Efficient nonlinear predictive error variance for highly parameterized models

Predictive error variance analysis attempts to determine how wrong predictions made by a calibrated model may be. Predictive error variance analysis is usually undertaken following calibration using a small number of parameters defined through a priori parsimony. In contrast, we introduce a method for investigating the potential error in predictions made by highly parameterized models calibrated using regularized inversion. Vecchia and Cooley (1987) describe a method of predictive error variance analysis that is constrained by calibration data. We extend this approach to include constraints on parameters that lie within the calibration null space. These constraints are determined by dividing parameter space into combinations of parameters for which estimates can be obtained and those for which they cannot. This enables the contribution to predictive error variance from parameterization simplifications required to solve the inverse problem to be quantified, in addition to the contribution from measurement noise. We also describe a novel technique that restricts the analysis to a strategically defined predictive solution subspace, enabling an approximate predictive error variance analysis to be completed efficiently. The method is illustrated using a synthetic and a real-world groundwater flow and transport model.

[1]  John Doherty,et al.  Ground Water Model Calibration Using Pilot Points and Regularization , 2003, Ground water.

[2]  John David Wilson,et al.  MODFLOW-2000, The U.S. Geological Survey Modular Ground-Water Model -- GMG Linear Equation Solver Package Documentation , 2004 .

[3]  Andrés Sahuquillo,et al.  Stochastic simulation of transmissivity fields conditional to both transmissivity and piezometric data 2. Demonstration on a synthetic aquifer , 1997 .

[4]  Catherine Moore.  The use of regularized inversion in groundwater model calibration and prediction uncertainty analysis , 2006 .

[5]  P. Kitanidis.  On the geostatistical approach to the inverse problem , 1996 .

[6]  John Fox,et al.  Applied Regression Analysis and Generalized Linear Models , 2008 .

[7]  Steen Christensen,et al.  Bias and uncertainty in regression-calibrated models of groundwater flow in heterogeneous media , 2006 .

[8]  S. P. Neuman,et al.  Estimation of Aquifer Parameters Under Transient and Steady State Conditions: 1. Maximum Likelihood Method Incorporating Prior Information , 1986 .

[9]  Richard L. Cooley,et al.  Simultaneous confidence and prediction intervals for nonlinear regression models with application to a groundwater flow model , 1987 .

[10]  G. Dagan,et al.  Stochastic identification of transmissivity and effective recharge in steady groundwater flow: 2. Case study , 1987 .

[11]  Andres Alcolea,et al.  Inverse problem in hydrogeology , 2005 .

[12]  W. Menke.  Geophysical data analysis , 1984 .

[13]  J. Gómez-Hernández,et al.  Stochastic conditional inverse modeling of subsurface mass transport: A brief review and the self-calibrating method , 2003 .

[14]  C. Vogel.  Computational Methods for Inverse Problems , 1987 .

[15]  Alberto Guadagnini,et al.  Nonlocal and localized analyses of conditional mean steady state flow in bounded, randomly nonuniform domains: 2. Computational examples , 1999 .

[16]  A. N. Tikhonov,et al.  Solutions of ill-posed problems , 1977 .

[17]  Paul P. Wang,et al.  MT3DMS: A Modular Three-Dimensional Multispecies Transport Model for Simulation of Advection, Dispersion, and Chemical Reactions of Contaminants in Groundwater Systems; Documentation and User's Guide , 1999 .

[18]  Clifford H. Thurber,et al.  Parameter estimation and inverse problems , 2005 .

[19]  N. Draper,et al.  Applied Regression Analysis , 1966 .

[20]  Catherine Certes,et al.  Application of the pilot point method to the identification of aquifer transmissivities , 1991 .

[21]  J. Doherty,et al.  The cost of uniqueness in groundwater model calibration , 2006 .

[22]  E. Haber,et al.  On optimization techniques for solving nonlinear inverse problems , 2000 .

[23]  Alberto Guadagnini,et al.  Nonlocal and localized analyses of conditional mean steady state flow in bounded, randomly nonuniform domains: 1. Theory and computational approach , 1999 .

[24]  S. P. Neuman,et al.  Estimation of aquifer parameters under transient and steady-state conditions: 2 , 1986 .

[25]  David W. Pollock,et al.  User's guide for MODPATH/MODPATH-PLOT, Version 3; a particle tracking post-processing package for MODFLOW, the U.S. Geological Survey finite-difference ground-water flow model , 1994 .

[26]  O. L. Franke,et al.  Summary of the hydrologic situation on Long Island, New York, as a guide to water-management alternatives , 1972 .

[27]  Minghui Jin,et al.  AN ITERATIVE STOCHASTIC INVERSE METHOD: CONDITIONAL EFFECTIVE TRANSMISSIVITY AND HYDRAULIC HEAD FIELDS , 1995 .

[28]  Keith Beven,et al.  The future of distributed models: model calibration and uncertainty prediction. , 1992 .

[29]  Ghislain de Marsily,et al.  Three‐dimensional interference test interpretation in a fractured aquifer using the Pilot Point Inverse Method , 2001 .

[30]  D. A. Zimmerman,et al.  A comparison of seven geostatistically based inverse approaches to estimate transmissivities for modeling advective transport by groundwater flow , 1998 .

[31]  Sean Hatch,et al.  Joint conditional simulations and the spectral approach for flow modeling , 1994 .

[32]  J. Doherty,et al.  A hybrid regularized inversion methodology for highly parameterized environmental models , 2005 .

[33]  S. P. Neuman,et al.  Inverse stochastic moment analysis of steady state flow in randomly heterogeneous media , 2006 .

[34]  D. Oliver,et al.  Markov chain Monte Carlo methods for conditioning a permeability field to pressure data , 1997 .

[35]  R. L. Cooley.  A theory for modeling ground-water flow in heterogeneous media , 2004 .

[36]  Steen Christensen,et al.  Evaluation of prediction intervals for expressing uncertainties in groundwater flow model predictions , 1999 .

[37]  Arlen W. Harbaugh,et al.  MODFLOW-2000, The U.S. Geological Survey Modular Ground-Water Model - User Guide to Modularization Concepts and the Ground-Water Flow Process , 2000 .

[38]  Clayton V. Deutsch,et al.  GSLIB: Geostatistical Software Library and User's Guide , 1993 .

[39]  George Kuczera,et al.  Monte Carlo assessment of parameter uncertainty in conceptual catchment models: the Metropolis algorithm , 1998 .

[40]  Andrés Sahuquillo,et al.  Coupled inverse modelling of groundwater flow and mass transport and the worth of concentration data , 2003 .

[41]  T. Ulrych,et al.  A full‐Bayesian approach to the groundwater inverse problem for steady state flow , 2000 .

[42]  W. Menke.  Geophysical data analysis : discrete inverse theory , 1984 .

[43]  J. Doherty,et al.  Role of the calibration process in reducing model predictive error , 2005 .

[44]  D. L. Simmons,et al.  Thickness and hydrogeology of aquifers and confining units below the upper glacial aquifer on Long Island, New York , 1987 .