI product, with differences exceeding 20 mm during the warm winter of 2020.
• Using the updated ESA CCI LC data (New LC) reduces the ISBA SWE mean bias by approximately 23% (from 6.4 mm to 4.9 mm) and the root-mean-square deviation (RMSD) from 34.7 mm to 32.9 mm.
• This reduction in SWE bias is primarily driven by specific land cover transitions, notably from grasslands to deciduous broadleaf forests (54.2% contribution to MB reduction) and grasslands to coniferous forests (10.3% contribution to MB reduction).
• Changes in SWE are also observed in areas where the dominant land cover remains unchanged, indicating that modifications in sub-grid vegetation fractions can affect snow-vegetation interactions and SWE bias.
• The updated LC has a very limited impact on the model's performance for land surface temperature (LST), with persistent daytime cold biases and nighttime warm biases in ISBA compared to CCI LST.
• ERA5 generally shows higher agreement with satellite-derived snow dynamics (R = 0.90, RMSD = 23.4 mm) than ISBA simulations (R = 0.84, RMSD = 32.9 mm for New LC).
• All datasets (CCI, ERA5, Old LC, New LC) consistently show predominantly negative SWE anomalies across much of Europe during the warm winter of 2019–2020.
• A known limitation of the CCI SWE product is its saturation for deep snowpacks (SWE > 150 mm), which may lead to an underestimation of actual SWE.

Contributions

• Provides a comprehensive evaluation of the impact of integrating updated ESA CCI land cover data (ECOCLIMAP-SG) into the ISBA land surface model for snow water equivalent (SWE) and land surface temperature (LST) simulations over Europe.
• Quantifies the reduction in SWE bias and RMSD attributable to the updated land cover, highlighting that specific vegetation transitions (e.g., grasslands to forests) are the primary drivers of improvement.
• Demonstrates that changes in sub-grid vegetation fractions, even without a change in dominant land cover, can significantly affect snow-vegetation interactions and SWE simulations.
• Benchmarks ISBA model performance against both satellite observations (ESA CCI) and reanalysis data (ERA5), providing insights into the strengths and limitations of each.
• Emphasizes the importance of accurate and updated land cover information for improving snow representation in land surface models, while also suggesting that further improvements require advances in atmospheric forcing, snow process representation, and data assimilation.

Funding

• European Space Agency (ESA Climate Change Initiative Climate Model User Group (ESA CCI CMUG), grant no. 4000125156/18/I-NB)
• Copernicus Climate Change Service Evolution (CERISE) project (grant no. 101082139)

Citation

@article{Muñoz2026Assessing,
  author = {Muñoz, Oscar Javier Rojas and Ardilouze, Constantin and Bonan, Bertrand and Tzanos, Diane and Ghent, Darren J. and Lamarche, Céline and Nägler, Thomas and Calvet, Jean‐Christophe},
  title = {Assessing the effect of land cover on ISBA snow water equivalent and land surface temperature simulations over Europe},
  journal = {˜The œcryosphere},
  year = {2026},
  doi = {10.5194/tc-20-4099-2026},
  url = {https://doi.org/10.5194/tc-20-4099-2026}
}