Taming the Invisible: A Comparative Study of All-Force Radiation Pressure Models for Multi-GNSS

We have developed three innovative, empirically-derived all-force radiation pressure models for multi-GNSS, addressing the shortcomings of traditional models that rely on simplified spacecraft geometries and extensive metadata. The models were created using ESA and CODE orbit products with different arc lengths (5-day and 10-day). This presentation will share the findings of a comparative study of these three models, based on a full year of high-precision GNSS data processing.

The results reveal significant variations in model performance across different GNSS constellations. A 5-day ESA-based model with a comprehensive parameter set (including a Z0 term and a Galileo-specific term) showed superior performance for Galileo and GPS. In contrast, a 5-day CODE-based model performed best for BeiDou, underscoring the importance of data source quality. Interestingly, a 10-day ESA-based model demonstrated lower performance, suggesting that longer arcs do not necessarily enhance model accuracy.

Further analysis indicated that while the models effectively reduced the variability in estimated ECOM parameters, consistent non-zero mean values for some parameters—particularly for BeiDou—highlight the presence of unmodeled forces. This is further supported by the anomalous behavior of certain BeiDou satellites, emphasizing the need for dedicated modeling strategies tailored to specific constellations.

This work underscores the critical role of data source quality, arc length selection, and careful parameterization in developing accurate and reliable all-force radiation pressure models for multi-GNSS. To further enhance the models, we will investigate these factors in greater detail, explore alternative parameterizations, and develop strategies to improve performance for challenging constellations, especially BeiDou.

The models developed in this study are freely available to the scientific community to support further research and enhance the accuracy of GNSS precise orbit determination.