Estimating model‐error covariances in nonlinear state‐space models using Kalman smoothing and the expectation–maximization algorithm

Specification and tuning of errors from dynamical models are important issues in data assimilation. In this work, we propose an iterative expectation–maximization (EM) algorithm to estimate the model‐error covariances using classical extended and ensemble versions of the Kalman smoother. We show tha...

Full description

Saved in:
Bibliographic Details
Published in:Quarterly journal of the Royal Meteorological Society Vol. 143; no. 705; pp. 1877 - 1885
Main Authors: Dreano, D., Tandeo, P., Pulido, M., Ait‐El‐Fquih, B., Chonavel, T., Hoteit, I.
Format: Journal Article
Language:English
Published: Chichester, UK John Wiley & Sons, Ltd 01.04.2017
Wiley Subscription Services, Inc
Wiley
Subjects:
Additives
Algorithms
Applications
Data assimilation
Data collection
Earth Sciences
Engineering Sciences
EnKF
EnKS
Errors
expectation–maximization
extended Kalman filter
Mathematical models
Meteorology
model error
Sciences of the Universe
Signal and Image processing
state‐space models
Statistics
ISSN:0035-9009, 1477-870X
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Specification and tuning of errors from dynamical models are important issues in data assimilation. In this work, we propose an iterative expectation–maximization (EM) algorithm to estimate the model‐error covariances using classical extended and ensemble versions of the Kalman smoother. We show that, for additive model errors, the estimate of the error covariance converges. We also investigate other forms of model error, such as parametric or multiplicative errors. We show that additive Gaussian model error is able to compensate for non‐additive sources of error in the algorithms we propose. We also demonstrate the limitations of the extended version of the algorithm and recommend the use of the more robust and flexible ensemble version. This article is a proof of concept of the methodology with the Lorenz‐63 attractor. We developed an open‐source Python library to enable future users to apply the algorithm to their own nonlinear dynamical models. We propose an iterative expectation–maximization algorithm to estimate the model‐error covariances using classical extended and ensemble versions of the Kalman smoother. We show convergence of the algorithm on the Lorenz‐63 model, including for non‐Gaussian model errors. We have developed an open‐source Python library to enable future users to apply the algorithm to their own nonlinear dynamical models.
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0035-9009
1477-870X
DOI:10.1002/qj.3048