Sun et al. (2026) An hourly 0.02° total precipitable water dataset for all-weather conditions over the Tibetan Plateau through the fusion of observations of geostationary and multi-source microwave satellites

Identification

• Journal: Earth system science data
• Year: 2026
• Date: 2026-01-14
• Authors: Qixiang Sun, Husi Letu, Yongqian Wang, Peng Zhang, Hong Liang, Chong Shi, Shuai Yin, Jiancheng Shi, Dabin Ji
• DOI: 10.5194/essd-18-371-2026

Research Groups

• State Key Laboratory of Remote Sensing and Digital Earth, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing, China
• University of Chinese Academy of Sciences, Beijing, China
• College of Resources and Environment, Chengdu University of Information Technology, Chengdu, China
• Meteorological Observation Center, China Meteorological Administration, Beijing, China
• State Key Laboratory of Environment Characteristics and Effects for Near-space, Beijing, China
• National Space Science Center, Chinese Academy of Sciences, Beijing, China

Short Summary

The Tibetan Plateau plays a vital role in Asia’s water cycle, but tracking water vapor in this mountainous region is difficult, especially under cloudy conditions. We developed a new satellite-based method to generate hourly water vapor data at 0.02-degree resolution from 2016 to 2022, now available at https://doi.org/10.11888/Atmos.tpdc.301518, which improves accuracy and reveals fine-scale moisture transport critical for understanding rainfall and extreme weather.

Objective

• To propose a novel multi-source remote sensing Total Precipitable Water (TPW) fusion framework that integrates observations from eight microwave satellites and the Himawari-8/9 (H8/9) geostationary satellite to produce an all-weather, high spatiotemporal resolution TPW dataset over the Tibetan Plateau.
• To address systematic biases among multi-source microwave observations and improve the accuracy and spatial continuity of fused TPW data under cloudy conditions over complex terrain.

Study Configuration

• Spatial Scale: Tibetan Plateau (25° to 40° N, 67° to 105° E), with a final dataset resolution of 0.02°.
• Temporal Scale: 2016 to 2022, with an hourly temporal resolution for the final dataset.

Methodology and Data

• Models used:
  • Multi-source remote sensing TPW fusion framework (overall).
  • Random Forest algorithm (for bias correction of microwave TPW using H8/9 as reference, and for spatial downscaling).
  • Adaptive correction method (for high-resolution TPW under cloudy conditions, based on clear-sky boundary differences).
  • Bilinear interpolation (for reprojection and spatial gap filling).
  • Linear interpolation (for temporal gap filling).
  • Savitzky–Golay filter (for smoothing reconstructed fields).
  • Neural Network (for H8/9 TPW retrieval, by Jiang et al., 2022a).
  • Cloud detection algorithm (by Shang et al., 2024).
• Data sources:
  • Satellite observations (inputs):
    • Eight microwave satellites: GCOM-W/AMSR2, NOAA-18/19 (MIRS), MetOp-A/B (MIRS), DMSP-F17/18 (MIRS), GPM Core (MIRS).
    • Himawari-8/9 (H8/9) Advanced Himawari Imager (AHI) clear-sky TPW.
  • Validation data:
    • GNSS TPW data (from CMONOC, 44 stations over TP).
    • Integrated Global Radiosonde Archive (IGRA) TPW data.
    • Scientific Expedition ground-based TPW observations: Microwave radiometer (MWR) observations, GNSS TPW observations in the central Himalayas.
  • Comparison/Reanalysis data:
    • Morphed Integrated Microwave Imagery at CIMSS TPW product version 2 (MIMIC-TPW2).
    • European Centre for Medium-Range Weather Forecasts Reanalysis v5 (ERA5) TPW.
  • Auxiliary data:
    • Tibetan Plateau vector boundary.
    • Elevation data (ETOPO2022, 15 arcsec resolution).
    • Geographic coordinates and time.

Main Results

• An all-weather TPW dataset for the Tibetan Plateau (2016-2022) was produced with hourly temporal and 0.02° spatial resolution.
• The fused TPW product (2017) showed an hourly Root Mean Square Error (RMSE) of 3.79 mm against GNSS TPW, which is 10.82% lower than MIMIC-TPW2 (4.25 mm) and 6.19% lower than ERA5 (4.04 mm).
• The bias of the fused TPW was -1.15 mm, smaller than MIMIC-TPW2 (-2.02 mm) and ERA5 (-2.22 mm).
• At daily and monthly scales, the fused TPW consistently exhibited lower RMSEs and biases compared to MIMIC-TPW2 and ERA5.
• The fused TPW demonstrated lower errors at most GNSS stations, particularly in station-sparse and complex terrain regions like the Himalayas, compared to ERA5 and MIMIC-TPW2.
• The dataset maintained stable accuracy across different years (2016-2022) and weather conditions (all-weather, clear-sky, cloudy-sky), with all-weather RMSE ranging from 3.37 to 4.24 mm and correlation coefficients from 0.91 to 0.92.
• The fused TPW provides a 12.5-fold improvement in spatial resolution compared to ERA5, enabling a finer depiction of water vapor transport processes and localized moisture enhancement over complex terrain (e.g., Yarlung Zsangbo Grand Canyon).
• The new fusion algorithm effectively reduced abnormal TPW overestimations observed in lake areas (e.g., Qinghai Lake) in MIMIC-TPW2.

Contributions

• Developed the first high spatiotemporal resolution multi-source remote sensing TPW fusion framework for the Tibetan Plateau based entirely on satellite observations, integrating all-weather microwave and high-resolution geostationary satellite data.
• Introduced a satellite-specific bias correction method using Himawari-8/9 TPW as a high-accuracy reference to address systematic discrepancies among microwave sensors.
• Proposed a novel adaptive correction scheme for high-resolution TPW under cloudy conditions, significantly improving accuracy and spatial continuity where clear-sky models typically fail.
• Generated and publicly released a unique all-weather, hourly, 0.02° spatial resolution TPW dataset for the Tibetan Plateau (2016-2022), offering superior spatial resolution, spatiotemporal continuity, and accuracy compared to existing products like MIMIC-TPW2 and ERA5.
• The dataset enables a clearer and more detailed depiction of water vapor gradients and transport pathways in complex terrain, which is crucial for regional atmospheric and hydrological studies, including water vapor flux estimation, precipitation forecasting, and extreme weather event analysis.

Funding

• National Natural Science Foundation of China (grant nos. 42025504, U2442214)
• National Key Research and Development Program of China (grant no. 2023YFB3907701)
• Second Tibetan Plateau Scientific Expedition and Research Program (grant no. 2019QZKK0206)
• Project “Theory, methods, and experimental validation of electromagnetic and geosphere interactions – Subtask I” (grant no. E4Z202021F)

Citation

@article{Sun2026hourly,
  author = {Sun, Qixiang and Letu, Husi and Wang, Yongqian and Zhang, Peng and Liang, Hong and Shi, Chong and Yin, Shuai and Shi, Jiancheng and Ji, Dabin},
  title = {An hourly 0.02° total precipitable water dataset for all-weather conditions over the Tibetan Plateau through the fusion of observations of geostationary and multi-source microwave satellites},
  journal = {Earth system science data},
  year = {2026},
  doi = {10.5194/essd-18-371-2026},
  url = {https://doi.org/10.5194/essd-18-371-2026}
}