Comparative analysis of single station-based and network-based VTEC modeling approaches in Nigeria using orthogonal transformation

Abstract

Vertical Total Electron Content (TEC) plays a crucial role in ionospheric dynamics, influencing radio signal propagation with wide-ranging applications in satellite communication, navigation, and space weather forecasting. Despite the importance of accurate TEC estimation, limitations in existing models, such as the Global Ionospheric Maps (GIM) provided by the International GNSS Service (IGS) and the International Reference Ionosphere (IRI), hinder precision in regions with sparse receiver networks, notably in Nigeria. To overcome these challenges, we attempt to develop and compare the performance of two innovative VTEC estimation approaches utilizing spherical harmonic expansion and orthogonal transformation. GNSS observation data in RINEX format sampled at 30 s intervals from nine multi-GNSS receivers in the Nigerian Geodetic Network was utilized for this study. To ensure data quality, we performed various preprocessing steps using both the Melbourne-Wubbena linear and the geometry-free linear combinations. Validation against established GIM models (IGS, CODE, and IRI-2020) demonstrates the effectiveness of both approaches. The single-receiver approach exhibits notable agreement with standard models, presenting a standard deviation range of 2.882 to 7.362 TECU and correlation coefficients ranging from 0.898 to 0.980 at various independent stations. The network-based model displays similar reliability, exhibiting standard deviations within 2.8 to 6.5 TECU and correlation coefficients exceeding 0.92. Notably, the single-receiver model provides a temporal resolution of 1 h, while the network-based model offers a higher resolution of 10 min. Both models perform well during periods of geomagnetic activity and were able to capture post-sunset VTEC enhancements. In addition, both approaches successfully estimated Differential Code Biases (DCBs) for GPS and GLONASS, aligning closely with CODE estimates. Our results suggest that the network-based model, is more suited for capturing rapid changes, while the single-receiver model is advised for applications requiring moderate temporal resolution, and may be a good choice for VTEC modeling, particularly in areas with sparse network coverage.

Keywords

• GNSS
• Orthogonal transformations
• Spherical harmonics function
• TEC
• Differential code biases
• Single station model