GPS satellite data calibrated for improved space weather research

For more than two decades, GPS has been quietly watching Earth’s radiation belts by measuring streams of energetic electrons. But the constellation’s satellites do not agree with one another, and those inconsistencies have made it difficult for scientists to trust a combined, long-term view of space weather. The result is a frustrating gap: valuable observations exist, yet researchers often hesitate to use them as a single, coherent dataset.

A new study tackles this problem head-on by performing the first systematic cross-calibration of energetic electron flux measurements from 25 GPS satellites. The work produces a unified record spanning two full solar cycles, turning GPS into what is essentially a multi-platform observatory for the medium-Earth-orbit environment where spacecraft are most vulnerable.

The stakes are high. In the outer radiation belt, relativistic electrons exceeding 1 MeV can intensify dramatically during geomagnetic storms, increasing risks to satellite electronics. These electrons can drive damaging internal charging and trigger electrostatic discharge, threatening both spacecraft health and mission continuity.

To harmonize measurements, researchers led by Beihang University and the Chinese Academy of Sciences used a reference satellite (NS59) selected for its long data record and strong overlap with other spacecraft. They matched observations at corresponding magnetic coordinates—specifically at the same ((Lm, B/B0)) positions—so that different satellites were effectively compared at the same physical locations in the radiation belt.

The calibration framework relies on a two-step cubic polynomial fit applied to log-transformed flux data. After an initial relationship was established, the lowest and highest 5% of points were removed to reduce the influence of outliers. A final fit then produced an improved cross-satellite calibration curve.

Performance improved substantially for test channels. For 2.0 MeV differential fluxes, root-mean-square deviation decreased by an average factor of 3.08, and correlation increased by 14%. For ≥2.0 MeV integral fluxes, RMSD improved by 1.68-fold.

Some satellites previously looked like outliers. Notably, NS48’s RMSD fell from 3.79 to 0.107 and its correlation rose from 0.255 to 0.992, despite carrying a different detector type. Meanwhile, NS74 remained problematic even after calibration, and the authors recommend excluding it from combined analyses.

The new calibrated dataset covers 2000–2020 and offers a more reliable long-term basis for radiation-belt modeling and space weather forecasting. The approach is designed to extend across all remaining energy channels and has already been applied to cross-calibrate GPS data with BeiDou and Van Allen Probes observations—potentially reshaping how scientists quantify and predict hazardous electron enhancements in near-Earth space.

Keywords

GPS, radiation belts, energetic electrons, cross-calibration, space weather forecasting, satellite anomaly detection