TY - GEN
T1 - Practical waveform inversion in anisotropic media: The natural combination of the data and image objectives
AU - Alkhalifah, Tariq Ali
AU - Wu, Zedong
N1 - KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: We thank KAUST for its support. We appreciate the Seismic wave analysis group (SWAG) for a collaborative environment.
PY - 2016/9
Y1 - 2016/9
N2 - Addressing anisotropy in full wavenumber inversion (FWI) is crucial to obtaining credible models, and it is extremely challenging considering the multi parameter nature of the inversion. A successful FWI in anisotropic media takes into account the sensitivity of the data (or the wave) to the long and short wavelength components of the anisotropic parameters. Considering the low sensitivity of FWI to the anellipticity parameter ? when parametrizing the acoustic transversely isotropic model with the horizontal velocity, η and ε, we develop a combined FWI and reflection waveform inversion (RWI) to invert for the anisotropic parameters that influence surface seismic data. This practical waveform inversion (PWI) separates the parameters to their resolvable scales, with information accessed from the data fitting (FWI) and the image focusing (RWI) objectives. With this parametrization, the RWI role is to obtain a smooth ηmodel, as well as velocity, while FWI focusses on the scattering potential of the horizontal velocity. The parameter η is used to produce the Born scattered wavefield for the RWI part and eventually fit the amplitude for the imperfect physics in the FWI part.
AB - Addressing anisotropy in full wavenumber inversion (FWI) is crucial to obtaining credible models, and it is extremely challenging considering the multi parameter nature of the inversion. A successful FWI in anisotropic media takes into account the sensitivity of the data (or the wave) to the long and short wavelength components of the anisotropic parameters. Considering the low sensitivity of FWI to the anellipticity parameter ? when parametrizing the acoustic transversely isotropic model with the horizontal velocity, η and ε, we develop a combined FWI and reflection waveform inversion (RWI) to invert for the anisotropic parameters that influence surface seismic data. This practical waveform inversion (PWI) separates the parameters to their resolvable scales, with information accessed from the data fitting (FWI) and the image focusing (RWI) objectives. With this parametrization, the RWI role is to obtain a smooth ηmodel, as well as velocity, while FWI focusses on the scattering potential of the horizontal velocity. The parameter η is used to produce the Born scattered wavefield for the RWI part and eventually fit the amplitude for the imperfect physics in the FWI part.
UR - http://hdl.handle.net/10754/625258
UR - http://library.seg.org/doi/10.1190/segam2016-13953250.1
UR - http://www.scopus.com/inward/record.url?scp=85019104586&partnerID=8YFLogxK
U2 - 10.1190/segam2016-13953250.1
DO - 10.1190/segam2016-13953250.1
M3 - Conference contribution
SP - 1315
EP - 1319
BT - SEG Technical Program Expanded Abstracts 2016
PB - Society of Exploration Geophysicists
ER -