 |
Center for Nonlinear Studies |
Earth’s radiation belts are very dynamic and undergo constant changes due to acceleration, loss, and transport processes of the energetic electrons. In this work, an ensemble data assimilation is applied to a radial diffusion model with the purpose of accurately estimating Earth’s radiation belt particle distribution. The model describes the phase space density (PSD) of relativistic electrons in the radiation belts. Together with observational data, gathered from a number of satellites, the ensemble data assimilation provides the best PSD estimate for the given data set. A particular concern in the data assimilation are model errors, which can adversely impact the solution of the assimilation. In the current work, model uncertainty is accounted in the data assimilation through a localized covariance inflation technique that assigns an appropriate uncertainty to the model. Numerical results from identical-twin experiments, where data is generated from the same model, as well as the assimilation of real observational data, are presented. The results show the improvement of the PSD solution due to the data assimilation and the proper inclusion of model errors.
Host: Kim Rasmussen, T-5