Cosmic-Ray-Enhanced LSTM for Geomagnetic Storm Prediction

This paper presents a novel approach to geomagnetic storm prediction by incorporating cosmic-ray flux modulation as a precursor signal within a physics-informed LSTM model. The use of cosmic-ray data, which can provide early warnings, is a significant contribution. The study demonstrates improved forecast skill, particularly for longer prediction horizons, highlighting the value of integrating physics knowledge with deep learning for space-weather forecasting. The results are promising for improving the accuracy and lead time of geomagnetic storm predictions, which is crucial for protecting technological infrastructure.