The Global Ionosphere Models

We found 5 different ionosphere models on the CDDIS site. We have looked quickly at all 5, but only display 2 in the attached figures to avoid clutter. It is clear that derivation of the ionosphere from GPS data is still not a precise art. The scatter in the results from the various models is on the order of 20-50%. The differences do not remain constant with time, although the ranking does typically last through the experiment -- the same model will always give the highest result etc. We are not especially familiar with the modeling methods, but guess that the differences are the result of different assumed vertical profiles and perhaps different allotment of delay to instrumental effects and ionosphere. Also, the receiver networks used are not always the same. We presume that the modeling methods will improve significantly over the next few years, so the ability to use the models to correct VLBI data should improve. One important improvement that is being studied is to add a third dimension to the models.

The following brief descriptions of the models examined are based on, or extracted from, the header information in the files.

• JPL: Global Ionospheric Maps (GIM) are generated on an hourly and daily basis at JPL using data from up to 100 GPS sites of the IGS and others institutions. The vertical TEC is modeled in a solar-geomagnetic reference frame using bi-cubic splines on a spherical grid. A Kalman filter is used to solve simultaneously for instrumental biases and VTEC on the grid (as stochastic parameters). Contact Address: gpsiono@cobra.jpl.nasa.gov .

  There were 97 stations contributing to the data set we used.

• CODE: The global ionosphere maps are generated on a daily basis by the Center for Orbit Determination in Europe (CODE), University of Berne, Switzerland. The TEC is modeled with a spherical harmonic expansion up to degree 12 and order 8 referring to a solar-geomagnetic reference frame. The 12 2-hour sets of 149 ionosphere parameters per day are derived from GPS data of the global IGS (International GPS Service) network. Contact address: stefan.schaer@aiub.unibe.ch.

  There were 80 stations contributing to the data set we used.

• EMRG: The NRCan ionosphere maps are generated daily by the Geodetic Survey Division (GSD) of Natural Resources Canada (NRCan). The grid point values are mean VTECs estimated in sun-fixed reference frame. Contact address: pheroux@nrcan.gc.ca.

  There were 43 stations contributing to the data set we used.

• ESAG: The header information for this file is a bit confusing. But there is more information at http://nng.esoc.esa.de/gps/ionmon.html. The model is from the ESOC Ionosphere Monitoring Facility (IONMON). That facility can fit for the height of the ionosphere using an assumed profile. But the models archived at CDDIS appear to be for a fixed height. This model tends to give the lowest TEC values.

  There were 49 stations contributing to the data set we used.

• UPCG: Modeled independently for each station with a tomographic model: 72x9x2 cells in local time, latitude and height The height boundaries are: 59-739-1419 km. The estimates have been interpolated with radial basis functions (spatial smooth: 2 pixels) See for instance: Hernandez-Pajares, Juan, Sanz and Sole, JGR, Vol.103, N.A9, 20789-20796, 1998 Contact e-mail: manuel@mat.upc.es Plots at: http://maite152.upc.es/ionex/gAGE_dip/gAGE_dip.html

  There were 83 stations contributing to the data set we used.

Note that the SATLOC models are based on approximately 15 receivers, which is much smaller than the above models, but they only attempt to cover the continental US rather than the whole globe.