SC Seminar Jean Francois Ripoll (CEA, Paris)

Modeling of the quiet decay of radiation belts electrons

Jean-François Ripoll, CEA, DAM, DIF, Arpajon, France

Modeling of the Quiet Decay of Radiation Belts Electrons

In this presentation, we address the questions on how to model the dynamics of the radiation belts during quiet geomagnetic times. We consider a broad hierarchy of models, from equilibrium (steady) model representation, to 1D reduced Fokker-Planck, then, full 3D Fokker-Planck formulations. We show how we can sometimes find analytically the solution or simplify some important terms, such as pitch angle diffusion, to the profit of lowering the computational cost while still keeping admissible accuracy.

We apply these models to the geomagnetic storm of March 1st 2013 and compute how fast the slot region forms gradually between the two radiation belts during long and quiet storm recovery, contributing to depopulate the close-Earth magnetosphere of the large amount of electrons injected by the storm. This scattering phenomenon by pitch angle diffusion is caused by wave-particle interactions from whistler hiss waves and is essential to the energy structure of the belts and slot region.

Here, pitch angle diffusion is computed from data-driven whistler mode hiss waves and ambient plasma observations from the NASA Van Allen Probes satellites. The high temporal and spatial resolution is meant to describe the nonlinear turbulent variability of the scattering and requires massively parallel simulations that will be briefly discussed. 3D Fokker-Planck simulations made with VERB-3D uses these data-driven pitch angle diffusion coefficients while the 1D reduced Fokker-Planck equation is based on losses computed from data-driven electron lifetimes that are fully consistent with the diffusion coefficients. Numerical results are compared to global observations from the Van Allen Probes using the Magnetic Electron and Ion Spectrometer (MagEIS) flux measurements of the belts. Different dedicated metrics are discussed and used to assess the models’ accuracy.