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 ClMCMC, ClHMC, MC-ClHMC, and MC-ClHMC are presented. ClMCMC uses a Gaussian proposal density to sample the posterior, and ClHMC is an extension to the Hamiltonian Monte-Carlo (HMC) sampling filter. MC-ClMCMC and MC-ClHMC are multi-chain versions of the cluster sampling filters ClMCMC and ClHMC 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. Numerical results demonstrate the usefulness of using GMMs to relax the Gaussian prior assumption especially in the HMC filtering paradigm. |
Year | DOI | Venue |
---|---|---|
2016 | 10.3390/atmos9060213 | ATMOSPHERE |
Keywords | Field | DocType |
data assimilation,ensemble filters,markov chain monte-carlo sampling,hamiltonian monte-carlo,gaussian mixture models | Meteorology,Mixture distribution,Likelihood function,Markov chain Monte Carlo,Algorithm,Posterior probability,Gaussian,Sampling (statistics),Prior probability,Statistics,Mixture model,Mathematics | Journal |
Volume | Issue | ISSN |
9 | 6 | 2073-4433 |
Citations | PageRank | References |
4 | 0.64 | 2 |
Authors | ||
3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Ahmed Attia | 1 | 13 | 3.80 |
Azam S. Zavar Moosavi | 2 | 14 | 4.12 |
Adrian Sandu | 3 | 325 | 58.93 |