Mid latitude ionospheric TEC modeling and the IRI model validation during the recent high solar activity (2013-2015)

Abstract This paper discusses the variability of the modelled Vertical Total Electron Content (VTEC) and performance of the latest versions of the International Reference Ionosphere (IRI) model in the estimation of TEC over the mid-latitude regions in the recent high solar activity (2013–2015) years. This is conducted by comparing the pattern of the variations of the VTEC obtained from five ground based Global Positioning System (GPS) receivers installed at different mid-latitude regions and the latest versions of the IRI model (IRI 2007, IRI 2012 and IRI 2016). It has been observed that the measured (GPS-derived) and modelled (IRI 2007, IRI 2012 and IRI 2016) monthly and seasonal diurnal variability of VTEC show the lowest values at around 10:00 UT (04:00 LT) and the highest values at around 20:00 UT (14:00 LT). Moreover, both the measured and modelled VTEC variations generally follow the pattern of the variation of the solar activity, showing enhancement in shifting from 2013 to 2014, and drop again in 2015, with some exceptional months. In the years 2013–2015, the highest measured and modelled seasonal arithmetic mean VTEC values are observed in the March equinox in 2014; while, the lowest measured and modelled VTEC values are observed in the December solstice and June solstice, respectively in 2015. It has also been shown that the IRI 2016 VTEC values are generally larger than those of the IRI 2007 and IRI 2012 VTEC values and tend to respond to variation of the GPS VTEC values better than others, especially in the equinoctial and June solstice months. Moreover, when compared to the IRI 2007 and IRI 2012 versions, the smallest root-mean-square deviations are observed in using the IRI 2016 version, showing that the IRI 2016 version is generally better in capturing the VTEC values with some exceptional months (especially in the December solstice months). Hence, in the December solstice months, the IRI 2007 and IRI 2012 versions are generally better in estimating the VTEC variations as compared to the IRI 2016 version. Moreover, the root-mean-square deviations obtained (either due to overestimations or underestimations) in the IRI 2007 and IRI 2012 versions are very close to each other, proving that the two versions show similar performance in TEC estimation.