On inferring the noise in probabilistic seismic AVO inversion using hierarchical Bayes
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Standard
On inferring the noise in probabilistic seismic AVO inversion using hierarchical Bayes. / Madsen, Rasmus Bødker; Zunino, Andrea; Hansen, Thomas Mejer.
SEG International Exposition and Annual Meeting, 24-29 September, Houston, Texas . Houston, Texas, 2017.Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Harvard
APA
Vancouver
Author
Bibtex
}
RIS
TY - GEN
T1 - On inferring the noise in probabilistic seismic AVO inversion using hierarchical Bayes
AU - Madsen, Rasmus Bødker
AU - Zunino, Andrea
AU - Hansen, Thomas Mejer
PY - 2017/8/24
Y1 - 2017/8/24
N2 - A realistic noise model is essential for trustworthy inversion of geophysical data. Sometimes, as in case of seismic data, quan- tification of the noise model is non-trivial. To remedy this, a hierarchical Bayes approach can be adopted in which proper- ties of the noise model, such as the amplitude of an assumed uncorrelated Gaussian noise model, can be inferred as part of the inversion. Here we demonstrate how such an approach can lead to substantial overfitting of noise when inverting a 1D re- flection seismic NMO data set. We then argue that usually the noise model is correlated, and suggest to infer the amplitude of a correlated Gaussian noise model. This provides better results than assuming an uncorrelated model. In general though, the results suggest that care should be taken using the hierarchical Bayes approach to infer the noise model.
AB - A realistic noise model is essential for trustworthy inversion of geophysical data. Sometimes, as in case of seismic data, quan- tification of the noise model is non-trivial. To remedy this, a hierarchical Bayes approach can be adopted in which proper- ties of the noise model, such as the amplitude of an assumed uncorrelated Gaussian noise model, can be inferred as part of the inversion. Here we demonstrate how such an approach can lead to substantial overfitting of noise when inverting a 1D re- flection seismic NMO data set. We then argue that usually the noise model is correlated, and suggest to infer the amplitude of a correlated Gaussian noise model. This provides better results than assuming an uncorrelated model. In general though, the results suggest that care should be taken using the hierarchical Bayes approach to infer the noise model.
UR - https://www.onepetro.org/conference-paper/SEG-2017-17725822
M3 - Article in proceedings
BT - SEG International Exposition and Annual Meeting, 24-29 September, Houston, Texas
CY - Houston, Texas
ER -
ID: 185407570