The final publication is available at Springer via http://dx.doi.org/10.1007/s00190-016-0977-7

By utilizing the numerical technique of principal component analysis (PCA), this work analyses temporal and spatial variations of the ionosphere under various solar conditions during the period 1999–2013. Applying the PCA technique to the time series of the global ionospheric total electron content (TEC) maps provides an efficient method for analyzing the main ionospheric variability on a global scale that is able to decompose periodic variations (e.g., annual and semiannual oscillations) while retaining the asymmetry in the temporal and spatial domains (e.g., seasonal and equator anomalies). The TEC series of different local times are processed separately at two time scales: (1) the whole 15 years of the period of study and (2) the individual years. In contrast with previous studies, the analysis of the dataset of the 15 years shows that dawn (e.g., LT4–6) and late morning (LT10–12) are the more remarkable characteristic times for ionospheric variability. This study also reveals a cyclic trend of the variability with respect to local times. The first two modes, which contain 80–90% of the total variance, represent spatial distributions and temporal variations with respect to the different stages of the solar cycle and local times. Annual and semiannual variations are demodulated from the first two modes, and the results show that these variations evidently have distinct trends for daytime and nighttime. An exception is that, under active solar conditions, extremely strong solar irradiance during the daytime has a residual effect on the variability of the nighttime.

Peer Reviewed