Martí et al. (2026) Implementation of a dry surface layer soil resistance in two contrasting semi-arid sites with SURFEX-ISBA V9.0

Identification

• Journal: Geoscientific model development
• Year: 2026
• Date: 2026-03-09
• Authors: Belén Martí, Jannis Groh, Guylaine Canut, Aaron Boone
• DOI: 10.5194/gmd-19-1991-2026

Research Groups

• Météo-France, CNRS, Université de Toulouse, CNRM, Toulouse, France
• Institute of Crop Science and Resource Conservations (INRES), Soil Science and Soil Ecology, University of Bonn, Bonn, Germany
• Institute of Bio- and Geosciences – Agrosphere (IBG-3), Forschungszentrum Jülich GmbH, Jülich, Germany
• Research Area 1 Landscape Functioning, Isotope Biogeochemistry and Gas Fluxes, Leibniz Centre for Agricultural Landscape Research (ZALF), Müncheberg, Germany

Short Summary

This study evaluates and improves the SURFEX-ISBA V9.0 land surface model's estimation of latent heat fluxes in semi-arid environments by implementing a dry surface layer (DSL) soil resistance. The DSL resistance successfully reduced the overestimation of bare soil evaporation, leading to a 29% to 32% reduction in the daily Root Mean Square Error of latent heat flux at two contrasting sites.

Objective

• To evaluate the performance of the ISBA land surface model within the SURFEX platform in estimating latent heat fluxes in semi-arid environments, particularly addressing the overestimation of soil evaporation.
• To implement and test a dry surface layer (DSL) soil resistance parameterization in ISBA to improve the simulation of bare soil evaporation and, consequently, the partitioning of evapotranspiration.

Study Configuration

• Spatial Scale: Two contrasting semi-arid sites in the northeastern Iberian Peninsula during the LIAISE (Land surface Interactions with the Atmosphere over the Iberian Semi-arid Environment) field experiment:
  • La Cendrosa: An irrigated alfalfa field.
  • Els Plans: A rainfed natural grassland.
• Temporal Scale:
  • La Cendrosa simulations: 1 July to 1 August (1 month).
  • Els Plans simulations: 17 June to 29 September (over 3 months).
  • LIAISE campaign: April 2021 to end of September 2021 (Long Observational Period).

Methodology and Data

• Models used:
  • SURFEX-ISBA V9.0 (Surface Externalisée – Interactions Soil-Biosphere-Atmosphere)
  • ISBA Multi-Energy Budget (MEB) model option with a multilayer diffusive soil scheme.
  • A-gs (Assimilation-stomatal conductance) scheme for vegetation processes.
  • Two soil resistance parameterizations tested: Sellers et al. (1992b) and a newly implemented Dry Surface Layer (DSL) resistance (based on Swenson and Lawrence, 2014).
• Data sources:
  • LIAISE field experiment data, including eddy-covariance systems for turbulent fluxes (latent heat, sensible heat).
  • Buried sensors for ground temperature and flux plates for ground heat flux.
  • Meteorological measurements: incident shortwave and longwave radiation fluxes, wind speed, temperature, specific humidity, pressure, atmospheric CO2 concentration, and rainfall rate (30 min time step).
  • In-situ measurements of soil and vegetation properties: Leaf Area Index (LAI), vegetation height, soil texture, soil hydraulic properties, volumetric water content, and soil temperature.
  • Microlysimeter measurements for evapotranspiration partitioning (La Cendrosa).

Main Results

• The default ISBA model (without soil resistance) generally overestimated evapotranspiration (ET), primarily due to an overestimation of bare soil evaporation.
• Implementing a Dry Surface Layer (DSL) soil resistance significantly improved the simulation of latent heat flux (LE) by reducing bare soil evaporation.
• The daily Root Mean Square Error (RMSE) of LE was reduced by 29% at the alfalfa site (La Cendrosa) and 32% at the natural grass site (Els Plans) using the DSL approach.
• Simulations of sensible heat flux (H) and net radiation (Rn) improved by approximately 10 W m⁻², while ground heat flux (G) simulations deteriorated by a similar magnitude.
• The DSL simulations reduced the overall global error compared to simulations without DSL resistance.
• A sensitivity analysis of DSL parameters suggested optimal values for the study sites: a DSL thickness (zdsl) of 0.01 m and a porosity coefficient (Kdsl) between 0.8 and 0.85, which are slightly different from global values found in previous studies.
• The model struggled to reproduce negative nocturnal LE values (dew formation or soil water vapor adsorption) observed at the dry natural grass site.
• Soil temperature simulations showed larger diurnal amplitudes than observed, and soil volumetric water content (VWC) simulations, while capturing trends, still exhibited biases, highlighting challenges in soil hydraulic property characterization.

Contributions

• First implementation and evaluation of a physically-based Dry Surface Layer (DSL) soil resistance parameterization within the SURFEX-ISBA V9.0 land surface model.
• Demonstrated a significant improvement in evapotranspiration (ET) partitioning and latent heat flux (LE) estimation in semi-arid environments, effectively mitigating the common overestimation of bare soil evaporation in Land Surface Models (LSMs).
• Provided a detailed local-scale evaluation of ISBA-MEB with specific vegetation and soil parameterizations for two contrasting semi-arid sites (irrigated alfalfa and rainfed natural grass), offering insights for global-scale approximations.
• Conducted a sensitivity analysis of the DSL parameters (Kdsl, zdsl) for ISBA, suggesting optimized values for the specific conditions studied.
• Highlighted the critical importance of accurate, site-specific vegetation (e.g., daily LAI evolution, photosynthesis parameters) and soil hydraulic property characterization for robust LSM performance in complex and heterogeneous environments.

Funding

• French National Agency for Research (ANR) through the project "HILIAISE: Human imprint on Land surface Interactions with the Atmosphere over the Iberian Semi-arid Environment" (grant no. ANR-19-CE01-0017).
• Deutsche Forschungsgemeinschaft (DFG, German Research Foundation) – project no. 460817082 (for Jannis Groh).

Citation

@article{Martí2026Implementation,
  author = {Martí, Belén and Groh, Jannis and Canut, Guylaine and Boone, Aaron},
  title = {Implementation of a dry surface layer soil resistance in two contrasting semi-arid sites with SURFEX-ISBA V9.0},
  journal = {Geoscientific model development},
  year = {2026},
  doi = {10.5194/gmd-19-1991-2026},
  url = {https://doi.org/10.5194/gmd-19-1991-2026}
}