Combination of swarm, Jason-3, and GNSS observations to construct a new modeling of global ionospheric maps

Abstract

By increasing various space geodetic observation techniques with different orbital altitudes the capability for ionosphere layer monitoring is enhanced. In this paper, in order to increase the accuracy and reliability of Global Ionosphere Maps (GIMs) in regions with no or sparse GNSS data coverage, the observational data derived from Swarm, consisting of three Low Earth Orbiter (LEO) satellite constellation missions and a satellite altimetry mission, was integrated with the Global Positioning System (GPS) observations. The vertical total electron content (VTEC) of GPS and Swarm LEO satellites was obtained by employing the modified single-layer mapping function (MSLM) on the slant total electron content (STEC) and then the VTECs derived from three techniques were represented by the Spherical Harmonics (SH) expansion function up to the degree and order 15 in a solar-geomagnetic frame. In regard to the combination of different observation techniques, the systematic biases between different data sources are required to be considered. Here, the systematic biases of Swarm and Jason-3 satellites were expanded in terms of SH function and regarded as unknown parameters for estimation. Moreover, in order to consider different accuracy levels of ionospheric data groups, the Helmert Variance Component Estimation (H-VCE) was used to determine appropriate relative weights of observation groups. The two-dimensional combined models were constructed during 10 days in the period of DOY 271–275, 2017 and DOY 003–007, 2018 by considering different high and low Kp-indexes from 1 to 7 values. The obtained results showed that after adding the Swarm and Jason-3 data to GPS observations, the reduction of mean standard deviation (STD) maps was in the range of about 46–56% in 10 days. The combined method improved the reliability and precision of GIMs in oceanic regions significantly. However, the impact of VTECs from Swarm satellites was higher than that of VTECs from satellite altimetry due to the appropriate data sampling and coverage of Swarm mission. The results are also validated against the VTECs derived from 5 IGS stations (that are not included in the modeling). Our results indicate that the Combined method's mean Root Mean Square Error (RMSE) compared to GNSS dual-frequency measurements were less than 1.6 TECU with an improvement of about 9.56–25.80% respect to the mean RMSE of GIMs constructed by only GPS measurements.