Abstract
A linear probabilistic waveform inversion strategy is proposed for crosshole tomographic data using machine learning algorithms that integrates a priori information described by Gaussian distributed slowness fields. A theoretical framework is outlined that combines an approximate linear forward operator describing the waveform with the ridge regression algorithm. The framework approximates arbitrary geostatistical a priori information with a Gaussian distribution, which makes it possible to obtain an analytic description of the posterior probability density, the solution of the inverse problem. The suggested inversion strategy is tested on synthetic tomographic crosshole ground-penetrating radar full-waveform data generated using multiple-point-based a priori information incorporating realistic noise. The benefits of the proposed strategy include: (I) Increased resolution of the posterior probability density compared to first arrival type inversion, due to using the wavefield. (II) Much faster execution than existing full waveform inversion algorithms based on Monte Carlo sampling or the adjoint method. (III) Greater efficiency in obtaining realizations from the posterior probability density. The main challenge in using the method is that it needs a quite large training data set from which the linear forward model can be inferred.
Similar content being viewed by others
References
Alumbaugh D, Chang PY, Paprocki L, Brainard JR, Glass RJ, Rautman CA (2002) Estimating moisture contents in the vadose zone using cross-borehole ground penetrating radar: a study of accuracy and repeatability. Water Resour Res 38(12):45-1
Annan AP (2005) GPR methods for hydrogeological studies. In: Rubin Y, Hubbard SS (eds) Hydrogeophysics. Springer, Berlin, pp 185–213
Binley A, Winship P, Middleton R, Pokar M, West J (2001) High-resolution characterization of vadose zone dynamics using cross-borehole radar. Water Resour Res 37(11):2639–2652
Bishop CM (2006) Pattern recognition and machine learning. Springer, Berlin
Buland A, Omre H (2003) Bayesian wavelet estimation from seismic and well databayesian wavelet estimation. Geophysics 68(6):2000–2009
Cordua KS, Hansen TM, Mosegaard K (2012) Monte carlo full-waveform inversion of crosshole GPR data using multiple-point geostatistical a priori information. Geophysics 77(2):H19–H31
Cordua KS, Looms MC, Nielsen L (2008) Accounting for correlated data errors during inversion of cross-borehole ground penetrating radar data. Vadose Zone J 7(1):263–271
Dafflon B, Irving J, Holliger K (2009) Use of high-resolution geophysical data to characterize heterogeneous aquifers: Influence of data integration method on hydrological predictions. Water Resour Res 45(9):W09407. https://doi.org/10.1029/2008WR007646
Dahlen FA, Hung SH, Nolet G (2000) Fréchet kernels for finite-frequency traveltimes—I. Theory. Geophys J Int 141(1):157–174
Day-Lewis FD, Lane JW (2004) Assessing the resolution-dependent utility of tomograms for geostatistics. Geophys Res Lett 31(7):L07503
Day-Lewis FD, Singha K, Binley AM (2005) Applying petrophysical models to radar travel time and electrical resistivity tomograms: resolution-dependent limitations. J Geophys Res Solid Earth 110(B8):B08206
Eppstein MJ, Dougherty DE (1998) Efficient three-dimensional data inversion: Soil characterization and moisture monitoring from cross-well ground-penetrating radar at a Vermont test site. Water Resour Res 34(8):1889–1900
Ernst JR, Green AG, Maurer H, Holliger K (2007) Application of a new 2D time-domain full-waveform inversion scheme to crosshole radar data. Geophysics 72(5):J53–J64
Ernst JR, Holliger K, Maurer H, Green A (2006) Realistic FDTD modelling of borehole georadar antenna radiation: methodolgy and application. Near Surface Geophys 4(1):19–30
Gloaguen E, Giroux B, Marcotte D, Dimitrakopoulos R (2007) Pseudo-full-waveform inversion of borehole GPR data using stochastic tomography. Geophysics 72(5):J43–J51
Hansen TM, Cordua KS (2017) Efficient Monte Carlo sampling of inverse problems using a neural network-based forward–applied to GPR crosshole traveltime inversion. Geophys J Int 211(3):1524–1533
Hansen TM, Cordua KS, Jacobsen BH, Mosegaard K (2014) Accounting for imperfect forward modeling in geophysical inverse problems—exemplified for crosshole tomography. Geophysics 79(3):H1–H21
Hansen TM, Journel AG, Tarantola A, Mosegaard K (2006) Linear inverse Gaussian theory and geostatistics. Geophysics 71(6):R101–R111
Hansen TM, Looms M, Nielsen L (2008) Inferring the subsurface structural covariance model using cross-borehole ground penetrating radar tomography. Vadose Zone J 7(7):249–262
Hansen TM, Vu LT, Bach T (2016) MPSLIB: a C++ class for sequential simulation of multiple-point statistical models. SoftwareX 5:127–133, ISSN 2352-7110
Hubbard SS, Peterson JE, Majer EL, Zawislanski PT, Williams KH, Roberts J, Wobber F (1997) Estimation of permeable pathways and water content using tomographic radar data. Lead Edge 16(11):1623–1630
Jacobsen BH, Hansen TM, Cordua KS (2010) Understanding why first-arrival travel times do not obey banana-doughnut sensitivity. In: EGU general assembly conference abstracts, volume 12 of EGU General Assembly Conference Abstracts. p 6480
Journel A, Zhang T (2006) The necessity of a multiple-point prior model. Math Geol 38(5):591–610
Kaipio J, Somersalo E (2007) Statistical inverse problems: discretization, model reduction and inverse crimes. J Comput Appl Math 198(2):493–504
Keskinen J, Klotzsche A, Looms MC, Moreau J, van der Kruk J, Holliger K, Stemmerik L, Nielsen L (2017) Full-waveform inversion of crosshole GPR data: Implications for porosity estimation in chalk. J Appl Geophys 140:102–116
Kowalsky MB, Finsterle S, Peterson J, Hubbard S, Rubin Y, Majer E, Ward A, Gee G (2005) Estimation of field-scale soil hydraulic and dielectric parameters through joint inversion of GPR and hydrological data. Water Resour Res 41(11):W11425
Lassen RN, Sonnenborg TO, Jensen KH, Looms MC (2015) Monitoring \(\text{ CO }_2\) gas-phase migration in a shallow sand aquifer using cross-borehole ground penetrating radar. Int J Greenh Gas Control 37:287–298
Linde N, Binley A, Tryggvason A, Pedersen LB, Revil A (2006) Improved hydrogeophysical characterization using joint inversion of cross-hole electrical resistance and ground-penetrating radar traveltime data. Water Resour Res 42(12):W12404
Menke W (1989) Geophysical data analysis: discrete inverse theory. Academic Press Inc, Cambridge
Molyneux J, Schmitt D (1999) First break timing: arrival onset times by direct correlation. Geophysics 64(5):1492–1501
Mosegaard K, Tarantola A (2002) Probabilistic approach to inverse problems. In: Lee W, Kanamori H, Jennings P, Kisslinger C, editors, International handbook of earthquake and engineering seismology, volume 81A, chapter 16, Academic Press, Cambridge. ISBN 978-0124406520, pp 237–265
Mosser L, Dubrule O, Blunt MJ (2018) Stochastic seismic waveform inversion using generative adversarial networks as a geological prior. arXiv preprint arXiv:1806.03720
Murphy KP (2012) Machine learning: a probabilistic perspective. The MIT Press, Cambridge ISBN 0262018020, 9780262018029
Peterson JE (2001) Pre-inversion corrections and analysis of radar tomographic data. J Environ Eng Geophys 6(1):1–18
Strébelle S (2001) Sequential simulation drawing structures from training images. Ph.D. thesis, Stanford University, an optional note
Strébelle S (2002) Conditional simulation of complex geological structures using multiple-point statistics. Math Geosci 34:1–21
Tanaka JS, Huba GJ (1989) A general coefficient of determination for covariance structure models under arbitrary GLS estimation. Br J Math Stat Psychol 42(2):233–239
Tarantola A (1984) Inversion of seismic reflection data in the acoustic approximation. Geophysics 49(8):1259–1266
Tarantola A (1987) Inverse problem theory. Methods for data fitting and model parameter estimation. Elsevier Science Publ. Co. Inc., Amsterdam
Tarantola A (2005) Inverse problem theory and methods for model parameter estimation, vol 89. SIAM, Philadelphia
Tarantola A, Mosegaard K (1995) Monte carlo sampling of solutions to inverse problems. J Geophys Res Solid Earth 100(B7):12431–12447
Tarantola A, Valette B (1982a) Generalized nonlinear inverse problems solved using the least squares criterion. Rev Geophys 20(2):219–232
Tarantola A, Valette B (1982b) Inverse problems= quest for information. J Geophys 50(3):150–170
Woodward MJ (1992) Wave-equation tomography. Geophysics 57(1):15–26
Acknowledgements
We thank everyone at the Climate and Computational Geophysics group at the Niels Bohr Institute for an encouraging and appealing research environment. Thank you to Jacques Ernst and ETH Zürich, for making the finite difference modeling code available.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Holm-Jensen, T., Hansen, T.M. Linear Waveform Tomography Inversion Using Machine Learning Algorithms. Math Geosci 52, 31–51 (2020). https://doi.org/10.1007/s11004-019-09815-7
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11004-019-09815-7