Efficient Kernel-Based Ensemble Gaussian Mixture Filtering

Handle URI:
http://hdl.handle.net/10754/597024
Title:
Efficient Kernel-Based Ensemble Gaussian Mixture Filtering
Authors:
Liu, Bo ( 0000-0001-6615-1096 ) ; Ait-El-Fquih, Boujemaa; Hoteit, Ibrahim ( 0000-0002-3751-4393 )
Abstract:
We consider the Bayesian filtering problem for data assimilation following the kernel-based ensemble Gaussian-mixture filtering (EnGMF) approach introduced by Anderson and Anderson (1999). In this approach, the posterior distribution of the system state is propagated with the model using the ensemble Monte Carlo method, providing a forecast ensemble that is then used to construct a prior Gaussian-mixture (GM) based on the kernel density estimator. This results in two update steps: a Kalman filter (KF)-like update of the ensemble members and a particle filter (PF)-like update of the weights, followed by a resampling step to start a new forecast cycle. After formulating EnGMF for any observational operator, we analyze the influence of the bandwidth parameter of the kernel function on the covariance of the posterior distribution. We then focus on two aspects: i) the efficient implementation of EnGMF with (relatively) small ensembles, where we propose a new deterministic resampling strategy preserving the first two moments of the posterior GM to limit the sampling error; and ii) the analysis of the effect of the bandwidth parameter on contributions of KF and PF updates and on the weights variance. Numerical results using the Lorenz-96 model are presented to assess the behavior of EnGMF with deterministic resampling, study its sensitivity to different parameters and settings, and evaluate its performance against ensemble KFs. The proposed EnGMF approach with deterministic resampling suggests improved estimates in all tested scenarios, and is shown to require less localization and to be less sensitive to the choice of filtering parameters.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Citation:
Efficient Kernel-Based Ensemble Gaussian Mixture Filtering 2015:151111131815001 Monthly Weather Review
Publisher:
American Meteorological Society
Journal:
Monthly Weather Review
Issue Date:
11-Nov-2015
DOI:
10.1175/MWR-D-14-00292.1
Type:
Article
ISSN:
0027-0644; 1520-0493
Sponsors:
The research reported in this publication was supported by the King Abdullah University of Science and Technology (KAUST).
Additional Links:
http://journals.ametsoc.org/doi/abs/10.1175/MWR-D-14-00292.1
Appears in Collections:
Articles; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorLiu, Boen
dc.contributor.authorAit-El-Fquih, Boujemaaen
dc.contributor.authorHoteit, Ibrahimen
dc.date.accessioned2016-02-23T14:20:36Zen
dc.date.available2016-02-23T14:20:36Zen
dc.date.issued2015-11-11en
dc.identifier.citationEfficient Kernel-Based Ensemble Gaussian Mixture Filtering 2015:151111131815001 Monthly Weather Reviewen
dc.identifier.issn0027-0644en
dc.identifier.issn1520-0493en
dc.identifier.doi10.1175/MWR-D-14-00292.1en
dc.identifier.urihttp://hdl.handle.net/10754/597024en
dc.description.abstractWe consider the Bayesian filtering problem for data assimilation following the kernel-based ensemble Gaussian-mixture filtering (EnGMF) approach introduced by Anderson and Anderson (1999). In this approach, the posterior distribution of the system state is propagated with the model using the ensemble Monte Carlo method, providing a forecast ensemble that is then used to construct a prior Gaussian-mixture (GM) based on the kernel density estimator. This results in two update steps: a Kalman filter (KF)-like update of the ensemble members and a particle filter (PF)-like update of the weights, followed by a resampling step to start a new forecast cycle. After formulating EnGMF for any observational operator, we analyze the influence of the bandwidth parameter of the kernel function on the covariance of the posterior distribution. We then focus on two aspects: i) the efficient implementation of EnGMF with (relatively) small ensembles, where we propose a new deterministic resampling strategy preserving the first two moments of the posterior GM to limit the sampling error; and ii) the analysis of the effect of the bandwidth parameter on contributions of KF and PF updates and on the weights variance. Numerical results using the Lorenz-96 model are presented to assess the behavior of EnGMF with deterministic resampling, study its sensitivity to different parameters and settings, and evaluate its performance against ensemble KFs. The proposed EnGMF approach with deterministic resampling suggests improved estimates in all tested scenarios, and is shown to require less localization and to be less sensitive to the choice of filtering parameters.en
dc.description.sponsorshipThe research reported in this publication was supported by the King Abdullah University of Science and Technology (KAUST).en
dc.language.isoenen
dc.publisherAmerican Meteorological Societyen
dc.relation.urlhttp://journals.ametsoc.org/doi/abs/10.1175/MWR-D-14-00292.1en
dc.rightsArchived with thanks to Monthly Weather Reviewen
dc.subjectForecastingen
dc.subjectEnsemblesen
dc.subjectForecasting techniquesen
dc.subjectProbability forecasts/models/distributionen
dc.subjectModels and modelingen
dc.subjectData assimilationen
dc.titleEfficient Kernel-Based Ensemble Gaussian Mixture Filteringen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.identifier.journalMonthly Weather Reviewen
dc.eprint.versionPublisher's Version/PDFen
dc.contributor.affiliationKing Abdullah University of Science and Technology (KAUST)en
kaust.authorLiu, Boen
kaust.authorAit-El-Fquih, Boujemaaen
kaust.authorHoteit, Ibrahimen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.