A Deep Learning‐Based Approach to Forecast the Onset of Magnetic Substorms. Issue 11 (11th November 2019)

• Record Type: Journal Article
• Authors: Maimaiti, M.; Kunduri, B.; Ruohoniemi, J. M.; Baker, J. B. H.; House, Leanna L.
• Is Part Of: Space weather. Volume 17:Issue 11(2019)
• Page Start: 1534
• Page End: 1552
• Publication Date: 2019-11-11
• Subjects: substorm onset forecasting -- deep learning -- machine learning -- substorm; Space environment -- Periodicals; 551.509992
• DOI: 10.1029/2019SW002251
• Languages: English
• ISSNs: 1542-7390

Abstract

Abstract: The auroral substorm has been extensively studied over the last six decades. However, our understanding of its driving mechanisms is still limited and so is our ability to accurately forecast its onset. In this study, we present the first deep learning‐based approach to predict the onset of a magnetic substorm, defined as the signature of the auroral electrojets in ground magnetometer measurements. Specifically, we use a time history of solar wind speed ( V x ), proton number density, and interplanetary magnetic field (IMF) components as inputs to forecast the occurrence probability of an onset over the next 1 hr. The model has been trained and tested on a data set derived from the SuperMAG list of magnetic substorm onsets and can correctly identify substorms ∼75% of the time. In contrast, an earlier prediction algorithm correctly identifies ∼21% of the substorms in the same data set. Our model's ability to forecast substorm onsets based on solar wind and IMF inputs prior to the actual onset time, and the trend observed in IMF B z prior to onset together suggest that a majority of the substorms may not be externally triggered by northward turnings of IMF. Furthermore, we find that IMF B z and V x have the most significant influence on model performance. Finally, principal component analysis shows a significant degree of overlap in the solar wind and IMF parameters prior to both substorm and nonsubstorm intervals, suggesting that solar wind and IMF alone may not be sufficient to forecast all substorms, and preconditioning of the magnetotail may be an important factor. Key Points: We present the first deep learning‐based time series model to forecast the onset of a magnetic substorm over the next 1 hr The model has been developed using a comprehensive list of onsets compiled between 1997 and 2017 and achieves 72% precision and 77% recall A significant overlap in solar wind and IMF conditions prior to minor substorms and non‐onsets reduces the classification accuracy