Data Assimilative Simulations Of Earth’s Radiation Belt Electron Dynamics Based On Multi-satellite Observations

The Earth’s radiation belts are filled with energetic particles,which poses a potential threat to the safety of in-orbit satellites and astronauts.These high energy electrons are captured by Earth’s magnetic field.Although satellite observations provide information of radiation belt electrons in different energy and pitch angle ranges,the measurements of satellite have limited spatial and temporal coverage.On the other hand,numerical models can simulate the dynamic processes of radiation belt electrons with full spatial and temporal coverage based on certain physical mechanisms.However,the simulation results can only approximate the real state and cannot replace the actual observation of satellites.In this study,the data assimilation method is used to combine physical models and actual measurements considering the uncertainties and limitations from both models and observations.Based on this method,we reconstruct the dynamic evolution of Earth’s radiation belt electrons,which provide favorable information for a deep understanding of the acceleration and loss mechanisms of radiation belt electrons.In this thesis,we focus on the study of electron phase space density variations observed by multi-satellite during storm time and data assimilation based on the Kalman filter.The main conclusions are summarized as follows:(1)By analyzing the electron flux measured simultaneously by two Van Allen Probes and three GPS navigation satellites during the entire month of March 2013,we found that radiation belt electron flux dropouts occurred significantly during the 17 March geomagnetic storm accompanied with southward IMF Bz and enhanced solar wind dynamic pressure.By converting the electron flux into the electron phase space density and investigating the electron PSD radial profiles and their differences for different pairs of the first and second adiabatic invariants,the non-adiabatic mechanisms attributed to the observed radiation belt electron variations are explored.During the main phase of the geomagnetic storms,electron PSDs drop quickly mainly owing to the effect of magnetopause shadowing and associated outward radial diffusion while electron PSDs increase rapidly during the storm recovery phase mainly due to the local acceleration processes.Overall,simultaneous measurements from multiple satellites with a broad spatial coverage facilitate in-depth understanding the dynamics of radiation belt electron.(2)The Kalman filter can combine observations of Van Allen Probes and GOES satellites to reconstruct the distribution of electron flux based on physics-based models,which include one-dimensional radial diffusion model,one-dimensional pitch angle diffusion model and two-dimensional radial-pitch angle mixed diffusion model in this thesis.The data assimilation method using one-dimensional radial diffusion model can well simulate the process that electrons are accelerated in the region of high L-shells during geomagnetic activitive conditions,but the electron fluxes at low L-shell region are seriously overestimated due to the lack of loss mechanism.The data assimilation method using one-dimensional pitch angle diffusion model can well simulate the loss of electrons during geomagnetic activity while the electron acceleration process in the low L-shell region cannot be reproduced due to the lack of acceleration mechanism.Including both the radial diffusion effect and the pitch angle diffusion effect,the data assimilation method using two-dimensional radial-pitch angle mixed diffusion model is introduced in this thesis to carefully reconstruct the evolution of radiation belt electron dynamics in different MLT ranges.The reliability of the adopted numerical model and the validity of Kalman filtering in reconstructing the complete images of the evolution of radiation belt electron dynamics are verified in comparison with the observation data.The two-dimensional electron radiation belt data assimilative model can help deepen the understanding of the electron acceleration and loss processes and the underlying physical mechanisms,and also creates favorable conditions for the simulation of the dynamic prediction of radiation belt electrons.