Reducing Exploration Cycle Time in the Campos Basin: Integrating Seismic Data
Acquisition and Processing.
Kerly Sanchez, Alex Cooke and Franck LeDiagon; WesternGeco, Rio de Janeiro-Brazil; WesternGeco Neptune OBP Staff;
Antonio Ortolan and Ionildo Veras, Petrobras-Brasil.
Copyright 2011, SBGf - Sociedade Brasileira de Geofsica
th
This paper was prepared for presentation during the 12 International Congress of the
Brazilian Geophysical Society held in Rio de Janeiro, Brazil, August 15-18, 2011.
Contents of this paper were reviewed by the Technical Committee of the 12th
International Congress of the Brazilian Geophysical Society and do not necessarily
represent any position of the SBGf, its officers or members. Electronic reproduction or
storage of any part of this paper for commercial purposes without the written consent
of the Brazilian Geophysical Society is prohibited.
____________________________________________________________________
Abstract
This paper presents the integrated approach to seismic
data acquisition and processing currently being used in the
Campos basin. An advanced onboard data processing
workflow is described that was designed to provide highquality data directly from the seismic vessel, enabling
earlier reservoir and exploration decisions.
Introduction
Since 2009 WesternGeco has been acquiring Q-Marine
surface-seismic data in the Campos basin for Petrobras
(Figure 1). To maximize the value of this data an integrated
approach to acquisition and processing was implemented
to provide near-final-quality products directly from the
vessel in a fraction of the usual delivery time.
Key data challenges in this area include variable waterdepths (from 60 to 3400 m), complex water-layer multiple
wavefields, strongly faulted and complex reservoir geology
and high cable feather due to strong and variable currents
in the area. In addition some of the data were acquired in
areas that were highly obstructed (for example, with
platforms and FPSOs), giving irregular data acquired with
several different source and receiver geometries (Figure 2).
These surveys form one of the largest time-lapse projects
in the world, which meant that the data also needed to be
optimized for 4D processing.
An onboard data processing sequence was designed that
met these challenges and requirements, using
technologies more usually reserved for in-house data
processing. This provided data of superior quality
compared to a conventional fast-track onboard product.
The enhanced data processing workflow was enabled by a
dedicated processing center onboard the vessel staffed
around the clock and connected by a high-bandwidth
satellite connection to Rio de Janeiro data processing
center, where data could be reviewed by WesternGeco and
Petrobras.
Figure 2. Post-plot map showing irregular coverage
in the northern part of this 1650 km2 survey.
Figure 1. Location of the survey acquired by M/V Western
Neptune 2009-2011.
Twelfth International Congress of the Brazilian Geophysical Society
Integration of the seismic acquisition and infield processing
________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
Data processing workflow
Highlights of the data processing sequence included;
Calibrated marine source (CMS) processing
A 3D surface related multiple elimination (3D
SRME) workflow
Compact Fourier interpolation (COMFI)
High-density prestack time migration
assumptions in 2D that lead to inaccuracies in the multiple
model. Firstly it is assumed in 2D SRME that the receivers
lie inline directly behind the source, and secondly that there
is no crossline dip. In the Campos basin acquisition, widetow acquisition arrays, strong feathering due to currents
and multi-vessel undershooting geometries compromise
the first assumption, and it is this aspect that was
addressed with a 3D SRME workflow. This workflow uses a
3D swath of input data to allow the multiples to be
predicted at true azimuth, rather than the zero-azimuth
assumption of 2D SRME (Figure 4).
The Q-Marine acquisition system brings several benefits for
surveys that are designed to be optimized for time-lapse
analysis. Some of these, such as dynamic spread control
(where source and receiver positions are optimized for
repeatability) were described by Le Diagon et al. (2010). In
addition to highly repeatable positioning, the CMS
technique allows the synthesis of far-field signatures from
near-field hydrophone measurements for every shotpoint.
During data processing a shot-by-shot operator is designed
that matches each of these individual signatures to a
common target signature, reducing non-repeatability due to
variation in pressure, gun timing and array geometry
(Figure 3).
Figure 4 In true-azimuth SRME, the multiple traveltime S-X-R
can be better predicted by the convolution of sub-events S-R1
and S2-R2 than can be predicted in 2D SRME where only a
single cable from a single inline is input.
Figure 3. Left: Input signature for CMS. Right: Output signature
after CMS.
Effective multiple attenuation is critical in this region where
multiple energy often interferes with primary reflection data
at the reservoir level introducing noise and uncertainty in
interpretation and amplitude analysis. The main challenges
in this area are surface multiples generated from sea-floor
and near-surface reflectors and diffractors. Peg-leg
multiples from deeper reflectors were also frequently
observed.
A combination of multiple attenuation techniques were
used, including Radon-domain multiple attenuation (which
discriminates multiples and primaries based on velocity),
tau-p deconvolution (which discriminates based on
periodicity) and SRME, a technique that produces an
estimate of the multiples directly from the data using an
auto-convolution technique (Verschuur et al. 1992).
SRME is commonly implemented in 2D in onboard
processing workflows, as it is efficient and can easily be
run in near real time. There are, however, two main
Figure 5. Left: Gather after Radon. Right: Gather after Radon
and SRME. The display show how was attenuated the far-offset
aliased and the near-surface multiples.
The true-azimuth SRME workflow is efficient and while it
does require data from multiple input sail lines, it does not
have to wait for a large swath of data to be acquired before
it can be run. Figure 5 show how the true-azimuth SRME
attenuates the far-offset aliased and near-surface multiples.
Twelfth International Congress of the Brazilian Geophysical Society
3
K. Sanchez, A. Cooke , WesternGeco Neptune OBP Staff; A. Ortolan, I. Macedo, F.LeDiagon.
________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
One of the assumptions of Kirchhoff prestack time
migration is that the input data are in regularly sampled
within common-offset groups. Due to The high levels of
infill and undershooting in this acquisition campaign, the
data were highly irregular, so an explicit regularization
technique was used to condition the data prior to migration.
Compact Fourier interpolation (Moore and Ferber, 2008)
was used to output the data in 80 common-offset groups
with data at bin centers on a 12.5 mx12.5 m grid. Full
Kirchhoff prestack timeqmigration was then run using a
migration velocity picked from migrated data onboard the
vessel (Figure 6).
Figure 7. Time slices through prestack time-migrated
volumes produced before and after acquisition and
processing of undershoot data.
Conclusion
Integrated acquisition and processing during the Campos
basin 4D seismic campaign significantly reduced the
turnaround time for the delivery of high-quality seismic
data, enabling earlier reservoir and exploration decisions.
Because high quality data was delivered 6 weeks after the
last shot.
Figure 6. Stack after prestack time migration.
Acquisition Quality assurance
In addition to the rapid delivery of high-quality seismic data
volumes, the additional flexibility allowed by integrating
acquisition and processing onboard the vessel also allowed
improved acquisition quality assurance. For example, it
was possible to judge the effect of noise (such as seismic
interference) on the final product, and to compare the effect
of additional campaigns of undershoot in the area, giving
confidence that the reservoirs were sufficiently illuminated
for interpretation and 4D analysis.
The data processing workflow developed for this area, was
customized for the different geophysical situations
encountered in each prospect, and provided high-quality
products that included high-resolution prestack timemigrated interpretation volumes, angle stacks for AVO
analysis and enhanced prestack gathers ready to be used
for depth imaging or further 4D processing.
In addition, dedicated and flexible data processing
resources provided enhanced acquisition quality
assurance.
Acknowledgments
Our special thanks to Marcus Holloway, William Griggs
and the other members of the M/V Western Neptune OBP
team for performing the onboard testing and processing.
References
Le Diagon, F, P. Waterson, J. Paulsen, R. Tovar, I.
Macedo and A. Ortolan. et al. 2010 Results seen from the
first application of a System for Dynamic Control of
Vessel, Source and Streamers in 4D acquisition off shore
Brazil. Proceedings of the Rio Oil and Gas Expo &
Conference.
Moore, I. and R.Ferber, 2008 Bandwidth Optimization for
th
Compact Fourier Interpolation, 70 EAGE conference &
exhibition, Extended abstract.
Twelfth International Congress of the Brazilian Geophysical Society
Integration of the seismic acquisition and infield processing
________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
Verschuur, D. A.J. Berkhout, and C.P.A. Wapenaar,
1992, Adaptive, surface-related multiple elimination,
Geophysics, 57, 1166-1177.
Twelfth International Congress of the Brazilian Geophysical Society
