Investigation of Ionospheric Electron Density Change During Two Partial Solar Eclipses and Its Comparison with Predictions of NeQuick 2 and IRI-2016 Models

Abstract

In the present study, the results obtained by incoherent scatter radar (ISR) and empirical models (NeQuick2 and IRI-2016) of the variations in mid-latitude ionospheric electron density during partial solar eclipses on March 29, 2006 and March 20, 2015 over Kharkiv (49.60° N, 36.30° E, Ukraine) are presented and the prediction performances of the models during these partial solar eclipses are compared. The electron density (Ne) values are obtained for seven (190, 210, 240, 290, 340, 410, and 490 km) different altitudes. The percent deviation of Ne values, statistical correlation coefficient and root mean square error (RMSE) during solar eclipses are applied to make this investigation. According to the percent deviation of Ne values measured by ISR, a decrease in electron density was observed at all heights during the eclipse of March 29, 2006, whereas there was a decrease in Ne values at only 190, 210, and 240 km altitudes during the eclipse of March 20, 2015. However, at other altitudes (290, 340, 410, and 490 km), the increase has occurred in the electron density during the eclipse on March 20, 2015. The IRI-2016 model predicted that Ne values increase at all altitudes in the solar eclipse on March 20, 2015, and a decrease in Ne at all altitudes occurs on March 29, 2006. It can be said that the NeQuick 2 model predicts a decrease in the first three heights and an increase in other heights. When the predictions of the empirical models were compared, according to the root mean square error (RMSE) calculations, the predictions of the IRI-2016 model can be said to be better than that of NeQuick 2 at all heights during eclipse of March 29, 2006. Besides, the IRI-2016 model's prediction can be stated to be better than NeQuick 2 model at the first three altitudes (190, 210 and 240 km), while the NeQuick 2 model's prediction can be said to be better than the IRI-2016 model at the last four (290, 340, 410, and 490 km) altitudes during March 20, 2015 eclipse. It can be stated that ionospheric models do not accurately reflect the ionospheric response to partial solar eclipses. The inclusion of ISR data in empirical models will help models accurately predict the ionospheric response under all conditions.