European Space Weather Week 2025

Machine Learning Approach for Prediction of Ionospheric Irregularities on ROTI Index over the Northern Anomaly Crest in Egypt during Solar Cycle 24

Not scheduled

20m

Idun

Poster CD1 - Combination of physics-based and data-driven methods for space weather forecasting CD1 - Combination of physics-based and data-driven methods for space weather forecasting

Speaker

Dr Hager M. Salah (1. Space Weather Monitoring Center, Helwan University, Egypt. 2. Engineering School, Canadian International College in Cairo, Egypt.)

Description

This study presents predictions of ionospheric irregularities using the ROTI index near the Northern peak of the equatorial ionization anomaly. A feedforward backpropagation Neural Network with a NARX time-series approach was applied, using dual-frequency GPS-SCINDA data from Helwan, Egypt (29.86°N, 31.32°E) over Solar Cycle 24 (2009–2017). The model incorporated diurnal, seasonal, solar, and geomagnetic parameters, with IRI-foF2 and IRI-hmF2 included to enhance learning of F-layer behavior.

Results show strong agreement between predicted and observed ROTI values, with an RMSE of 0.106 TECU/min and a prediction efficiency of 95%. Regression reached 0.89, with irregularities more frequent during equinoxes. Seasonal RMSEs remained below 0.05 TECU/min across solar activity levels. The findings confirm a clear solar cycle dependency in irregularity occurrence, with predicted occurrence percentages closely matching observed values.