Prediction Covariance Estimation using machine learning

SACCO, MAXIMILIANO A.; RUIZ J. J.; TANDEO P; PULIDO M

Moulin Mer, Logonna-Daoulas, Finist\`ere, Francia.

Workshop; Machine learning and uncertainties in climate simulations.; 2022

Henri Lebesgue Center de Mathematiques

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.