Relativistic Electron Model in the Outer Radiation Belt Using a Neural Network Approach. Issue 12 (14th December 2021)

• Record Type: Journal Article
• Authors: Chu, Xiangning; Ma, Donglai; Bortnik, Jacob; Tobiska, W. Kent; Cruz, Alfredo; Bouwer, S. Dave; Zhao, Hong; Ma, Qianli; Zhang, Kun; Baker, Daniel N.; Li, Xinlin; Spence, Harlan; Reeves, Geoff
• Is Part Of: Space weather. Volume 19:Issue 12(2021)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-14
• Subjects: machine learning -- neural network -- radiation belt -- energetic electron fluxes -- Van Allen Probes -- forecast; Space environment -- Periodicals; 551.509992
• DOI: 10.1029/2021SW002808
• Languages: English
• ISSNs: 1542-7390

Abstract

Abstract: We present a machine‐learning‐based model of relativistic electron fluxes >1.8 MeV using a neural network approach in the Earth's outer radiation belt. The Outer RadIation belt Electron Neural net model for Relativistic electrons (ORIENT‐R) uses only solar wind conditions and geomagnetic indices as input. For the first time, we show that the state of the outer radiation belt can be determined using only solar wind conditions and geomagnetic indices, without any initial and boundary conditions. The most important features for determining outer radiation belt dynamics are found to be AL, solar wind flow speed and density, and SYM‐H indices. ORIENT‐R reproduces out‐of‐sample relativistic electron fluxes with a correlation coefficient of 0.95 and an uncertainty factor of ∼2. ORIENT‐R reproduces radiation belt dynamics during an out‐of‐sample geomagnetic storm with good agreement to the observations. In addition, ORIENT‐R was run for a completely out‐of‐sample period between March 2018 and October 2019 when the AL index ended and was replaced with the predicted AL index (lasp.colorado.edu/home/personnel/xinlin.li ). It reproduces electron fluxes with a correlation coefficient of 0.92 and an out‐of‐sample uncertainty factor of ∼3. Furthermore, ORIENT‐R captured the trend in the electron fluxes from low‐earth‐orbit (LEO) SAMPEX, which is a completely out‐of‐sample data set both temporally and spatially. In sum, the ORIENT‐R model can reproduce transport, acceleration, decay, and dropouts of the outer radiation belt anywhere from short timescales (i.e., geomagnetic storms) and very long timescales (i.e., solar cycle) variations. Plain Language Summary: The Earth's radiation belts consist of energetic particles. During periods of intense space weather, the energy and density of radiation belt particles can increase significantly and pose a danger to astronauts, spacecraft, and even technologies on the ground. This study presents a machine‐learning‐based model (ORIENT‐R) that calculates the energetic radiation belt electron fluxes. It uses solar wind observations and geomagnetic activity indices as input, without the need for boundary conditions such as other satellite measurements. The ORIENT‐R model can determine the energetic electrons with a high Pearson correlation of 0.95 and a small uncertainty factor of ∼2. Furthermore, even when geomagnetic index AL is not available, the ORIENT‐R model can still estimate with high accuracy of Pearson correlation of 0.92 and a small uncertainty factor of ∼3. Thus, the ORIENT‐R model has great value and wide application in the space physics community and space weather industry. Key Points: A neural network model was developed to forecast relativistic electron fluxes with energies >1.8 MeV in the outer radiation belt (ORIENT‐R) ORIENT‐R model reproduces the relativistic electron fluxes with high out‐of‐sample accuracy: Pearson r ∼ 0.95, uncertainty of a factor of ∼2 The ORIENT‐R model reproduces short‐ and long‐term dynamics of the outer radiation belt such as transport, acceleration, decay, and dropouts