Reconstruction of ionospheric TEC maps using singular value decomposition in both space and time
Advances in Space Research, vol.78, no.1, pp.299-315, 2026 (SCI-Expanded, Scopus)
- Publication Type: Article / Article
- Volume: 78 Issue: 1
- Publication Date: 2026
- Doi Number: 10.1016/j.asr.2025.11.087
- Journal Name: Advances in Space Research
- Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus, Artic & Antarctic Regions, Compendex, INSPEC, MEDLINE, Academic Search Ultimate (EBSCO), Engineering Source (EBSCO)
- Page Numbers: pp.299-315
- Keywords: Image reconstruction, Ionosphere, PDF estimation, Singular Value Decomposition (SVD), Total Electron Content (TEC)
- Hacettepe University Affiliated: Yes
Abstract
The ionosphere affects the performance of shortwave and satellite communications, and space-based navigation and positioning systems. Understanding the ionospheric structure is crucial for enhancing the performance of these systems. Total Electron Content (TEC) is a key parameter for characterizing the ionospheric structure. TEC can typically be estimated at only a limited number of spatial points for most parts of the globe. Consequently, accurate, reliable, and robust TEC mapping/interpolation methods are required. In this study, a basis expansion algorithm is proposed to represent the underlying physics of TEC using singular value decomposition (SVD) for 2-D reconstruction. Extracting the significant singular values corresponding to signal subspace allows a reduced number of contributing basis vectors. Utilizing the TEC estimated from 32% of the grid, the least squares optimization based algorithm allows reconstruction in computationally efficient closed form. For the preparation of Model Matrices, interpolated Global Ionospheric Maps (GIM) can be utilized for user defined space–time resolution for the two-dimensional (2-D) reconstruction of TEC even in regions where GPS receivers are either non-existent or sparse such as over oceans and deserts. In this study, an example is provided for reconstruction of TEC for European region where Model Matrices are generated considering diverse solar activity levels, hourly-monthly trend variations, and geomagnetic indices. We provide examples of reconstruction over a wide range of geomagnetic activity from 2000 to 2024, covering almost two solar cycles. Of the 145,152 reconstructed TEC values, 99.757% demonstrate an error below 3 TECU relative to Jet Propulsion Laboratory (JPL) TEC data. The Probability Density Function (PDF) of the differences between JPL-TEC and the reconstructed maps is estimated. The estimated PDF parameters can be used to determine the optimal signal subspace for practical applications. The proposed reconstruction algorithm is suitable for near-real time estimation and prediction of TEC maps.