Ionospheric TEC from the Turkish Permanent GNSS Network (TPGN) and comparison with ARMA and IRI models

Abstract

The present study investigates the ionospheric Total Electron Content (TEC) variations in the lower mid-latitude Turkish region from the Turkish Permanent GNSS Network (TPGN) and International GNSS Services (IGS) observations during the year 2016. The corresponding vertical TEC (VTEC) predicted by Auto Regressive Moving Average (ARMA) and International Reference Ionosphere 2016 (IRI-2016) models are evaluated to realize their effectiveness over the region. The spatial, diurnal and seasonal behavior of VTEC and the relative VTEC variations are modeled with Ordinary Least Square Estimator (OLSE). The spatial behavior of modeled result during March equinox and June solstice indicates an inverse relationship of VTEC with the longitude across the region. On the other hand, the VTEC variation during September equinox and December solstice including March equinox and June solstice are decreasing with increase in latitude. The GNSS observed and modeled diurnal variation of the VTEC show that the VTEC slowly increases with dawn, attains a broader duration of peak around 09.00 to 12.00 UT, and thereafter decreases gradually reaching minimum around 21.00 UT. The seasonal variation of VTEC shows an annual mode, maxima in equinox and minima in solstice. The average value of VTEC during the June solstice is with slightly higher value than the March equinox though variations during the latter season is more. Moreover, the study shows minimum average value during December solstice compared to June solstice at all stations. The comparative analysis demonstrates the prediction errors by OLSE, ARMA and IRI remaining between 0.23 to 1.17%, 2.40 to 4.03% and 24.82 to 25.79% respectively. Also, the observed VTEC seasonal variation has good agreement with OLSE and ARMA models whereas IRI-VTEC often underestimated the observed value at each location. Hence, the deviations of IRI estimated VTEC compared to ARMA and OLSE models claim further improvements in IRI model over the Turkish region. Although IRI estimations are well accepted over the mid-latitudes but the performance over the lower mid-latitudes is not satisfactory and needs further improvement. The long-term TEC data from the TPGN network can be incorporated in the IRI under laying database with appropriate calibration for further improvement of estimation accuracy over the region.