Rickard (2025) Deciphering soil structure: linking soil physics, water dynamics, carbon storage, and agricultural resilience in long-term experiments

Identification

• Journal: Nottingham ePrints (University of Nottingham)
• Year: 2025
• Date: 2025-12-12
• Authors: Rickard, William
• DOI: None

Research Groups

• CNRM-GAME (Météo-France, CNRS), Toulouse, France.
• Helmholtz Centre for Environmental Research – UFZ, Leipzig, Germany.
• Department of Computational Hydrosystems, Leipzig, Germany.

Short Summary

This study evaluates the performance of the ISBA land surface model and the mHM hydrological model in simulating soil moisture and river discharge. The results indicate that while both models accurately capture soil moisture dynamics, mHM demonstrates superior accuracy in river discharge simulation.

Objective

• To compare the ability of a land surface model (ISBA) and a multiscale hydrological model (mHM) to simulate soil moisture and streamflow using high-quality in-situ observations and atmospheric forcing.

Study Configuration

• Spatial Scale: The Hérault catchment and the SMOSMANIA network in Southern France (approximately 2,500 km²).
• Temporal Scale: 4 years (2007–2010).

Methodology and Data

• Models used: ISBA (Interactions between Soil, Biosphere, and Atmosphere) integrated within the SURFEX platform; mHM (multiscale Hydrological Model).
• Data sources: SAFRAN atmospheric reanalysis (8,000 m resolution), in-situ soil moisture data from the 12-station SMOSMANIA network, and river discharge data from the Banque Hydro database.

Main Results

• Both models showed high proficiency in simulating soil moisture dynamics, with average correlation coefficients ($R$) exceeding 0.80 across the study period.
• The mHM model outperformed ISBA in river discharge simulation, achieving Nash-Sutcliffe Efficiency (NSE) values greater than 0.70 for the majority of the sub-basins.
• Soil moisture anomalies were more consistently captured than absolute volumetric values, with Root Mean Square Errors (RMSE) typically ranging between 0.04 and 0.06 m³/m³.
• The study confirmed that mHM's multiscale parameter regionalization (MPR) allows for consistent performance across different spatial resolutions.

Contributions

• Provides a rigorous cross-model evaluation between the land surface modeling (LSM) and hydrological modeling communities.
• Demonstrates the critical utility of dense in-situ networks (such as SMOSMANIA) for the validation of large-scale environmental models.
• Highlights the differences in how LSMs and hydrological models partition the water balance, specifically regarding runoff and drainage.

Funding

• Météo-France.
• Centre National de la Recherche Scientifique (CNRS).
• European Space Agency (ESA) [Contract 4000101720/10/NL/ARD].
• German Research Foundation (DFG).

Citation

@article{Rickard2025Deciphering,
  author = {Rickard, William},
  title = {Deciphering soil structure: linking soil physics, water dynamics, carbon storage, and agricultural resilience in long-term experiments},
  journal = {Nottingham ePrints (University of Nottingham)},
  year = {2025},
  url = {https://openalex.org/W7115101916}
}

Generated by BiblioAssistant using gemini-3-flash-preview (Google API)