Machine Learning Based Storm Time Modeling Of Ionospheric Vertical Total Electron Content Over Ethiopia

A recent study has found that applying machine learning algorithms specifically to model the low-latitude African ionospheric region can improve predictive accuracy during geomagnetic storms. The study utilized deep neural networks and light gradient boosting machine techniques to predict TEC data from thirteen GPS receiver stations in Ethiopia. The results showed that these models outperformed the IRI 2020 global empirical model in predicting TEC during geomagnetic disturbed conditions in the low-latitude ionospheric region.