Full-waveform Inversion on 2D Slanted Streamer Data – A Case Study


One of the recent developments in prestack seismic imaging is
the use of full-waveform inversion (FWI) in velocity model
building and updating. FWI computes the difference between
synthetic waveforms generated from the current P-wave
velocity (Vp) model and the actual recorded (observed) data
and then back propagates the difference to update the Vp
model. It achieves convergence when there is minimum
difference between the two datasets. The inversion proceeds
from lower to higher frequencies to obtain global convergence
first before updating the localized velocity trend. Therefore,
very low-frequency signal plays a critical role in the success of
In the marine environment, a data acquisition technique that
can provide more low-frequency signal than a conventional
method is a slanted streamer technique where the streamer is
towed at varying depths. A comparative seismic test survey
over a shallow-water environment in Asia was recently
conducted using slanted- and constant-depth streamers. We
processed both data sets independently using similar
processing techniques except for cable deghosting. The aim of
this study is to evaluate the benefit of this acquisition and
processing technique for FWI model building.
We applied acoustic anisotropic FWI to both data sets to
analyze the benefit of the additional low-frequency signal
gained in the slanted streamer data. In this paper, I will show
that FWI on the slanted streamer data produce a more
geologically plausible velocity model. Also the slanted
streamer image has better stack continuity and simpler geologic
structure compared to the conventional 8m depth flat streamer
Key words: FWI, slanted, deghosting.