Echo of ring current storms in the ionosphere.

The ionospheric weather Vσ index maps are generated from GIM-TEC produced at the Jet Propulsion Laboratory (JPL) with the denser grids of 1 × 1° in latitude and longitude, respectively. Using sliding-window statistical analysis, moving TEC median μ for 15 preceding days with estimated variance bounds is obtained for each grid pixel of GIM-TEC. The ionosphere variability index, Vσ, is ΔTEC deviation from the median normalized by the standard deviation σ. Vσ index segmentation is introduced specifying TEC storm if an instant TEC is outside the bound of μ±2σ. Comparisons of Vσ variability of 1 × 1° GIM-TEC maps with normal GIM-TECs of 2.5 × 5° in latitude and longitude demonstrates the greater TEC storm activity with denser grids maps. The superposed epoch analysis is performed for 77 intense storms (Dst ≤ −100 nT) and 246 moderate storms (−100 < Dst ≤ −50 nT) during 2000–2017, with zero time, t 0 , at Dst storm peak. The probabilities of TEC storm occurrence both positive (Vσp index) and negative (Vσn) are evaluated for 24 h prior to t 0 and 48 h after t 0 for global scale and three specified latitude zones: North High (NH) magnetic latitudes λ > 50°, South High (SH) λ < −50°, and Equatorial Ionization Anomaly (EIA) Band between ±35° magnetic latitudes. The equinoctial positive Vσp index shows the maxima on March and September, Dst intensity peaks on April and October, and the negative Vσn peaks on May and November. The amplitude of Vσp for the intense storms exceeds that for the moderate storm by 2 times and exceeds the amplitude of Vσn by 1.5 times. Storm Enhanced Density (SED) is evident with the largest amplitude of Vσp at EIA Band exceeding Vσp for NH and SH zones by 3 times. The negative Vσn occurrence observed for selected zones during the intense storms is decreased by 5 times for the moderate storms. The normalized cross-correlation function is used to define lag (time delay) between peak of Vσp and Vσn and t 0 of Dst peak. The majority of positive Vσp indices (57.5%) are observed prior to Dst storm (lag varies from −1 to −6 h with Dst from −50 to −400 nT) so the TEC enhancement can serve as a precursor of Dst storm. The time delay for the negative Vσn is inversely related with Dst storm intensity (lag varies from 11 to 8 h with Dst from −50 to −400 nT). • Superiority of denser global TEC maps (1 × 1 deg grids) over standard IONEX GIM-TECs is demonstrated. • Hourly-seasonal hierarchy is established with the positive TEC storms preceeding Dst storm echoed by the negative TEC storm. • Storm Enhanced Density is dominant at Equatorial zone over North and South high latitude zones. • Intensity of ionosphere storms is dropped by 3–5 times from intense to moderate Dst storms. [ABSTRACT FROM AUTHOR]