Gaussian process emulators for quantifying uncertainty in CO2 spreading predictions in heterogeneous media

Abstract We explore the use of Gaussian process emulators (GPE) in the numerical simulation of CO 2 injection into a deep heterogeneous aquifer. The model domain is a two-dimensional, log-normally distributed stochastic permeability field. We first estimate the cumulative distribution functions (CDFs) of the CO 2 breakthrough time and the total CO 2 mass using a computationally expensive Monte Carlo (MC) simulation. We then show that we can accurately reproduce these CDF estimates with a GPE, using only a small fraction of the computational cost required by traditional MC simulation. In order to build a GPE that can predict the simulator output from a permeability field consisting of 1000s of values, we use a truncated Karhunen-Loeve (K-L) expansion of the permeability field, which enables the application of the Bayesian functional regression approach. We perform a cross-validation exercise to give an insight of the optimization of the experiment design for selected scenarios: we find that it is sufficient to use 100s values for the size of training set and that it is adequate to use as few as 15 K-L components. Our work demonstrates that GPE with truncated K-L expansion can be effectively applied to uncertainty analysis associated with modelling of multiphase flow and transport processes in heterogeneous media.

[1]  A. O'Hagan,et al.  Bayesian inference for the uncertainty distribution of computer model outputs , 2002 .

[2]  Michael L. Stein,et al.  Interpolation of spatial data , 1999 .

[3]  Jared T. Freiburg,et al.  Understanding the (co)variance in petrophysical properties of CO2 reservoirs comprising sedimentary architecture , 2016 .

[4]  A. Niemi,et al.  Assessing CO2 storage capacity in the Dalders Monocline of the Baltic Sea Basin using dynamic models of varying complexity , 2015 .

[5]  David Crevillen Garcia,et al.  Uncertainty quantification for flow and transport in porous media , 2016 .

[6]  Dan Cornford,et al.  Quantifying Simulator Discrepancy in Discrete-Time Dynamical Simulators , 2011 .

[7]  Neil D. Lawrence,et al.  Kernels for Vector-Valued Functions: a Review , 2011, Found. Trends Mach. Learn..

[8]  F. James,et al.  Monte Carlo theory and practice , 1980 .

[9]  Michael Kühn,et al.  The impact of heterogeneity on the distribution of CO2: Numerical simulation of CO2 storage at Ketzin , 2010 .

[10]  Integrated simulations of CO2 spreading and pressure response in the multilayer saline aquifer of South Scania Site, Sweden , 2016 .

[11]  T. J. Mitchell,et al.  Exploratory designs for computational experiments , 1995 .

[12]  Jason E. Heath,et al.  Evaluation of CO2 Storage Mechanisms in CO2 Enhanced Oil Recovery Sites: Application to Morrow Sandstone Reservoir , 2016 .

[13]  M. D. McKay,et al.  A comparison of three methods for selecting values of input variables in the analysis of output from a computer code , 2000 .

[14]  W. Illman,et al.  Geostatistical reduced‐order models in underdetermined inverse problems , 2013 .

[15]  Auli Niemi,et al.  Upscaling of the constitutive relationships for CO2 migration in multimodal heterogeneous formations , 2013 .

[16]  Amanda Bordelon,et al.  Potential chemical impacts of CO2 leakage on underground source of drinking water assessed by quantitative risk analysis , 2016 .

[17]  Klaus Ritter,et al.  Multilevel Monte Carlo Approximation of Distribution Functions and Densities , 2015, SIAM/ASA J. Uncertain. Quantification.

[18]  Christine Doughty,et al.  Modeling geologic storage of carbon dioxide: Comparison of non-hysteretic and hysteretic characteristic curves , 2006 .

[19]  Front dynamics of supercritical non‐Boussinesq gravity currents , 2006 .

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

[21]  Zunsheng Jiao,et al.  Simulation of industrial-scale CO2 storage: Multi-scale heterogeneity and its impacts on storage capacity, injectivity and leakage , 2012 .

[22]  D. Crevillén-García,et al.  Gaussian process modelling for uncertainty quantification in convectively-enhanced dissolution processes in porous media , 2017 .

[23]  Christopher F. Brown,et al.  Modeling the impact of carbon dioxide leakage into an unconfined, oxidizing carbonate aquifer , 2016 .

[24]  Mathieu Sauvageau,et al.  Adapted sequential Gaussian simulations with Bayesian approach to evaluate the CO2 storage potential in low porosity environment , 2014 .

[25]  Weon Shik Han,et al.  Effects of permeability on CO2 trapping mechanisms and buoyancy‐driven CO2 migration in saline formations , 2010 .

[26]  A. O'Hagan,et al.  Predicting the output from a complex computer code when fast approximations are available , 2000 .

[27]  Aki Vehtari,et al.  Bayesian Modeling with Gaussian Processes using the GPstuff Toolbox , 2012, 1206.5754.

[28]  Frances Y. Kuo,et al.  Quasi-Monte Carlo methods for elliptic PDEs with random coefficients and applications , 2011, J. Comput. Phys..

[29]  Mark White,et al.  CO2 Accounting and Risk Analysis for CO2 Sequestration at Enhanced Oil Recovery Sites. , 2016, Environmental science & technology.

[30]  Dongxiao Zhang,et al.  Effect of spatial heterogeneity on plume distribution and dilution during CO2 sequestration , 2011 .

[31]  A. Niemi,et al.  Effects of permeability heterogeneity on CO2 injectivity and storage efficiency coefficient , 2016 .

[32]  George Kuczera,et al.  Understanding predictive uncertainty in hydrologic modeling: The challenge of identifying input and structural errors , 2010 .

[33]  C. Doughty,et al.  Modeling Supercritical Carbon Dioxide Injection in Heterogeneous Porous Media , 2004, Vadose Zone Journal.

[34]  Zhenxue Dai,et al.  Pre-site characterization risk analysis for commercial-scale carbon sequestration. , 2014, Environmental science & technology.

[35]  Karsten Pruess,et al.  Multiphase flow dynamics during CO2 disposal into saline aquifers , 2002 .

[36]  Matthew Flett,et al.  Heterogeneous saline formations for carbon dioxide disposal: Impact of varying heterogeneity on containment and trapping , 2007 .

[37]  Jan M. Nordbotten,et al.  Uncertainties in practical simulation of CO2 storage , 2012 .

[38]  Ruben Juanes,et al.  Impact of relative permeability hysteresis on geological CO2 storage , 2006 .

[39]  Mohamad Reza Soltanian,et al.  Influence of small‐scale fluvial architecture on CO2 trapping processes in deep brine reservoirs , 2014, 1407.0369.

[40]  S. Lakshminarasimhan,et al.  Geologic factors controlling CO2 storage capacity and permanence: case studies based on experience with heterogeneity in oil and gas reservoirs applied to CO2 storage , 2008 .

[41]  Jonny Rutqvist,et al.  A comparative review of hydrologic issues involved in geologic storage of CO2 and injection disposal of liquid waste , 2008 .

[42]  Karsten Pruess,et al.  ECO2N – A fluid property module for the TOUGH2 code for studies of CO2 storage in saline aquifers , 2007 .