Ionospheric data assimilation methods for geodetic applications

One of the major limiting factors in geodetic applications of the Global Positioning System (GPS) is lack of knowledge of the propagation delays imposed by the ionosphere. Single frequency, differential carrier phase measurements are limited to baselines with lengths less than the correlation size of the ionosphere (typically 10-20 km). Extending these measurements to longer distances requires accurate estimates of the slant total electron content (TEC) from a receiver to all observable GPS satellites. While dual frequency carrier phase measurements permit an ionosphere-free linear combination, accurate estimates of the double difference in integrated TEC between pairs of satellites and receivers provide an important constraint for accurate and rapid carrier phase ambiguity resolution. To achieve these accuracy requirements various approaches to the assimilation of ground-based GPS data from the CORS network and the mathematical representation of the ionospheric electron density field have been studied. The model presented uses a Kalman filter algorithm to assimilate data in various forms and an optional mapping function to alter the representation of the state vector in terms of a set of discrete radial empirical orthonormal functions (EOF's). Initial results from local networks show agreement with ambiguity-fixed double-differenced ionosphere delays of a few tenths of a TEC. The advantages of the various approaches and additional results will be discussed.