Assessing and Improving the Robustness of Bayesian Evidential Learning in One Dimension for Inverting Time-Domain Electromagnetic Data: Introducing a New Threshold Procedure - 2024
Assessing and Improving the Robustness of Bayesian Evidential Learning in One Dimension for Inverting Time-Domain Electromagnetic Data: Introducing a New Threshold Procedure
BEL1D; saltwater intrusion; SimPEG; TDEM; uncertainty; Bayesian; Geophysical methods; Hydrogeological; One dimension; Saltwater intrusion; Stochastics; Time-domain electromagnetic data; Uncertainty; Biochemistry; Geography, Planning and Development; Aquatic Science; Water Science and Technology
Abstract :
[en] Understanding the subsurface is of prime importance for many geological and hydrogeological applications. Geophysical methods offer an economical alternative for investigating the subsurface compared to costly borehole investigations. However, geophysical results are commonly obtained through deterministic inversion of data whose solution is non-unique. Alternatively, stochastic inversions investigate the full uncertainty range of the obtained models, yet are computationally more expensive. In this research, we investigate the robustness of the recently introduced Bayesian evidential learning in one dimension (BEL1D) for the stochastic inversion of time-domain electromagnetic data (TDEM). First, we analyse the impact of the accuracy of the numerical forward solver on the posterior distribution, and derive a compromise between accuracy and computational time. We also introduce a threshold-rejection method based on the data misfit after the first iteration, circumventing the need for further BEL1D iterations. Moreover, we analyse the impact of the prior-model space on the results. We apply the new BEL1D with a threshold approach on field data collected in the Luy River catchment (Vietnam) to delineate saltwater intrusions. Our results show that the proper selection of time and space discretization is essential for limiting the computational cost while maintaining the accuracy of the posterior estimation. The selection of the prior distribution has a direct impact on fitting the observed data and is crucial for a realistic uncertainty quantification. The application of BEL1D for stochastic TDEM inversion is an efficient approach, as it allows us to estimate the uncertainty at a limited cost.
Disciplines :
Geological, petroleum & mining engineering
Author, co-author :
Ahmed, Arsalan ; Department of Geology, Ghent University, Krijgslaan, Belgium
Aigner, Lukas; Research Unit Geophysics, Department of Geodesy and Geoinformation, TU Wien, Vienna, Austria
Michel, Hadrien ; Université de Liège - ULiège > Département ArGEnCo > Géophysique appliquée
Deleersnyder, Wouter ; Department of Geology, Ghent University, Krijgslaan, Belgium ; Department of Physics, KU Leuven Campus Kortrijk—KULAK, Kortrijk, Belgium
Dudal, David ; Department of Physics, KU Leuven Campus Kortrijk—KULAK, Kortrijk, Belgium ; Department of Physics and Astronomy, Ghent University, Gent, Belgium
Flores Orozco, Adrian ; Research Unit Geophysics, Department of Geodesy and Geoinformation, TU Wien, Vienna, Austria
Hermans, Thomas ; Department of Geology, Ghent University, Krijgslaan, Belgium
Language :
English
Title :
Assessing and Improving the Robustness of Bayesian Evidential Learning in One Dimension for Inverting Time-Domain Electromagnetic Data: Introducing a New Threshold Procedure
HEC - Higher Education Commission UGent - Ghent University FWO - Fonds Wetenschappelijk Onderzoek Vlaanderen KU Leuven - Catholic University of Leuven F.R.S.-FNRS - Fonds de la Recherche Scientifique
Funding text :
AA: was funded by the Higher Education Commission (HEC) of Pakistan, with the reference number 50028442/PM/UESTPs/HRD/HEC/2022/December 5, 2022, and from the Bijzonder Onderzoeksfonds (BOF) of Ghent University, with grant number BOF.CDV.2023.0058.01. WD was funded by the Fund for Scientific Research (FWO) in Flanders, with grants 1113020N and 1113022N and the KU Leuven Postdoctoral Mandate PDMt223065. HM was funded by the F.R.S,-FNRS, with grants 32905391 and 40000777.
Kemna A. Nguyen F. Gossen S. On Linear Model Uncertainty Computation in Electrical Imaging Proceedings of the SIAM Conference on Mathematical and Computational Issues in the Geosciences Santa Fe, NM, USA 19–22 March 2007
Hermans T. Vandenbohede A. Lebbe L. Nguyen F. A Shallow Geothermal Experiment in a Sandy Aquifer Monitored Using Electric Resistivity Tomography Geophysics 2012 77 B11 B21 10.1190/geo2011-0199.1
Parsekian A.D. Grombacher D. Uncertainty Estimates for Surface Nuclear Magnetic Resonance Water Content and Relaxation Time Profiles from Bootstrap Statistics J. Appl. Geophys. 2015 119 61 70 10.1016/j.jappgeo.2015.05.005
Aster R. Borchers B. Thurber C. Parameter Estimation and Inverse Problems Academic Press Cambridge, MA, USA 2013
Sambridge M. Monte Carlo Methods in Geophysical Inverse Problems Rev. Geophys. 2002 40 3-1 3-29 10.1029/2000RG000089
Yeh W.W.-G. Review of Parameter Identification Procedures in Groundwater Hydrology: The Inverse Problem Water Resour. Res. 1986 22 95 108 10.1029/WR022i002p00095
McLaughlin D. Townley L.R. A Reassessment of the Groundwater Inverse Problem Water Resour. Res. 1996 32 1131 1161 10.1029/96WR00160
Hidalgo J. Slooten L.J. Carrera J. Alcolea A. Medina A. Inverse Problem in Hydrogeology Hydrogeol. J. 2005 13 206 222 10.1007/s10040-004-0404-7
Zhou H. Gómez-Hernández J.J. Li L. Inverse Methods in Hydrogeology: Evolution and Recent Trends Adv. Water Resour. 2014 63 22 37 10.1016/j.advwatres.2013.10.014
Sen M.K. Stoffa P.L. Bayesian Inference, Gibbs’ Sampler, and Uncertainty Estimation in Geophysical Inversion Geophys. Prospect. 1996 44 313 350 10.1111/j.1365-2478.1996.tb00152.x
Tarantola A. Inverse Problem Theory and Methods for Model Parameter Estimation Society for Industrial and Applied Mathematics Philadelphia, PA, USA 2005
Irving J. Singha K. Stochastic Inversion of Tracer Test and Electrical Geophysical Data to Estimate Hydraulic Conductivities Water Resour. Res. 2010 46 W11514 10.1029/2009WR008340
Trainor-Guitton W. Hoversten G.M. Stochastic Inversion for Electromagnetic Geophysics: Practical Challenges and Improving Convergence Efficiency Geophysics 2011 76 F373 F386 10.1190/geo2010-0223.1
De Pasquale G. Linde N. On Structure-Based Priors in Bayesian Geophysical Inversion Geophys. J. Int. 2017 208 1342 1358 10.1093/gji/ggw458
Ball L.B. Davis T.A. Minsley B.J. Gillespie J.M. Landon M.K. Probabilistic Categorical Groundwater Salinity Mapping from Airborne Electromagnetic Data Adjacent to California’s Lost Hills and Belridge Oil Fields Water Resour. Res. 2020 56 e2019WR026273 10.1029/2019WR026273
Bobe C. Van De Vijver E. Keller J. Hanssens D. Van Meirvenne M. De Smedt P. Probabilistic 1-D Inversion of Frequency-Domain Electromagnetic Data Using a Kalman Ensemble Generator IEEE Trans. Geosci. Remote Sens. 2020 58 3287 3297 10.1109/TGRS.2019.2953004
Tso C.-H.M. Iglesias M. Wilkinson P. Kuras O. Chambers J. Binley A. Efficient Multiscale Imaging of Subsurface Resistivity with Uncertainty Quantification Using Ensemble Kalman Inversion Geophys. J. Int. 2021 225 887 905 10.1093/gji/ggab013
Michel H. Hermans T. Kremer T. Elen A. Nguyen F. 1D Geological Imaging of the Subsurface from Geophysical Data with Bayesian Evidential Learning. Part 2: Applications and Software Comput. Geosci. 2020 138 104456 10.1016/j.cageo.2020.104456
Michel H. Hermans T. Nguyen F. Iterative Prior Resampling and Rejection Sampling to Improve 1D Geophysical Imaging Based on Bayesian Evidential Learning (BEL1D) Geophys. J. Int. 2023 232 958 974 10.1093/gji/ggac372
Scheidt C. Renard P. Caers J. Prediction-Focused Subsurface Modeling: Investigating the Need for Accuracy in Flow-Based Inverse Modeling Math. Geosci. 2015 47 173 191 10.1007/s11004-014-9521-6
Satija A. Scheidt C. Li L. Caers J. Direct Forecasting of Reservoir Performance Using Production Data without History Matching Comput. Geosci. 2017 21 315 333 10.1007/s10596-017-9614-7
Hermans T. Nguyen F. Klepikova M. Dassargues A. Caers J. Uncertainty Quantification of Medium-Term Heat Storage from Short-Term Geophysical Experiments Using Bayesian Evidential Learning Water Resour. Res. 2018 54 2931 2948 10.1002/2017WR022135
Athens N.D. Caers J.K. A Monte Carlo-Based Framework for Assessing the Value of Information and Development Risk in Geothermal Exploration Appl. Energy 2019 256 113932 10.1016/j.apenergy.2019.113932
Thibaut R. Compaire N. Lesparre N. Ramgraber M. Laloy E. Hermans T. Comparing Well and Geophysical Data for Temperature Monitoring within a Bayesian Experimental Design Framework Water Resour. Res. 2022 58 e2022WR033045 10.1029/2022WR033045
Park J. Caers J. Direct Forecasting of Global and Spatial Model Parameters from Dynamic Data Comput. Geosci. 2020 143 104567 10.1016/j.cageo.2020.104567
Yin Z. Strebelle S. Caers J. Automated Monte Carlo-Based Quantification and Updating of Geological Uncertainty with Borehole Data (AutoBEL v1.0) Geosci. Model Dev. 2020 13 651 672 10.5194/gmd-13-651-2020
Tadjer A. Bratvold R.B. Managing Uncertainty in Geological CO2 Storage Using Bayesian Evidential Learning Energies 2021 14 1557 10.3390/en14061557
Fang J. Gong B. Caers J. Data-Driven Model Falsification and Uncertainty Quantification for Fractured Reservoirs Engineering 2022 18 116 128 10.1016/j.eng.2022.04.015
Thibaut R. Laloy E. Hermans T. A New Framework for Experimental Design Using Bayesian Evidential Learning: The Case of Wellhead Protection Area J. Hydrol. 2021 603 126903 10.1016/j.jhydrol.2021.126903
Yang H.-Q. Chu J. Qi X. Wu S. Chiam K. Bayesian Evidential Learning of Soil-Rock Interface Identification Using Boreholes Comput. Geotech. 2023 162 105638 10.1016/j.compgeo.2023.105638
Hermans T. Oware E. Caers J. Direct Prediction of Spatially and Temporally Varying Physical Properties from Time-Lapse Electrical Resistance Data Water Resour. Res. 2016 52 7262 7283 10.1002/2016WR019126
Pradhan A. Mukerji T. Seismic Bayesian Evidential Learning: Estimation and Uncertainty Quantification of Sub-Resolution Reservoir Properties Comput. Geosci. 2020 24 1121 1140 10.1007/s10596-019-09929-1
Heagy L.J. Cockett R. Kang S. Rosenkjaer G.K. Oldenburg D.W. A Framework for Simulation and Inversion in Electromagnetics Comput. Geosci. 2017 82 WB9 WB19 10.1016/j.cageo.2017.06.018
Cockett R. Kang S. Heagy L.J. Pidlisecky A. Oldenburg D.W. SimPEG: An Open Source Framework for Simulation and Gradient Based Parameter Estimation in Geophysical Applications Comput. Geosci. 2015 85 142 154 10.1016/j.cageo.2015.09.015
Chongo M. Christiansen A.V. Tembo A. Banda K.E. Nyambe I.A. Larsen F. Bauer-Gottwein P. Airborne and Ground-Based Transient Electromagnetic Mapping of Groundwater Salinity in the Machile–Zambezi Basin, Southwestern Zambia Near Surf. Geophys. 2015 13 383 396 10.3997/1873-0604.2015024
Mohamaden M. Araffa S.A. Taha A. AbdelRahman M.A. El-Sayed H.M. Sharkawy M.S. Geophysical Techniques and Geomatics-Based Mapping for Groundwater Exploration and Sustainable Development at Sidi Barrani Area, Egypt Egypt. J. Aquat. Res. 2023 in press 10.1016/j.ejar.2023.12.001
Rey J. Martínez J.M. Mendoza R. Sandoval S. Tarasov V. Kaminsky A. Morales K. Geophysical Characterization of Aquifers in Southeast Spain Using ERT, TDEM, and Vertical Seismic Reflection Appl. Sci. 2020 10 7365 10.3390/app10207365
Siemon B. Christiansen A.V. Auken E. A Review of Helicopter-Borne Electromagnetic Methods for Groundwater Exploration Near Surf. Geophys. 2009 7 629 646 10.3997/1873-0604.2009043
Vilhelmsen T.N. Marker P.A. Foged N. Wernberg T. Auken E. Christiansen A.V. Bauer-Gottwein P. Christensen S. Høyer A.-S. A Regional Scale Hydrostratigraphy Generated from Geophysical Data of Varying Age, Type, and Quality Water Resour. Manag. 2018 33 539 553 10.1007/s11269-018-2115-1
Goebel M. Knight R. Halkjær M. Mapping Saltwater Intrusion with an Airborne Electromagnetic Method in the Offshore Coastal Environment, Monterey Bay, California J. Hydrol. Reg. Stud. 2019 23 100602 10.1016/j.ejrh.2019.100602
Auken E. Foged N. Larsen J.J. Lassen K.V.T. Maurya P.K. Dath S.M. Eiskjær T.T. tTEM—A Towed Transient Electromagnetic System for Detailed 3D Imaging of the Top 70 m of the Subsurface Geophysics 2019 84 E13 E22 10.1190/geo2018-0355.1
Lane J.W. Jr. Briggs M.A. Maurya P.K. White E.A. Pedersen J.B. Auken E. Terry N. Minsley B.K. Kress W. LeBlanc D.R. et al. Characterizing the Diverse Hydrogeology Underlying Rivers and Estuaries Using New Floating Transient Electromagnetic Methodology Sci. Total Environ. 2020 740 140074 10.1016/j.scitotenv.2020.140074 32927542
Aigner L. Högenauer P. Bücker M. Flores Orozco A. A Flexible Single Loop Setup for Water-Borne Transient Electromagnetic Sounding Applications Sensors 2021 21 6624 10.3390/s21196624 34640941
Viezzoli A. Christiansen A.V. Auken E. Sørensen K. Quasi-3D Modeling of Airborne TEM Data by Spatially Constrained Inversion Geophysics 2008 73 F105 F113 10.1190/1.2895521
Linde N. Ginsbourger D. Irving J. Nobile F. Doucet A. On Uncertainty Quantification in Hydrogeology and Hydrogeophysics Adv. Water Resour. 2017 110 166 181 10.1016/j.advwatres.2017.10.014
Josset L. Ginsbourger D. Lunati I. Functional Error Modeling for Uncertainty Quantification in Hydrogeology Water Resour. Res. 2015 51 1050 1068 10.1002/2014WR016028
Scholer M. Irving J. Looms M.C. Nielsen L. Holliger K. Bayesian Markov-Chain-Monte-Carlo Inversion of Time-Lapse Crosshole GPR Data to Characterize the Vadose Zone at the Arrenaes Site, Denmark Vadose Zone J. 2012 11 vzj2011.0153 10.2136/vzj2011.0153
Calvetti D. Ernst O. Somersalo E. Dynamic Updating of Numerical Model Discrepancy Using Sequential Sampling Inverse Probl. 2014 30 114019 10.1088/0266-5611/30/11/114019
Köpke C. Irving J. Elsheikh A.H. Accounting for Model Error in Bayesian Solutions to Hydrogeophysical Inverse Problems Using a Local Basis Approach Adv. Water Resour. 2018 116 195 207 10.1016/j.advwatres.2017.11.013
Deleersnyder W. Dudal D. Hermans T. Multidimensional Surrogate Modelling for Airborne TDEM Data Accepted Pending Revision Comput. Geosci. 2024 10.48550/arXiv.2311.13998
Khu S.-T. Werner M.G.F. Reduction of Monte-Carlo Simulation Runs for Uncertainty Estimation in Hydrological Modelling Hydrol. Earth Syst. Sci. 2003 7 680 692 10.5194/hess-7-680-2003
Brynjarsdóttir J. O’Hagan A. Learning About Physical Parameters: The Importance of Model Discrepancy Inverse Probl. 2014 30 114007 10.1088/0266-5611/30/11/114007
Hansen T.M. Cordua K.S. Jacobsen B.H. Mosegaard K. Accounting for Imperfect Forward Modeling in Geophysical Inverse Problems—Exemplified for Crosshole Tomography Geophysics 2014 79 H1 H21 10.1190/geo2013-0215.1
Deleersnyder W. Maveau B. Hermans T. Dudal D. Inversion of Electromagnetic Induction Data Using a Novel Wavelet-Based and Scale-Dependent Regularization Term Geophys. J. Int. 2021 226 1715 1729 10.1093/gji/ggab182
Deleersnyder W. Dudal D. Hermans T. Novel Airborne EM Image Appraisal Tool for Imperfect Forward Modelling Remote Sens. 2022 14 5757 10.3390/rs14225757
Bai P. Vignoli G. Hansen T.M. 1D Stochastic Inversion of Airborne Time-Domain Electromagnetic Data with Realistic Prior and Accounting for the Forward Modeling Error Remote Sens. 2021 13 3881 10.3390/rs13193881
Minsley B.J. Foks N.L. Bedrosian P.A. Quantifying Model Structural Uncertainty Using Airborne Electromagnetic Data Geophys. J. Int. 2021 224 590 607 10.1093/gji/ggaa393
Michel H. pyBEL1D: Latest Version of pyBEL1D version 1.1.0 Zenodo Genève, Switzerland 14 July 2022 10.5281/zenodo.6833249
Krzanowski W.J. Principles of Multivariate Analysis: A User’s Perspective Oxford University Press New York, NY, USA 2000
Thibaut R. Machine Learning for Bayesian Experimental Design in the Subsurface Ph.D. Dissertation Ghent University, Faculty of Sciences Ghent, Belgium 2023 10.22541/essoar.168057575.58936022/v1
Wand M.P. Jones M.C. Comparison of Smoothing Parameterizations in Bivariate Kernel Density Estimation J. Am. Stat. Assoc. 1993 88 520 10.1080/01621459.1993.10476303
Devroye L. An Automatic Method for Generating Random Variates with a Given Characteristic Function SIAM J. Appl. Math. 1986 46 698 719 10.1137/0146046
Werthmüller D. An Open-Source Full 3D Electromagnetic Modeler for 1D VTI Media in Python: Empymod Geophysics 2017 82 WB9 WB19 10.1190/geo2016-0626.1
Yee K. Numerical Solution of Initial Boundary Value Problems Involving Maxwell’s Equations in Isotropic Media IEEE Trans. Antennas Propag. 1966 14 302 307 10.1109/TAP.1966.1138693
Haber E. Computational Methods in Geophysical Electromagnetics Society for Industrial and Applied Mathematics Philadelphia, PA, USA 2014 10.1137/1.9781611973808
Hyman J. Morel J. Shashkov M. Steinberg S. Mimetic Finite Difference Methods for Diffusion Equations Comput. Geosci. 2002 6 333 352 10.1023/A:1021282912658
Hyman J.M. Shashkov M. Mimetic Discretizations for Maxwell’s Equations J. Comput. Phys. 1999 151 881 909 10.1006/jcph.1999.6225
Diep C.-T. Linh P.D. Thibaut R. Paepen M. Ho H.H. Nguyen F. Hermans T. Imaging the Structure and the Saltwater Intrusion Extent of the Luy River Coastal Aquifer (Binh Thuan, Vietnam) Using Electrical Resistivity Tomography Water 2021 13 1743 10.3390/w13131743
Dieu L.P. Cong-Thi D. Segers T. Ho H.H. Nguyen F. Hermans T. Groundwater Salinization and Freshening Processes in the Luy River Coastal Aquifer, Vietnam Water 2022 14 2358 10.3390/w14152358
Cong-Thi D. Dieu L.P. Caterina D. De Pauw X. Thi H.D. Ho H.H. Nguyen F. Hermans T. Quantifying Salinity in Heterogeneous Coastal Aquifers through ERT and IP: Insights from Laboratory and Field Investigations J. Contam. Hydrol. 2024 262 104322 10.1016/j.jconhyd.2024.104322
Aigner L. Werthmüller D. Flores Orozco A. Sensitivity Analysis of Inverted Model Parameters from Transient Electromagnetic Measurements Affected by Induced Polarization Effects J. Appl. Geophys. 2024 223 105334 10.1016/j.jappgeo.2024.105334
Hermans T. Lesparre N. De Schepper G. Robert T. Bayesian Evidential Learning: A Field Validation Using Push-Pull Tests Hydrogeol. J. 2019 27 1661 1672 10.1007/s10040-019-01962-9
