li et al. (2026) Daily drought prediction in the Huaihe River Basin using VMD-informer-LSTM

Identification

• Journal: Hydrology and earth system sciences
• Year: 2026
• Date: 2026-06-26
• Authors: Ming Ou, Yao yuhang
• DOI: 10.5194/hess-30-3925-2026

Research Groups

• College of Hydraulic Science and Engineering, Yangzhou University
• National Key Laboratory of Water Disaster Prevention, Hohai University

Short Summary

This study proposes a novel drought prediction framework that integrates Variational Mode Decomposition (VMD), Informer, and Long Short-Term Memory (LSTM) networks to enhance hydrological drought forecasting in the Huaihe River Basin, China.

Objective

• To develop an accurate and robust drought prediction model for regional water resource management and drought risk assessment

Study Configuration

• Spatial Scale: The study focuses on the Huaihe River Basin in China, covering approximately 270,000 km^2.
• Temporal Scale: The dataset spans from 1980 to 2020.

Methodology and Data

• Models used:
  • Variational Mode Decomposition (VMD)
  • Informer
  • Long Short-Term Memory (LSTM) networks
• Data sources:
  • ERA5 reanalysis data for daily actual evapotranspiration, potential evapotranspiration, surface pressure, cloud cover, maximum temperature, mean temperature, wind speed, and precipitation.
  • Land use data from the Data Center for Resources and Environmental Sciences, Chinese Academy of Sciences.

Main Results

• The VMD-informer-LSTM model outperforms other models (LSTM, Transformer-LSTM, Informer-LSTM) in predicting drought events in the Huaihe River Basin.
• The model achieves an improvement of 28.4% in R^2, 46.2% in RMSE, 46.5% in MAE, and 50.8% in MAPE compared to the baseline LSTM.

Contributions

• This study provides a novel drought prediction framework that integrates VMD, Informer, and LSTM networks.
• The model demonstrates superior performance in capturing long-range climate trends and short-term fluctuations in complex non-stationary drought time series.

Funding

• National Key Laboratory of Water Disaster Prevention, Hohai University (project code: XXXXXX)
• Yangzhou University (project code: YZU-XXXXX)

Citation

@article{li2026Daily,
  author = {li, min and Ou, Ming and yuhang, Yao and Yin, Changman},
  title = {Daily drought prediction in the Huaihe River Basin using VMD-informer-LSTM},
  journal = {Hydrology and earth system sciences},
  year = {2026},
  doi = {10.5194/hess-30-3925-2026},
  url = {https://doi.org/10.5194/hess-30-3925-2026}
}