Rapid sensing of atmospheric water vapor with timely service of the GNSS satellite clock error

• Published in: Advances in space research Vol. 75; no. 6; pp. 4600 - 4612
• Main Authors: Li, XiaoMing, Li, HaoJun, Li, Zhicheng
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 15.03.2025
• Subjects: Parameterization estimation; Precipitable water vapor; Real time service; Satellite clock error; Precipitable water vapor; Real time service; Parameterization estimation; Satellite clock error
• ISSN: 0273-1177
• DOI: 10.1016/j.asr.2024.12.071

The real-time retrieval of atmospheric water vapor is vital for accurate weather nowcasting. However, current global navigation satellite systems (GNSS) rely on high-precision real-time service (RTS) for detecting atmospheric water vapor, which suffers from delays due to computational overhead and network latency. To tackle this, a rapid parameter estimation method for GNSS satellite clock error model coefficients is proposed. This method efficiently estimates the clock error model coefficients presented by quadratic polynomial functions, leveraging 1-second sampling epoch-differenced phase observations. The real-time GNSS satellite clock error values at any given time can be derived from the estimated quadratic polynomial coefficients, whose computation time is reduced, improving timeliness. The estimated coefficients are then used to generate timely RTS for satellite clock error, adequate for retrieving real-time zenith tropospheric delay (ZTD) and precipitable water vapor (PWV), updated every minute. Results show that the recovered 1 Hz RT satellite clock error achieves average STDs of 0.057, 0.048, and 0.056 ns for GPS, BDS-3, and Galileo respectively, meeting accuracy requirements for ZTD and PWV retrieval. In scenarios of short-term time delays, the proposed method outperforms IGS RTS, with average RMS of ZTD reaching 12.4, 13.7, and 14.2 mm for GPS, BDS-3, and Galileo, and RMS values of 2.4, 3.1, and 3.2 mm for PWV conversion. These accuracies meet high-frequency numerical weather prediction needs, enhancing rapid sensing capabilities.