Specifications of the F-region variations for quiet and disturbed conditions

Prediction of global positioning system total electron content using Neural Networks over South Africa

“…This means that there was completely no representation of all parameters that affect TEC during this period and the NN experienced a difficulty in learning the trend of the data and making an informed generalization. Poor prediction in September may also be due to the fact that TEC varies strongly during the equinoxes (Kouris et al, 1999). The relative corrections in Table 2 give the average percentage level of the NN's prediction accuracy during equinoxes and solstices.…”

Pre-storm electron density enhancements at middle latitudes

“…As for (b), Kouris et al (1999) analysed the foF2 variability at six European stations for the period 1985-1995. They found for quiet conditions that at European low middle latitudes (e.g., Rome), foF2 was within 710% of monthly median for more than 90% of time, and even at Chilton/Slough (magnetic latitude 54.11N) foF2 was within 715% of monthly median for more than 90% of time.…”

Time-dependent prediction degredation assessment of neural-networks-based TEC forecasting models

“…It is also possible that the cause of this variation in the prediction error is linked to the approach of training the neural network on TEC from one solar cycle and predicting on a consecutive one (Feichter and Leitinger, 1997). On the other hand, this could be due to the fact that both TEC and foF2 exhibit stronger variability during the equinoxes (Kouris et al, 1999). It is also known that the foF2 and TEC variabilities differ, to some extent at least due to the fact that the topside ionosphere and influences from the plasmasphere above the F2-region are important contributors to the TEC (Ciraolo and Spalla, 1997).…”