- Title
- A limited-memory acceleration strategy for MCMC sampling in hierarchical Bayesian calibration of hydrological models
- Creator
- Kuczera, George; Kavetski, Dmitri; Renard, Benjamin; Thyer, Mark
- Relation
- Water Resources Research Vol. 46
- Publisher Link
- http://dx.doi.org/10.1029/2009WR008985
- Publisher
- American Geophysical Union
- Resource Type
- journal article
- Date
- 2010
- Description
- Hydrological 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.
- Subject
- nonlinear equations; approximation algorithms; Bayesian networks; calibration; catchments; groundwater
- Identifier
- http://hdl.handle.net/1959.13/930946
- Identifier
- uon:10965
- Identifier
- ISSN:0043-1397
- Language
- eng
- Full Text
- Reviewed
- Hits: 2308
- Visitors: 2666
- Downloads: 479
Thumbnail | File | Description | Size | Format | |||
---|---|---|---|---|---|---|---|
View Details Download | ATTACHMENT01 | Publisher version (open access) | 255 KB | Adobe Acrobat PDF | View Details Download |