Cluster Sampling Filters for Non-Gaussian Data Assimilation
Abstract
This paper presents a fully non-Gaussian filter for sequential data assimilation. The filter is named the “cluster sampling filter”, and works by directly sampling the posterior distribution following a Markov Chain Monte-Carlo (MCMC) approach, while the prior distribution is approximated using a Gaussian Mixture Model (GMM). Specifically, a clustering step is introduced after the forecast phase of the filter, and the prior density function is estimated by fitting a GMM to the prior ensemble. Using the data likelihood function, the posterior density is then formulated as a mixture density, and is sampled following an MCMC approach. Four versions of the proposed filter, namely C ℓ MCMC , C ℓ HMC , MC- C ℓ HMC , and MC- C ℓ HMC are presented. C ℓ MCMC uses a Gaussian proposal density to sample the posterior, and C ℓ HMC is an extension to the Hamiltonian Monte-Carlo (HMC) sampling filter. MC- C ℓ MCMC and MC- C ℓ HMC are multi-chain versions of the cluster sampling filters C ℓ MCMC and C ℓ HMC respectively. The multi-chain versions are proposed to guarantee that samples are taken from the vicinities of all probability modes of the formulated posterior. The new methodologies aremore »
- Authors:
-
- Argonne National Lab. (ANL), Argonne, IL (United States)
- Virginia Polytechnic Inst. and State Univ. (Virginia Tech), Blacksburg, VA (United States)
- Publication Date:
- Research Org.:
- Argonne National Laboratory (ANL), Argonne, IL (United States)
- Sponsoring Org.:
- Air Force Research Laboratory (AFRL), Air Force Office of Scientific Research (AFOSR); National Science Foundation (NSF); Virginia Polytechnic Institute, Dept of Computer Science; USDOE
- OSTI Identifier:
- 1463676
- Grant/Contract Number:
- AC02-06CH11357
- Resource Type:
- Accepted Manuscript
- Journal Name:
- Atmosphere (Basel)
- Additional Journal Information:
- Journal Name: Atmosphere (Basel); Journal Volume: 9; Journal Issue: 6; Journal ID: ISSN 2073-4433
- Publisher:
- MDPI
- Country of Publication:
- United States
- Language:
- English
- Subject:
- 97 MATHEMATICS AND COMPUTING; data assimilation; ensemble filters; gaussian mixture models; hamiltonian monte-carlo; markov chain monte-carlo sampling
Citation Formats
Attia, Ahmed, Moosavi, Azam, and Sandu, Adrian. Cluster Sampling Filters for Non-Gaussian Data Assimilation. United States: N. p., 2018.
Web. doi:10.3390/atmos9060213.
Attia, Ahmed, Moosavi, Azam, & Sandu, Adrian. Cluster Sampling Filters for Non-Gaussian Data Assimilation. United States. https://doi.org/10.3390/atmos9060213
Attia, Ahmed, Moosavi, Azam, and Sandu, Adrian. Thu .
"Cluster Sampling Filters for Non-Gaussian Data Assimilation". United States. https://doi.org/10.3390/atmos9060213. https://www.osti.gov/servlets/purl/1463676.
@article{osti_1463676,
title = {Cluster Sampling Filters for Non-Gaussian Data Assimilation},
author = {Attia, Ahmed and Moosavi, Azam and Sandu, Adrian},
abstractNote = {This paper presents a fully non-Gaussian filter for sequential data assimilation. The filter is named the “cluster sampling filter”, and works by directly sampling the posterior distribution following a Markov Chain Monte-Carlo (MCMC) approach, while the prior distribution is approximated using a Gaussian Mixture Model (GMM). Specifically, a clustering step is introduced after the forecast phase of the filter, and the prior density function is estimated by fitting a GMM to the prior ensemble. Using the data likelihood function, the posterior density is then formulated as a mixture density, and is sampled following an MCMC approach. Four versions of the proposed filter, namely C ℓ MCMC , C ℓ HMC , MC- C ℓ HMC , and MC- C ℓ HMC are presented. C ℓ MCMC uses a Gaussian proposal density to sample the posterior, and C ℓ HMC is an extension to the Hamiltonian Monte-Carlo (HMC) sampling filter. MC- C ℓ MCMC and MC- C ℓ HMC are multi-chain versions of the cluster sampling filters C ℓ MCMC and C ℓ HMC respectively. The multi-chain versions are proposed to guarantee that samples are taken from the vicinities of all probability modes of the formulated posterior. The new methodologies are tested using a simple one-dimensional example, and a quasi-geostrophic (QG) model with double-gyre wind forcing and bi-harmonic friction. Here, numerical results demonstrate the usefulness of using GMMs to relax the Gaussian prior assumption especially in the HMC filtering paradigm.},
doi = {10.3390/atmos9060213},
journal = {Atmosphere (Basel)},
number = 6,
volume = 9,
place = {United States},
year = {Thu May 31 00:00:00 EDT 2018},
month = {Thu May 31 00:00:00 EDT 2018}
}
Web of Science
Works referenced in this record:
Estimating the Dimension of a Model
journal, March 1978
- Schwarz, Gideon
- The Annals of Statistics, Vol. 6, Issue 2
Hybrid Monte Carlo
journal, September 1987
- Duane, Simon; Kennedy, A. D.; Pendleton, Brian J.
- Physics Letters B, Vol. 195, Issue 2
The Maximum Likelihood Ensemble Filter as a non‐differentiable minimization algorithm
journal, April 2008
- Zupanski, Milija; Navon, I. Michael; Zupanski, Dusanka
- Quarterly Journal of the Royal Meteorological Society, Vol. 134, Issue 633
Sequential data assimilation with a nonlinear quasi-geostrophic model using Monte Carlo methods to forecast error statistics
journal, January 1994
- Evensen, Geir
- Journal of Geophysical Research, Vol. 99, Issue C5
Analysis methods for numerical weather prediction
journal, October 1986
- Lorenc, A. C.
- Quarterly Journal of the Royal Meteorological Society, Vol. 112, Issue 474
Data Assimilation Using an Ensemble Kalman Filter Technique
journal, March 1998
- Houtekamer, P. L.; Mitchell, Herschel L.
- Monthly Weather Review, Vol. 126, Issue 3
Evaluation of probabilistic prediction systems for a scalar variable
journal, July 2005
- Candille, G.; Talagrand, O.
- Quarterly Journal of the Royal Meteorological Society, Vol. 131, Issue 609
Distance-Dependent Filtering of Background Error Covariance Estimates in an Ensemble Kalman Filter
journal, November 2001
- Hamill, Thomas M.; Whitaker, Jeffrey S.; Snyder, Chris
- Monthly Weather Review, Vol. 129, Issue 11
A deterministic formulation of the ensemble Kalman filter: an alternative to ensemble square root filters
journal, January 2008
- Sakov, Pavel; Oke, Peter R.
- Tellus A: Dynamic Meteorology and Oceanography, Vol. 60, Issue 2
Cluster ensemble Kalman filter
journal, January 2007
- Smith, Keston W.
- Tellus A: Dynamic Meteorology and Oceanography, Vol. 59, Issue 5
The Ensemble Kalman Filter: theoretical formulation and practical implementation
journal, November 2003
- Evensen, Geir
- Ocean Dynamics, Vol. 53, Issue 4
A Sequential Ensemble Kalman Filter for Atmospheric Data Assimilation
journal, January 2001
- Houtekamer, P. L.; Mitchell, Herschel L.
- Monthly Weather Review, Vol. 129, Issue 1
An Iterative Ensemble Kalman Filter for Multiphase Fluid Flow Data Assimilation
journal, November 2007
- Gu, Yaqing; Oliver, Dean S.
- SPE Journal, Vol. 12, Issue 04
Auxiliary Variable Methods for Markov Chain Monte Carlo with Applications
journal, June 1998
- Higdon, David M.
- Journal of the American Statistical Association, Vol. 93, Issue 442
A Hybrid Monte Carlo Sampling Filter for Non-Gaussian Data Assimilation
journal, January 2015
- Attia, Ahmed; Sandu, Adrian
- AIMS Geosciences, Vol. 1, Issue 1
Analysis Scheme in the Ensemble Kalman Filter
journal, June 1998
- Burgers, Gerrit; Jan van Leeuwen, Peter; Evensen, Geir
- Monthly Weather Review, Vol. 126, Issue 6
Construction of correlation functions in two and three dimensions
journal, January 1999
- Gaspari, Gregory; Cohn, Stephen E.
- Quarterly Journal of the Royal Meteorological Society, Vol. 125, Issue 554
A Method for Producing and Evaluating Probabilistic Forecasts from Ensemble Model Integrations
journal, July 1996
- Anderson, Jeffrey L.
- Journal of Climate, Vol. 9, Issue 7
A Monte Carlo Implementation of the Nonlinear Filtering Problem to Produce Ensemble Assimilations and Forecasts
journal, December 1999
- Anderson, Jeffrey L.; Anderson, Stephen L.
- Monthly Weather Review, Vol. 127, Issue 12
A New Approach to Linear Filtering and Prediction Problems
journal, March 1960
- Kalman, R. E.
- Journal of Basic Engineering, Vol. 82, Issue 1
Ensemble Data Assimilation without Perturbed Observations
journal, July 2002
- Whitaker, Jeffrey S.; Hamill, Thomas M.
- Monthly Weather Review, Vol. 130, Issue 7
New Results in Linear Filtering and Prediction Theory
journal, March 1961
- Kalman, R. E.; Bucy, R. S.
- Journal of Basic Engineering, Vol. 83, Issue 1
The reduced-order hybrid Monte Carlo sampling smoother: The reduced-order hybrid Monte Carlo sampling smoother
journal, June 2016
- Attia, Ahmed; Ştefănescu, Răzvan; Sandu, Adrian
- International Journal for Numerical Methods in Fluids, Vol. 83, Issue 1
Riemann manifold Langevin and Hamiltonian Monte Carlo methods: Riemann Manifold Langevin and Hamiltonian Monte Carlo Methods
journal, March 2011
- Girolami, Mark; Calderhead, Ben
- Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 73, Issue 2
A Hybrid Monte-Carlo sampling smoother for four-dimensional data assimilation: HMC SAMPLING SMOOTHER FOR 4D-DA
journal, June 2016
- Attia, Ahmed; Rao, Vishwas; Sandu, Adrian
- International Journal for Numerical Methods in Fluids, Vol. 83, Issue 1
Ensemble Square Root Filters*
journal, July 2003
- Tippett, Michael K.; Anderson, Jeffrey L.; Bishop, Craig H.
- Monthly Weather Review, Vol. 131, Issue 7
Measures of multivariate skewness and kurtosis with applications
journal, January 1970
- Mardia, K. V.
- Biometrika, Vol. 57, Issue 3
Construction of correlation functions in two and three dimensions
journal, January 1999
- Gaspari, Gregory; Cohn, Stephen E.
- Quarterly Journal of the Royal Meteorological Society, Vol. 125, Issue 554
Auxiliary Variable Methods for Markov Chain Monte Carlo with Applications
journal, June 1998
- Higdon, David M.
- Journal of the American Statistical Association, Vol. 93, Issue 442
Spatial Statistics and Bayesian Computation
journal, September 1993
- Besag, Julian; Green, Peter J.
- Journal of the Royal Statistical Society: Series B (Methodological), Vol. 55, Issue 1
An Iterative EnKF for Strongly Nonlinear Systems
journal, June 2012
- Sakov, Pavel; Oliver, Dean S.; Bertino, Laurent
- Monthly Weather Review, Vol. 140, Issue 6
Maximum Likelihood Ensemble Filter: Theoretical Aspects
journal, June 2005
- Zupanski, Milija
- Monthly Weather Review, Vol. 133, Issue 6
Covariance Tapering for Interpolation of Large Spatial Datasets
journal, September 2006
- Furrer, Reinhard; Genton, Marc G.; Nychka, Douglas
- Journal of Computational and Graphical Statistics, Vol. 15, Issue 3
Evaluation of probabilistic prediction systems for a scalar variable
journal, July 2005
- Candille, G.; Talagrand, O.
- Quarterly Journal of the Royal Meteorological Society, Vol. 131, Issue 609
A deterministic formulation of the ensemble Kalman filter: an alternative to ensemble square root filters
journal, March 2008
- Sakov, Pavel; Oke, Peter R.
- Tellus A
Geometrically Tempered Hamiltonian Monte Carlo
preprint, January 2016
- Nishimura, Akihiko; Dunson, David
- arXiv
Works referencing / citing this record:
Tuning Covariance Localization Using Machine Learning
book, June 2019
- Moosavi, Azam; Attia, Ahmed; Sandu, Adrian
- Computational Science – ICCS 2019: 19th International Conference, Faro, Portugal, June 12–14, 2019, Proceedings, Part IV, p. 199-212