Zhao et al. (2026) The impact of numerical error on hydrological model performance

Identification

• Journal: Journal of Hydrology
• Year: 2026
• Date: 2026-09-29
• Authors: Jianfei Zhao, Yanan Duan, Jialu HUANG, Yiming Hu, Binquan Li, Jun Wang, Zhongmin Liang
• DOI: 10.1016/j.jhydrol.2026.136510

Research Groups

• College of Hydrology and Water Resources, Hohai University
• Jiangsu Province Water Conservancy Engineering Sci-tech Consulting Co., Ltd.

Short Summary

This study investigates the impact of numerical error on hydrological model performance using the Xinanjiang model as a case study. The results show that controlling numerical error leads to systematic performance improvements, underscoring its non-negligible contribution to model uncertainty.

Objective

• Investigate the effect of numerical error on hydrological model performance and explore methods for mitigating this impact

Study Configuration

• Spatial Scale: Watersheds from the CAMELS dataset (547 watersheds)
• Temporal Scale: Daily time step

Methodology and Data

• Models used: Xinanjiang model in both discrete equation form (DTE-XAJ) and differential equation form (ODE-XAJ)
• Data sources: CAMELS dataset, satellite remote sensing data, and in situ observations

Main Results

• The DTE-XAJ model converges to the ODE-XAJ model under time step refinement, demonstrating structural consistency.
• Numerical error significantly degrades Xinanjiang model performance during both calibration and validation periods.
• Controlling numerical error leads to systematic performance improvements.

Contributions

• This study proposes a convergence validation method for assessing structural consistency in hydrological models.
• The findings provide empirical evidence of the non-negligible contribution of numerical error to model uncertainty.

Funding

• This research was funded by the National Natural Science Foundation of China (Grant No. 52109233) and the Jiangsu Province Water Conservancy Engineering Sci-tech Consulting Co., Ltd.

Citation

@article{Zhao2026impact,
  author = {Zhao, Jianfei and Duan, Yanan and HUANG, Jialu and Hu, Yiming and Li, Binquan and Wang, Jun and Liang, Zhongmin},
  title = {The impact of numerical error on hydrological model performance},
  journal = {Journal of Hydrology},
  year = {2026},
  doi = {10.1016/j.jhydrol.2026.136510},
  url = {https://doi.org/10.1016/j.jhydrol.2026.136510}
}