A limited memory acceleration strategy for MCMC sampling in hierarchical Bayesian calibration of hydrological models

Simple item page
dc.contributor.authorKuczera, G.
dc.contributor.authorKavetski, D.
dc.contributor.authorRenard, B.
dc.contributor.authorThyer, M.
dc.date.issued2010
dc.description.abstractHydrological calibration and prediction using conceptual models is affected by forcing/response data uncertainty and structural model error. The Bayesian Total Error Analysis methodology uses a hierarchical representation of individual sources of uncertainty. However, it is shown that standard multiblock “Metropolis-within-Gibbs” Markov chain Monte Carlo (MCMC) samplers commonly used in Bayesian hierarchical inference are exceedingly computationally expensive when applied to hydrologic models, which use recursive numerical solutions of coupled nonlinear differential equations to describe the evolution of catchment states such as soil and groundwater storages. This note develops a “limited-memory” algorithm for accelerating multiblock MCMC sampling from the posterior distributions of such models using low-dimensional jump distributions. The new algorithm exploits the decaying memory of hydrological systems to provide accurate tolerance-based approximations of traditional “full-memory” MCMC methods and is orders of magnitude more efficient than the latter.
dc.description.statementofresponsibilityGeorge Kuczera, Dmitri Kavetski, Benjamin Renard and Mark Thyer
dc.identifier.citationWater Resources Research, 2010; 46(7):1-6
dc.identifier.doi10.1029/2009WR008985
dc.identifier.issn0043-1397
dc.identifier.issn1944-7973
dc.identifier.orcidKavetski, D. [0000-0003-4966-9234]
dc.identifier.orcidThyer, M. [0000-0002-2830-516X]
dc.identifier.urihttp://hdl.handle.net/2440/65580
dc.language.isoen
dc.publisherAmer Geophysical Union
dc.rightsCopyright 2010 by the American Geophysical Union
dc.source.urihttps://doi.org/10.1029/2009wr008985
dc.titleA limited memory acceleration strategy for MCMC sampling in hierarchical Bayesian calibration of hydrological models
dc.typeJournal article
pubs.publication-statusPublished

Files

Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
hdl_65580.pdf
Size:
262.46 KB
Format:
Adobe Portable Document Format
Description:
Published version
Download

Collections

Civil and Environmental Engineering publications
Environment Institute publications