Zeinolabedini et al. (2026) Enhancing high-flow estimation in rainfall-runoff models using resampling techniques: applications to GR4J and Sacramento models

Identification

• Journal: Environmental Modelling & Software
• Year: 2026
• Date: 2026-06-22
• Authors: Maryam Zeinolabedini, Fiona Johnson, Lucy Amanda Marshall
• DOI: 10.1016/j.envsoft.2026.107092

Research Groups

• Water Research Centre, School of Civil and Environmental Engineering, University of New South Wales, Kensington, New South Wales, Australia
• ARC Training Centre Data Analytics for Resources and Environments, School of Life and Environmental Sciences, The University of Sydney, Camperdown, New South Wales, Australia
• The University of Sydney, Camperdown, New South Wales, Australia
• School of Environment and Science, Griffith University, Queensland, Australia

Short Summary

This study proposes a new method to improve high-flow estimation in rainfall-runoff models using resampling techniques. A modified Kling-Gupta efficiency (KGE) objective function incorporating over-sampling (OS) and hybrid-sampling (HS) is developed and applied to the Genie Rural a 4 Parameters Journalier (GR4J) and Sacramento Soil Moisture Accounting Model (SAC). The results show that resampling enhances high-flow estimates in approximately 60% of the catchments, with improved performance for both models.

Objective

• To investigate the effectiveness of resampling techniques on high-flow predictions of rainfall-runoff models.
• To develop a modified KGE objective function incorporating OS and HS to improve high-flow estimation.

Study Configuration

• Spatial Scale: Catchment scale (168 Hydrologic Reference Stations across Australia)
• Temporal Scale: Daily time series data from 1970-2014

Methodology and Data

• Models used: GR4J and SAC conceptual rainfall-runoff models
• Data sources: Australian Bureau of Meteorology (daily rainfall, potential evapotranspiration, and streamflow)

Main Results

• Resampling techniques enhance high-flow estimates in approximately 60% of the catchments.
• The modified KGE objective function incorporating OS and HS improves high-flow estimation for both GR4J and SAC models.
• The SAC model exhibits better overall performance than GR4J when using the modified KGE objective function.

Contributions

• This study proposes a new method to improve high-flow estimation in rainfall-runoff models using resampling techniques.
• The results show that resampling enhances high-flow estimates in approximately 60% of the catchments, with improved performance for both models.

Funding

• ARC Training Centre Data Analytics for Resources and Environments (project code: [insert project code])
• Australian Research Council (grant number: [insert grant number])

Citation

@article{Zeinolabedini2026Enhancing,
  author = {Zeinolabedini, Maryam and Johnson, Fiona and Marshall, Lucy Amanda},
  title = {Enhancing high-flow estimation in rainfall-runoff models using resampling techniques: applications to GR4J and Sacramento models},
  journal = {Environmental Modelling & Software},
  year = {2026},
  doi = {10.1016/j.envsoft.2026.107092},
  url = {https://doi.org/10.1016/j.envsoft.2026.107092}
}