Skip to Content Site Navigation Tree Accessibility



The ESA Technical Research Programme (TRP) Study “GNSS Contribution to Next Generation Global Ionospheric Monitoring” was conducted from March 2009 to January 2010. Within a study team of four partners (Hewlett-Packard GmbH, Rüsselsheim, Germany; QinetiQ Ltd, Malvern, U.K.; Technical University of Catalonia (UPC), Barcelona, Spain; German Aerospace Centre (DLR), Neustrelitz, Germany) recommendations for a new ionosphere monitoring system were identified and formulated, meeting the requirements of different types of potential users, e.g. for operational now/forecasting but on the other hand also for scientific applications. The study was performed in three steps: 1) Description of the main physical processes driving the ionosphere; overview of GNSS and non-GNSS ground- and space-based methods to monitor the ionosphere. 2) Description of different methods within and outside the study team to model ionospheric Total Electron Content (TEC), their concepts and philosophy; selection of reference models; definition and establishment of a test dataset to validate the different methods. 3) Performance of the tests defined in step two in order to identify the strengths and weaknesses of the tested methods and, based on the outcome of these tests, recommendations for a new ionosphere model meeting the requirements of many different potential users (practically orientated, science orientated).

The results of the study are documented in a final report (Ref. [1]) and an associated technical note (Ref. [2]). A paper about the test results obtained in step three has recently been published in the AGU Radio Science magazine (Ref. [4]).


In the first part of the study the typical features and characteristics of the ionosphere were described: its layered structure, its main geographical regions, its dependency on solar activity, season and latitude. In addition, the ionization processes, the disturbed ionosphere, its close coupling with magnetosphere and thermosphere and the physical processes standing behind these phenomena were explained (Ref. [2]).

Since 1998 ESOC is active as Analysis Centre in the International GNSS Service (IGS) Ionosphere Working Group. Daily TEC maps displaying the global ionosphere with a 2-hours time resolution are routinely delivered to the IGS. In parallel to the 2-hours processing, a 1-hour processing is currently performed in an experimental phase and will replace the 2-hours processing in future. Also test runs with 15-minutes time resolution were successful. For details about ESOC’s ionosphere processing see Ionosphere Webpage. The establishment of 3D reconstructions of the ionosphere will be a very important task of future ESOC activities see IONMON Webpage, and this aspect played also a major role in this study.


There exist different global and regional GNSS networks allowing to monitor the ionosphere from ground through TEC observables derived from dual-frequency GNSS data:

  • IGS: about 400 sites, data availability with 1 day, 1 hour and 15 minutes latency, Figure 1 top.
  • EUREF: about 220 sites (some at the same time IGS sites), 1 day, 1 hour and 15 minutes latency.
  • CORS: Continental US and some in the Pacific, Caribbean, Middle East, 1-2 hours latency.
  • GEONET: more than 1200 sites in Japan, about 6 hours latency.

In addition: DORIS, currently about 60 sites, 2 days latency if interest is indicated.



Figure 1: The global IGS GPS Network (top); ways of ionosphere monitoring with GNSS on LEOs (bottom, DLR Neustrelitz) 

Space based GNSS can be used to:

  • Derive radio occultation electron density profiles from LEO altitude down to the bottom of the ionosphere, (1) in Figure 1 bottom.
  • Record TEC data from the LEO up to the GNSS spacecraft, (2) in Figure 1 bottom.


  • About 1000 GNSS ground sites and about 90 GNSS spacecraft.
  • Enhanced global coverage.
  • More frequencies.


 Figure 2: Receiver chain to follow a LEO beacon pass (Ref. [2])

In addition, there exist several non-GNSS methods to monitor the ionosphere:

  • TEC from LEO beacons transmitting phase coherent signals at 150 and 400 MHz, about a dozen satellites. Receiver chains along a beacon spacecraft pass are well suited for ionospheric tomography, Figure 2.
  • Vertical TEC (vTEC) from dual-frequency altimeters, e.g. JASON, provide vTEC values over the oceans where no GNSS sites can be placed.
  • Ground-based ionosondes, data access to about 30 ionosondes worldwide, Figures 3 & 4.

The working principle of an ionosonde is the following: Pulses are transmitted over a range of frequencies into the ionosphere and the signal return times are recorded (whereby the pulse frequency must be below the ionosphere’s critical frequency, otherwise the pulse signal would not be reflected by the ionosphere). With this information, an ionogram can be build up, showing a plot of pulse time of flight (or virtual height) against radio frequency, Figure 3. Multiple returns are possible at each transmission frequency due to the multi-layer structure of the ionosphere and magneto-ionic splitting. The annotations in Figure 3 indicate some of the key ionospheric parameters that are derived from the ionosonde measurement. Frequencies can be transformed into equivalent electron densities.

 Figure 3: Example ionogram (from Galkin, 2009, personal communication, see Ref. [2])

As an example of an ionosonde network, Figure 4 shows the digisonde network.

 Figure 4: Digisonde station map (


For navigation applications, the following major ionospheric features must be captured with sufficient temporal and spatial resolution:

  • The major regions of the ionosphere: equatorial anomaly, midlatitude trough, auroral oval, polar regions.
  • The typical variations of ionized layers: diurnal, seasonal, with solar cycle, latitudinal.
  • The ionosphere’s layered structure including the plasmasphere.
  • Ionospheric disturbances: storms, Travelling Ionospheric Disturbances (LSTIDs & MSTIDs), scintillations.
  • The ionosphere’s coupling with other geo-spheres: namely magnetosphere, thermosphere, but also lithosphere, hydrosphere and atmosphere.

In turn, GNSS systems could be used for:

  • Solar flare monitoring.
  • TIDs monitoring.
  • Investigations into the so called 2nd and higher order ionospheric terms.


In order to identify optimal modelling methods for a New European Ionosphere Monitoring System, the different ionosphere modelling techniques of the study team were validated against an agreed test dataset and against NeQuick (e.g. Ref. [5]), Figure 5, and the IGS TEC Maps, which served as 3D and 2D reference.

Figure 5: Global TEC map based on NeQuick model computations for the 8 February 2000

The study team models are (details in Ref. [1]):

  • Electron Density Assimilative Model (EDAM), QinetiQ – different types of ionospheric observation data are assimilated into a background model. Currently, IRI2007 (Ref. [3]) is used as background.
  • IONosphere MONintoring Facility (IONMON) Version 2, ESOC – Closed function approach describing the ionospheric structures by vertical profiles combined with horizontal surface functions, no background, TEC and electron density data are processed.
  • TOMographic IONosphere model (TOMION), UPC – The ionosphere is represented by two or more layers of voxels. In each voxel, electron density is assumed to be constant, no background, fed with different types of ionospheric observation data.
  • Neustrelitz TEC Model (NTCM), DLR – This is a family of empirical TEC models developed for European and Polar cap regions, and recently also for global TEC. The TEC reconstructions used in this study were obtained by assimilating ground based TEC measurements into the European TEC model NTCM-EU.

The study team models and NeQuick and the IGS TEC Maps were validated against (details in Ref. [1]):

  • Differential slant TEC (dSTEC) data.
  • Vertical TEC (vTEC) from TOPEX and JASON.
  • Ionosonde data (only EDAM, TOMION and NeQuick).
  • Electron density measured along the CHAMP orbit (only EDAM, TOMION and NeQuick).

Two test periods: May 2002, solar maximum, and December 2006, solar minimum.

The models performance can in brief be summarized as follows (details in Ref. [1]):

  • Generally the models display similar performance, having RMS values between 1.5 – 3.5 TECU for all stations (apart from some isolated equatorial sites).
  • IONMON Version 2 showed a worse performance (model is still experimental, and could not be properly evaluated during the study, see also IONMON Webpage).

A detailed presentation of the dSTEC test results was recently published in Ref. [4].


A new system should combine the advantages of different methods. Important appears to be a background model to create a median ionosphere. This median ionosphere will then be upgraded by an assimilation technique with actual observation data, where available. The new system should describe the ionosphere in 3D and inhere to a certain extent some physics and allow for future extensions into the direction of a real physics-based model. The new system should be able to run in real-time and allow for ionospheric predictions.

The new system should be suitable for the following two major groups of potential users:

1)      Practically oriented users: GNSS users in general, oil industry, airlines, transport companies, …, who are primarily interested in getting high-precision ionospheric corrections for their GNSS data.

2)      Scientific users: In order to earn more knowledge about the physics of the ionosphere, its coupling with the magnetosphere and the Sun-Earth system, the availability of a real physics-based first-principles ionosphere model would be desirable.

The new system should be able to process all observation data types listed in Sections 3 & 4 and new sources of observation data types that may arise in the future.


In the ESA TRP Study “GNSS Contribution to Next Generation Global Ionospheric Monitoring”, based on an analysis of current ionosphere modelling and observation techniques and on tests and validations of existing ionosphere modelling methods, recommendations were formulated on how a future ionosphere monitoring system could be realized. Principally, the development of a New European Ionosphere Monitoring System should follow two strategic lines:

I)       Development of a pragmatic solution for near-real-time data provision, based on current near-real-time GNSS measurements.

II)     Preparation of future-oriented physics-based modelling techniques.

Both activities should be established and supported by ESA and European Commission simultaneously.


[1] DOPS-SYS-RP-5001-OPS-GN: Recommendations for a New European Ionosphere Monitoring System, Iss. 1/0, 20/01/2010.

[2]  DOPS-SYS-TN-0017-OPS-GN: Analysis of the State of the Art Ionosphere Modelling and Observation Techniques, Iss. 1/0, 26/06/2009.

[3] Bilitza, D., and B. W. Reinisch (2008), International Reference Ionosphere 2007: Improvements and new parameters, J. Adv. Space Res., 42(4), 599-609

[4]  Feltens, J., M. Angling, N. Jackson-Booth, N. Jakowski, M. Hoque, M. Hernández-Pajares, A. Aragón-Angel, R. Orús and R. Zandbergen (2011): “Comparative testing of four ionospheric models driven with GPS measurements”, Radio Science, Vol. 46, RS0D12, doi: 10.1029/2010RS004584, 2011.

[5] Nava B., P. Coisson, and S. M. Radicella (2008), A new version of the NeQuick ionosphere electron density model, JASTP, 1856-1862.