CIMA 09099
CENTRO DE INVESTIGACIONES DEL MAR Y LA ATMOSFERA
Unidad Ejecutora - UE
congresos y reuniones científicas
Título:
Inference of stochastic parameters for model error representation with nested EnKFs
Autor/es:
SCHEFFLER GUILLERMO; RUIZ JUAN JOSE; PULIDO, MANUEL
Lugar:
Kobe
Reunión:
Workshop; RIKEN International Workshop on Uncertainty Quantification; 2018
Institución organizadora:
RIKEN
Resumen:
Model errors can be represented by the incorporation of an additive stochastic forcing[1]. However, the variance of such processes cannot be estimated via state augmentation within the ensemble Kalman filter (EnKF). We propose a novel technique based on nested EnKFs to estimate model error parameters, which can also be extended to filter hyper-parameters like multiplicative covariance inflation.
