PreMevE Update: Forecasting Ultra‐Relativistic Electrons Inside Earth's Outer Radiation Belt

• Published in: Space Weather Vol. 19; no. 9
• Main Authors: Sinha, Saurabh, Chen, Yue, Lin, Youzuo, Pires de Lima, Rafael
• Format: Journal Article
• Language: English
• Published: Washington John Wiley & Sons, Inc 01.09.2021; American Geophysical Union
• Subjects: ASTRONOMY AND ASTROPHYSICS; Computer architecture; Electron flux; Electron precipitation; Electronics; Ensemble forecasting; Forecasting; Forecasting techniques; Geosynchronous orbits; Heliospheric and Magnetospheric Physics; In situ measurement; Light speed; Machine learning; Mathematical models; Monitors; Neural networks; NOAA satellites; Orbits; Outer radiation belt; predicting ultra‐relativistic electrons; Prediction models; Probes; Radiation; Radiation belts; Relativistic effects; Satellites; Solar wind; Solar wind velocity; supervised machine learning; Upstream; Van Allen electron radiation belt; Wind speed
• ISSN: 1542-7390; 1539-4964; 1542-7390
• DOI: 10.1029/2021SW002773

Energetic electrons inside Earth's Van Allen belts pose a major radiation threat to space‐borne electronics that often play vital roles in modern society. Ultra‐relativistic electrons with energies greater than or equal to two megaelectron‐volt (MeV) are of particular interest, and thus forecasting these ≥2 MeV electrons has a significant meaning to all space sectors. Here, we update the latest development of the predictive model for MeV electrons in the outer radiation belt. The new version, called PREdictive MEV Electron (PreMevE)‐2E, forecasts ultra‐relativistic electron flux distributions across the outer belt, with no need for in situ measurements of the trapped MeV electron population except at the geosynchronous orbit (GEO). Model inputs include precipitating electrons observed in low‐Earth‐orbits by NOAA satellites, upstream solar wind speeds and densities from solar wind monitors, as well as ultra‐relativistic electrons measured by one Los Alamos GEO satellite. We evaluated 32 supervised machine learning models that fall into four different classes of linear and neural network architectures, and successfully tested ensemble forecasting by using groups of top‐performing models. All models are individually trained, validated, and tested by in situ electron data from NASA's Van Allen Probes mission. It is shown that the final ensemble model outperforms individual models at most L‐shells, and this PreMevE‐2E model can provide 25‐h (∼1‐day) and 50‐h (∼2‐day) forecasts with high mean performance efficiency and correlation values. Our results also suggest that this new model is dominated by nonlinear components at L‐shells <∼4 for ultra‐relativistic electrons, different from the dominance of linear components for 1 MeV electrons as previously discovered. Plain Language Summary High‐speed electrons inside Earth's outer Van Allen belt pose a major radiation threat to man‐made satellites by causing malfunction of space‐borne electronics. These electrons, especially those traveling at nearly the speed of light, are of particular interest due to their high penetrating ability, and thus forecasting ultra‐relativistic (with energies greater than or equal to 2 MeV) electrons has a significant meaning to all space sectors. Here, we update the latest development of the predictive model for megaelectron‐volt (MeV) electrons inside the Earth's outer radiation belt. This new version, called PREdictive MEV Electron (PreMevE)‐2E, focuses on forecasting ultra‐relativistic electron flux distributions across the outer radiation belt, with no need for local measurements of the trapped MeV electrons except at the geosynchronous orbit (GEO). Model inputs include electrons observed in low‐Earth‐orbits by NOAA satellites, upstream solar wind conditions from solar wind monitors, and ultra‐relativistic electrons measured by one Los Alamos GEO satellite. We evaluated 32 supervised machine learning models in four different classes, and successfully demonstrated the performance of the ensemble forecasting technique. All models are trained, validated, and tested by electron data from NASA's Van Allen Probes mission. This new PreMevE‐2E model provides 1‐ and 2‐day forecasts of incoming nearly light‐speed electrons with high statistical fidelity. Key Points Machine‐learning based PREdictive MEV Electron (PreMevE) model is extended to predict ultra‐relativistic electron flux distributions during megaelectron‐volt (MeV) electron events This new PreMevE‐2E model makes reliable 1‐ and 2‐day ensemble forecasts of ≥2 MeV electrons inside Earth's outer radiation belt Nonlinear components play a major role in this new model at small L‐shells (<∼4) in contrast to previous PreMevE for 1 MeV electrons