Ensemble Kalman Filter for Assimilation in a Ring Current Model

Over the recent years there has been significant progress to develop data assimilation techniques to improve the predictive capabilities of space weather models. As a consequence, data assimilation methods have been widely adopted into several fields of space weather, including specification and forecasting of the ionosphere-thermosphere, ring currents, radiation belts, and solar photosphere to name a few. We will present a new assimilation method based on the ensemble Kalman filter that projects into the space of dominant model dynamics and perform the assimilation in this new space. The new assimilation method is applied to the Ring Current-Atmosphere Interactions Model with Self-Consistent Magnetic field (RAM-SCB) model using particle flux data from the Van Allen Probes. The results provide a dramatic improvement over other versions of the Kalman filter techniques, and provides a solution with a reduced forecast error.