Energetic Electron Flux Predictions in the Near‐Earth Plasma Sheet From Solar Wind Driving. Issue 11 (3rd November 2022)

• Record Type: Journal Article
• Authors: Swiger, B. M.; Liemohn, M. W.; Ganushkina, N. Y.; Dubyagin, S. V.
• Is Part Of: Space weather. Volume 20:Issue 11(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-11-03
• Subjects: plasma sheet -- solar wind -- neural network -- electron flux -- machine learning; Space environment -- Periodicals; 551.509992
• DOI: 10.1029/2022SW003150
• Languages: English
• ISSNs: 1542-7390

Abstract

Abstract: Suprathermal electrons in the near‐Earth plasma sheet are important for inner magnetosphere considerations. They are the source population for outer radiation belt electrons and they pose risks to geosynchronous satellites through their contribution to surface charging. We use empirical modeling to address relationships between solar driving parameters and plasma sheet electron flux. Using Time History of Events and Macroscale Interactions during Substorms, OMNI, and Flare Irradiance Spectral Model Version 2 data, we develop a neural network model to predict differential electron flux from 0.08 to 93 keV in the plasma sheet, at distances from 6 to 12 R E . Driving parameters include solar wind (SW) density and speed, interplanetary magnetic field (IMF) B Z and B Y, solar extreme ultraviolet flux, IMF B Z ultra‐low frequency (ULF) wave power, SW‐magnetosphere coupling functions P α 1 and NXCF, and the 4‐hr time history of these parameters. Our model predicts overall plasma sheet electron flux variations with correlation coefficients of between 0.59 and 0.77, and median symmetric accuracy in the 41%–140% range (depending on energy). We find that short time‐scale electron flux variations are not reproduced using short time‐scale inputs. Using a recently published technique to extract information from neural networks, we determine the most important drivers impacting model prediction are V SW, VB S, and IMF B Z . SW‐magnetosphere coupling functions that include IMF clock angle, IMF B Z ULF wave power, and IMF B Y have little impact in our model of electron flux in the near‐Earth plasma sheet. The new model, built directly on differential flux, outperforms an existing model that derives fluxes from plasma moments, with the performance improvement increasing with increasing energy. Plain Language Summary: In near‐Earth space, electrons are energized and transported toward the Earth, where they pose hazards to spacecraft or travel to the upper atmosphere to generate aurora. Previous studies have shown that much of the variations are attributed to changes in the upstream solar wind (SW), the driving conditions from the Sun. Yet, it has been difficult to determine which variations in the SW are most predictive of changes in the near‐Earth electrons. To help confront these difficulties, we have developed a machine learning neural network model that predicts the variations of electrons using inputs of the SW and sunlight levels. We report details of the development of this model and assess the model's performance. The model well reproduces global variations of near‐Earth electrons, yet does not fully reproduce flux variations during all space weather events. By ranking the importance of the model inputs, we show that individual SW parameters are more important to predictions than using some previously defined and investigated coupling functions calculated using those same parameters. Key Points: New model predicts plasma sheet electron flux (0.08–93 keV) from solar wind (SW) within a factor of two of observed based on median symmetric accuracy metric Input analysis of new model supports that the most impactful drivers are SW speed, electric field, and north‐south component of interplanetary magnetic field Short‐term changes of electron flux (<1 hr) in the plasma sheet are not predicted from high resolution (5 min) SW inputs