Zhang et al. (2026) Comparison of standalone and hybrid algorithms for SWH model parameterization: Reliable evapotranspiration simulation and partitioning via an ensemble selection mechanism

Identification

• Journal: Agricultural Water Management
• Year: 2026
• Date: 2026-07-13
• Authors: Hao Zhang, Baozhong Zhang, Xin Han, Junfeng Dai
• DOI: 10.1016/j.agwat.2026.110636

Research Groups

• College of Environmental Science and Engineering, Guilin University of Technology
• State Key Laboratory of Water Cycle and Water Security, China Institute of Water Resources and Hydropower Research
• College of Water Conservancy Engineering, Tianjin Agricultural University

Short Summary

This study compares standalone and hybrid algorithms for calibrating the Shuttleworth-Wallace-Hu (SWH) model parameters to simulate evapotranspiration (ET) and its components. The results show that hybrid algorithms outperform standalone algorithms in terms of accuracy and robustness.

Objective

• To develop a reliable method for calibrating SWH model parameters using global heuristic algorithms and local refinement optimizer Fmincon.
• To evaluate the performance of different calibration strategies on summer maize in the North China Plain.

Study Configuration

• Spatial Scale: Local scale (study site: Daxing Water-Saving Irrigation Experimental Station)
• Temporal Scale: Daily and half-hourly scales

Methodology and Data

• Models used: Shuttleworth-Wallace-Hu (SWH) model, Bayesian Optimization (BO), Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Sparrow Search Algorithm (SSA), Fmincon local refinement optimizer.
• Data sources: Eddy covariance system data, meteorological data from the study site.

Main Results

• The hybrid algorithms (BO-Fmincon, GA-Fmincon, PSO-Fmincon, SSA-Fmincon) outperform standalone algorithms in terms of accuracy and robustness.
• The best-performing algorithm is BO-Fmincon, which achieves a test-set R² of 0.934 (half-hourly) and 0.904 (daily), with RMSE of 0.06 mm⋅h⁻¹ and 0.44 mm⋅d⁻¹ , respectively.

Contributions

• This study provides a reliable method for calibrating SWH model parameters using hybrid algorithms.
• The results highlight the importance of combining global heuristic algorithms with local refinement optimizers to improve model accuracy and robustness.

Funding

• This research was funded by [project name], [program name], and [reference code].

Citation

@article{Zhang2026Comparison,
  author = {Zhang, Hao and Zhang, Baozhong and Han, Xin and Dai, Junfeng},
  title = {Comparison of standalone and hybrid algorithms for SWH model parameterization: Reliable evapotranspiration simulation and partitioning via an ensemble selection mechanism},
  journal = {Agricultural Water Management},
  year = {2026},
  doi = {10.1016/j.agwat.2026.110636},
  url = {https://doi.org/10.1016/j.agwat.2026.110636}
}