Combined Estimation of Hydrogeologic Conceptual Model, Parameter, and Scenario Uncertainty with Application to Uranium Transport at the Hanford Site 300 Area

Meyer, Philip D.; Ye, Ming; Rockhold, Mark; Neuman, Shlomo P.; Cantrell, Kirk J.

doi:10.2172/921264

Cited by 43 publications

(52 citation statements)

References 82 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…It has been recently suggested that uncertainties in groundwater model predictions are largely dominated by this type of uncertainty (conceptual model uncertainty) and that accounting for parametric uncertainty solely does not allow compensating for conceptual model uncertainty in model predictions (Bredehoeft, 2003;Neuman and Wierenga, 2003;Neuman, 2003;Ye et al, 2004;Højberg and Refsgaard, 2005;Poeter and Anderson, 2005;Bredehoeft, 2005;Refsgaard et al, 2006Refsgaard et al, , 2007Meyer et al, 2007;Seifert et al, 2008;Rojas et al, 2008). This is especially important for the case when predicted variables are not included in the data used for calibration (Højberg and Refsgaard, 2005;Troldborg et al, 2007).…”

Section: Introduction and Scopementioning

confidence: 99%

“…Harrar et al, 2003;Højberg and Refsgaard, 2005;Poeter and Anderson, 2005;Meyer et al, 2007;Rojas et al, 2008;Ijiri et al, 2009). Whereas predictions based on a single conceptualization are more likely to be biased and under-dispersive, estimates based on an ensemble of models are less (artificially) conservative and are more likely to capture the unknown true predicted value (Neuman, 2003;Rojas et al, 2008).…”

Section: Introduction and Scopementioning

confidence: 99%

See 1 more Smart Citation

Assessment of conceptual model uncertainty for the regional aquifer Pampa del Tamarugal – North Chile

Rojas

Batelaan

Feyen

et al. 2010

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract. In this work we assess the uncertainty in modelling the groundwater flow for the Pampa del Tamarugal Aquifer (PTA) -North Chile using a novel and fully integrated multi-model approach aimed at explicitly accounting for uncertainties arising from the definition of alternative conceptual models. The approach integrates the Generalized Likelihood Uncertainty Estimation (GLUE) and Bayesian Model Averaging (BMA) methods. For each member of an ensemble M of potential conceptualizations, model weights used in BMA for multi-model aggregation are obtained from GLUE-based likelihood values. These model weights are based on model performance, thus, reflecting how well a conceptualization reproduces an observed dataset D. GLUEbased cumulative predictive distributions for each member of M are then aggregated obtaining predictive distributions accounting for conceptual model uncertainties. For the PTA we propose an ensemble of eight alternative conceptualizations covering all major features of groundwater flow models independently developed in past studies and including two recharge mechanisms which have been source of debate for several years. Results showed that accounting for heterogeneities in the hydraulic conductivity field (a) reduced the uncertainty in the estimations of parameters and state variables, and (b) increased the corresponding model weights used for multi-model aggregation. This was more noticeable when the hydraulic conductivity field was conditioned on available hydraulic conductivity measurements. ContriCorrespondence to: R. Rojas (rodrigo.rojas@jrc.ec.europa.eu) bution of conceptual model uncertainty to the predictive uncertainty varied between 6% and 64% for ground water head estimations and between 16% and 79% for ground water flow estimations. These results clearly illustrate the relevance of conceptual model uncertainty.

show abstract

Section: Introduction and Scopementioning

confidence: 99%

Section: Introduction and Scopementioning

confidence: 99%

Assessment of conceptual model uncertainty for the regional aquifer Pampa del Tamarugal – North Chile

Rojas

Batelaan

Feyen

et al. 2010

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

show abstract

“…The saturated thickness is variable over the site, ranging from about 5 m to 8 m, with daily and seasonal fluctuations of the water table in response to changes in the river stage. The prior estimates of hydraulic conductivity are 1000-100 000 m/day for the Hanford formation and 0.01-3.00 m/day for the Ringold formation (Meyer et al, 2007).…”

Section: Site and Experiments Descriptionmentioning

confidence: 99%

“…The aquifer in the Hanford formation has been very difficult to characterize, since typical methods, such as permeameter tests and slug tests, do not provide reliable results due to the coarse-grained and highly permeable nature of the aquifer (Meyer et al, 2007;Newcomer, 2008;Vermeul et al, 2009). In addition, traditional analysis of pumping and injection tests yields only the averaged property over a large domain, since the zone-of-influence expands very rapidly.…”

Section: Site and Experiments Descriptionmentioning

confidence: 99%

“…The prior distributions of each parameter sufficiently cover possible values from the historical data at the site (Meyer et al, 2007) or literature values for similar geological formations (Rubin, 2003). The uniform distributions are considered to be less informative than Gaussian distributions, which have been commonly used in the Bayesian geostatistical inversion (Li et al, 2005).…”

Section: Prior Distribution For Mad Inversionmentioning

confidence: 99%

See 1 more Smart Citation

Bayesian approach for three-dimensional aquifer characterization at the Hanford 300 Area

Murakami

Chen

Hahn

et al. 2010

Hydrol. Earth Syst. Sci.

View full text Add to dashboard Cite

Abstract.This study presents a stochastic, threedimensional characterization of a heterogeneous hydraulic conductivity field within the Hanford 300 Area, Washington, USA, by assimilating large-scale, constant-rate injection test data with small-scale, three-dimensional electromagnetic borehole flowmeter (EBF) measurement data. We first inverted the injection test data to estimate the transmissivity field, using zeroth-order temporal moments of pressure buildup curves. We applied a newly developed Bayesian geostatistical inversion framework, the method of anchored distributions (MAD), to obtain a joint posterior distribution of geostatistical parameters and local log-transmissivities at multiple locations. The unique aspects of MAD that make it suitable for this purpose are its ability to integrate multiscale, multi-type data within a Bayesian framework and to compute a nonparametric posterior distribution. After we combined the distribution of transmissivities with depthdiscrete relative-conductivity profile from the EBF data, we inferred the three-dimensional geostatistical parameters of the log-conductivity field, using the Bayesian model-based geostatistics. Such consistent use of the Bayesian approach throughout the procedure enabled us to systematically incorporate data uncertainty into the final posterior distribution.

show abstract

Assessment of parametric uncertainty for groundwater reactive transport modeling

Shi

Curtis

et al. 2014

Water Resour. Res.

Self Cite

View full text Add to dashboard Cite

The validity of using Gaussian assumptions for model residuals in uncertainty quantification of a groundwater reactive transport model was evaluated in this study. Least squares regression methods explicitly assume Gaussian residuals, and the assumption leads to Gaussian likelihood functions, model parameters, and model predictions. While the Bayesian methods do not explicitly require the Gaussian assumption, Gaussian residuals are widely used. This paper shows that the residuals of the reactive transport model are non-Gaussian, heteroscedastic, and correlated in time; characterizing them requires using a generalized likelihood function such as the formal generalized likelihood function developed by Schoups and Vrugt (2010). For the surface complexation model considered in this study for simulating uranium reactive transport in groundwater, parametric uncertainty is quantified using the least squares regression methods and Bayesian methods with both Gaussian and formal generalized likelihood functions. While the least squares methods and Bayesian methods with Gaussian likelihood function produce similar Gaussian parameter distributions, the parameter distributions of Bayesian uncertainty quantification using the formal generalized likelihood function are non-Gaussian. In addition, predictive performance of formal generalized likelihood function is superior to that of least squares regression and Bayesian methods with Gaussian likelihood function. The Bayesian uncertainty quantification is conducted using the differential evolution adaptive metropolis (DREAM ( zs ) ) algorithm; as a Markov chain Monte Carlo (MCMC) method, it is a robust tool for quantifying uncertainty in groundwater reactive transport models. For the surface complexation model, the regression-based local sensitivity analysis and Morris-and DREAM (ZS) -based global sensitivity analysis yield almost identical ranking of parameter importance. The uncertainty analysis may help select appropriate likelihood functions, improve model calibration, and reduce predictive uncertainty in other groundwater reactive transport and environmental modeling.

show abstract

Combined Estimation of Hydrogeologic Conceptual Model, Parameter, and Scenario Uncertainty with Application to Uranium Transport at the Hanford Site 300 Area

Cited by 43 publications

References 82 publications

Assessment of conceptual model uncertainty for the regional aquifer Pampa del Tamarugal – North Chile

Assessment of conceptual model uncertainty for the regional aquifer Pampa del Tamarugal – North Chile

Bayesian approach for three-dimensional aquifer characterization at the Hanford 300 Area

Assessment of parametric uncertainty for groundwater reactive transport modeling

Contact Info

Product

Resources

About