Search

The small-scale electron density irregularities in the ionosphere have a significant impact on the interruptions of Global Navigation Satellite System (GNSS) navigation and the accuracy of GNSS positioning techniques. The sporadic ionospheric E (Es) layer significantly contributes to the transient interruptions of signals (loss of lock) for GNSS tracking loops. These effects on the GNSS radio occultation (RO) signals can be used to derive the global location and intensity of Es layers as a complement to ground-based observations. Here we conduct statistical analyses of the intensity of Es layers, based on the scintillation index S4max from the FORMOSAT-3/COSMIC during the period 2006–2014. In comparison with simultaneous observations from an ionosonde network of five low-to-middle latitude ionosondes, the S4max indices from COSMIC, especially the small values, are linearly related to the critical frequency of Es layers (foEs). An accumulated period of less than 1 h is required to derive the short-term variations in real-time ionospheric Es layers. A total of 30.22%, 69.57% and 98.13% coincident hourly foEs values have a relative difference less than 10%, 30% and 100%. Overall, the GNSS RO measurements have the potential to provide accurate hourly observations of Es layers. Observations with S4max oEs

We developed a parameterized ionospheric electron density model based on the IRI-2012 model by spherical harmonic expansions in the horizontal and empirical orthogonal functions in the vertical. Then, after assimilating the monthly multisource total electron content (TEC) data from ground-based GPS, LEO radio occultation (RO), and the oceanic altimeter during magnetically quiet time into the model, we reanalyzed the monthly global ionospheric electron density TEC and other key parameters such as foF2 and NmF2. Both the reanalyzed and IRI-2012 model results were compared to the TEC measurements, the monthly median foF2 in a middle-latitude ionosonde station, and the global TEC map from CODE. The comparisons showed that both the reanalyzed and IRI results are consistent with those observations and the reanalyzed results perform better than the IRI model. Furthermore, the reanalyzed results are also consistent with the retrieved maps of HmF2, NmF2, and TEC from COSMIC RO observations. In summary, our method can reanalyze the global TEC and electron density using multisource TEC data assimilated into our model and improve the performance of IRI model.

The mapping function is commonly used to convert slant to vertical total electron content (TEC) based on the assumption that the ionospheric electrons concentrate in a layer. The height of the layer is called ionospheric effective height (IEH) or shell height. The mapping function and IEH are generally well understood for ground-based global navigation satellite system (GNSS) observations, but they are rarely studied for the low earth orbit (LEO) satellite-based TEC conversion. This study is to examine the applicability of three mapping functions for LEO-based GNSS observations. Two IEH calculating methods, namely the centroid method based on the definition of the centroid and the integral method based on one half of the total integral, are discussed. It is found that the IEHs increase linearly with the orbit altitudes ranging from 400 to 1400 km. Model simulations are used to compare the vertical TEC converted by these mapping functions and the vertical TEC directly calculated by the model. Our results illustrate that the F&K (Foelsche and Kirchengast) geometric mapping function together with the IEH from the centroid method is more suitable for the LEO-based TEC conversion, though the thin layer model along with the IEH of the integral method is more appropriate for the ground-based vertical TEC retrieval.

The global positioning system (GPS) differential code biases (DCB) provided by the International GNSS Service (IGS) show solar-cycle-like variation during 2002---2013. This study is to examine whether this variation of the GPS DCBs is associated with ionospheric variability. The GPS observations from low earth orbit (LEO) satellites including CHAMP, GRACE and Jason-1 are used to address this issue. The GPS DCBs estimated from the LEO-based observations at different orbit altitudes show a similar tendency as the IGS DCBs. However, this solar-cycle-like dependency is eliminated when the DCBs of 13 continuously operating GPS satellites are constrained to zero-mean. Our results thus revealed that ionospheric variation is not responsible for the long-term variation of the GPS DCBs. Instead, it is attributed to the GPS satellite replacement with different satellite types and the zero-mean condition imposed on all satellite DCBs.

Radio occultation (RO) has been proven to be a powerful technique for ionospheric electron density profile (EDP) retrieval. The Abel inversion currently used in RO EDP retrieval has degraded performance in regions with large horizontal gradients because of an assumption of spherical symmetry as indicated by many studies. Some alternative methods have been proposed in the past; the global ionospheric map (GIM)-aided Abel inversion is most frequently studied. Since the number of RO observations will likely increase rapidly in the near future, it is worthwhile to continue to improve retrieval method. In this study, both the simulations and the real data test have been done to evaluate the GIM-aided Abel inversion method. It is found that the GIM-aided Abel inversion can significantly improve upon the standard Abel inversion in either the F or the E region if an accurate GIM is available. However, the current IGS GIM does not appear accurate enough to improve retrieval results significantly, because of the spherical symmetry assumption and sparse global navigation satellite system (GNSS) stations used in its creation. Generating accurate GIM based on dense GNSS network to aid the Abel inversion might be an alternative method.

We analyzed the effect of the Abel inversion on the wave number 4 (WN4) structure from the GPS radio occultation (RO)---measured electron densities by using the FORMOSAT-3/COSMIC (F-3/C) observations under the equinox condition. The Abel-retrieved electron density from both the F-3/C observations and the simulated results by an empirical model with an imposed WN4 structure in the F layer are investigated. It is found that the Abel inversion can reproduce the real WN4 structure well in the F2 layer. However, it will result in pseudo and reversed-phase WN4 structure in the lower altitude (F1 and E layers). Quantitatively, relative ±15% WN4 signature in the F2 layer can produce ±40% artificial WN4 in the E and F1 layers. Analysis on the F-3/C data shows about ±15% WN4 signature in the F2 layer and ±50% WN4 with reversed-phase in the E and F1 layers. The F-3/C-observed WN4 structure in the E and F1 layers might be the combinations of the real WN4 signature and the artificial effects of Abel retrieval.

On the basis of previous work, we develop a middle and low latitude theoretical ionospheric model in this paper, named Theoretical Ionospheric Model of the Earth in the Institute of Geology and Geophysics, Chinese Academy of Sciences (TIME-IGGCAS). TIME-IGGCAS solves the equations of mass continuity, motion and energy of electron and ions self-consistently and uses an eccentric dipole field approximation to the Earth’s magnetic field. We combine the Eulerian and Lagrangian approaches in the model and take account of the plasma E×B drift velocity. Calculation results reveal that the model is steady and credible and can reproduce most large-scale features of ionosphere. By using TIME-IGGCAS, we carried out an observation system data assimilation experiment. Assimilation results show that the E×B drift velocity can be accurately estimated by ingesting the observed foF2 and hmF2 into the model applying nonlinear least-square fit method. We suggest that this work is of great significance in the development of ionospheric data assimilation model to give better nowcast and forecast of ionosphere.

Daytime half-hourly values of the critical frequency of the ionospheric E-layer, f oE, obtained at Wuhan Ionospheric Observatory (geographic 114.4°E, 30.6°N; 45.2°dip), China, during the whole interval of 1957–1991 and 1999–2004 have been used to develop an empirical model. The model, including variations with local time, day number and solar cycle, is in agreement with the observations. A comparison between our model and IRI and Titheridge’s model has also been made. Statistically, our model gives a better performance than IRI and Titheridge’s model because data set is obtained with our own station. Both the IRI and Titheridge’s model overestimate f oE especially in May to September months. Combing with past investigations, we suggest that overestimation of ionospheric parameters by IRI may be a common feature in East Asia. This result is very helpful for both the correction of IRI in East Asia and the development of Chinese Reference Ionosphere (CRI) model.