Allowance for the solar-magnetospheric connections in a problem of predicting the critical frequency of the subauroral ionosphere

Abstract

The cause-and-effect relation between variations in the critical frequency of the F 2 subauroral ionospheric layer and the key solar-magnetospheric parameters is studied. The features of the dependence of the critical frequency of the subauroral ionosphere on the value and components of the interplanetary magnetic field, the solar-wind parameters, the X-ray and ultraviolet radiation intensities, the solar zenith angle, and the intensity of precipitation of low-energy particles are established. Within the framework of this study, codes allowing one to solve the problems of predicting the critical frequency of the ionosphere by the method of artificial neural networks has been developed. The neural-network forecasting experiments reveal the characteristic times of the ionospheric response to magnetospheric disturbancs. The found optimal architecture of the neural network allows us to predict the time series of the critical frequency of the high-latitude ionosphere for the interval 0.5-3 h with an accuracy of up to 93%. The obtained linear and nonlinear dependences of the critical frequency of the ionospheric F 2 layer on the geophysical parameters can be used for developing ionospheric models required for determining the maximum usable frequency of HF radio communication.

Spatio-temporal correlation of the maximum observable frequency on midlatitude radio lines

“…This parameter is most effective in the daytime and makes it possible to improve the training quality [7]. The introduction of the mentioned additional parameter allowed the quality of training of an artificial neural network to be increased up to 97% (Fig.…”