The Global Atmospheric Model IGGatm Based on CRA-40 and Polynomial Fitting Formulas

Beibei Yu, Wei Li, Hui Liu

Peer Reviewed

Abstract:	Aiming to improve the service performance of the BeiDou system and multi-GNSS integration under low-elevation conditions, the IGGatm model is developed based on China’s self-developed atmospheric reanalysis dataset, CRA-40. The model explicitly accounts for latitudinal variations of the tropopause as well as the vertical distribution characteristics of key meteorological parameters. By combining a non-uniform vertical grid with polynomial fitting, IGGatm provides a three-dimensional empirical representation of temperature, pressure, and water vapor pressure from the surface to the lower stratosphere, with parameterization comparable to that of GPT3-type models. Based on the modelled atmospheric field, accurate slant tropospheric delays and mapping functions are derived using a revised piecewise linear ray-tracing method. IGGatm can achieve computational efficiency comparable to traditional empirical models while delivering improved positioning accuracy and performance close to that of real-time mapping functions, providing an effective solution for tropospheric delay correction in BeiDou high-precision applications. Validation experiments using global radiosonde data and IGS station observations from 2022 demonstrate that the IGGatm model exhibits strong consistency and stability from the surface to the upper atmosphere. The estimated global mean RMS error of the equivalent hydrostatic slant delay (SHD) is 1.8 cm, corresponding to an improvement of approximately 22% over the GPT3 model and accuracy comparable to that of the discrete VMF1/VMF3 models. In positioning experiments, the global mean three-dimensional RMS values achieved by IGGatm are 1.51 cm under equal weighting and 1.49 cm under elevation-dependent weighting, showing performance comparable to VMF3 and improvements over GPT3 of about 24% and 13%, respectively. In low-latitude regions, accuracy improvements of up to 43% and 22% are observed, indicating enhanced robustness, particularly under low-elevation angles and complex terrain conditions. Overall, the IGGatm model and its accompanying mapping function algorithm provide an accurate and computationally efficient approach for tropospheric delay correction across a wide range of GNSS applications, from ground-based positioning to high-altitude platforms and LEO satellite–enhanced positioning. The proposed framework offers a precise, reliable, and fully autonomous solution for BeiDou system applications and global GNSS users.
Published in:	Proceedings of the ION 2026 Pacific PNT Meeting April 13 - 16, 2026 Hilton Waikiki Beach Honolulu, Hawaii
Pages:	11 - 32
Cite this article:	Yu, Beibei, Li, Wei, Liu, Hui, "The Global Atmospheric Model IGGatm Based on CRA-40 and Polynomial Fitting Formulas," Proceedings of the ION 2026 Pacific PNT Meeting, Honolulu, Hawaii, April 2026, pp. 11-32. https://doi.org/10.33012/2026.20569
Full Paper:	ION Members/Non-Members: 1 Download Credit Sign In