Maximum likelihood Bayesian averaging of spatial variability models in unsaturated fractured tuff

Ming Ye, Shlomo P Neuman, Philip D. Meyer

Research output: Contribution to journalArticle

154 Citations (Scopus)

Abstract

Hydrologic analyses typically rely on a single conceptual-mathematical model. Yet hydrologic environments are open and complex, rendering them prone to multiple interpretations and mathematical descriptions. Adopting only one of these may lead to statistical bias and underestimation of uncertainty. Bayesian model averaging (BMA) [Hoeting et al., 1999] provides an optimal way to combine the predictions of several competing models and to assess their joint predictive uncertainty. However, it tends to be computationally demanding and relies heavily on prior information about model parameters. Neuman [2002, 2003] proposed a maximum likelihood version (MLBMA) of BMA to render it computationally feasible and to allow dealing with cases where reliable prior information is lacking. We apply MLBMA to seven alternative variogram models of log air permeability data from single-hole pneumatic injection tests in six boreholes at the Apache Leap Research Site (ALRS) in central Arizona. Unbiased ML estimates of variogram and drift parameters are obtained using adjoint state maximum likelihood cross validation [Samper and Neuman, 1989a] in conjunction with universal kriging and generalized least squares. Standard information criteria provide an ambiguous ranking of the models, which does not justify selecting one of them and discarding all others as is commonly done in practice. Instead, we eliminate some of the models based on their negligibly small posterior probabilities and use the rest to project the measured log permeabilities by kriging onto a rock volume containing the six boreholes. We then average these four projections and associated kriging variances, using the posterior probability of each model as weight. Finally, we cross validate the results by eliminating from consideration all data from one borehole at a time, repeating the above process and comparing the predictive capability of MLBMA with that of each individual model. We find that MLBMA is superior to any individual geostatistical model of log permeability among those we consider at the ALRS.

Original languageEnglish (US)
JournalWater Resources Research
Volume40
Issue number5
StatePublished - May 2004

Fingerprint

tuff
Maximum likelihood
kriging
Boreholes
permeability
borehole
variogram
uncertainty
Air permeability
air permeability
rendering
Pneumatics
least squares
ranking
mathematical models
rocks
Rocks
Mathematical models
injection

Keywords

  • Conceptual model uncertainty
  • Cross validation
  • Drift
  • Predictive performance
  • Predictive uncertainty
  • Stochastic continuum

ASJC Scopus subject areas

  • Environmental Science(all)
  • Environmental Chemistry
  • Aquatic Science
  • Water Science and Technology

Cite this

Maximum likelihood Bayesian averaging of spatial variability models in unsaturated fractured tuff. / Ye, Ming; Neuman, Shlomo P; Meyer, Philip D.

In: Water Resources Research, Vol. 40, No. 5, 05.2004.

Research output: Contribution to journalArticle

@article{0222d4e4834d479391f70488333c7ca0,
title = "Maximum likelihood Bayesian averaging of spatial variability models in unsaturated fractured tuff",
abstract = "Hydrologic analyses typically rely on a single conceptual-mathematical model. Yet hydrologic environments are open and complex, rendering them prone to multiple interpretations and mathematical descriptions. Adopting only one of these may lead to statistical bias and underestimation of uncertainty. Bayesian model averaging (BMA) [Hoeting et al., 1999] provides an optimal way to combine the predictions of several competing models and to assess their joint predictive uncertainty. However, it tends to be computationally demanding and relies heavily on prior information about model parameters. Neuman [2002, 2003] proposed a maximum likelihood version (MLBMA) of BMA to render it computationally feasible and to allow dealing with cases where reliable prior information is lacking. We apply MLBMA to seven alternative variogram models of log air permeability data from single-hole pneumatic injection tests in six boreholes at the Apache Leap Research Site (ALRS) in central Arizona. Unbiased ML estimates of variogram and drift parameters are obtained using adjoint state maximum likelihood cross validation [Samper and Neuman, 1989a] in conjunction with universal kriging and generalized least squares. Standard information criteria provide an ambiguous ranking of the models, which does not justify selecting one of them and discarding all others as is commonly done in practice. Instead, we eliminate some of the models based on their negligibly small posterior probabilities and use the rest to project the measured log permeabilities by kriging onto a rock volume containing the six boreholes. We then average these four projections and associated kriging variances, using the posterior probability of each model as weight. Finally, we cross validate the results by eliminating from consideration all data from one borehole at a time, repeating the above process and comparing the predictive capability of MLBMA with that of each individual model. We find that MLBMA is superior to any individual geostatistical model of log permeability among those we consider at the ALRS.",
keywords = "Conceptual model uncertainty, Cross validation, Drift, Predictive performance, Predictive uncertainty, Stochastic continuum",
author = "Ming Ye and Neuman, {Shlomo P} and Meyer, {Philip D.}",
year = "2004",
month = "5",
language = "English (US)",
volume = "40",
journal = "Water Resources Research",
issn = "0043-1397",
publisher = "American Geophysical Union",
number = "5",

}

TY - JOUR

T1 - Maximum likelihood Bayesian averaging of spatial variability models in unsaturated fractured tuff

AU - Ye, Ming

AU - Neuman, Shlomo P

AU - Meyer, Philip D.

PY - 2004/5

Y1 - 2004/5

N2 - Hydrologic analyses typically rely on a single conceptual-mathematical model. Yet hydrologic environments are open and complex, rendering them prone to multiple interpretations and mathematical descriptions. Adopting only one of these may lead to statistical bias and underestimation of uncertainty. Bayesian model averaging (BMA) [Hoeting et al., 1999] provides an optimal way to combine the predictions of several competing models and to assess their joint predictive uncertainty. However, it tends to be computationally demanding and relies heavily on prior information about model parameters. Neuman [2002, 2003] proposed a maximum likelihood version (MLBMA) of BMA to render it computationally feasible and to allow dealing with cases where reliable prior information is lacking. We apply MLBMA to seven alternative variogram models of log air permeability data from single-hole pneumatic injection tests in six boreholes at the Apache Leap Research Site (ALRS) in central Arizona. Unbiased ML estimates of variogram and drift parameters are obtained using adjoint state maximum likelihood cross validation [Samper and Neuman, 1989a] in conjunction with universal kriging and generalized least squares. Standard information criteria provide an ambiguous ranking of the models, which does not justify selecting one of them and discarding all others as is commonly done in practice. Instead, we eliminate some of the models based on their negligibly small posterior probabilities and use the rest to project the measured log permeabilities by kriging onto a rock volume containing the six boreholes. We then average these four projections and associated kriging variances, using the posterior probability of each model as weight. Finally, we cross validate the results by eliminating from consideration all data from one borehole at a time, repeating the above process and comparing the predictive capability of MLBMA with that of each individual model. We find that MLBMA is superior to any individual geostatistical model of log permeability among those we consider at the ALRS.

AB - Hydrologic analyses typically rely on a single conceptual-mathematical model. Yet hydrologic environments are open and complex, rendering them prone to multiple interpretations and mathematical descriptions. Adopting only one of these may lead to statistical bias and underestimation of uncertainty. Bayesian model averaging (BMA) [Hoeting et al., 1999] provides an optimal way to combine the predictions of several competing models and to assess their joint predictive uncertainty. However, it tends to be computationally demanding and relies heavily on prior information about model parameters. Neuman [2002, 2003] proposed a maximum likelihood version (MLBMA) of BMA to render it computationally feasible and to allow dealing with cases where reliable prior information is lacking. We apply MLBMA to seven alternative variogram models of log air permeability data from single-hole pneumatic injection tests in six boreholes at the Apache Leap Research Site (ALRS) in central Arizona. Unbiased ML estimates of variogram and drift parameters are obtained using adjoint state maximum likelihood cross validation [Samper and Neuman, 1989a] in conjunction with universal kriging and generalized least squares. Standard information criteria provide an ambiguous ranking of the models, which does not justify selecting one of them and discarding all others as is commonly done in practice. Instead, we eliminate some of the models based on their negligibly small posterior probabilities and use the rest to project the measured log permeabilities by kriging onto a rock volume containing the six boreholes. We then average these four projections and associated kriging variances, using the posterior probability of each model as weight. Finally, we cross validate the results by eliminating from consideration all data from one borehole at a time, repeating the above process and comparing the predictive capability of MLBMA with that of each individual model. We find that MLBMA is superior to any individual geostatistical model of log permeability among those we consider at the ALRS.

KW - Conceptual model uncertainty

KW - Cross validation

KW - Drift

KW - Predictive performance

KW - Predictive uncertainty

KW - Stochastic continuum

UR - http://www.scopus.com/inward/record.url?scp=2942700220&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=2942700220&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:2942700220

VL - 40

JO - Water Resources Research

JF - Water Resources Research

SN - 0043-1397

IS - 5

ER -