NGUYEN et al. (2026) Multi-station rainfall–runoff–operation model for assessing flood characteristics in a cascading dam system under the impacts of climate change and sediment dynamics

Identification

• Journal: Journal of Hydrology
• Year: 2026
• Date: 2026-07-01
• Authors: Phuoc Sinh NGUYEN, Sameh A. Kantoush, Sohei Kobayashi, Amirreza Tabataba Vakili, Vahid Nourani
• DOI: 10.1016/j.jhydrol.2026.135929

Research Groups

• Disaster Prevention Research Institute, Kyoto University
• Department of Urban Management, Kyoto University
• School of Civil Engineering, Shaoxing University
• Center of Excellence in Hydroinformatics and Faculty of Civil Engineering, University of Tabriz
• Faculty of Civil and Environmental Engineering, Near East University
• Water Resources Research Center, Disaster Prevention Research Institute, Kyoto University

Short Summary

This study aimed to develop a multi-station deep learning rainfall–runoff–operation (DL-RRO) model for cascading dam systems and apply it through the large ensemble dataset for Policy Decision-Making for Future Climate Change (d4PDF) to assess flood characteristic changes under climate change and sediment dynamics.

Objective

• Develop a DL-RRO model that couples simulated rainfall–runoff processes, dam operations, and dynamic sedimentation.
• Apply the model to the upper Chikugo River Basin (CRB) using the d4PDF dataset to assess changes in flood characteristics driven by climate change and sediment accumulation.

Study Configuration

• Spatial Scale: The study focuses on the upper CRB, a river basin located in southwestern Japan, with a spatial scale of approximately 533 km².
• Temporal Scale: The study uses hourly data from 1986 to 2025 for training, validation, and testing, and applies future climate scenarios (2031–2090 and 2051–2110) using the d4PDF dataset.

Methodology and Data

• Models used: Long Short-Term Memory (LSTM) network was used as the DL-RRO model.
• Data sources: Hourly rainfall data from 38 rain gauges, discharge and water level observations from Ohira hydrometry station, and dam operation records from Shimouke and Matsubara dams.

Main Results

• The DL-RRO model showed high performance when trained, validated, and tested with 40 years of hourly hydrological observations (NSE = 0.89, RMSE = 18.90 m³/s).
• Analysis of 8,725 simulated flood events revealed increased flood frequency and notable changes in hydrograph characteristics under future climate scenarios.
• Projected 100-year return flood may increase by +53% at Shimouke and +165% at Matsubara under future scenarios, exceeding existing spillway design capacities.

Contributions

• This study provides a comprehensive assessment of the impacts of climate change and sediment dynamics on flood characteristics in cascading dam systems.
• The DL-RRO model developed in this study can be applied to other cascading dam systems after re-tuning.

Funding

• This research was supported by the Japan Ministry of Education, Culture, Sports, Science and Technology (MEXT) through the d4PDF project.
• Additional funding was provided by the Disaster Prevention Research Institute, Kyoto University.

Citation

@article{NGUYEN2026Multistation,
  author = {NGUYEN, Phuoc Sinh and Kantoush, Sameh A. and Kobayashi, Sohei and Vakili, Amirreza Tabataba and Nourani, Vahid},
  title = {Multi-station rainfall–runoff–operation model for assessing flood characteristics in a cascading dam system under the impacts of climate change and sediment dynamics},
  journal = {Journal of Hydrology},
  year = {2026},
  doi = {10.1016/j.jhydrol.2026.135929},
  url = {https://doi.org/10.1016/j.jhydrol.2026.135929}
}