Hybrid Levenberg–Marquardt and weak-constraint ensemble Kalman smoother method

The ensemble Kalman smoother (EnKS) is used as a linear least-squares solver in the Gauss–Newton method for the large nonlinear least-squares system in incremental 4DVAR. The ensemble approach is naturally parallel over the ensemble members and no tangent or adjoint operators are needed. Furthermore...

Full description

Saved in:
Bibliographic Details
Published inNonlinear processes in geophysics Vol. 23; no. 2; pp. 59 - 73
Main Authors Mandel, J, Bergou, E, Gürol, S, Gratton, S, Kasanický, I
Format Journal Article
LanguageEnglish
Published Gottingen Copernicus GmbH 01.01.2016
European Geosciences Union (EGU)
Copernicus Publications
Subjects
Aquatic reptiles
Data assimilation
Geophysics
Least squares
Linear algebra
Localization
Mathematics
Methods
Newton methods
Nonlinear systems
Optimization and Control
Regularization
filter
covariance matrices
least-squares
variational data assimilation
systems
analysis scheme
Online AccessGet full text

Cover

More Information
Summary:The ensemble Kalman smoother (EnKS) is used as a linear least-squares solver in the Gauss–Newton method for the large nonlinear least-squares system in incremental 4DVAR. The ensemble approach is naturally parallel over the ensemble members and no tangent or adjoint operators are needed. Furthermore, adding a regularization term results in replacing the Gauss–Newton method, which may diverge, by the Levenberg–Marquardt method, which is known to be convergent. The regularization is implemented efficiently as an additional observation in the EnKS. The method is illustrated on the Lorenz 63 model and a two-level quasi-geostrophic model.
ISSN:1607-7946
1023-5809
1607-7946
DOI:10.5194/npg-23-59-2016