Online machine-learning forecast uncertainty estimation for sequential data assimilation - IMT Atlantique Accéder directement au contenu
Pré-Publication, Document De Travail Année : 2023

Online machine-learning forecast uncertainty estimation for sequential data assimilation

Résumé

Quantifying forecast uncertainty is a key aspect of state-of-the-art numerical weather prediction and data assimilation systems. Ensemble-based data assimilation systems incorporate state-dependent uncertainty quantification based on multiple model integrations. However, this approach is demanding in terms of computations and development. In this work a machine learning method is presented based on convolutional neural networks that estimates the state-dependent forecast uncertainty represented by the forecast error covariance matrix using a single dynamical model integration. This is achieved by the use of a loss function that takes into account the fact that the forecast errors are heterodastic. The performance of this approach is examined within a hybrid data assimilation method that combines a Kalman-like analysis update and the machine learning based estimation of a state-dependent forecast error covariance matrix. Observing system simulation experiments are conducted using the Lorenz'96 model as a proof-of-concept. The promising results show that the machine learning method is able to predict precise values of the forecast covariance matrix in relatively high-dimensional states. Moreover, the hybrid data assimilation method shows similar performance to the ensemble Kalman filter outperforming it when the ensembles are relatively small.
Fichier principal
Vignette du fichier
online_machine-learning_forecast_uncertainty_estim.pdf (1008.4 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-04195361 , version 1 (04-09-2023)

Identifiants

Citer

Maximiliano Sacco, Manuel Pulido, Juan Ruiz, Pierre Tandeo. Online machine-learning forecast uncertainty estimation for sequential data assimilation. 2023. ⟨hal-04195361⟩
16 Consultations
9 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More