Modeling the Dynamic Variability of Sub‐Relativistic Outer Radiation Belt Electron Fluxes Using Machine Learning. Issue 8 (11th August 2022)

• Record Type: Journal Article
• Authors: Ma, Donglai; Chu, Xiangning; Bortnik, Jacob; Claudepierre, Seth G.; Tobiska, W. Kent; Cruz, Alfredo; Bouwer, S. Dave; Fennell, Joseph F.; Blake, J. Bernard
• Is Part Of: Space weather. Volume 20:Issue 8(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-08-11
• Subjects: machine learning -- radiation belts -- electron flux; Space environment -- Periodicals; 551.509992
• DOI: 10.1029/2022SW003079
• Languages: English
• ISSNs: 1542-7390

Abstract

Abstract: We present a set of neural network models that reproduce the dynamics of electron fluxes in the range of 50 keV ∼1 MeV in the outer radiation belt. The Outer Radiation belt Electron Neural net model for Medium energy electrons uses only solar wind conditions and geomagnetic indices as input. The models are trained on electron flux data from the Magnetic Electron Ion Spectrometer instrument onboard Van Allen Probes, and they can reproduce the dynamic variations of electron fluxes in different energy channels. The model results show high coefficient of determination ( R 2 ∼ 0.78–0.92) on the test data set, an out‐of‐sample 30‐day period from 25 February to 25 March in 2017, when a geomagnetic storm took place, as well as an out‐of‐sample one year period after March 2018. In addition, the models are able to capture electron dynamics such as intensifications, decays, dropouts, and the Magnetic Local Time dependence of the lower energy (∼<100 keV) electron fluxes during storms. The models have reliable prediction capability and can be used for a wide range of space weather applications. The general framework of building our model is not limited to radiation belt fluxes and could be used to build machine learning models for a variety of other plasma parameters in the Earth's magnetosphere. Plain Language Summary: The Earth's radiation belts consist of energetic particles trapped by the geomagnetic field. This radiation environment is known to be particularly hazardous to spacecrafts and difficult to predict given the complex dynamics of electrons at different energy states. This paper presents a set of neural‐network‐based models that use measurements of geomagnetic and solar activities as drivers to reconstruct radiation belt electron fluxes ranging from 50 keV to 1 MeV. The models can determine the flux with high accuracy and capture the electron dynamics with long‐ and short‐term time scales. The models provide reliable prediction capability and can be used for a wide range of space weather applications. The approach through which our models are built is not limited to radiation belt fluxes and can be generalized for a variety of other plasma parameters in the Earth's magnetosphere. Key Points: A set of neural network models was developed to reconstruct the 50 keV‐1 MeV electron fluxes in the outer radiation belt The models reproduce fluxes with a high overall accuracy ( R 2 ∼ 0.78–0.92) during out‐of‐sample periods, including long‐ and short‐term dynamics The models reproduce the time‐varying Magnetic Local Time dependence exhibited during storms by the lower energy electrons (∼<100 keV)