Wei et al. (2025) A time-varying weighted merging method for integrating multisource precipitation data considering error variations

Identification

• Journal: Journal of Hydrology
• Year: 2025
• Date: 2025-11-07
• Authors: Linyong Wei, S. S. Jiang, Liliang Ren, Bin Yong, Zulin Hua, Linqi Zhang, Liping Zeng, Zheng Duan, Chong‐Yu Xu
• DOI: 10.1016/j.jhydrol.2025.134547

Research Groups

• Key Laboratory of Integrated Regulation and Resource Development on Shallow Lake, Ministry of Education, College of Environment, Hohai University, Nanjing, China
• State Key Laboratory of Water Disaster Prevention, Hohai University, Nanjing, China
• College of Hydrology and Water Resources, Hohai University, Nanjing, China
• Meteorological Service Center of Guizhou Province, Guiyang, China
• Department of Physical Geography and Ecosystem Science, Lund University, Lund, Sweden
• Department of Geosciences, University of Oslo, Oslo, Norway

Short Summary

This study developed a Time-varying Weighted (TVW) merging method to integrate multi-source precipitation data, demonstrating its superior performance over parent datasets and state-of-the-art products in mainland China, even with reduced station density.

Objective

• To develop and validate a Time-varying Weighted (TVW) merging method that accounts for error variations to integrate multi-source precipitation data, aiming to produce high-quality precipitation estimates, especially in ungauged areas.

Study Configuration

• Spatial Scale: Mainland China
• Temporal Scale: Daily observations from 2010 to 2015

Methodology and Data

• Models used:
  • Time-varying Weighted (TVW) merging method (proposed)
  • MEIP (Merged ERA5, IMERG-F, and PERSIANN-CDR)
  • MEIC (Merged ERA5, IMERG-F, and CPC)
  • Comparison with: APHRODITE, CMFD, MSWEP
• Data sources:
  • Daily observations from 2278 ground stations (70% for weight determination, 30% for validation)
  • ERA5 (fifth-generation of the European Centre for Medium Range Weather Forecasts atmospheric reanalysis)
  • IMERG-F (Integrated Multi-satellitE Retrievals for Global Precipitation Measurement Final Run)
  • PERSIANN-CDR (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks- Climate Data Record)
  • CPC (Climate Prediction Center) global unified gauge-based analysis

Main Results

• Both MEIP and MEIC products, generated by the TVW method, significantly outperformed their individual parent datasets across all evaluated metrics (correlation, relative bias, absolute difference, and detection probability).
• The TVW method maintained robust performance even when the reference station density was reduced to 10% of the total stations.
• MEIC generally outperformed state-of-the-art precipitation products such as APHRODITE, CMFD, and MSWEP.

Contributions

• Proposed a novel Time-varying Weighted (TVW) merging method that dynamically considers error variations for multi-source precipitation data integration.
• Demonstrated the effectiveness and robustness of the TVW method in producing high-quality precipitation estimates, particularly in data-scarce (ungauged) and large-scale regions.
• Showed that the TVW method can outperform existing state-of-the-art precipitation products, even with limited ground station data.

Funding

Not specified in the provided text.

Citation

@article{Wei2025timevarying,
  author = {Wei, Linyong and Jiang, S. S. and Ren, Liliang and Yong, Bin and Hua, Zulin and Zhang, Linqi and Zeng, Liping and Duan, Zheng and Xu, Chong‐Yu},
  title = {A time-varying weighted merging method for integrating multisource precipitation data considering error variations},
  journal = {Journal of Hydrology},
  year = {2025},
  doi = {10.1016/j.jhydrol.2025.134547},
  url = {https://doi.org/10.1016/j.jhydrol.2025.134547}
}