Bayesian full-waveform tomography with application to crosshole ground penetrating radar data

Research output: Contribution to journalArticle

Standard

Bayesian full-waveform tomography with application to crosshole ground penetrating radar data. / Hunziker, Jürg; Laloy, Eric; Linde, Niclas.

In: Geophysical Journal International, Vol. 218, No. 2, 01.08.2019, p. 913-931.

Research output: Contribution to journalArticle

Harvard

Hunziker, J, Laloy, E & Linde, N 2019, 'Bayesian full-waveform tomography with application to crosshole ground penetrating radar data', Geophysical Journal International, vol. 218, no. 2, pp. 913-931. https://doi.org/10.1093/gji/ggz194

APA

Vancouver

Author

Hunziker, Jürg ; Laloy, Eric ; Linde, Niclas. / Bayesian full-waveform tomography with application to crosshole ground penetrating radar data. In: Geophysical Journal International. 2019 ; Vol. 218, No. 2. pp. 913-931.

Bibtex - Download

@article{11e7fce8d21e4d7f9157828e036753b7,
title = "Bayesian full-waveform tomography with application to crosshole ground penetrating radar data",
abstract = "We present a probabilistic full-waveform inversion (FWI) approach that infers a geostatistical model along with the subsurface structure. Probabilistic FWI with Markov chainMonte Carlo (MCMC) allows for uncertainty quantification and removes the requirement of having a starting model in the cone of attraction of the assumed correct global minimum. We demonstrate our approach on a synthetic and a field data set. For the latter, we compare the results with deterministic FWI and a cone penetration test. Our results compare equally well with the cone penetration test as the deterministic results do. This is a positive result as for the deterministic inversion almost seven times more data were used than for the probabilistic inversion. Furthermore, the probabilistic FWI was able to converge to the posterior distribution starting from randomly drawn models. However, our uncertainty estimates are too narrow, because the necessarily shortMarkov chains implied by a computationally costly forward problem and the global nature of the model proposal scheme prevented a full exploration of the posterior probability density function.Without prior information such as borehole logs, the algorithm is only able to infer relative electric conductivity values, because the unknown amplitude of the wavelet and the mean of the conductivity are strongly correlated. This study clearly demonstrates the feasibility of probabilistic FWI and highlights the advantages and disadvantages of the approach.",
keywords = "Hydrogeophysics, Ground penetrating radar, Statistical methods, Tomography;, Waveform inversion",
author = "J{\"u}rg Hunziker and Eric Laloy and Niclas Linde",
note = "Score=10",
year = "2019",
month = "8",
day = "1",
doi = "10.1093/gji/ggz194",
language = "English",
volume = "218",
pages = "913--931",
journal = "Geophysical Journal International",
issn = "0956-540X",
publisher = "Oxford University Press",
number = "2",

}

RIS - Download

TY - JOUR

T1 - Bayesian full-waveform tomography with application to crosshole ground penetrating radar data

AU - Hunziker, Jürg

AU - Laloy, Eric

AU - Linde, Niclas

N1 - Score=10

PY - 2019/8/1

Y1 - 2019/8/1

N2 - We present a probabilistic full-waveform inversion (FWI) approach that infers a geostatistical model along with the subsurface structure. Probabilistic FWI with Markov chainMonte Carlo (MCMC) allows for uncertainty quantification and removes the requirement of having a starting model in the cone of attraction of the assumed correct global minimum. We demonstrate our approach on a synthetic and a field data set. For the latter, we compare the results with deterministic FWI and a cone penetration test. Our results compare equally well with the cone penetration test as the deterministic results do. This is a positive result as for the deterministic inversion almost seven times more data were used than for the probabilistic inversion. Furthermore, the probabilistic FWI was able to converge to the posterior distribution starting from randomly drawn models. However, our uncertainty estimates are too narrow, because the necessarily shortMarkov chains implied by a computationally costly forward problem and the global nature of the model proposal scheme prevented a full exploration of the posterior probability density function.Without prior information such as borehole logs, the algorithm is only able to infer relative electric conductivity values, because the unknown amplitude of the wavelet and the mean of the conductivity are strongly correlated. This study clearly demonstrates the feasibility of probabilistic FWI and highlights the advantages and disadvantages of the approach.

AB - We present a probabilistic full-waveform inversion (FWI) approach that infers a geostatistical model along with the subsurface structure. Probabilistic FWI with Markov chainMonte Carlo (MCMC) allows for uncertainty quantification and removes the requirement of having a starting model in the cone of attraction of the assumed correct global minimum. We demonstrate our approach on a synthetic and a field data set. For the latter, we compare the results with deterministic FWI and a cone penetration test. Our results compare equally well with the cone penetration test as the deterministic results do. This is a positive result as for the deterministic inversion almost seven times more data were used than for the probabilistic inversion. Furthermore, the probabilistic FWI was able to converge to the posterior distribution starting from randomly drawn models. However, our uncertainty estimates are too narrow, because the necessarily shortMarkov chains implied by a computationally costly forward problem and the global nature of the model proposal scheme prevented a full exploration of the posterior probability density function.Without prior information such as borehole logs, the algorithm is only able to infer relative electric conductivity values, because the unknown amplitude of the wavelet and the mean of the conductivity are strongly correlated. This study clearly demonstrates the feasibility of probabilistic FWI and highlights the advantages and disadvantages of the approach.

KW - Hydrogeophysics

KW - Ground penetrating radar

KW - Statistical methods

KW - Tomography;

KW - Waveform inversion

UR - http://ecm.sckcen.be/OTCS/llisapi.dll/open/36584809

U2 - 10.1093/gji/ggz194

DO - 10.1093/gji/ggz194

M3 - Article

VL - 218

SP - 913

EP - 931

JO - Geophysical Journal International

JF - Geophysical Journal International

SN - 0956-540X

IS - 2

ER -

ID: 5878343