Ding et al. (2026) Joint prediction of river discharge and suspended sediment load through a physics-informed capacity-supply decomposition framework

Identification

• Journal: Journal of Hydrology
• Year: 2026
• Date: 2026-09-26
• Authors: Can Ding, Shicheng Li, Nanbo Tang, Zhong Tian
• DOI: 10.1016/j.jhydrol.2026.136501

Research Groups

• Water and Environmental Engineering, Department of Built Environment, Aalto University
• Department of Civil and Architectural Engineering, KTH Royal Institute of Technology
• CHN ENERGY Group Jinsha River Xulong Hydropower Co., Ltd.
• CHINA Energy Dadu River Hydropower Dev Co., LTD.
• State Key Lab of Hydraulics and Mountain River Engineering, Sichuan University

Short Summary

This study proposes a physics-informed capacity-supply decomposition (PCSD) framework for joint prediction of river discharge (RD) and suspended sediment load (SSL). The framework maintains high RD prediction accuracy while achieving the best point-estimate performance for SSL among evaluated models.

Objective

• Develop a physically interpretable approach with strong predictive performance for coupled hydro-sediment prediction.
• Reformulate SSL prediction as the product of a flow-dependent transport-capacity term and a supply-related modulation term.

Study Configuration

• Spatial Scale: The study focuses on the Beijiang River Basin, China.
• Temporal Scale: Daily time scale with a 7-day window for RD and a 12-day window for SSL.

Methodology and Data

• Models used: Physics-informed capacity-supply decomposition (PCSD) framework.
• Data sources: Hydro-meteorological data from the Guangdong Research Institute of Water Resources and Hydropower, including daily mean discharge, daily sediment load, rainfall records, evapotranspiration, and soil moisture.

Main Results

• The PCSD framework maintains high RD prediction accuracy while achieving the best point-estimate performance for SSL among evaluated models.
• For one-step SSL prediction, PCSD achieved the best overall performance with RMSE of 195 kg/s, NSE of 0.89, R2 of 0.90, and KGE of 0.94.

Contributions

• The study contributes to the development of a physically interpretable approach for coupled hydro-sediment prediction.
• The PCSD framework provides a novel decomposition strategy for SSL prediction, separating flow-dependent transport capacity from supply-related modulation.

Funding

• This research was funded by the State Key Lab of Hydraulics and Mountain River Engineering, Sichuan University.

Citation

@article{Ding2026Joint,
  author = {Ding, Can and Li, Shicheng and Tang, Nanbo and Tian, Zhong},
  title = {Joint prediction of river discharge and suspended sediment load through a physics-informed capacity-supply decomposition framework},
  journal = {Journal of Hydrology},
  year = {2026},
  doi = {10.1016/j.jhydrol.2026.136501},
  url = {https://doi.org/10.1016/j.jhydrol.2026.136501}
}