Linear time series modeling of GPS-derived TEC observations over the Indo-Thailand region

Abstract

This paper proposes a linear time series model to represent the climatology of the ionosphere and to investigate the characteristics of hourly averaged total electron content (TEC). The GPS–TEC observation data at the Bengaluru international global navigation satellite system (GNSS) service (IGS) station (geographic $$13.02^{\circ }\hbox {N}$$ , $$77.57^{\circ }\hbox {E}$$ ; geomagnetic latitude $$4.4^{\circ }\hbox {N}$$ ) have been utilized for processing the TEC data during an extended period (2009–2016) in the $$24{\mathrm{th}}$$ solar cycle. Solar flux F10.7p index, geomagnetic Ap index, and periodic oscillation factors have been considered to construct a linear TEC model. It is evident from the results that solar activity effect on TEC is high. It reaches the maximum value ( $$\sim $$ 40 TECU) during the high solar activity (HSA) year (2014) and minimum value ( $$\sim $$ 15 TECU) during the low solar activity (LSA) year (2009). The larger magnitudes of semiannual variations are observed during the HSA periods. The geomagnetic effect on TEC is relatively low, with the highest being $$\sim $$ 4 TECU (March 2015). The magnitude of periodic variations can be seen more significantly during HSA periods (2013–2015) and less during LSA periods (2009–2011). The correlation coefficient of 0.89 between the observations and model-based estimations has been found. The RMSE between the observed TEC and model TEC values is 4.0 TECU (linear model) and 4.21 TECU (IRI2016 Model). Further, the linear TEC model has been validated at different latitudes over the northern low-latitude region. The solar component (F10.7p index) value decreases with an increase in latitude. The magnitudes of the periodic component become less significant with the increase in latitude. The influence of geomagnetic component becomes less significant at Lucknow GNSS station $$(26.76^{\circ }\hbox {N}, 80.88^{\circ }\hbox {E})$$ when compared to other GNSS stations. The hourly averaged TEC values have been considered and ionospheric features are well recovered with linear TEC model.