Observation-Based Noise Calibration: An Efficient Dynamics for the Ensemble Kalman Filter

Benjamin Dufée (1, 2) , Etienne Mémin (1, 2) , Dan Crisan (3)

1 ODYSSEY - Océan Dynamique Observations Analyse
2 IRMAR - Institut de Recherche Mathématique de Rennes
3 Department of Mathematics [Imperial College London]

Abstract We investigate the calibration of the stochastic noise in order to guide the realizations towards the observational data used for the assimilation. This is done in the context of the stochastic parametrization under Location Uncertainty (LU) and data assimilation. The new methodology is rigorously justified by the use of the Girsanov theorem, and yields significant improvements in the experiments carried out on the Surface Quasi Geostrophic (SQG) model, when applied to Ensemble Kalman filters. The particular test case studied here shows improvements of the peak MSE from 85% to 93%.

Domains: Probability [math.PR]

Benjamin Dufée, Etienne Mémin, Dan Crisan. Observation-Based Noise Calibration: An Efficient Dynamics for the Ensemble Kalman Filter. Stochastic Transport in Upper Ocean Dynamics, 10, Springer International Publishing, pp.43-56, 2023, Mathematics of Planet Earth, ⟨10.1007/978-3-031-18988-3_4⟩. ⟨hal-03910764⟩