1887
  1. Home
  2. A-Z Publications
  3. Geophysical Prospecting
  4. Volume 70, Issue 5
  5. Article
Volume 70, Issue 5
  • E-ISSN: 1365-2478
PDF

Abstract

ABSTRACT

Casting a geophysical inverse problem into a Bayesian setting is often discouraged by the computational workload needed to run many forward modelling evaluations. Here we present probabilistic inversions of electrical resistivity tomography data in which the forward operator is replaced by a trained residual neural network that learns the non‐linear mapping between the resistivity model and the apparent resistivity values. The use of this specific architecture can provide some advantages over standard convolutional networks as it mitigates the vanishing gradient problem that might affect deep networks. The modelling error introduced by the network approximation is properly taken into account and propagated onto the estimated model uncertainties. One crucial aspect of any machine learning application is the definition of an appropriate training set. We draw the models forming the training and validation sets from previously defined prior distributions, while a finite element code provides the associated datasets. We apply the approach to two probabilistic inversion frameworks: A Markov chain Monte Carlo algorithm is applied to synthetic data, while an ensemble‐based algorithm is employed for the field measurements. For both the synthetic and field tests, the outcomes of the proposed method are benchmarked against the predictions obtained when the finite element code constitutes the forward operator. Our experiments illustrate that the network can effectively approximate the forward mapping even when a relatively small training set is created. The proposed strategy provides a forward operator that is three orders of magnitude faster than the accurate but computationally expensive finite element code. Our approach also yields the most likely solutions and uncertainty quantifications comparable to those estimated when the finite element modelling is employed. The presented method allows solving the Bayesian electrical resistivity tomography with a reasonable computational cost and limited hardware resources.

© 2022 European Association of Geoscientists & Engineers. © 2022 The Authors. Geophysical Prospecting published by John Wiley & Sons Ltd on behalf of European Association of Geoscientists & Engineers.
Loading

Article metrics loading...

/content/journals/10.1111/1365-2478.13189
2022-05-18
2024-04-20

  • From This Site

    /content/journals/10.1111/1365-2478.13189
    dcterms_title,dcterms_subject,pub_keyword
    -contentType:Journal -contentType:Contributor -contentType:Concept -contentType:Institution
    10
    5
Loading full text...

Full text loading...

/deliver/fulltext/gpr/70/5/gpr13189.html?itemId=/content/journals/10.1111/1365-2478.13189&mimeType=html&fmt=ahah

References

  1. Aleardi, M. (2020) Combining discrete cosine transform and convolutional neural networks to speed up the Hamiltonian Monte Carlo inversion of pre‐stack seismic data. Geophysical Prospecting, 68(9), 2738–2761.
     [Google Scholar]
  2. Aleardi, M. (2021) A gradient‐based Markov chain Monte Carlo algorithm for elastic pre‐stack inversion with data and model space reduction. Geophysical Prospecting, 69(5), 926–948.
     [Google Scholar]
  3. Aleardi, M. and Salusti, A. (2020) Hamiltonian Monte Carlo algorithms for target‐and interval‐oriented amplitude versus angle inversions. Geophysics, 85(3), R177–R194.
     [Google Scholar]
  4. Aleardi, M. and Salusti, A. (2021) Elastic prestack seismic inversion through discrete cosine transform reparameterization and convolutional neural networks. Geophysics, 86(1), R129–R146.
     [Google Scholar]
  5. Aleardi, M., Vinciguerra, A. and Hojat, A. (2020) A geostatistical Markov chain Monte Carlo inversion algorithm for electrical resistivity tomography. Near Surface Geophysics, 19(1), 7–26.
     [Google Scholar]
  6. Aleardi, M., Vinciguerra, A. and Hojat, A. (2021a) A convolutional neural network approach to electrical resistivity tomography. Journal of Applied Geophysics, 193, 104434.
     [Google Scholar]
  7. Aleardi, M., Vinciguerra, A. and Hojat, A. (2021b) Ensemble‐based electrical resistivity tomography with data and model space compression. Pure and Applied Geophysics, 178, 1781–1803.
     [Google Scholar]
  8. Araya‐Polo, M., Jennings, J., Adler, A. and Dahlke, T. (2018) Deep‐learning tomography. The Leading Edge, 37(1), 58–66.
     [Google Scholar]
  9. Aster, R.C., Borchers, B. and Thurber, C.H. (2018) Parameter Estimation and Inverse Problems. Cambridge: Elsevier.
     [Google Scholar]
  10. Bièvre, G., Oxarango, L., Günther, T., Goutaland, D. and Massardi, M. (2018) Improvement of 2D ERT measurements conducted along a small earth‐filled dyke using 3D topographic data and 3D computation of geometric factors. Journal of Applied Geophysics, 153, 100–112.
     [Google Scholar]
  11. Biswas, R. and Sen, M. (2017) 2D full‐waveform inversion and uncertainty estimation using the reversible jump Hamiltonian Monte Carlo. In SEG Technical Program Expanded Abstracts 2017, pp. 1280–1285.
     [Google Scholar]
  12. Dahlin, T. (2020) Geoelectrical monitoring of embankment dams for detection of anomalous seepage and internal erosion – experiences and work in progress in Sweden. Fifth International Conference on Engineering Geophysics (ICEG), Al Ain, UAE, https://doi.org/10.1190/iceg2019‐053.1.
  13. Dosso, S.E., Holland, C.W. and Sambridge, M. (2012) Parallel tempering for strongly nonlinear geoacoustic inversion. The Journal of the Acoustical Society of America, 132(5), 3030–3040.
     [Google Scholar]
  14. Emerick, A.A. and Reynolds, A.C. (2013) Ensemble smoother with multiple data assimilation. Computers & Geosciences, 55, 3–15.
     [Google Scholar]
  15. Fichtner, A., Zunino, A. and Gebraad, L. (2019) Hamiltonian Monte Carlo solution of tomographic inverse problems. Geophysical Journal International, 216(2), 1344–1363.
     [Google Scholar]
  16. Gebraad, L., Boehm, C. and Fichtner, A. (2020) Bayesian elastic full‐waveform inversion using Hamiltonian Monte Carlo. Journal of Geophysical Research: Solid Earth, 125(3), e2019JB018428.
     [Google Scholar]
  17. Goodfellow, I., Bengio, Y. and Courville, A. (2016), Deep Learning. Cambridge: MIT Press.
     [Google Scholar]
  18. Grana, D., de Figueiredo, L. and Mosegaard, K. (2021) Markov chain Monte Carlo for petrophysical inversion. Geophysics, 87(1), 1–60.
     [Google Scholar]
  19. Grana, D., Passos de Figueiredo, L. and Azevedo, L. (2019) Uncertainty quantification in Bayesian inverse problems with model and data dimension reduction. Geophysics, 84(6), M15–M24.
     [Google Scholar]
  20. Hahnloser, R.H., Sarpeshkar, R., Mahowald, M.A., Douglas, R.J. and Seung, H.S. (2000) Digital selection and analogue amplification coexist in a cortex‐inspired silicon circuit. Nature, 405(6789), 947–951.
     [Google Scholar]
  21. Hansen, T.M. and Cordua, K.S. (2017) Efficient Monte Carlo sampling of inverse problems using a neural network‐based forward: applied to GPR crosshole traveltime inversion. Geophysical Journal International, 211(3), 1524–1533.
     [Google Scholar]
  22. He, K., Zhang, X., Ren, S. and Sun, J. (2015) Delving deep into rectifiers: Surpassing human‐level performance on imagenet classification. In Proceedings of the IEEE International Conference on Computer Vision, pp. 1026–1034.
  23. He, K., Zhang, X., Ren, S. and Sun, J. (2016) Deep residual learning for image recognition. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 770–778.
  24. Hermans, T. and Paepen, M. (2020) Combined inversion of land and marine electrical resistivity tomography for submarine groundwater discharge and saltwater intrusion characterization. Geophysical Research Letters, 47(3), e2019GL085877.
     [Google Scholar]
  25. Hojat, A., Arosio, D., Di Luch, I., Ferrario, M., Ivanov, V.I., Longoni, L.et al. (2019a) Testing ERT and fiber optic techniques at the laboratory scale to monitor river levees. 25th European Meeting of Environmental and Engineering Geophysics, The Hague, Netherlands, https://doi.org/10.3997/2214‐4609.201902440.
  26. Hojat, A., Arosio, D., Longoni, L., Papini, M., Tresoldi, G. and Zanzi, L. (2019b) Installation and validation of a customized resistivity system for permanent monitoring of a river embankment. EAGE‐GSM 2nd Asia Pacific Meeting on Near Surface Geoscience and Engineering, Kuala Lumpur, Malaysia. https://doi.org/10.3997/2214‐4609.201900421.
  27. Hojat, A., Tresoldi, G. & Zanzi, L. (2021) Correcting the effect of the soil covering buried electrodes for permanent electrical resistivity tomography monitoring systems. 4th Asia Pacific Meeting on Near Surface Geoscience & Engineering, Ho Chi Minh, Vietnam, Online, DOI: 10.3997/2214‐4609.202177070.
  28. Karaoulis, M., Revil, A., Tsourlos, P., Werkema, D.D. and Minsley, B.J. (2013) IP4DI: a software for time‐lapse 2D/3D DC‐resistivity and induced polarization tomography. Computers & Geosciences, 54, 164–170.
     [Google Scholar]
  29. Kingma, D.P. and Ba, J. (2014) Adam: a method for stochastic optimization. arXiv:1412.6980.
  30. Krizhevsky, A., Sutskever, I. and Hinton, G.E. (2017) Imagenet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84–90.
     [Google Scholar]
  31. LeCun, Y., Bottou, L., Bengio, Y. and Haffner, P. (1998) Gradient‐based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278–2324.
     [Google Scholar]
  32. LeCun, Y., Bengio, Y. and Hinton, G. (2015) Deep learning. Nature, 521(7553), 436–444.
     [Google Scholar]
  33. Liu, B., Guo, Q., Li, S., Liu, B., Ren, Y., Pang, Y. and, et al. (2020) Deep learning inversion of electrical resistivity data. IEEE Transactions on Geoscience and Remote Sensing, 58(8), 5715–5728.
     [Google Scholar]
  34. Loke, M.H., Papadopoulos, N., Wilkinson, P.B., Oikonomou, D., Simyrdanis, K. and Rucker, D.F. (2020) The inversion of data from very large three‐dimensional electrical resistivity tomography mobile surveys. Geophysical Prospecting, 68(8), 2579–2597.
     [Google Scholar]
  35. Luo, X., Bhakta, T., Jakobsen, M. and Nævdal, G. (2018) Efficient big data assimilation through sparse representation: a 3D benchmark case study in petroleum engineering. PloS One, 13(7), e0198586.
     [Google Scholar]
  36. Menke, W. (2018) Geophysical Data Analysis: Discrete Inverse Theory. Orlando, FL: Academic Press.
     [Google Scholar]
  37. Moghadas, D. (2020) One‐dimensional deep learning inversion of electromagnetic induction data using convolutional neural network. Geophysical Journal International, 222(1), 247–259.
     [Google Scholar]
  38. Monajemi, H., Donoho, D.L. and Stodden, V. (2016) Making massive computational experiments painless. In 2016 IEEE International Conference on Big Data, pp. 2368–2373.
  39. Moradipour, M., Ranjbar, H., Hojat, A., Karimi‐Nasab, S. and Daneshpajouh, S. (2016) Laboratory and field measurements of electrical resistivity to study heap leaching pad no. 3 at Sarcheshmeh copper mine. 22nd European Meeting of Environmental and Engineering Geophysics, https://doi.org/10.3997/2214‐4609.201602140.
  40. Norooz, R., Olsson, P.I., Dahlin, T., Günther, T. and Bernston, C. (2021) A geoelectrical pre‐study of Älvkarleby test embankment dam: 3D forward modelling and effects of structural constraints on the 3D inversion model of zoned embankment dams. Journal of Applied Geophysics, 191, 104355.
     [Google Scholar]
  41. Park, M.J. and Sacchi, M.D. (2020) Automatic velocity analysis using convolutional neural network and transfer learning. Geophysics, 85(1), V33–V43.
     [Google Scholar]
  42. Puzyrev, V. (2019) Deep learning electromagnetic inversion with convolutional neural networks. Geophysical Journal International, 218(2), 817–832.
     [Google Scholar]
  43. Ray, A., Kaplan, S., Washbourne, J. and Albertin, U. (2017) Low frequency full waveform seismic inversion within a tree based Bayesian framework. Geophysical Journal International, 212(1), 522–542.
     [Google Scholar]
  44. Richardson, A. (2018) Seismic full‐waveform inversion using deep learning tools and techniques. arXiv:1801.07232.
  45. Rucker, D.F., Fink, J.B., Loke, M.H. (2011) Environmental monitoring of leaks using time‐lapsed long electrode electrical resistivity. Journal of Applied Geophysics, 74(4), 242–254.
     [Google Scholar]
  46. Sætrom, J. and Omre, H. (2013) Uncertainty quantification in the ensemble Kalman filter. Scandinavian Journal of Statistics, 40(4), 868–885.
     [Google Scholar]
  47. Sambridge, M. and Mosegaard, K. (2002) Monte Carlo methods in geophysical inverse problems. Reviews of Geophysics, 40(3), 3–1.
     [Google Scholar]
  48. Santurkar, S., Tsipras, D., Ilyas, A. and Madry, A. (2018) How does batch normalization help optimization?. Advances in Neural Information Processing Systems, 31, 2483–2493.
     [Google Scholar]
  49. Sen, M.K. and Stoffa, P.L. (2013) Global Optimization Methods in Geophysical Inversion. Cambridge: Cambridge University Press.
     [Google Scholar]
  50. Sun, J., Slang, S., Elboth, T., Larsen Greiner, T., McDonald, S. and Gelius, L.J. (2020) A convolutional neural network approach to deblending seismic data. Geophysics, 85(4), WA13–WA26.
     [Google Scholar]
  51. Tarantola, A. (2005) Inverse Problem Theory and Methods for Model Parameter Estimation. Philadelphia, PA: Society for Industrial and Applied Mathematics.
     [Google Scholar]
  52. Uhlemann, S., Wilkinson, P.B., Chambers, J.E., Maurer, H., Merritt, A.J., Gunn, D.A. and Meldrum, P.I. (2015) Interpolation of landslide movements to improve the accuracy of 4D geoelectrical monitoring. Journal of Applied Geophysics, 121, 93–105.
     [Google Scholar]
  53. Vinciguerra, A., Aleardi, M., Hojat, A. and Stucchi, E. (2021) Discrete cosine transform for parameter space reduction in Bayesian electrical resistivity tomography. Geophysical Prospecting, 70(1), 193–209. https://doi.org/10.1111/1365‐2478.13148.
     [Google Scholar]
  54. Vrugt, J.A. (2016) Markov chain Monte Carlo simulation using the DREAM software package: theory, concepts, and MATLAB implementation. Environmental Modelling & Software, 75, 273–316.
     [Google Scholar]
  55. Vu, M.T. and Jardani, A. (2021) Convolutional neural networks with SegNet architecture applied to three‐dimensional tomography of subsurface electrical resistivity: CNN‐3D‐ERT. Geophysical Journal International, 225(2), 1319–1331.
     [Google Scholar]
  56. Waldeland, A.U., Jensen, A.C., Gelius, L.J. and Solberg, A.H.S. (2018) Convolutional neural networks for automated seismic interpretation. The Leading Edge, 37(7), 529–537.
     [Google Scholar]
  57. Wang, B., Zhang, N., Lu, W. and Wang, J. (2019) Deep‐learning‐based seismic data interpolation: a preliminary result. Geophysics, 84(1), V11–V20.
     [Google Scholar]
  58. Whiteley, J., Chambers, J.E. and Uhlemann, S., (2017) Integrated monitoring of an active landslide in Lias group Mudrocks, North Yorkshire, UK. In: Hoyer, S. (Ed.) GELMON 2017: 4th International Workshop on GeoElectrical Monitoring, 27.
  59. Wu, Y. and McMechan, G.A. (2019) Parametric convolutional neural network‐domain full‐waveform inversion. Geophysics, 84(6), R881–R896.
     [Google Scholar]
  60. Zhao, Z. and Sen, M.K. (2021) A gradient‐based Markov chain Monte Carlo method for full‐waveform inversion and uncertainty analysis. Geophysics, 86(1), R15–R30.
     [Google Scholar]
http://instance.metastore.ingenta.com/content/journals/10.1111/1365-2478.13189
Loading
Probabilistic inversions of electrical resistivity tomography data with a machine learning‐based forward operator
Geophysical Prospecting 70, 938 (2022); https://doi.org/10.1111/1365-2478.13189
/content/journals/10.1111/1365-2478.13189
/content/journals/10.1111/1365-2478.13189
Loading

Data & Media loading...

  • Article Type: Research Article
Keyword(s): Electrical resistivity tomography; Inversion

Most Cited This Month Most Cited RSS feed

This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error