On Ensemble Nonlinear Kalman Filtering with Symmetric Analysis Ensembles
The ensemble square root filter (EnSRF) [1, 2, 3,4] is a popular method for data assimilation in high dimensional systems (e.g., geophysics models). Essentially the EnSRF is a Monte Carlo implementation of the conventional Kalman filter (KF) [5, 6]. It is mainly different from the KF at the predicti...
| Main Authors: | , , |
|---|---|
| Other Authors: | |
| Format: | Conference item |
| Published: |
AMER INST PHYSICS
2010
|
| _version_ | 1826278352943579136 |
|---|---|
| author | Luo, X Hoteit, I Moroz, I |
| author2 | Simos, T |
| author_facet | Simos, T Luo, X Hoteit, I Moroz, I |
| author_sort | Luo, X |
| collection | OXFORD |
| description | The ensemble square root filter (EnSRF) [1, 2, 3,4] is a popular method for data assimilation in high dimensional systems (e.g., geophysics models). Essentially the EnSRF is a Monte Carlo implementation of the conventional Kalman filter (KF) [5, 6]. It is mainly different from the KF at the prediction steps, where it is some ensembles, rather then the means and covariance matrices, of the system state that are propagated forward. In doing this, the EnSRF is computationally more efficient than the KF, since propagating a covariance matrix forward in high dimensional systems is prohibitively expensive. In addition, the EnSRF is also very convenient in implementation. By propagating the ensembles of the system state, the EnSRF can be directly applied to nonlinear systems without any change in comparison to the assimilation procedures in linear systems.However, by adopting the Monte Carlo method, the EnSRF also incurs certain sampling errors. One way to alleviate this problem is to introduce certain symmetry to the ensembles, which can reduce the sampling errors and spurious modes in evaluation of the means and covariances of the ensembles [7]. In this contribution, we present two methods to produce symmetric ensembles. One is based on the unscented transform [8, 9], which leads to the unscented Kalman filter (UKF) [8, 9] and its variant, the ensemble unscented Kalman filter (EnUKF) [7]. The other is based on Stirling's interpolation formula (SIF), which results in the divided difference filter (DDF) [10]. Here we propose a simplified divided difference filter (sDDF) in the context of ensemble filtering. The similarity and difference between the sDDF and the EnUKF will be discussed. Numerical experiments will also be conducted to investigate the performance of the sDDF and the EnUKF, and compare them to a well-established EnSRF, the ensemble transform Kalman filter (ETKF) [2]. |
| first_indexed | 2024-03-06T23:42:39Z |
| format | Conference item |
| id | oxford-uuid:6fd7f4f8-a460-4daa-b5c0-4931b1455c25 |
| institution | University of Oxford |
| last_indexed | 2024-03-06T23:42:39Z |
| publishDate | 2010 |
| publisher | AMER INST PHYSICS |
| record_format | dspace |
| spelling | oxford-uuid:6fd7f4f8-a460-4daa-b5c0-4931b1455c252022-03-26T19:33:15ZOn Ensemble Nonlinear Kalman Filtering with Symmetric Analysis EnsemblesConference itemhttp://purl.org/coar/resource_type/c_5794uuid:6fd7f4f8-a460-4daa-b5c0-4931b1455c25Symplectic Elements at OxfordAMER INST PHYSICS2010Luo, XHoteit, IMoroz, ISimos, TPsihoyios, GTsitouras, CThe ensemble square root filter (EnSRF) [1, 2, 3,4] is a popular method for data assimilation in high dimensional systems (e.g., geophysics models). Essentially the EnSRF is a Monte Carlo implementation of the conventional Kalman filter (KF) [5, 6]. It is mainly different from the KF at the prediction steps, where it is some ensembles, rather then the means and covariance matrices, of the system state that are propagated forward. In doing this, the EnSRF is computationally more efficient than the KF, since propagating a covariance matrix forward in high dimensional systems is prohibitively expensive. In addition, the EnSRF is also very convenient in implementation. By propagating the ensembles of the system state, the EnSRF can be directly applied to nonlinear systems without any change in comparison to the assimilation procedures in linear systems.However, by adopting the Monte Carlo method, the EnSRF also incurs certain sampling errors. One way to alleviate this problem is to introduce certain symmetry to the ensembles, which can reduce the sampling errors and spurious modes in evaluation of the means and covariances of the ensembles [7]. In this contribution, we present two methods to produce symmetric ensembles. One is based on the unscented transform [8, 9], which leads to the unscented Kalman filter (UKF) [8, 9] and its variant, the ensemble unscented Kalman filter (EnUKF) [7]. The other is based on Stirling's interpolation formula (SIF), which results in the divided difference filter (DDF) [10]. Here we propose a simplified divided difference filter (sDDF) in the context of ensemble filtering. The similarity and difference between the sDDF and the EnUKF will be discussed. Numerical experiments will also be conducted to investigate the performance of the sDDF and the EnUKF, and compare them to a well-established EnSRF, the ensemble transform Kalman filter (ETKF) [2]. |
| spellingShingle | Luo, X Hoteit, I Moroz, I On Ensemble Nonlinear Kalman Filtering with Symmetric Analysis Ensembles |
| title | On Ensemble Nonlinear Kalman Filtering with Symmetric Analysis Ensembles |
| title_full | On Ensemble Nonlinear Kalman Filtering with Symmetric Analysis Ensembles |
| title_fullStr | On Ensemble Nonlinear Kalman Filtering with Symmetric Analysis Ensembles |
| title_full_unstemmed | On Ensemble Nonlinear Kalman Filtering with Symmetric Analysis Ensembles |
| title_short | On Ensemble Nonlinear Kalman Filtering with Symmetric Analysis Ensembles |
| title_sort | on ensemble nonlinear kalman filtering with symmetric analysis ensembles |
| work_keys_str_mv | AT luox onensemblenonlinearkalmanfilteringwithsymmetricanalysisensembles AT hoteiti onensemblenonlinearkalmanfilteringwithsymmetricanalysisensembles AT morozi onensemblenonlinearkalmanfilteringwithsymmetricanalysisensembles |