Statistical Analysis of Global and Regional Ionospheric Total Electron Content (TEC) Using Extreme Value Distributions

Satellite-based radio communication and navigation systems rely mostly on transionospheric propagation of radio signals concerning changes by total electron content (TEC). Therefore, it is necessary to understand the changing behavior of structural variations of ionospheric TEC for high-frequency applications besides mitigating the risks associated with the space weather impacts for navigation and aviation applications. The reference thresholds identified for moderate and severe levels of threshold activity for TEC by the International Civil Aviation Organization (ICAO) are 125 and 175 TECU, respectively. In view of this, statistical analysis of long-term ground-based global navigation satellite system (GNSS) TEC data of 25 years (1997–2021) using extreme value theory (EVT) is performed for both magnetically quiet and disturbed day conditions for four different regions – India (5-45°N), Japan low (20-30°N), Japan mid-latitudes (30–50°N), and global regions to identify the extreme ionospheric TEC events that take place once in 11, 22, 44, 66, 88, and 110 years with 95% confidence intervals. In the present work, both generalized extreme value (GEV) using annual maxima and generalized Pareto distribution (GPD) using peak-over-threshold (PoT) analysis are performed. A likelihood test and PoT constraint in addition to adjustment of shape parameter illustrate that GPD works better than GEV to identify return periods of extreme events. The recommended TEC moderate and severe thresholds for India, Japan low, Japan mid-latitude, and global regions are 121, 101, 90, and 139 TECU and 131, 135, 98, and 157 TECU, respectively. The analysis would be helpful in developing risk assessment and mitigation strategies for critical (GNSS) space weather systems.