Zhou et al. (2026) Climate-driven optimal canal water allocation in the Changgang Irrigation District, Northeast China, using LSTM (Long Short-Term Memory)-iTransformer forecasting and bilevel fractional programming

Identification

• Journal: Agricultural Water Management
• Year: 2026
• Date: 2026-10-13
• Authors: Huiying Zhou, Qiang Fu, Yaowen Xu, Mo Li, Tianxiao Li, Xuechen Yang, Renjie Hou
• DOI: 10.1016/j.agwat.2026.110848

Research Groups

• College of Hydraulic Science and Engineering, Northeast Agricultural University, Harbin, Heilongjiang, 150030, China
• Joint Laboratory for International Cooperation on Cold Region Black Soil Habitat Health of Ministry of Education, Harbin, Heilongjiang, 150030, China
• Key Laboratory of Effective Utilization of Agricultural Water Resources of Ministry of Agriculture, Northeast Agricultural University, Harbin, Heilongjiang, 150030, China
• Heilongjiang Provincial Key Laboratory of Water Resources and Water Conservancy Engineering in Cold Region, Northeast Agricultural University, Harbin, Heilongjiang, 150030, China

Short Summary

This study develops an integrated framework that combines the LSTM-iTransformer model for water supply-demand forecasting with the BFP model for optimal canal scheduling to enhance water allocation efficiency in rice irrigation districts under future climate change.

Objective

• To integrate climate-driven forecasting of water supply and demand with hierarchical canal water allocation optimization to support refined scheduling in irrigation districts.
• To develop a comprehensive analytical framework that combines LSTM-iTransformer prediction model with BFP-based optimal canal scheduling model.

Study Configuration

• Spatial Scale: Regional scale, focusing on the Changgang Irrigation District in Heilongjiang Province.
• Temporal Scale: Daily time step for water supply and demand forecasting, and hourly time step for canal scheduling optimization.

Methodology and Data

• Models used:
  • LSTM-iTransformer hybrid model for multi-step forecasting of regional water supply and demand.
  • BFP-based optimal irrigation water scheduling model for hierarchical decision-making in irrigation districts.
• Data sources:
  • Meteorological data from CMIP6 climate scenarios (SSP-RCP).
  • Hydrological data from the Changgang Irrigation District.

Main Results

• The LSTM-iTransformer model achieves high accuracy in predicting runoff and ET0, with R2 values of 0.8946 and 0.8415, respectively.
• Optimization results show that the BFP model effectively coordinates canal system operational efficiency with field irrigation benefits, reducing total water allocation time by 16.19% and canal water conveyance loss by 23.76%.

Contributions

• This study provides a comprehensive analytical framework for integrating climate-driven forecasting of water supply and demand with hierarchical canal water allocation optimization.
• The proposed framework can enhance water allocation efficiency in rice irrigation districts under future climate change.

Funding

• This research was funded by the National Natural Science Foundation of China (Grant No. 52109251).
• The Heilongjiang Provincial Key Laboratory of Water Resources and Water Conservancy Engineering in Cold Region also provided financial support for this project.

Citation

@article{Zhou2026Climatedriven,
  author = {Zhou, Huiying and Fu, Qiang and Xu, Yaowen and Li, Mo and Li, Tianxiao and Yang, Xuechen and Hou, Renjie},
  title = {Climate-driven optimal canal water allocation in the Changgang Irrigation District, Northeast China, using LSTM (Long Short-Term Memory)-iTransformer forecasting and bilevel fractional programming},
  journal = {Agricultural Water Management},
  year = {2026},
  doi = {10.1016/j.agwat.2026.110848},
  url = {https://doi.org/10.1016/j.agwat.2026.110848}
}