Liu et al. (2026) A health-score-based framework for quality assessment and spatial reconstruction of rainfall monitoring data in reservoir watersheds

Identification

• Journal: Scientific Reports
• Year: 2026
• Date: 2026-09-11
• Authors: Binbin Liu, Mingming Wang, Xiaolei Zhu, Wanbo Zhang
• DOI: 10.1038/s41598-026-69607-y

Research Groups

• Binbin Liu (Corresponding Author)
• Mingming Wang
• Xiaolei Zhu
• Wanbo Zhang

Short Summary

A health-score-based framework is developed to diagnose and reconstruct rainfall monitoring data from the Meishan Reservoir network in Anhui, China. The framework combines rule-based anomaly detection with a station health score that guides donor selection for inverse distance weighting reconstruction.

Objective

• Develop an operational pipeline for quality screening and gap reconstruction in reservoir rainfall monitoring.
• Evaluate the performance of the proposed framework against four alternative interpolation methods.

Study Configuration

• Spatial Scale: Watershed scale, covering 100 automatic telemetry rain gauge stations distributed across the Meishan Reservoir watershed and surrounding administrative areas.
• Temporal Scale: Continuous 2-year monitoring period (2023–2024) with up to 17,520 hourly records per station.

Methodology and Data

• Models used: Rule-based anomaly detection algorithm, health score metric, and inverse distance weighting reconstruction pipeline.
• Data sources: Rainfall values, timestamps, and station metadata from the Meishan Reservoir network.

Main Results

• The proposed framework detected 82 affected stations with anomalies, including 11 cfva events, 29 ipa events, and 24 nta events.
• The health score distribution showed that 21 stations were poor or critical quality, while 70 stations were eligible for donor selection.
• In leave-one-out cross-validation, the IDW method produced an RMSE of 0.281 mm, an MAE of 0.182 mm, and a NSE of 0.979.

Contributions

• The proposed framework provides a practical basis for quality screening and gap reconstruction in reservoir rainfall monitoring.
• The health score metric offers a comprehensive evaluation of station data quality, guiding donor selection for spatial reconstruction.

Funding

• This research was supported by the National Natural Science Foundation of China (Grant No. 52109201) and the Anhui Provincial Key Research and Development Program (Grant No. 202104j06020004).

Citation

@article{Liu2026healthscorebased,
  author = {Liu, Binbin and Wang, Mingming and Zhu, Xiaolei and Zhang, Wanbo},
  title = {A health-score-based framework for quality assessment and spatial reconstruction of rainfall monitoring data in reservoir watersheds},
  journal = {Scientific Reports},
  year = {2026},
  doi = {10.1038/s41598-026-69607-y},
  url = {https://doi.org/10.1038/s41598-026-69607-y}
}