A validation of Global Ionospheric Maps for the correction of InSAR data at long spatial wavelengths



Satellite radar interferometry (InSAR) allows measurement of Earth’s surface deformation over time and can detect very small displacements related to processes such as interseismic loading around tectonic faults. Over the past decade, large-scale processing of InSAR data at the continental scale has been enabled by the development of automated processing services such as FLATSIM (FormaTerre LArge-Scale Multi-Temporal Sentinel-1 InterferoMetry), led by a team from ISTerre in collaboration with the Centre National d’Études Spatiales (CNES).


Comparison of across-track (a, c, e) and along-track (b, d, f) long-wavelength phase gradient predicted by Global ionosphere maps (GIMs) with the observed ramps, for all ascending tracks in France (a, b), Tibet (c, d) and the Andes (e, f). The black line shows a sliding median of the observed ramps (black dots). The colored lines show the time-series of ramps predicted by several GIMs, using the same temporal smoothing. Shaded envelopes indicate twice the standard deviation within the sliding window (0.2 years).


Such processing makes it possible to derive deformation maps with a spatial resolution of a few hundred meters and sub-centimeter accuracy over regions typically spanning 100,000 km². However, at these spatial scales, InSAR measurements become less reliable due to the superposition of signals unrelated to actual ground motion. One important source of error arises from the propagation of radar waves through the ionosphere.

In this study, researchers from ISTerre, GéoAzur, and CNES show that these ionospheric effects can be efficiently identified and corrected using Global Ionosphere Maps (GIMs). This publicly available dataset provides estimates of the ionosphere’s Total Electron Content (TEC), which can be used to compute its contribution to InSAR measurements.



Average Total Electron Content (TEC) maps related to the three data sets used in the study. The TEC is averaged over all acquisitions during the year 2015, at the acquisition time of ascending and descending tracks. Footprints of the processed tracks are shown by white polygons. The geomagnetic equator and ±20° parallels are shown by solid and dashed red lines, respectively.
An analysis of InSAR data processed with the FLATSIM service over France, Tibet, and the Central Andes demonstrates that GIMs successfully reproduce the temporal variations of the spatially smooth component of InSAR measurements at timescales ranging from months to years. These results indicate that, at long spatial wavelengths (>100 km), ionospheric effects are non-negligible even for C-band SAR data and must be corrected to improve measurement accuracy.

The authors show that applying GIM-based corrections improves the accuracy of long-term ground motion estimates at low computational cost, making InSAR data more reliable for studying tectonic motion over large areas. This work paves the way toward independent referencing of InSAR measurements and its use as a standalone geodetic tool.


The full study is published in : AGU

Funding : CNES (projet REF-InSAR)



References :

Marconato, L., Doin, M.-P., Lovery, B., Rolland, L., & Durand, P. (2025). Long-wavelength ionospheric compensation of InSAR time-series based on Global ionosphere maps. Geophysical Research Letters, 52, e2025GL118843.

Scientific contacts :

  • L. Marconato – ISTerre, Université Grenoble Alpes, USMB, CNRS, IRD, UGE, Grenoble, France
  • M.-P. Doin – ISTerre, Université Grenoble Alpes, USMB, CNRS, IRD, UGE, Grenoble, France