Gimeno et al. (2025) Hydropedological clustering: improving the representation of low streamflows in a semi-distributed hydrological model

Identification

• Journal: Journal of Hydrology
• Year: 2025
• Date: 2025-12-20
• Authors: Fernando Gimeno, Mauricio Zambrano-Bigiarini, Camila Álvarez-Garretón, Mauricio Galleguillos
• DOI: 10.1016/j.jhydrol.2025.134787

Research Groups

• Doctorate Program in Natural Resources Sciences, Universidad de la Frontera, Chile
• Center for Climate and Resilience Research, Universidad de Chile, Chile
• Department of Civil Engineering, Universidad de la Frontera, Chile
• Faculty of Engineering and Sciences, Universidad Adolfo Ibáñez, Chile
• Data Observatory Foundation, Chile

Short Summary

This study evaluated how hydropedological clustering, based on soil hydraulic properties, improves the simulation of low streamflows and soil water content in the SWAT+ model, finding significant enhancements compared to traditional soil datasets and demonstrating its criticality over soil map resolution.

Objective

• To evaluate how different soil datasets and classification approaches, particularly a new hydropedological clustering method, affect the performance of the semi-distributed, physically based SWAT+ model in simulating low streamflows and soil water content (SWC) in a Mediterranean catchment.

Study Configuration

• Spatial Scale: Cauquenes catchment, central Chile.
• Temporal Scale: Not explicitly specified in the abstract.

Methodology and Data

• Models used: SWAT+ (Soil and Water Assessment Tool Plus), a semi-distributed, physically based hydrological model.
• Data sources:
  • Global soil datasets: HWSDv1.2, DSOLMap.
  • Locally derived soil products: CLSoilMapsTex, CLSoilMapsCl (the latter using a new hydropedological clustering approach).
  • Hydropedological clustering based on saturated hydraulic conductivity (Ks), available water capacity (AWC), and pore-size distribution index (α).
  • Streamflow observations for low streamflow simulation evaluation.
  • Soil water content (SWC) observations for model evaluation.

Main Results

• Hydropedological clustering (CLSoilMapsCl) substantially improved low streamflow simulations, achieving a Kling-Gupta Efficiency (KGE_lf) of 0.67, which is 44 % higher than simulations using HWSDv1.2.
• The clustering approach more accurately reproduced hydrological signatures.
• Hydropedological clustering reduced calibration runtime by 29 %.
• SWC simulations using hydropedological clustering showed a balanced fit with a correlation coefficient (r) of 0.69 and a Root Mean Square Error (RMSE) of 0.059.
• The study found that hydropedological clustering is more critical than soil map resolution for accurate low streamflow simulation.

Contributions

• Developed and demonstrated a new hydropedological clustering approach based on key soil hydraulic properties (Ks, AWC, α) for improved hydrological modeling.
• Highlighted that integrating hydropedological information significantly enhances the representation of soil-water interactions within the SWAT+ model.
• Provided evidence that this approach leads to more reliable low streamflow modeling and water-resource assessments, particularly in Mediterranean catchments.
• Showed that the proposed clustering method is more impactful than increasing soil map resolution for low streamflow simulation and reduces calibration effort.

Funding

• ANID Technology Center No. DO210001 (Data Observatory Foundation).

Citation

@article{Gimeno2025Hydropedological,
  author = {Gimeno, Fernando and Zambrano-Bigiarini, Mauricio and Álvarez-Garretón, Camila and Galleguillos, Mauricio},
  title = {Hydropedological clustering: improving the representation of low streamflows in a semi-distributed hydrological model},
  journal = {Journal of Hydrology},
  year = {2025},
  doi = {10.1016/j.jhydrol.2025.134787},
  url = {https://doi.org/10.1016/j.jhydrol.2025.134787}
}