@article{fdi:010064958,
  title = {{E}fficient 3-{D} frequency-domain mono-parameter full-waveform inversion of ocean-bottom cable data : application to {V}alhall in the visco-acoustic vertical transverse isotropic approximation},
  author = {{O}perto, {S}. and {M}iniussi, {A}. and {B}rossier, {R}. and {C}ombe, {L}. and {M}{\'e}tivier, {L}. and {M}onteiller, {V}. and {R}ibodetti, {A}lessandra and {V}irieux, {J}.},
  editor = {},
  language = {{ENG}},
  abstract = {{C}omputationally efficient 3-{D} frequency-domain full waveform inversion ({FWI}) is applied to ocean-bottom cable data from the {V}alhall oil field in the visco-acoustic vertical transverse isotropic ({VTI}) approximation. {F}requency-domain seismic modelling is performed with a parallel sparse direct solver on a limited number of computer nodes. {A} multiscale imaging is performed by successive inversions of single frequencies in the 3.5-10 {H}z frequency band. {T}he vertical wave speed is updated during {FWI} while density, quality factor {Q}({P}) and anisotropic {T}homsen's parameters delta and epsilon are kept fixed to their initial values. {T}he final {FWI} model shows the resolution improvement that was achieved compared to the initial model that was built by reflection traveltime tomography. {T}his {FWI} model shows a glacial channel system at 175 m depth, the footprint of drifting icebergs on the palaeo-seafloor at 500 m depth, a detailed view of a gas cloud at 1 km depth and the base cretaceous reflector at 3.5 km depth. {T}he relevance of the {FWI} model is assessed by frequency-domain and time-domain seismic modelling and source wavelet estimation. {T}he agreement between the modelled and recorded data in the frequency domain is excellent up to 10 {H}z although amplitudes of modelled wavefields propagating across the gas cloud are overestimated. {T}his might highlight the footprint of attenuation, whose absorption effects are underestimated by the homogeneous background {Q}({P}) model ({Q}({P}) = 200). {T}he match between recorded and modelled time-domain seismograms suggests that the inversion was not significantly hampered by cycle skipping. {H}owever, late arrivals in the synthetic seismograms, computed without attenuation and with a source wavelet estimated from short-offset early arrivals, arrive 40ms earlier than the recorded seismograms. {T}his might result from dispersion effects related to attenuation. {T}he repeatability of the source wavelets inferred from data that are weighted by a linear gain with offset is dramatically improved when they are estimated in the {FWI} model rather than in the smooth initial model. {T}he two source wavelets, estimated in the {FWI} model from data with and without offset gain, show a 40ms time-shift, which is consistent with the previous analysis of the time-domain seismograms. {T}he computational efficiency of our frequency-domain approach is assessed against a recent time-domain {FWI} case study performed in a similar geological environment. {T}his analysis highlights the efficiency of the frequency-domain approach to process a large number of sources and receivers with limited computational resources, thanks to the efficiency of the substitution step performed by the direct solver. {T}his efficiency can be further improved by using a block-low rank version of the multifrontal solver and by exploiting the sparsity of the source vectors during the substitution step. {F}uture work will aim to update attenuation and density at the same time of the vertical wave speed.},
  keywords = {{I}nverse theory ; {C}ontrolled source seismology ; {B}ody waves ; {S}eismic anisotropy ; {C}omputational seismology ; {W}ave propagation},
  booktitle = {},
  journal = {{G}eophysical {J}ournal {I}nternational},
  volume = {202},
  numero = {2},
  pages = {1362--1391},
  ISSN = {0956-540{X}},
  year = {2015},
  DOI = {10.1093/gji/ggv226},
  URL = {https://www.documentation.ird.fr/hor/fdi:010064958},
}