Estimation of Ionospheric Critical Plasma Frequencies From GNSS-TEC Measurements Using Artificial Neural Networks

This paper describes a new neural network-based approach to estimate ionospheric critical plasma frequencies (f0F2) from Global Navigation Satellite Systems (GNSS)-vertical total electron content (TEC) measurements. The motivation for this work is to provide a method that is realistic and accurate for using GNSS receivers (which are far more commonly available than ionosondes) to acquire f0F2 data. Neural networks were employed to train vertical TEC and corresponding f0F2 observations respectively obtained from closely located GNSS receivers and ionosondes in various parts of the globe. Available data from 52 pairs of ionosonde-GNSS receiver stations for the 17-year period from 2000 to 2016 were used. Results from this work indicate that the relationship between f0F2 and TEC is mostly affected by the seasons, followed by the level of solar activity, and then the local time. Geomagnetic activity was the least significant of the factors investigated. The relationship between f0F2 and TEC was also shown to exhibit spatial variation; the variation is less conspicuous for closely located stations. The results also show that there is a good correlation between the f0F2 and TEC parameters. The f0F2/TEC ratio was generally observed to be lower during enhanced ionospheric ionizations in the day time and higher during reduced ionospheric ionizations in the nights and early mornings. The analysis of errors shows that the model developed in this work (known as the NNT2F2 model) can be used to estimate the f0F2 from GNSS-TEC measurements with accuracies of less than 1 MHz. The new approach described in this paper to obtain f0F2 based on GNSS-TEC data represents an important contribution in space weather prediction.