Implicit sampling combined with reduced order modeling for the inversion of vadose zone hydrological data

Abstract Bayesian inverse modeling techniques are computationally expensive because many forward simulations are needed when sampling the posterior distribution of the parameters. In this paper, we combine the implicit sampling method and generalized polynomial chaos expansion (gPCE) to significantly reduce the computational cost of performing Bayesian inverse modeling. There are three steps in this approach: (1) find the maximizer of the likelihood function using deterministic approaches; (2) construct a gPCE-based surrogate model using the results from a limited number of forward simulations; and (3) efficiently sample the posterior distribution of the parameters using implicit sampling method. The cost of constructing the gPCE-based surrogate model is further decreased by using sparse Bayesian learning to reduce the number of gPCE coefficients that have to be determined. We demonstrate the approach for a synthetic ponded infiltration experiment simulated with TOUGH2. The surrogate model is highly accurate with mean relative error that is 0.035 % in predicting saturation and 0.25 % in predicting the likelihood function. The posterior distribution of the parameters obtained using our proposed technique is nearly indistinguishable from the results obtained from either an implicit sampling method or a Markov chain Monte Carlo method utilizing the full model.

[1]  Stefan Finsterle,et al.  Joint Hydrological-Geophysical Inversion for Soil Structure Identification , 2006 .

[2]  Dongbin Xiu,et al.  The Wiener-Askey Polynomial Chaos for Stochastic Differential Equations , 2002, SIAM J. Sci. Comput..

[3]  M. Yousuff Hussaini,et al.  Parametric uncertainty quantification in the Rothermel model with randomised quasi-Monte Carlo methods , 2015 .

[4]  Clayton V. Deutsch,et al.  Geostatistical Software Library and User's Guide , 1998 .

[5]  A. Chorin,et al.  Implicit particle filtering for models with partial noise, and an application to geomagnetic data assimilation , 2011, 1109.3664.

[6]  Van Genuchten,et al.  A closed-form equation for predicting the hydraulic conductivity of unsaturated soils , 1980 .

[7]  J. Vrugt,et al.  A formal likelihood function for parameter and predictive inference of hydrologic models with correlated, heteroscedastic, and non‐Gaussian errors , 2010 .

[8]  Neil J. Gordon,et al.  A tutorial on particle filters for online nonlinear/non-Gaussian Bayesian tracking , 2002, IEEE Trans. Signal Process..

[9]  George Eastman House,et al.  Sparse Bayesian Learning and the Relevance Vector Machine , 2001 .

[10]  D. Higdon,et al.  Accelerating Markov Chain Monte Carlo Simulation by Differential Evolution with Self-Adaptive Randomized Subspace Sampling , 2009 .

[11]  Yuqiong Liu,et al.  Uncertainty in hydrologic modeling: Toward an integrated data assimilation framework , 2007 .

[12]  W. Riley,et al.  A Hybrid Reduced‐Order Model of Fine‐Resolution Hydrologic Simulations at a Polygonal Tundra Site , 2016 .

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

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

[15]  L. A. Richards.  Capillary conduction of liquids through porous mediums , 1931 .

[16]  Matthias Morzfeld,et al.  Parameter estimation by implicit sampling , 2013, 1308.4640.

[17]  A. Ramm,et al.  Inverse Problems: Mathematical and Analytical Techniques with Applications to Engineering , 2004 .

[18]  Matthias Morzfeld,et al.  Implicit particle filters for data assimilation , 2010, 1005.4002.

[19]  Daniel Foreman-Mackey,et al.  emcee: The MCMC Hammer , 2012, 1202.3665.

[20]  Nando de Freitas,et al.  An Introduction to MCMC for Machine Learning , 2004, Machine Learning.

[21]  Matthias Morzfeld,et al.  A survey of implicit particle filters for data assimilation , 2013 .

[22]  James Martin,et al.  A Stochastic Newton MCMC Method for Large-Scale Statistical Inverse Problems with Application to Seismic Inversion , 2012, SIAM J. Sci. Comput..

[23]  Dongxiao Zhang,et al.  On importance sampling Monte Carlo approach to uncertainty analysis for flow and transport in porous media , 2003 .

[24]  Jonathan R Goodman,et al.  Ensemble samplers with affine invariance , 2010 .

[25]  Yalchin Efendiev,et al.  Bayesian uncertainty quantification for flows in heterogeneous porous media using reversible jump Markov chain Monte Carlo methods , 2010 .

[26]  D. Xiu.  Numerical Methods for Stochastic Computations: A Spectral Method Approach , 2010 .

[27]  Matthias Morzfeld,et al.  A random map implementation of implicit filters , 2011, J. Comput. Phys..

[28]  Maarten Speekenbrink,et al.  A tutorial on particle filters , 2016 .

[29]  Francisco Duarte Moura Neto,et al.  An Introduction to Inverse Problems with Applications , 2012 .

[30]  Stefan Finsterle,et al.  Robust estimation of hydrogeologic model parameters , 1998 .

[31]  A. Sahuquillo,et al.  Stochastic simulation of transmissivity fields conditional to both transmissivity and piezometric data—I. Theory , 1997 .

[32]  Bryan A. Tolson,et al.  Review of surrogate modeling in water resources , 2012 .

[33]  M. Marietta,et al.  Pilot Point Methodology for Automated Calibration of an Ensemble of conditionally Simulated Transmissivity Fields: 1. Theory and Computational Experiments , 1995 .

[34]  Yaning Liu,et al.  Accurate construction of high dimensional model representation with applications to uncertainty quantification , 2016, Reliab. Eng. Syst. Saf..

[35]  Peter E. Thornton,et al.  DIMENSIONALITY REDUCTION FOR COMPLEX MODELS VIA BAYESIAN COMPRESSIVE SENSING , 2014 .

[36]  Michael E. Tipping,et al.  Fast Marginal Likelihood Maximisation for Sparse Bayesian Models , 2003 .

[37]  Jun S. Liu,et al.  Sequential Monte Carlo methods for dynamic systems , 1997 .

[38]  Ioannis Kyriakides,et al.  Bayesian Compressive Sensing in Radar Systems , 2013 .

[39]  George Shu Heng Pau,et al.  Reduced order modeling in iTOUGH2 , 2014, Comput. Geosci..

[40]  A. Chorin,et al.  Implicit Particle Methods and Their Connection with Variational Data Assimilation , 2012, 1205.1830.

[41]  J. C. Ramírez,et al.  Estimation of aquifer parameters under transient and steady-state conditions , 1984 .

[42]  Aria Abubakar,et al.  Inversion algorithms for large-scale geophysical electromagnetic measurements , 2009 .

[43]  Brian D. Ripley,et al.  Stochastic Simulation , 2005 .

[44]  S. Kabanikhin.  Definitions and examples of inverse and ill-posed problems , 2008 .

[45]  Albert Tarantola,et al.  Inverse problem theory - and methods for model parameter estimation , 2004 .

[46]  Aggelos K. Katsaggelos,et al.  Bayesian Compressive Sensing Using Laplace Priors , 2010, IEEE Transactions on Image Processing.

[47]  H. Rabitz,et al.  Efficient input-output model representations , 1999 .

[48]  A. Pettitt,et al.  Introduction to MCMC , 2012 .

[49]  Ming Ye,et al.  Evaluating two sparse grid surrogates and two adaptation criteria for groundwater Bayesian uncertainty quantification , 2016 .

[50]  A. Chorin,et al.  Implicit sampling for particle filters , 2009, Proceedings of the National Academy of Sciences.