Forecasting the Radiation Belt Environment With Data Assimilation

We present initial steps toward a forecast model of the radiation belt electron environment, which assimilates available space-based observations into the framework of a physics-based evolutionary model. The model is designed to assimilate particle flux data from spacecraft that span a broad range of L-shells in an orbit near the equatorial plane. For each pass through the modeling region, spatial analysis techniques are used to incorporate observations into the physical model, using the statistical error structures of the model and data to produce an optimal estimate. The analysis algorithm makes use of an empirical magnetic field model, parameterized by forecast solar wind inputs, to convert measured flux into phase space density. The forecast model then solves the radial diffusion equation to evolve the phase space density in time. The assimilation algorithm is validated in a series of experiments, using artificial storm-time particle flux data fabricated by an independent model. The simulated data is compared to the model over the entire modeling region at each time step, providing a global assessment of the model's accuracy over a wide range of magnetospheric activity levels.