The ionosphere scale height is one of the most significant ionospheric parameters, which contains information about the ion and electron temperatures and dynamics in upper ionosphere. In this paper, an empirical orthogonal function (EOF) analysis method is applied to process all the ionospheric radio occultations of GPS/COSMIC (Constellation Observing System for Meteorology, Ionosphere, and Climate) from the year 2007 to 2011 to reconstruct a global ionospheric scale height model. This monthly medium model has spatial resolution of 5° in geomagnetic latitude (−87.5° ~ 87.5°) and temporal resolution of 2 h in local time. EOF analysis preserves the characteristics of scale height quite well in the geomagnetic latitudinal, anural, seasonal, and diurnal variations. In comparison with COSMIC measurements of the year of 2012, the reconstructed model indicates a reasonable accuracy. In order to improve the topside model of International Reference Ionosphere (IRI), we attempted to adopt the scale height model in the Bent topside model by applying a scale factor q as an additional constraint. With the factor q functioning in the exponent profile of topside ionosphere, the IRI scale height should be forced equal to the precise COSMIC measurements. In this way, the IRI topside profile can be improved to get closer to the realistic density profiles. Internal quality check of this approach is carried out by comparing COSMIC realistic measurements and IRI with or without correction, respectively. In general, the initial IRI model overestimates the topside electron density to some extent, and with the correction introduced by COSMIC scale height model, the deviation of vertical total electron content (VTEC) between them is reduced. Furthermore, independent validation with Global Ionospheric Maps VTEC implies a reasonable improvement in the IRI VTEC with the topside model correction.

The Global Navigation Satellite System radio occultation technique provides materials for the validation of a mathematical description of the topside ionosphere up to the low Earth orbit (LEO) satellite altitude. An attempt to represent the topside electron density profile is using α-Chapman function with a continuously varying scale height. In this study, the Vary-Chap scale height profiles are obtained based on Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) electron density profiles from 1 January 2008 to 31 December 2013 and fitted by a shape function composed of two weighted patterns representing the ion and electron contributions of lower and higher altitudes. The associated fitting parameters a and b are analyzed to reveal their temporal and spatial features and variations along with enhancement of solar activity. The prominent dependence on latitudes, longitudes, the local time, the season, and the solar cycle shown in the parameters inspires future modeling of the Vary-Chap scale height in constructing empirical topside ionospheric models.

Increasing total electron content (TEC) measurements from the low Earth orbiting satellites to Global Positioning System satellites flourish the exploration of the ionosphere and plasmasphere for decades. This paper indicates a method that 3-D Var is applied to assimilate precise orbit determination antenna TEC measurements of Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) satellites into the background global core plasma model (GCPM). The slant TEC data archived in the COSMIC Data Analysis and Archive Center from 500 km to 20,200 km are used to reconstruct a new electron density model. This model has a temporal resolution of 2 h and spatial resolutions of 2.5°in geomagnetic latitude, 5°in longitude, 50 km in the upper ionosphere, and several hundred kilometers in the plasmasphere. Preliminary results show that the data assimilation modifies the initial GCPM forecast to be better coincident with actual COSMIC measurements in internal quality check. Furthermore, independent validation with upper ionosphere-retrieved electron density and TEC of global ionosphere maps implies a reasonable improvement in the estimation of plasmaspheric electron density after the assimilation.