V017
` Intelligent Infill  for Cost Effective 3D Seismic
`Marine Acquisitions
`P. Capelle* (Total E&P Nigeria) & P. Matthews (Total E&P Nigeria)
`
`SUMMARY
`TOTAL E&P NIGERIA (TEPNG) presents a new approach to infill management by leveraging the power
`of interpolation available during processing, the real time and offline coverage maps and availability of
`steerable streamers to obtain more cost effective 3D seismic marine acquisitions.
`The approach was fully implemented by TEPNG over high density (6.25 x 12.5m bin size) surveys
`(EGINA, PREOWEI), a long offset survey (GTB), leading to significant cost savings and illustrated
`through the EGINA 3D HD survey located in the Deep Offshore Nigerian waters where fan mode
`shooting was applied for the first time worldwide.
`
`
`
`
`
`71st EAGE Conference & Exhibition — Amsterdam, The Netherlands, 8 - 11 June 2009
`
`WESTERNGECO Exhibit 2114, pg. 1
`PGS v. WESTERNGECO
`IPR2014-01478
`
`

`
`
`
`Introduction
`
`There is a natural limit to the resolution of the 3D seismic image that can be made of the subsurface.
`Knowing the physics of this, the bin (or pixel) size of the 3D acquisition grid can be determined in
`order to adequately sample the subsurface.
`As the high frequencies are attenuated at longer offset and depth, the bin size can be increased with
`offset and depth without damaging the quality of the images. However, current seismic processing
`algorithms require a constant bin size. So, seismic acquisition grids are always designed to properly
`image the shallowest and nearest offsets to be recorded. This results in a massive over-sampling at
`depth and farthest offsets.
`In practice, feathering of the cables, linked to unpredictability of water currents, means all bins cannot
`be acquired with all offsets with a single pass of the survey vessel. Consequently, additional infill
`passes are required.
`The challenge is to minimize infill rate by acquiring only relevant traces leading to optimized quality,
`turnaround and cost. It is most important now due to the high increase of 3D seismic marine
`acquisition rates the last 4 years (between 200 and 300 %).
`Day and Rekdal, 2005 described an infill assessment methodology where the coverage requirements
`for any hole size was evaluated by modelling the effects of coverage holes in migrated data,
`methodology implemented on a survey acquired along pre plots lines (Strand et al., 2008).
`In this paper, TOTAL E&P NIGERIA (TEPNG) presents a new approach to infill management by
`leveraging the power of interpolation available during processing, the real time and offline coverage
`maps and availability of steerable streamers to obtain more cost effective 3D seismic marine
`acquisitions.
`The approach was fully implemented by TEPNG over high density (6.25 x 12.5m bin size) surveys
`(EGINA, PREOWEI), a long offset survey (GTB), leading to significant cost savings and illustrated
`through the EGINA 3D HD survey located in the Deep Offshore Nigerian waters where “fan mode
`shooting” was applied for the first time worldwide.
`
`New Approach to infill management
`
`Potential of trace interpolation at processing
`Based on the optimized acquisition variable grid versus offset and that one empty bin can easily be
`interpolated at processing, we have considered (confirmed by intensive testing on real data) that one
`to several nominal empty bins versus offset can be interpolated at processing stage (Figure 1).
`
`
`
`In line
`In line
`
`Cross line
`Cross line
`
`
`2 Redundancy Edits2 Redundancy Edits
`
`1 trace per bin and only one is 1 trace per bin and only one is
`
`keptkept
`
`
`
`44
`
`
`Bad quality Bad quality
`
`interpolated traces interpolated traces
`
`editionedition
`
`Removal of flagged tracesRemoval of flagged traces
`
`
`
`11
`
`InputInput
`
`
`Pre-processing (Desig, Noise Pre-processing (Desig, Noise
`
`Attenuation, SRME 2D, Q comp, Attenuation, SRME 2D, Q comp,
`
`SP/RCV/Cable terms attenuation)SP/RCV/Cable terms attenuation)
`
`3 Bin centering + trace 3 Bin centering + trace
`
`
`
`
`missing restorationmissing restoration
`
`All traces with a poor quality All traces with a poor quality
`
`factor are flagged (red points)factor are flagged (red points)
`
`5 Trace interpolation5 Trace interpolation
`
`
`
`
`Interpolation of edited tracesInterpolation of edited traces
`
`NEAR
`
`1
`
`NOMINAL acquisition grid
`1
`1
`0
`1
`
`1
`
`In line
`In line
`
`FAR
`
`0
`
`0
`
`0
`
`0
`
`1
`
`1
`
`Cross lineCross line
`OPTIMIZED acquisition grid (variable versus offset)
`NEAR
`1
`1
`0
`1
`1
`1
`
`FAR
`
`1
`
`0
`
`0
`
`0
`
`0
`
`1
`
`NOMINAL grid after interpolation at processing
`NEAR
`1
`1
`1
`1
`1
`1
`
`FAR
`
`1
`
`1
`
`1
`
`1
`
`1
`
`1
`
`
`
`
`
`Figure 1: Left), potential of interpolation at processing versus offset. Right) Methodology of
`regularization at processing applied at least by CGGVeritas and WesternGeco.
`
`
`
`71st EAGE Conference & Exhibition — Amsterdam, The Netherlands, 8 - 11 June 2009
`
`
`
`WESTERNGECO Exhibit 2114, pg. 2
`PGS v. WESTERNGECO
`IPR2014-01478
`
`

`
`
`
`
`New coverage specifications and displays
`Coverage specifications have been set up based on the number of empty columns of bins versus
`specific single offset planes driven by bin size characteristics and the potential of the trace
`interpolation performed at processing. For the EGINA HD survey, the maximum number of empty
`columns of bins accepted was ranging between 2 columns for near offset plane to 5 columns for the
`far offset plane 5425-5500m. These values correspond to a bin expansion from 200 % to 500% while
`conventional specifications were ranging from 0 to 100 %!
`ONLINE, driven by these new specifications, new real time coverage displays (figure 2) have been
`used for optimization of the steering when shooting for coverage. A single offset class (fold 1) is now
`displayed instead of a multi offset classes. That binary information provides very precise information
`through direct access to the number of empty columns leading to sharp steering and an optimization
`of the overlapping of the near or far offsets.
`OFFLINE, single offset displays before and after flex binning simulation (figure 2) driven by these
`new specifications offer a real and perfect idea of the quality of the coverage offset by offset. The
`infill program can then be easily optimized.
`
`
`Traditional display
`Multi offset class 2475-3600 m
`
`New display
`Single offset class 3225-3300 m
`
`Single offset display
` Before bin expansion
`
`Single offset display
` After bin expansion
`
`Inline
`Inline
`
`
`
`CrosslineCrossline
`
`
`Figure 2: left) ONLINE - Comparison between traditional and new coverage displays used for
`steering. Right) OFFLINE - Single offset display before and after bin expansion.
`
`Potential of specific seismic acquisition equipment- “Fan mode shooting”
`For the first time worldwide, the ability of steering the streamers with the “Q marine” system of
`WesternGeco, was used partially over the 2007 EGINA HD for simulating a variable streamer
`separation with increasing offset. That specific acquisition spread provides an optimized grid of bins
`variable with offsets which adequately sample the subsurface tuned to the seismic signal behaviour.
`That method of acquisition was named “fan mode shooting” (figure 3).
`
`
`
`500 m
`
`CONVENTIONAL geometry
`Pass - 2
`Pass - 1
`
`FAN geometry
`Pass - 1
`Pass - 2
`
`Fars
`
`Fars
`
`50m
`
`50 m
`
`750 m
`
`75 m
`
`75 m
`
`75 m
`
`Figure 3: left) Schema of “Fan mode shooting” method and online coverage display with fan mode
`spread. Right) Impact of streamer geometry on far offsets coverage with slight feather mismatch.
`
`
`71st EAGE Conference & Exhibition — Amsterdam, The Netherlands, 8 - 11 June 2009
`
`
`
`WESTERNGECO Exhibit 2114, pg. 3
`PGS v. WESTERNGECO
`IPR2014-01478
`
`

`
`
`
`Main Results (EGINA HD survey)
`
`The “Fan mode shooting” method, which is part of the infill management approach, has contributed
`sensibly to reduce the infill rate and produce a more uniform coverage distribution. The increase in
`footprint of the far offsets gives a more even coverage (figure 4) where feather mismatches occur and
`permit the steering point to be brought in nearly all instances closer to the front of the streamers.
`Thereby, when shooting for coverage, it reduces the cross-line corrections required (figure 4) and the
`amount of overlapping of nears coverage when greater feather lines are shot alongside lesser feather
`lines.
`
`Conventional mode shooting
`
`
`
`Fan mode shooting
`
`
`
`Figure 4: Coverage map of a far single offset for conventional mode shooting and fan mode shooting.
`
`The final coverage maps have shown a high average coverage percentage, with a low percentage of
`infill in spite of the unpredictability of the currents. The interpolation at processing has handled the
`one to five empty columns of bins with efficiency (figure 5).
`A fte r red u n d a c y e d it & b efo re re g u la ris atio n
`A fte r reg u larisa tio n & in te rp o la tio n
`
`
`
`Figure 5: Comparison before and after interpolation – Cross line – offset 4000 m – Shallow part.
`
`The provisional migrated images of the 2008 3D HD processing are outstanding, mainly due to the
`HD acquisition, with no evidence of degradation due to the infill strategy applied when compared to
`the final migrated image of the 2005 3D exploration reprocessing (figure 6).
`Exploration 3D 2005 reprocessing
`
`HD 3D 2008 processing
`
`
`
`Figure 6: Final 3D original exploration reprocessing (random line) versus 3D HD provisional
`processing (inline).
`
`71st EAGE Conference & Exhibition — Amsterdam, The Netherlands, 8 - 11 June 2009
`
`
`
`WESTERNGECO Exhibit 2114, pg. 4
`PGS v. WESTERNGECO
`IPR2014-01478
`
`

`
`
`
`Benefits & Conclusion
`
`With this new approach to infill management (figure 7), the infill rate was minimized by acquiring
`only relevant traces leading to optimized quality, turnaround and cost
`
`
`
`ACQUISITION
`
`AFTER PRIME
`
`AFTER INFILL
`
`2 0 0 0 0 1 1
`2 0 0 0 0 1 1
`2 0 0 0 0 1 1
`2 0 0 0 0 1 1
`2 0 0 0 0 1 1
`2 0 0 0 0 1 1
`2 0 0 0 0 1 1
`2 0 0 0 0 1 1
`2 0 0 0 0 1 1
`2 0 0 0 0 1 1
`Crossline
`
`AFTER PRIMEAFTER PRIME
`
`Inline
`
`
`1 1 0 0 0 1 11 1 0 0 0 1 1
`
`1 1 0 0 0 1 11 1 0 0 0 1 1
`
`1 1 0 0 0 1 11 1 0 0 0 1 1
`
`1 1 0 0 0 1 11 1 0 0 0 1 1
`
`1 1 0 0 0 1 11 1 0 0 0 1 1
`
`1 11 1 0 0 0 1 1
`1 1 0 0 0 1 1
`1 1 0 0 0 1 1
`
`BIN
`
`No Data
`
`BEFORE
`
`NOW
`
`PROCESSING
`
`Same images
`
`AFTER INFILL
`Bin expansion 100 %
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1
`1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`
`AFTER PRIME
`Bin expansion 300%
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1
`1
`1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`
`1 1 1 1 1 1 11 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`1 1 1 1 1 1 1
`
`3 1 1 1 0 1 1
`3 1 1 1 0 1 1
`3 1 1 1 0 1 1
`3 1 1 1 0 1 1
`3 1 1 1 0 1 1
`3 1 1 1 0 1 1
`3 1 1 1 0 1 1
`3 1 1 1 0 1 1
`3 1 1 1 0 1 1
`3 1 1 1 0 1 1
`
`INFILL
`
`NO INFILL
`
`Data Acquired
`
`Data Redundant
`
`
`
`Data interpolatedData interpolated
`
`Figure 7: Synopsis of new approach to infill management. (Zoom of one single offset grid with holes).
`
`The cost saving to the EGINA HD and GTB long offset surveys is estimated at 1.4 & 6.7 M USD
`respectively. We estimate that this new approach to infill management, results in 30% less infill
`required to fulfill a survey objectives.
`
`We are convinced that this new approach to infill management, implemented successfully over
`several surveys in Nigeria, will generate significant cost savings for future 3D marine seismic surveys
`and will be further enhanced by the increased availability of steerable streamers.
`
`Acknowledgments
`
`The authors would like to thank M. Erbetta & P. Charron (Total E&P Nigeria) for their advice, trust
`and support towards the approach, R. Rushby & P. Melody (RPS QC reps) who have executed the
`implementation of the methodology on the field, WesternGeco for their close cooperation and our
`partners of OML 130 (NAPIMS, CNOOC, Petrobras & Sapetro) for their permission to show this
`work.
`
`References
`
`Strand, C., Buchan, I., Mostavan, A., Ross, J. and Monk, D. [2008] Evaluating infill requirements
`when acquiring a marine 3D seismic survey along pre-plots lines. 78th annual SEG meeting expanded
`abstracts, 70-74.
`Day, A and Rekdal, T. [2005] Determining infill specifications based on geophysical criteria. 75th
`annual SEG meeting expanded abstracts, 80-83.
`
`71st EAGE Conference & Exhibition — Amsterdam, The Netherlands, 8 - 11 June 2009
`
`
`
`WESTERNGECO Exhibit 2114, pg. 5
`PGS v. WESTERNGECO
`IPR2014-01478