Reconstruction of electron radiation belts using numerical modeling, data assimilation and machine learning

Alexander Drozdov, University of California, Los Angeles - 12:00

This seminar presented an overview of numerical modeling, data assimilation, and machine learning in understanding Earth's electron radiation belts. Central to the discussion is the Versatile Electron Radiation Belt (VERB) code, a powerful tool for solving the Fokker-Planck equation. The VERB code accounts for radial diffusion, advection, and local wave-particle interactions, including pitch-angle, energy, and mixed diffusion. This presentation will show development evolution and capabilities of the VERB code, and integration of data assimilation, presenting how satellite measurements are utilized for reconstructing the radiation belts and the ring current. Special emphasis will be placed on the innovative application of machine learning techniques, highlighting a reconstruction of radiation belt electron fluxes using data assimilation with low-Earth-orbiting Polar Orbiting Environmental Satellites (POES) measurements mapped to near equatorial regions.