TY - GEN
T1 - Using well information to verify and improve waveform inversion results in transversely isotropic media
AU - Li, Y.
AU - Alkhalifah, Tariq Ali
AU - Gu, H.
N1 - KAUST Repository Item: Exported on 2020-10-01
Acknowledgements: We thank China University of Geosciences and King Abdullah University of Science and Technology for the financial support. We thank Ramzi Djebbi and all SWAG members for the useful discussion.
PY - 2018/10/16
Y1 - 2018/10/16
N2 - There is a fundamental problem with trade-offs between parameters in multiparameter full waveform inversion (FWI), including in transversely isotropic media with vertical axis of symmetry (VTI). In the parameterization using normal moveout (NMO) velocity $v_n$ and anisotropic parameters $\delta$ and $\eta$, the trade-off between $\delta$ (which relates the NMO velocity to depth) and the depth may cause shifts in the inverted NMO velocity. With the availability of well check-shot velocities, which often represent the vertical velocity, we develop a method to estimate the shifts at the well location and calculate a background $\delta$ model for additional iterations of FWI using this $\delta$ model. We apply the method to the VTI Marmousi II model and invert simultaneously for $v_{n}$, $\delta$ and $\eta$ parameters. Results show that we can use the assumed well velocities to estimate the required vertical shift and calculate a good background $\delta$ model. Using the background $\delta$ model, we can get a reliable NMO velocity model.
AB - There is a fundamental problem with trade-offs between parameters in multiparameter full waveform inversion (FWI), including in transversely isotropic media with vertical axis of symmetry (VTI). In the parameterization using normal moveout (NMO) velocity $v_n$ and anisotropic parameters $\delta$ and $\eta$, the trade-off between $\delta$ (which relates the NMO velocity to depth) and the depth may cause shifts in the inverted NMO velocity. With the availability of well check-shot velocities, which often represent the vertical velocity, we develop a method to estimate the shifts at the well location and calculate a background $\delta$ model for additional iterations of FWI using this $\delta$ model. We apply the method to the VTI Marmousi II model and invert simultaneously for $v_{n}$, $\delta$ and $\eta$ parameters. Results show that we can use the assumed well velocities to estimate the required vertical shift and calculate a good background $\delta$ model. Using the background $\delta$ model, we can get a reliable NMO velocity model.
UR - http://hdl.handle.net/10754/663454
UR - http://www.earthdoc.org/publication/publicationdetails/?publication=92637
UR - http://www.scopus.com/inward/record.url?scp=85083938132&partnerID=8YFLogxK
U2 - 10.3997/2214-4609.201801250
DO - 10.3997/2214-4609.201801250
M3 - Conference contribution
SN - 9789462822542
BT - 80th EAGE Conference and Exhibition 2018
PB - EAGE Publications BV
ER -