Speaker
Description
Given the critical impact of accurate radiation belt modeling on space radiation environment restitution and forecasting, data assimilation has been employed to enhance physical models estimations since their earliest days. The method, allows the correction of the theoretical description of a given model, thanks to the ingestion of in-situ observations. Precisely, the widely adopted approach applies the correction solely at a model’s output level, with no impact on the model’s parameters or more generally on its physical description. Yet, the model's output biases are typically rooted in an inaccurate estimation of those same parameters within the underlying physical description. Moreover, one can explore within the same data assimilation framework, the correction of the model’s biases directly on the model’s faulty parameters, thanks to the state vector extension operation.
In this talk, we will present the principle of the state vector extension and how it can be leveraged in the radiation belt modeling context. For that, we will rely on twin experiments deployed on a realistic Fokker-Planck based model of the electron radiation belts. These numerical tests will help us evaluate the potential of the method as a new way to extract theoretical inferences from satellite observations or to detect the model’s trajectory deviation.
| Do you plan to attend in-person or online? | In-person |
|---|
