Ensemble Filters in Space Weather Forecasting

Ensemble filters have grown in popularity as a data assimilation tool for geophysical systems in recent years. This is largely due to their ability to handle highly nonlinear models without requiring detailed knowledge of the underlying dynamical equations. The Data Assimilation Research Testbed (DART) at NCAR is a powerful and flexible ensemble filter-based data assimilation framework whose well-defined programming interface and modular design provide a straight-forward means to apply advanced data assimilation algorithms to almost any dynamical model. We demonstrate several DART capabilities using both a relatively simple (but operationally relevant) radiation belt model, and by assimilating a limited portion of the Lyon-Fedder-Mobarry (LFM) magnetospheric state, namely electron precipitation generated at the MHD inner boundary.