Ionosphere/Plasmasphere sounding with ground and spacebased GNSS observations Guillermo GonzálezCasado, J. Miguel Juan, Jaume Sanz and Yixie Shao Research Group of Astronomy and Geomatics, Technical University of Catalonia (gAGE/UPC) Barcelona, Spain guillermo.gonzalez@upc.edu ! I. INTRODUCTION Usually, in Global Navigation Satellite System (GNSS) applications, the ionosphere is modeled as a single layer located at a fixed altitude representative of that region [1]. However, this approach disregards the contribution to slant total electron content measurements from the plasmasphere, a region extending toward heights of several thousands of kilometers. Previous studies have shown that this contribution is not negligible, especially during the night [2]. Hence, the characterization of the separated electron content from the ionosphere and the plasmasphere is of great importance to achieve a precise modeling of the GNSS signal delays. Recently ([3], [4]), a method that selfconsistently combines radio occultations (ROs) from the COSMIC/FORMOSAT3 (CF3) constellation and global ionospheric maps (GIMs) [5] from the International GNNS Service (IGS) has been developed to achieve a reliable determination of the electron contents in the ionospheric and plasmaspheric regions separately. In this study, we present preliminary results showing that the method can be used to derive some general trends characterizing the ionosphereplasmasphere interplay, which can be reproduced by means of a simple parametric model. II. DATA SAMPLE AND METHOD The details of the methodology applied in this study can be found in [3] and [4], we only summarize here the details required to understand how the ionospheric and plasmaspheric electron contents are calculated and the data required to implement the method. First, the improved Abel transform inversion ([3], [6]) is used to retrieve electron density profiles from a data sample of RO measurements taken by the CF3 constellation. This improved retrieval technique takes into account the horizontal gradients in the electron density by using externally provided vertical total electron content (VTEC) values. RO observations sampled altitudes from 200 to 700 km at least. The data sampled used in this study amounts to more than 70,000 ROs covering a period of one year, 2007, that was part of a particularly long and deep solar minimum, lasting over 4 years (from 2007 to 2010). On the other hand, VTEC values for the time intervals corresponding to different RO measurements were collected from the GIMs provided by IGS and subsequently interpolated to the different geographical locations covered by each RO in the sample. The data sample provides a reasonably dense coverage of all geographic locations around the globe. After deriving the altitude variation of the electron density from every RO, the topsideelectron density profile (altitude higher than 400 km typically) was modeled by means of the simplified topside ionosphere plus protonosphere (STIP) model [3]. The STIP model fit was used to extract the contribution from the bottomside plasmaspheric electron density to the total electron density profile. After the bottomside plasmaspheric electron density was separated, the remaining electron density profile was integrated to obtain the ionospheric electron content, ion . Finally, after having calculated ion , the plasmaspheric electron content, pl , is calculated as the difference [4]: pl = – ion (1) where is the vertical total electron content at the longitude and latitude associated to the RO and having been previously used to retrieve the electron density profile. III. IONOSPHEREPLASMASPHERE INTERPLAY The relationship between pl and ion is of great interest to characterize the coupling and interplay of the ionosphere plasmasphere system. This relationship has been analyzed using the methodology described in the previous section. In