Assessing clustering strategies for Gaussian mixture filtering a subsurface contaminant model

Handle URI:
http://hdl.handle.net/10754/595580
Title:
Assessing clustering strategies for Gaussian mixture filtering a subsurface contaminant model
Authors:
Liu, Bo ( 0000-0001-6615-1096 ) ; El Gharamti, Mohamad ( 0000-0002-7229-8366 ) ; Hoteit, Ibrahim ( 0000-0002-3751-4393 )
Abstract:
An ensemble-based Gaussian mixture (GM) filtering framework is studied in this paper in term of its dependence on the choice of the clustering method to construct the GM. In this approach, a number of particles sampled from the posterior distribution are first integrated forward with the dynamical model for forecasting. A GM representation of the forecast distribution is then constructed from the forecast particles. Once an observation becomes available, the forecast GM is updated according to Bayes’ rule. This leads to (i) a Kalman filter-like update of the particles, and (ii) a Particle filter-like update of their weights, generalizing the ensemble Kalman filter update to non-Gaussian distributions. We focus on investigating the impact of the clustering strategy on the behavior of the filter. Three different clustering methods for constructing the prior GM are considered: (i) a standard kernel density estimation, (ii) clustering with a specified mixture component size, and (iii) adaptive clustering (with a variable GM size). Numerical experiments are performed using a two-dimensional reactive contaminant transport model in which the contaminant concentration and the heterogenous hydraulic conductivity fields are estimated within a confined aquifer using solute concentration data. The experimental results suggest that the performance of the GM filter is sensitive to the choice of the GM model. In particular, increasing the size of the GM does not necessarily result in improved performances. In this respect, the best results are obtained with the proposed adaptive clustering scheme.
KAUST Department:
Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division; Earth Science and Engineering Program
Citation:
Assessing clustering strategies for Gaussian mixture filtering a subsurface contaminant model 2016 Journal of Hydrology
Publisher:
Elsevier BV
Journal:
Journal of Hydrology
Issue Date:
3-Feb-2016
DOI:
10.1016/j.jhydrol.2016.01.048
Type:
Article
ISSN:
00221694
Additional Links:
http://linkinghub.elsevier.com/retrieve/pii/S0022169416000664
Appears in Collections:
Articles; Earth Science and Engineering Program; Computer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division

Full metadata record

DC FieldValue Language
dc.contributor.authorLiu, Boen
dc.contributor.authorEl Gharamti, Mohamaden
dc.contributor.authorHoteit, Ibrahimen
dc.date.accessioned2016-02-04T13:26:22Zen
dc.date.available2016-02-04T13:26:22Zen
dc.date.issued2016-02-03en
dc.identifier.citationAssessing clustering strategies for Gaussian mixture filtering a subsurface contaminant model 2016 Journal of Hydrologyen
dc.identifier.issn00221694en
dc.identifier.doi10.1016/j.jhydrol.2016.01.048en
dc.identifier.urihttp://hdl.handle.net/10754/595580en
dc.description.abstractAn ensemble-based Gaussian mixture (GM) filtering framework is studied in this paper in term of its dependence on the choice of the clustering method to construct the GM. In this approach, a number of particles sampled from the posterior distribution are first integrated forward with the dynamical model for forecasting. A GM representation of the forecast distribution is then constructed from the forecast particles. Once an observation becomes available, the forecast GM is updated according to Bayes’ rule. This leads to (i) a Kalman filter-like update of the particles, and (ii) a Particle filter-like update of their weights, generalizing the ensemble Kalman filter update to non-Gaussian distributions. We focus on investigating the impact of the clustering strategy on the behavior of the filter. Three different clustering methods for constructing the prior GM are considered: (i) a standard kernel density estimation, (ii) clustering with a specified mixture component size, and (iii) adaptive clustering (with a variable GM size). Numerical experiments are performed using a two-dimensional reactive contaminant transport model in which the contaminant concentration and the heterogenous hydraulic conductivity fields are estimated within a confined aquifer using solute concentration data. The experimental results suggest that the performance of the GM filter is sensitive to the choice of the GM model. In particular, increasing the size of the GM does not necessarily result in improved performances. In this respect, the best results are obtained with the proposed adaptive clustering scheme.en
dc.language.isoenen
dc.publisherElsevier BVen
dc.relation.urlhttp://linkinghub.elsevier.com/retrieve/pii/S0022169416000664en
dc.rightsNOTICE: this is the author’s version of a work that was accepted for publication in Journal of Hydrology. Changes resulting from the publishing process, such as peer review, editing, corrections, structural formatting, and other quality control mechanisms may not be reflected in this document. Changes may have been made to this work since it was submitted for publication. A definitive version was subsequently published in Journal of Hydrology, 3 February 2016. DOI: 10.1016/j.jhydrol.2016.01.048en
dc.subjectEnsemble Kalman filteringen
dc.subjectParticle filteringen
dc.subjectGaussian mixturesen
dc.subjectClusteringen
dc.subjectSubsurface contaminant modelen
dc.titleAssessing clustering strategies for Gaussian mixture filtering a subsurface contaminant modelen
dc.typeArticleen
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Divisionen
dc.contributor.departmentEarth Science and Engineering Programen
dc.identifier.journalJournal of Hydrologyen
dc.eprint.versionPost-printen
dc.contributor.institutionMohn-Sverdrup Center for Global Ocean Studies and Operational Oceanography, Nansen Environmental and Remote Sensing Center (NERSC), Bergen 5006, Norwayen
dc.contributor.affiliationKing Abdullah University of Science and Technology (KAUST)en
kaust.authorLiu, Boen
kaust.authorEl Gharamti, Mohamaden
kaust.authorHoteit, Ibrahimen
All Items in KAUST are protected by copyright, with all rights reserved, unless otherwise indicated.