Professional Documents
Culture Documents
org/
Ac
qui
s
i
t
i
on
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Expanded Abstracts
2016 Technical Program
SEG International Exposition
and Eighty-Sixth Annual Meeting
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Abstracts
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
ACQ 2
ACQ 3
ACQ 4
ACQ 5
ACQ E-P1
ACQ P1
Land and Marine Acquisition ..................... Tuesday p.m., Oct .18 ................. 245
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 1
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
x j t R j x t j
(1)
Fj(f ) R j F(f )e
i 2f j
(2)
i 2f1
R 2e
i 2f 2
R 3e
i 2f 3
(4)
2016 SEG
SEG International Exposition and 86th Annual Meeting
H0
m
1.5
V1
m/s
1066
H1
m
11.0
V2
m/s
1478
H2
m
21.0
V3
m/s
1716
Remarks
Uphole
The shooting depths for the test were 3, 5, 7, 9, 12, 15, 34,
and 47m respectively; the charge size was 3kg (Ning et al.,
2011). Figure 3 shows the shot gather records of the test
point: the shooting depth is 3m on the left and 34m on the
right. It is easy to find that the visual frequency is low and
the interference energy is strong on left; while the
interference is weaker obviously and the visual frequency is
high on the right. For the time window 1400ms-4000ms of
the100-140th trace, the inter-trace spectrum analysis is
made. Figure 4 shows the spectrum of three different
shooting depths. You can see that with the increasing of the
shooting depth, the high-frequency energy decreases,
Page 2
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 3
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
The authors wish to thank Crew 217 and Crew 2256 for
their time and hard work spent on the test acquisition. And
thanks BGP, CNPC for permitting us to publish the related
results.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 4
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Li, T. S., H. Q. Gao, Z. X. Liu, X. X. Liu, S. H. Qian, and Z. X. Zeng, 1997, The acquisition of ghost
information and the study of its generation and frequency response: Geophysical Prospecting for
Petroleum, 36, 3844.
Ling, Y., 2001, Analysis of charge size and charge type: Oil Geophysical Prospecting, 36, 584590.
Liu, Z. X., C. W. Ding, X. X. Liu, W. F. Pan, and S. H. Qian, 1999, A new technique for high resolution
3-D seismic data acquisition: Oil Geophysical Prospecting, 34, 3744.
Lv, G. H., 2002, Measurement of the ghosting interface of seismic survey and its application: Oil
Geophysical Prospecting, 37, 295299.
Ning, H. X., Y. H. Zhang, D. R. Zhang, Y. X. Jiang, and W. L. Jiao, 2011, Seismic source optimization in
the Qilian Mountain area: Oil Geophysical Prospecting, 46, 370373.
Qian, R. J., 2003, Analysis of the explosive effect of the dynamite source: Oil Geophysical Prospecting,
38, 583588.
Qian, S. H., J. P. Liu, Y. X. Gu, T. S. Li, and Z. X. Liu, 1998, The research on the explosive mechanism
and excitation condition of dynamite source: Geophysical Prospecting for Petroleum, 37, 114.
Tang, D. L., F. Yan, and X. Q. Wang, 2005, Determination of shooting depth under water table: Progress
in Exploration Geophysics, 28, 3641.
Wang, W. H., 1999, Analysis of the dynamite source pattern in compressional wave exploration: Oil
Geophysical Prospecting, 34, 249259.
Zhang, D. R., J. H. Zhang, Y. Y. An, Y. X. Jiang, and W. Z. Lu, 2006, Study on shooting parameters of
seismic acquisition in Kulongshan mountainous area: Oil Geophysical Prospecting, 41, 249252.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 5
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
After almost 40 years of production, Prudhoe Bay remains
one of the largest oil fields in North America with
additional drilling activity planned for the coming years. In
order to support continued drilling it was decided to acquire
new seismic data in 2015 over key parts of the field, see
figure 1, replacing suboptimal legacy data.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 6
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2009 Survey
3 fleets of 4
4 seconds
6-84Hz Linear
110 / 770- 880
110 / 1320-1430
16
0.8 million
10Hz / 3
ISS Survey
12 single
32 seconds
2-96Hz
-3dB/octave nonlinear
110 / 110
220 / 660-770
32
>12 million
5 Hz / single
Benefits of ISS
This ISS survey area covered an offshore area and a coastal
plain including tundra, lakes, rivers and sea ice. Floating
sea-ice covered 46% of the survey area and 10% was
frozen rivers and lakes. The weight of the single vibrator
compared to a fleet of vibrators allowed safe access to
floating ice and areas that legacy surveys left incomplete.
A single vibrator with a single sweep delivered the required
energy as has been reported by others (Thacker et al. 2014,
Ourabah et al 2015). The initial comparison of receiver
field gathers will favor fleets of vibrators, but once the
aggregate data-density is realized after processing, the
single source, dense vibrator technique produces superior
signal to noise and illumination. Denser sampling and high
fold data is well suited for modern noise removal
techniques suppressing different types of noise (source
generated, cultural and random noise).
The vertical sampling was improved by broadening the
bandwidth of the data. The low frequencies were increased
by starting the sweep at 2Hz and using 5Hz geophones. The
sweep used was a 32 second non-linear sweep (low dwell).
The increased bandwidth reduced side-lobes of the wavelet
(Yilmaz 2001) compared to legacy data and resulted in
cleaner images. Figure 3a and 3b clearly shows the
suppression of side-lobes and reduction in the peak to
trough ratio of the wavelet extracted from the ISS data
compared to the legacy wavelet. Figure 3c and 3d shows
the effect broader bandwidth has on the data by suppressing
artificial reflections from side lobes, increasing the vertical
resolution and making the data more believable and simpler
to interpret. Finer spatial sampling reduces the acquisition
footprint. The example shown in Figure 3e and 3f is a
comparison of shallow depth slices. The ISS has better
spatial continuity while the legacy data is noisier, making it
difficult to recognize continuous events.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 7
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 8
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 6. RMS amplitude extraction at the reservoir comparing the legacy sparse data (left) to the dense ISS data (right).
Conclusion
The ISS method combined with autonomous node
recording technology offers operational advantages
delivering an order of magnitude increase in trace density
and source productivity (over 11,000 VPs on the best day).
This efficiency improvement translated to less HSE
exposure hours, lower costs and allowed us to complete the
survey in fewer days during the short operating season.
This method also proved to be robust and tolerant to high
wind conditions and cultural noise in the busy oilfield. This
acquisition method provides data that produce a significant
uplift in imaging quality. We are hopeful that ISS in the
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 9
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Baeten, G. J. M., V. Belougne, L. Combee, E. Kragh, A. Laake, J. E. Martin, J. Orban, A. zbek, and P.
L. Vermeer, 2000, Acquisition and processing of point receiver measurements in land seismic:
70th Annual International Meeting, SEG, Expanded Abstracts, 4144,
http://dx.doi.org/10.1190/1.1816083.
Baeten, G. J. M., V. Belougne, M. Daly, B. Jeffryes, and J. E. Martin, 2000, Acquisition and processing
of point source measurements in land seismic: 70th Annual International Meeting, SEG,
Expanded Abstracts, 4548, http://dx.doi.org/10.1190/1.1816095.
Ellis, D., 2013, Simultaneous source acquisition Achievements and challenges: 75th Annual
International Conference and Exhibition, EAGE, Extended Abstracts, Th-08-01,
http://dx.doi.org/10.3997/2214-4609.20130080.
Howe, D., M. Foster, T. Allen, B. Taylor, and I. Jack, 2008, Independent simultaneous sweeping A
method to increase the productivity of land seismic: 78th Annual International Meeting, SEG,
Expanded Abstracts, 28262828, http://dx.doi.org/10.1190/1.3063932
Ourabah, A., J. Keggin, C. Brooks, D. Ellis, and J. Etgen, 2015, Seismic acquisition, what really matters?
85th Annual International Meeting, SEG, Expanded Abstracts, 69,
http://dx.doi.org/10.1190/segam2015-5844787.1.
Thacker, P., D. Harger, and D. Iverson, 2014, An evaluation of single vibrator, single sweep, 3D seismic
acquisition in the Western Canada Sedimentary Basin: CSEG Recorder, 39.
Trupp, R., J. Hastings, S. Cheadle, and R. Vesely, 2009, Seismic in arctic environments: Meeting the
challenge: The Leading Edge, 28, 936942, http://dx.doi.org/10.1190/1.3192840.
Yilmaz, O., 1987, Seismic data processing: SEG.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 10
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
The study area, located near Fort McMurray, Alberta, overlies a portion of McMurray oil sands that will be developed using
Steam Assisted Gravity Drainage (SAGD) techniques. Steam injection into the reservoir will be periodically monitored using
time-lapse 3C/3D seismic data to understand why, where and how the heat moves.
Near-surface, low-velocity, heterogeneous layers can pose significant detrimental impacts on the quality of land seismic data. In
some parts of Northern Canada a type of bog consisting of water and partly decomposed organic material, called muskeg, occurs
at the surface. This muskeg can become thick enough to attenuate some PP and most PS seismic waves.
In this study area the muskeg has been measured up to 8m thick and was detrimental to an existing PP and PS 3C/3D surface
seismic dataset. Figure 1 shows an example of PS converted-wave data from the 3C/3D dynamite surface seismic survey. The
center part of the line, where the thick muskeg exists, demonstrates the severe absorptive effect on the shear-waves being
recorded at the surface. The P-waves were also affected but to a lesser degree than the S-waves.
Time-lapse joint PP/PS pre-stack inversion has demonstrated its value in locating steam chambers and mobile bitumen within
heavy oil reservoirs surrounding this study area (Gray et al., 2016; Zhang and Larson, 2016). Therefore, 4D joint inversion
techniques are expected to form an integral part of the reservoir management for this SAGD project. These expectations led to
the decision to investigate whether the quality of the seismic data could be substantially improved by planting dynamite sources
and permanent buried receivers below the thick muskeg layer. In the winter and summer of 2015 a 3C/2D seismic line was
acquired with buried receivers with the purpose of testing whether future high-quality 3C/3D time-lapse seismic monitoring was
a possibility.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 11
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Source
1. 22 lines orthogonal to the receiver lines with two buried sources (0.125 kg dynamite @ 9m) every 20m
2. 1 line parallel to the receiver lines with two buried sources (0.125 kg dynamite @ 9m) every 20m
Figure 2 shows a map view of the layout of the central 2D line with receiver stations in red along with inline source stations and
the source stations along orthogonal lines shown in green.
Front end preparations included widening an existing 3D receiver line to accommodate a Low Impact Seismic (LIS) drill for
geophone installation. The LIS drill assisted with installing 0.125kg dynamite sources on both the grid and the 2D line. The
duplicate source points employed 4 PVC pipes as both a monument and to protect the leads against curious wildlife.
A few analog and digital receiver planting poles were manufactured from three meter sections of steel pipe and fastened together
with pinned steel couplers. An orientation tool was attached at the top and custom fitted cups attached to the bottom. The planting
pole design proved cumbersome for the 3m and 9m buried receivers but was manageable in weather down to -10oC. When
temperatures plunged to -25oC significant challenges were encountered with wet muskeg quickly freezing to the loading pole
cups and couplers.
Sand points were attached to the sensors before installation to increase the coupling success rate and to aid in decoupling the tool
from the sensor (Figure 3). 4 PVC pipe and colored caps were used at the surface as both a monument and to protect the cable
and connectors against the elements and curious wildlife.
Wireless recording equipment was deployed and the sensors went through a series of QCs to ensure they were operating
correctly before acquisition. The recording equipment was collected in a methodical manner as to not confuse data from different
sensors at different depths. The crew returned in the summer, using Argos in the extreme wet terrain (Figure 4), to re-deploy the
recording equipment and re-acquire the data from the buried geophones.
Processing
Figure 5 shows a comparison of a dynamite shot gather being recorded into the vertical component of analog receivers at the
surface, 3m depth and 9m depth. In several ways, Figure 5 illustrates the successful outcome of the buried-receiver experiment
since the data recorded by the 9m receiver data is obviously far superior to the surface receivers and the 3m deep receivers. Not
only are there much larger static delays present on the surface and 3m receivers than on the 9m receivers, but also there is much
more variation in frequency content from trace to trace on the surface and 3m receivers. From Figure 5 it is evident that a good
deal of the reflected energy returning to the surface is being absorbed by the muskeg layer. After observing the pre-stack gathers,
it was decided not to process the 3m receivers since the 9m receivers were much better quality.
The first acquisition of the 3C/2D line took place in January, 2015, and the second took place in June, 2015. Figure 6 shows a
comparison of the PS radial-component asymptotic-conversion point stacks from the surface receivers versus the 9m deep
receivers from the winter acquisition. The zone of interest (the McMurray oil sands) is approximately at 400ms PS time on the
9m deep receiver section. The PS stack from the 9m deep receivers shows good quality, but it is impossible to interpret the
surface-receiver PS section due to its poor quality.
One concern during acquisition was that the 3C down-hole receivers might move or twist in position over time. Analysis of the
orientation of the 3C receivers between winter and summer indicated that the receivers had stayed in place in the 9m deep holes.
It was unexpectedly discovered that several of the 9m deep receivers had inadvertently been placed down the hole in reversepolarity orientation. This was an important finding for improving future acquisitions.
Based on comparisons such as those in Figure 6, it was decided to perform time-lapse processing of winter and summer datasets
using only the 9m deep receivers, not the surface or 3m deep receivers. Although no production took place between winter and
summer acquisitions, it was considered important to test the repeatability of the buried receivers and to test whether the
processing flow could compensate for any changes in the seismic response between winter and summer. The decrease in NRMS
from raw data through post-stack migration during the AVO-compliant processing steps of the PP 9m deep receiver data was
from 130% to 20%. The decrease in NRMS to 20% after post-stack migration indicates that the non-repeatable factors were
reduced to a level where time-lapse differences could be reliably detected, if SAGD production had occurred. Figure 7 shows the
winter, summer and difference sections for the 9m deep receiver dynamite PP data.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 12
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
The decrease in NRMS of the radial converted-wave PS data through the major AVO-compliant processing steps was from 130%
to 40%. The decrease in NRMS down to 40% (compared to 20% for the PP data) is to be expected for the noisier, narrow-band
PS data. Figure 8 shows the winter, summer and difference sections for the 9m deep receiver, dynamite PS data.
Conclusions
The acquisition and processing of the 3C/2D test line successfully demonstrates that high-quality 3C images can be obtained if
both the sources and receivers are buried beneath an extremely absorptive muskeg layer.
Acknowledgements
The authors would like to thank Brion Energy and Arcis Seismic Solutions/TGS for permission to publish these results.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 13
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 5: The vertical component of a dynamite shot gather recorded by 9m deep receivers (left), receivers at the surface (center)
and 3m deep receivers (right). The highest quality, most coherent, broadest bandwidth reflections were recorded on the 9m deep
receivers since the down and upgoing raypaths do not pass through the absorptive muskeg layer at the surface.
Figure 6: Radial PS converted-wave stacks recorded by 9m deep receivers (left) and by receivers at the surface (right).
Figure 7: Final migrated PP stacks recorded with 9m deep receivers from the winter (left), summer (center) and difference (right).
Figure 8: Final migrated PS stacks recorded with 9m deep receivers from the winter (left), summer (center) and difference (right).
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 14
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Gray, F. D., K. A. Wagner, and D. J. Naidu, 2016, 3C-4D locates mobile bitumen in oil sands reservoirs:
CSEG Geoconvention.
Pullin, N., L. Matthews and K. Hirsche, 1987, Techniques used to obtain very high resolution 3-D seismic
imaging at an Athabasca oil sands thermal project: The Leading Edge, 6, no. 12, 1015,
http://dx.doi.org/10.1190/1.1439353.
Zhang, J., and G. Larson, 2016, Time-lapse PP-PS joint inversion over an active SAGD Project: CSEG
Geoconvention.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 15
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
For mini-SOSIE, the autocorrelation characteristics of
scanning signal, especially the quality of reference channel
signal will directly affect the quality of seismic profiles. So
the key issue is to design and develop high quality scanning
signal. Barbier (1976) proposed mini-SOSIE seimic
exploration, its main idea is to make the vibrator run as
random as possible by varying the engine speed
(Barbier,1976), but there exists serious background noise,
which affect the quality of correlation seismic record.
Cunningham(1979) first applied pseudorandom coding
theory into the design of vibrosies (Cunningham,1979),
which could avoid side lobe interference generated by
linear or nonlinear sweeping signal(Coupillaud,1976), but
there was serious coherent noise in seismic profile. Shaun
Strong (2004), Hayan Nasreddin(2012), John J Sallas(2008,
2012), T. Dean(2014), etc.al. applied pseudorandom coding
in vibrosies (Shaun Strong, 2004; Hayan Nasreddin,2012;
John J Sallas, 2008, 2012; T. Dean,2014), and pointed that
pseudorandom scheme method can reduce interference,
improve resolution, thus improve the effect of seismic
exploration. With further study, pseudorandom coding is
widely used in vibroseis (M arc Becquey,2002; Timothy
Dean, 2012; ). But due to particularity of mini-SOSIE
signal, its output is equal to the CFP scanning signal,
compared with natural vibration, it can improve the
resolution of seismic record, and reduce side lobe.
In order to realize better result, here we propose VFP
coding scanning technique based on CFP coding scanning
technique, and analyze the principle of the two methods
detailedly and compared the simulation results.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 16
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2 1, j 0(mod N )
R(j)
1 , j 0(mod N ) (2)
Its period is N, it is approximately a pulse function.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 17
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
VFP scanning technique
In order to increase the frequency range of the signal and
reduce side lobe, in this paper we put forward VFP coding
scanning technique. So this scheme is based on
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 18
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Natural Science
No. 41304139),
project of Jilin
SinoProbe-Deep
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 19
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Barbier, M. G., P. Bondon, R. Mellinger, and J. R. Viallix, 1976, Mini-Sosie for Land Seimology:
Geophysical Prospecting, 24, 518527, http://dx.doi.org/10.1111/j.1365-2478.1976.tb00952.x.
Barbier, M. G., and Villix J. R., 1974, Pulse coding in seismic prospecting sosie and seiscode,
Geophysical Prospecting, 22, 153-175.
Becquey, M., 2002, Pseudorandom coded simultaneous vibroseismics: 72nd Annual International
Meeting, SEG, Expanded Abstracts, http://dx.doi.org/10.1190/1.1817375.
Cunningham, A. B, 1979, Some alternate vibrator signals: Geophysics, 44, 19011921,
http://dx.doi.org/10.1190/1.1440947.
Dean, T., 2012, Establishing the limits of vibrator performance - experiments with pseudorandom sweeps:
Annual International Meeting, SEG, Expanded Abstracts.
Dean,T., 2014, The use of pseudorandom sweeps for vibroseis surveys, Geophysical Prospecting, 62, 5074.
Goupillaud, P.,1976, Signal Design in the Vibroseis Technique, Geophysics, 41, 1291-1304.
http://dx.doi.org/10.1190/1.1440680.
Nasreddin, H., T. Dean, and K. Iranpour, 2012, The use of pseudorandom to reduce interference noise in
simultaneous vibrosies surveys: 22nd International Geophysical Conference and Exhibition.
Sallas, J., J. Gibson, F. Lin, O. Winter, B. Montgomery, and P. Nagarajappa, 2008, Broadband vibroseis
using simultaneous pseudorandom sweeps: Annual International Meeting, SEG, Expanded
Abstracts.
Sallas, J., J. Gibson, P. Maxwell, and F. Lin, 2011, Pseudorandom sweeps for simultaneous sourcing and
low-frequency generation: The Leading Edge, 30, 11621172,
http://dx.doi.org/10.1190/1.3657077.
Sarwate, D., and M. Pursley, 1980, Crosscorrelation properties of pseudorandom and related sequences,
Proceedings of the IEEE, 68, 593-619.
Singh, C., 1987, Generating Random Sampling Numbers: Defence Science Journal, 37, 361366,
http://dx.doi.org/10.14429/dsj.37.5923.
Stasev, Y., A. Kuznetsov, and A. Nosik, 2007, Formation of pseudorandom sequences with improved
autocorrelation properties: Cybernetics and Systems Analysis, 43, 111,
http://dx.doi.org/10.1007/s10559-007-0021-2.
Strong, S. and S. Hearn, 2004, Numerical modeling of pseudo-random land seismic sources: ASEG 17th
Annual Geophysics Conference and Exhibition.
Wei, Z., M. A. Hall, and T. F. Phillips, 2012, Geophysical benefits from an improved seismic vibration:
Geophysical Prospecting, 60, 466479, http://dx.doi.org/10.1111/j.1365-2478.2011.01008.x.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 20
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
A seismic acquisition technology to improve the production efficiency in big sand dune area
Zhang Beibei *, E Dianliang, Hou Chengfu, Qi Yongfei, Liang Junfeng, Li Kanghu and Bai Jie, BGP, CNPC
Summary
In big sand dune area, the complex topography and surface
mobility make a great deal of difficulties for stake-less
vibrator operation and layout. The COG (Center of gravity of
a source array) is often not accurate and has the high
likelihood to exceed survey requirement, which will cause
additional acquisition . In an attempt to solve these issues,
BGP developed a new system named Digital-Seis System
(DSS) which can realize autonomous navigation and real-time
QC in sand area operation . Through several indoor and
outdoor simulation, we use shot point projection technology to
realize the goal. Before one program line producton start, we
set up a safe distance, it can reach a timely reminder of
obstacle avoidance purposes. The application of DSS can not
only improve operation efficiency and reduce safety risks, but
also significantly decrease the likelihood of COG exceeding
survey requirement. In a word, the application of DSS plays
an important role in high efficiency and safety operation.
Introduction
2016 SEG
SEG International Exposition and 86th Annual Meeting
5 vibrators are per pattern for the project. Based on work area
topographical features, the technology procedure requires
theoretical source points offset maximum 500 meters in
Cross-line direction and 2.5 meters in In-line direction. There
are two source pattern dimensions: one is in flat terrain, the
vibrators can be placed vertical line direction, namely Crossline combinations, the distance between vibrator pad is 8
meters; Another is in the sand dunes and other special areas,
the source can only move along with the pathway of the
source, namely In- line combinations, the pad to pad distance
is 20 meters.
Page 21
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Project Difficulties
24 hours production as a calendar day. According to the
mode of project in the field, for the field production, in order
to improve the production efficiency we must save every
second.
Work in big sand area, its very difficult for vibrators
movement. The acquisition will be carried out twice or more
times as the results cannot meet the requirement. The mainly
reason is the COG value beyond the limited and also this is
the key factor that waste production time. As the offset case,
because of the certain angle between the vibrator pathway and
program lines, Sercel navigation system cannot be well for 5
vibrators proper navigation, in order to meet the requirements
value of COG, the accuracy position can be only adjusted by
field work experience. For the current version of Sercel
seismic acquisition instruments, the acquisition data can be
transmitted after the shooting finished and then the recorder
section check the qualification. If the qualification cannot
meet the requirement, the vibratos should return back to shoot
again. These processes are consuming time, restricting
production efficiency.
The vibrator pattern type conversion together with the surface
conditions change between the dunes and gentle. Compare
with the huge volume of vibrators and the small source COG
value requirements.Its a plenty of time cost durning the
pattern change and reposition, to complete this process the
instrument terminal need to change the pattern mode, and then
transfer the pattern parameters to each of work vibrator.
Occasional case encounter signal blind spots, it will need to
manually set up the repeater in the highest point for the signal
transmission.This process is time consuming.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 22
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 23
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
Operation Efficiency Comparison
Operation efficiency was calculated by the effect operation
time and production. Comparison of two years' statistical data,
it can improve around 3-5 shots per hour after use of the DSS
system. From the figure 7, the overall efficiency performs an
upward trend.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 24
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
No references.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 25
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 26
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 27
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 28
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 29
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Berkhout, A. J., 2008, Changing the mindset in seismic data acquisition: The Leading Edge, 27, 924938,
http://dx.doi.org/10.1190/1.2954035.
Berkhout, A. J., G. Blacquire, and D. J. Verschuur, 2009, The concept of double blending: Combining
incoherent shooting with incoherent sensing: Geophysics, 74, no. 4, A59A62,
http://dx.doi.org/10.1190/1.3141895.
Berkhout, A. J., 2012, Blended acquisition with dispersed source arrays: Geophysics, 77, no. 4, A19
A23, http://dx.doi.org/10.1190/geo2011-0480.1.
Tsingas, C., Y. Kim, and J. Yoo, 2015, Broadband Acquisition, Deblending and Imaging Employing
Dispersed Source Arrays: EAGE Workshop on Broadband Seismic - A Broader View for the
Middle East, EAGE, BS27.
Phillips, T., Z. Wei, and R. Chen, 2013, Method of seismic vibratory limits control at low frequencies: US
Patent application, US 2013/0201789.
Wei, Z., 2015a, A new generation low frequency seismic vibrator: 85th Annual International Meeting,
SEG, Expanded Abstracts, 211 215.
Wei, Z., 2015b, Extending the vibroseis acquisition bandwidth with s newly designed low frequency
seismic vibrator: EAGE Workshop on Broadband Seismic - A Broader View for the Middle East,
EAGE, BS06.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 30
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Microseismic surface patch array: Modeling and velocity estimation using ambient noise
Tianxia Jia1*, Carl Regone2, Jianhua Yu2, Abhijit Gangopadhyay1, Robert Pool1, Colin Melvin1, Scott Michell1
1: BP; 2: Formerly with BP
Summary
Microseismic events have been widely used for the
monitoring of hydraulic fracturing (Duncan & Eisner,
2010). Common methods of microseismic monitoring use
downhole or buried arrays, both of which are expensive and
can be operationally difficult to deploy. Surface arrays,
therefore, have the potential to mitigate such issues. In this
paper, we used finite difference modeling based on the
SEAM II model to test a surface patch arrays effectiveness
in locating and imaging microseismic events. We detected
the modeled microseismic events on both noise-free and
noisy data using different layout geometries of the patches.
The results show that surface patch arrays could generate
reasonably good locations of the modeled microseismic
events. Based on the modeling results, BP acquired their
first microseimic dataset for fracture monitoring using a
surface patch array. The array reasonably recorded the
microseismic events. Beyond the traditional event
locations, we also turned the ambient noise recorded by the
surface patch array into signal using the Spatial AutoCoherency (SPAC) technique (Jia, 2011). We noted that the
Rayleigh wave phase velocity calculated from ambient
noise correlates very well with the shallow subsurface
geology.
Introduction
The industry widely uses microseismic events from
hydraulic fracture stimulations to understand the
effectiveness of stimulation (Duncan & Eisner, 2010).
Traditionally, downhole and buried arrays are two popular
ways to record microseismic data. Downhole arrays
(Maxwell et al. 2010), or putting geophones into a
neighboring borehole, is the least noisy, but is limited by
the locations of observation wells, and often by the distance
from the observation well/s to the treatment well. A
shallow, near surface, or buried array can mitigate some of
these problems. Although they are noisier than a downhole
array, they have better observations because of the wide
azimuthal coverage. However, they are quite expensive,
require a longer term investment, and take longer to deploy.
A surface array is another alternative. It could have similar
azimuthal coverage compared to a buried array. It
maintains the observations without the long term capital
investment. Although the data may be noisier compared
with both downhole and buried array, surface arrays are
quick to deploy. Surface arrays come with different layout
geometries. The star pattern, which is a set of linearly
2016 SEG
SEG International Exposition and 86th Annual Meeting
total stations
3600
1764
900
324
1400
1224
total stations
6400
3236
1600
576
total stations
900
Page 31
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Microseismic surface patch array: Modeling and velocity estimation using ambient noise
2016 SEG
SEG International Exposition and 86th Annual Meeting
We observed that using both noise free and noisy data, the
surface patch array could generate reasonably good
microseismic event locations. Imaging is more sensitive to
patch spacing. For the same number of patches, the closer
the spacing between patches, the better the results. The
closer station spacing within patches is preferred. The
modeling favors either a 10 x 10 or 7 x 7 patch.
Page 32
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Microseismic surface patch array: Modeling and velocity estimation using ambient noise
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 33
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Microseismic surface patch array: Modeling and velocity estimation using ambient noise
Acknowledgments
We thank BP for the permission to publish this work,
Dawson Geophysical for the acquisition and Schlumberger
for the data processing.
Conclusions
Using finite difference modeling, we demonstrate that
surface patch arrays can be used to image and locate
microseismic events in the subsurface. The main advantage
of the surface patch array is the flexibility of location of
each individual array. We were able to move the patches
around the planned location due to permits and land access
constraints. This allowed us to significantly reduce the
cycle time and costs associated with obtaining the permits,
potential damage claims and get access to areas that are not
easily permitted, thus dramatically improving the overall
efficiency of the acquisition. We successfully imaged
microseismic events using surface patches to monitor the
hydraulic fracturing of two wells. We were also able to
calculate the Rayleigh wave phase velocity from ambient
noise recorded by the surface patch array using SPAC
technique. The phase velocity map correlates well with
shallow subsurface geology. Rayleigh wave phase velocity
can be further inverted to shear wave velocity, which could
be used for velocity model building. Ground roll
attenuation, a critical step in land data processing, could
benefit hugely from the accurate estimation of Rayleigh
wave phase velocity. The SPAC technique can also be
applied to traditional active source 3D land seismic
acquisition, where a great amount of ambient noise data
recorded by geophones are discarded or treated as noise
and therefore attenuated during pre-processing.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 34
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Aki, K., 1957, Space and time spectra of stationary stochastic waves, with special reference to
microtremors: Bulletin of the Earthquake Research Institute, 35, 415456.
Duncan, P. M., and L. Eisner, 2010, Reservoir characterization using surface microseismic monitoring:
Geophysics, 75, no. 5, 75A13975A146, http://dx.doi.org/10.1190/1.3467760.
Jia, T., 2011, Advanced analysis of complex seismic waveforms to characterize the subsurface earth
structure: Ph.D. thesis, Columbia University.
Maxwell, S. C., Rutledge, J., Jones, R. and Fehler, M., 2010, Petroleum reservoir characterization using
downhole Microseismic monitoring: Geophysics, 75, no. 5, 75A12975A137,
http://dx.doi.org/10.1190/1.3477966.
Oristaglio, M., 2012, SEAM Phase IILand seismic challenges: The Leading Edge, 31, 264266,
http://dx.doi.org/10.1190/1.3694893.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 35
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 36
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 1: Receiver line in blue and active source points for the
experiment in red
Analysis Methods
The first level of analysis focuses on ambient data
recordings made over six different days. Comparing the
spectra of the two input geophone group types, we might
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 37
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Random
noise
Reflection
signal
Random
noise
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 38
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
Data recorded on groups with 6 geophones in the southern
Midland Basin, Texas have higher reflection signal-torandom noise than groups with three geophones as
measured on FK filtered stacked data and as measured by
autopicking confidence values on stacked data volumes.
The improvements of the signal-to-noise were in the 1719% range on the stacked lines with 6 geophones versus the
three-geophone stacks.
The improvement in pick
confidence in the interpretation tests were in the 4-7%
range.
Acknowledgements
Thanks to Pioneer Natural Resources management for
letting me to conduct the experiment. Dawson Geophysical
carried out the experiment in the field, and GeoEnergy
provided data processing services.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 39
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Beresford, G., and P. Johnston, 2007, Slant Arrays to reduce ambient noise in Southern Bangladesh: 77th
Annual International Meeting, SEG, Expanded Abstracts, 1620,
http://dx.doi.org/10.1190/1.2792373.
Sheriff, R. E., 2002, Encyclopedic Dictionary of Applied Geophysics, 4th ed.: SEG,
http://dx.doi.org/10.1190/1.9781560802969.
Watts, H. J., J. Gibson, R. Burnett, and S. Ronen, 2005, Evaluation of 3C sensor coupling using ambient
noise measurements: 75th Annual International Meeting, SEG, Expanded Abstracts, 912919,
http://dx.doi.org/10.1190/1.2148309.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 40
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Assessing marine 3D seismic acquisition with new technology: A case history from Suriname
Peter Aaron*, Grant Byerley and David Monk, Apache Corporation
Summary
Before acquiring data offshore Suriname in 2016 a survey
assessment was performed to compare various contractor
methods of de-blending and de-ghosting. In this paper
we will present the model data that was used to make this
assessment and the results of the test comparisons.
Introduction
Over the last 10 years both broadband acquisition and
processing techniques and simultaneous source acquisition,
have played a major role in advancing both the data quality
and the efficiency of 3D marine seismic surveys.
The aim of broadband acquisition and processing is to
remove the unwanted effects on the phase and amplitude
spectra of the source and receiver ghosts, often referred to
as de-ghosting. This is obtained either through a
combination of acquisition and processing solutions, such
as multi-sensor streamer (Carlson et al. 2007) and multilevel source acquisition and variable depth streamers
(Soubaras 2010) or may be achieved using processing
solutions alone (King et al. 2015).
Simultaneous source acquisition typically refers to data that
have been acquired with several sources which overlap in
time. This can either be where shots are fired within a
known and usually relatively short firing window of each
other (Beasley et al 2012), or through the extraction of shot
records from a continuously recorded data set, where the
extracted record length is longer than the interval between
shots. Allowing overlap in the sources is typically done in
order to improve acquisition efficiency and to obtain a
denser dataset which may be better sampled in midpoint,
offset and azimuth. The improved sampling comes with the
trade-off of unwanted interference from one or more other
sources that must be handled in processing using an
appropriate de-blending technique (for example, Akerberg
et al. 2008). Therefore, the ability to separate out, or deblend, in processing the seismic shots must be taken into
account as part of any acquisition design.
This paper uses both real and simulated model datasets in
order to assess current industry receiver de-ghosting and
de-blending capabilities to aid with acquisition design
offshore Suriname.
Background
In 2016 Apache began an assessment of acquisition
methodologies for a survey to be acquired in Suriname.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 41
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 42
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 43
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
In this paper we have described how a test of both deblending and de-ghosting was conducted prior to
acquisition of new data offshore Suriname.
While it is easy to accept that new technology can improve
both the quality and efficiency of acquisition of marine
seismic data, it is necessary to validate the potential
improvement and understand any potential risk before
embarking on acquisition. The tests carried out here
provided us with an insight into current industry
capabilities for both de-blending and de-ghosting using
models which were directly applicable to our new survey.
In the case of de-ghosting, we were able to see how well
the 8 and 18m tow depths could be matched in processing.
The de-blending results provided a quantitative assesment
of any potential trade-off in data quality when acquiring
with a triple source configuration and an insight into how
this could affect our main target levels and goals.
The models described were given to a number of
contractors, and as part of the presentation of the paper,
results will be shown.
Acknowledgements
At Apache Corporation we would like to thank the New
Venture group for pushing the envelope in how they
acquired data, and Polarcus, who acquired the test data in
2013, and the production data in 2016. As a consideration
to those companies who participated in the test evaluation,
we have not identified the individual contributions.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 44
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Akerberg, P., G. Hampson, J. Rickett, H. Martin, and J. Cole, 2008, Simultaneous source separation by
sparse Radon transform: 78th Annual International Meeting, SEG, Expanded Abstracts, 2801
2805, http://dx.doi.org/10.1190/1.3063927.
Baardman, R., and R. Borselen, 2013, Separating sources in marine simultaneous shooting acquisition
Method & Applications: 83rd Annual International Meeting, SEG, Expanded Abstracts, 15,
http://dx.doi.org/10.1190/segam2012-0992.1.
Beasley, C., I. Moore, D. Monk, and L. Hansen, 2012, Simultaneous sources: The inaugural full-field,
marine seismic case: 82nd Annual International Meeting, SEG, Expanded Abstracts, 15,
http://dx.doi.org/10.1190/segam2012-0834.1.
Carlson, D., A. Long, W. Sllner, H. Tabti, R. Tenghamn, and N. Lunde, 2007, Increased resolution and
penetration from a towed dual-sensor streamer: First Break, 25, 7177.
King, S., and G. Poole, 2015, Hydrophone-only receiver deghosting using variable sea surface datum:
85th Annual International Meeting, SEG, Expanded Abstracts, 46104614,
http://dx.doi.org/10.1190/segam2015-5891123.1.
Soubaras, R., 2010, Deghosting by joint deconvolution of a migration and a mirror migration: 80th
Annual International Meeting, SEG, Expanded Abstracts, 34063410,
http://dx.doi.org/10.1190/1.3513556.
Wolfarth, S., D. Priyambodo, T. Manning, T. Septyana, and S. Putri, 2016, Where are we today with ISS
de-blending processing capability? Results from Shallow Water OBC Data, Indonesia: 78th
Annual International Conference and Exhibition, EAGE, Extended Abstracts.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 45
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
Models and measurements
Figure 1 shows a cross section of a typical air-gun. Before
firing, the forces exerted by the pressurized air in the
operating chamber and the fire chamber act on the two faces
of the shuttle to keep it in the sealed position. When the
solenoid valve is actuated by an electrical signal it initiates
an air flow that shifts the net force, accelerating the shuttle
to the right. This exposes a progressively larger area of the
ports, allowing air to escape from the fire chamber into the
surrounding water.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 46
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 47
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Main peak
The main peak is formed when the shuttle moves past the
ports, allowing most of the air in the fire chamber to escape
into the surrounding water. In a standard air-gun, the shuttle
accelerates to a high speed before reaching the ports,
resulting in the steeply rising flank of Figure 5 (green line,
right inset). In contrast, the new air-gun releases air more
gradually, resulting in a signature with a gentler slope and
reduced peak amplitude (Figure 6). As the bubble expands,
its motion is opposed by the local hydrostatic pressure and
by the inertia of the surrounding water, reducing the rate of
volume growth and the resulting acoustic pressure. The
existing precursor bubble also affects the injection of the
main bubble, typically damping the primary peak and
reducing the amount of high-frequency energy.
Figure 6: As Figure 5, but for the new air-gun at 150 cu. in.
Free-bubble oscillation
After the shuttle has re-sealed the ports, gun dynamics no
longer directly influence the bubble. Nevertheless, the
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 48
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
Signature stability
Although air-gun output is remarkably stable in the seismic
band, it is still subject to three main sources of variability:
short-term shot variation, manufacturing tolerances, and
normal long-term wear. We measured shot variation
physically, with the high-frequency system described
above. It would be impractical to physically characterize
the remaining two sources of variation to a reasonable
statistical confidence, or to simulate in CFD due to the long
computation times. We therefore developed a statistical
meta-model that can replace the CFD model in predicting
the output spectrum, training the Kriging model (Sacks et
al., 1989) by performing CFD simulations over the design
space. We then used Monte Carlo methods to predict the
expected range of the output as a result of random inputs.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 49
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Coste, E., D. Gerez, H. Groenaas, J.-F. Hopperstad, O.-P. Larsen, R. Laws, J. Norton, M. Padula, and M.
Wolfstirn, 2014, Attenuated high-frequency emission from a new design of airgun: SEG
Technical Program Expanded Abstracts, 132137, http://dx.doi.org/10.1190/segam2014-0445.1.
Cox, E., A. Pearson, J. R. Blake, and S. R. Otto, 2004, Comparison of methods for modelling the
behavior of bubbles produced by marine seismic airguns: Geophysical Prospecting, 52, 461477,
http://dx.doi.org/10.1111/j.1365-2478.2004.00425.x.
Gerez, D., H. Groenaas, O. P. Larsen, M. Wolfstirn, and M. Padula, 2015, Controlling air-gun output to
optimize seismic content while reducing unnecessary high-frequency emissions: 85th Annual
International Meeting, SEG, Expanded Abstracts, 154158,
http://dx.doi.org/10.1190/segam2015-5843413.1.
Giles, B. F., and R. C. Johnston, 1973, System approach to air-gun array design: Geophysical
Prospecting, 21, 77101, http://dx.doi.org/10.1111/j.1365-2478.1973.tb00016.x.
Landr, M., L. Amundsen, and D. Barker, 2011, High-frequency signals from air-gun arrays: Geophysics,
76, no. 4, Q19Q27, http://dx.doi.org/10.1190/1.3590215.
Laws, R. M., L. Hatton, and M. Haartsen, 1990, Computer modelling of clustered airguns: First Break, 8,
331338, http://dx.doi.org/10.3997/1365-2397.1990017.
Rayleigh, O. M., 1917, On the pressure developed in a liquid during the collapse of a spherical cavity:
Philosophical Magazine, 34, 9498, http://dx.doi.org/10.1080/14786440808635681.
Sacks, J., W. J. Welch, T. J. Mitchell, and H. P. Wynn, 1989, Design and analysis of computer
experiments: Statistical Science, 4, 409423, http://dx.doi.org/10.1214/ss/1177012413.
Vaage, S., B. Ursin, and K. Haugland, 1984, Interaction between airguns: Geophysical Prospecting, 32,
676689, http://dx.doi.org/10.1111/j.1365-2478.1984.tb01713.x.
Ziolkowski, A., G. Parkes, L. Hatton, and T. Haugland, 1982, The signature of an air gun array:
Computation from near-field measurements including interactions: Geophysics, 47, 14131421,
http://dx.doi.org/10.1190/1.1441289.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 50
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
In situ test results obtained in the Mediterranean Sea with a novel marine seismic acquisition
system (FreeCableTM)
Luc Haumont*, Laurent Velay, and Michel Manin, Kietta
Summary
A new patented marine acquisition system was tested in the
Mediterranean Sea in October 2014. The purpose of the
paper is to explain the test setup, to describe the acquisition
parameters and to present the tests results.
Introduction
The new marine acquisition system under test is an
innovative system based on classical principles of
reflection seismology: one or several source vessels are
used as seismic sources, and a large number of seismic
sensors fastened to a set of submerged cables measure the
reflected waves (receivers). The main originality of the
system relies in the following points:
4 ARVs
a control room installed onboard a master vessel
launch and recovery equipment
Test setup
The test campaign was the first in situ survey for this new
acquisition method. All system components were tested in
real conditions.
The system under test consists of:
two SMACs
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 51
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2500
2000
1500
1000
500
SL
RCV
0
-500
-1000
-1500
-2000
-2500
-3000
-2000
-1000
1000
2000
3000
500
5
0
4
-500
2
-1000
-1500 -1000
-500
500
1000
1500
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 52
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Conclusions
The first at-sea tests for a new marine acquisition system
were conducted in the Mediterranean Sea in 2014. The tests
proved that the performance of the system was in line with
the preliminary simulations. The key aspects of the system
were validated: the operability of the system, the capacity
to produce data efficiently in a controlled manner, and the
capability to deliver high quality 4C seismic data. The
seismic data exhibits high signal to noise ratios and low
noise on all components, full bandwidth is obtained at any
depth between 5 and 100m, and the technology is able to
offer true 4C data with usable inline geophone signals. All
technical aspects were qualified by this test campaign. A
subsequent paper will cover the first fully operational
survey performed in the fall of 2015.
Page 53
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 12: raw 4C data receiver gather (crossline shooting); from left to right: hydrophone, inline geophone, transverse geophone #1, and
transverse geophone #2.
Figure 13: geophone records before rotation (receiver gather during inline shooting); from left to right: inline geophone, transverse geophone #1,
and transverse geophone #2.
Figure 14: geophone records after rotation (receiver gather during inline shooting); from left to right: inline geophone, crossline geophone, and
vertical geophone.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 54
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Haumont, L., L. Velay, and M. Manin, 2016, Improving subsalt imaging with a new seismic acquisition
system: Presented at the EAGE/SPE Subsalt Imaging Workshop, http://dx.doi.org/10.3997/22144609.201600435.
Manin, M., 2012, Enhanced method and device for aquatic seismic prospecting: French Patent 2,940,838.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 55
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Fredrik Andersson , Lund University; Kurt Eggenberger, Seismic Apparition GmbH; Dirk-Jan van Manen and
Johan O. A. Robertsson, ETH-Zurich / Seismic Apparition GmbH; Lasse Amundsen, NTNU
SUMMARY
Seismic apparition is a technology for decoding simultaneously source acquisition where a periodic or aperiodic filter
is applied to the different sources in a way that they can be distinguished by means of aliasing effects. In the case of dense
sampling on the source side, this allows for direct separation
of the sources. As the sampling is reduced, aliasing effects
can appear in the reconstructions unless additional measures
are taken. In this paper we will make use of directionality
penalties to dealias the decoded data. The directionality of the
wavefield is measured using the lower frequency components
where the aliasing effects are absent. The decoding procedure
can then be formulated as a least squares problem using the
estimated directionality structure.
INTRODUCTION
Simultaneous shooting is a major trend within the seismic acquisition industry which has the potential to increase the rate
at which seismic data can be acquired and as well improve
subsurface sampling by increased shot density (Ellis, 2013).
Originally established for land seismic data acquisition (Howe
et al., 2008; Bouska et al., 2009), the idea is to trigger two
or more (encoded) sources sufficiently close together in time
so that the recorded signal energy interferes. The interference of signals is handled in data processing to decode or separate the information generated from each source. To date,
the main principle in marine seismic multishooting has been
to shoot with random dithering for one or several sources acquiring seismic data simultaneously. The random dithers are
known and can be removed in processing to generate seismic data where all reflections generated by that source are coherent (e.g., in the common-offset domain) whereas the signals from the other source(s) have a random time distribution. One popular method to decode such simultaneous-source
data is to consider the data separation to be an underdetermined inverse problem, which can be solved through an iterative procedure, assuming additional constraints, like sparsity and coherency. Moore et al. (2012) reported a narrow
azimuth survey with two source arrays firing simultaneously,
one source-array being time-dithered. The sources were separated using a modelling and inversion algorithm (Ji et al.,
2012). Langhammer and Bennion (2015) reported on triplesource simultaneous shooting to achieve higher density seismic. They used an adaptive subtraction method for source separation. Wavefield signal apparition or seismic apparition is a
new method to sample time-discrete signals that allows for the
separation of interfering signals from multiple sources. The
theory of wavefield signal apparition is discussed in Robertsson et al. (2016). In essence, by changing a well sampled conventional source sequence where the wavefield in the spectral
2016 SEG
SEG International Exposition and 86th Annual Meeting
(1)
Page 56
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
the recorded wave field at this depth is a solution to the homogeneous wave equation with a velocity c. The wavefields
are generated at discrete values of x which we assume to be
equally spaced, i.e. of the form x = 4x k.
We now assume that the two sources are used simultaneously,
in such a way that their combination takes the form
d(t, k) = f1 (t, k4x ) + f2 (t 4t (1)k , k4x )
k
= F ( fb1 )(t, k4x ) + F ( fb2 (, )e2i(1) 4t )(t, k4x ),
i.e., the recorded data are now modeled as a sum of the two
functions, but where one of them has been subject to a periodic time shift. In a more general version more general filtering operations than time shifts can be applied. Let ak be
filter operators (acting on the time variable) where the k dependence is such that it only depends on if k is odd or even,
i.e., that ak = ak(mod2) .
Figure 1: The function D1 (, ) from (3) for a set of apparition simultaneously source synthetic data with noise. The right
panel shows in green the region described by (1), while the
central green part in the right panel shows the diamond shaped
conditions (4), and the aliasing/mixing effects from (3) and (5)
is highlighted in red color.
Now, due the assumption of conic support of fb1 and fb2 it holds
that if
||
1
+ 1 , (4)
| | <
, || < c
,
||
<
c
c
4x
4x
f (k) =
k=
k=
fb(k)
f (k)e2i k =
k=
k=
k=
f ()e2i (k)
fb( + k).
k=
k
1 X b
k
f +
e2ix0 4x
4x
4x
k=
k=
X
k
k
b
b
f1 , +
+
f2 , +
4x
24x
k=
k=
1
ab0 ()e2i4t + (1)k ab1 ()e2i4t
.
(3)
2
=
2016 SEG
SEG International Exposition and 86th Annual Meeting
ab ()e
D1 (, ) = fb1 (, ) + fb2 (, ) 0
ab0 () + ab1 ()
= fb1 cos(24t ) (, ) + fb2 (, )
.
2
(5)
This implies that for each pair (, ) satisfying (4), the values
of fb1 (, ) and fb2 (, ) can be obtained by solving the linear
system of equations
1
cos(24t )
ab0 ()e2i4t +b
a1 ()e2i4t
2
ab0 ()+b
a1 ()
2
!
fb1 (, )
D1 (, )
.
=
D2 (, )
fb2 (, )
(6)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
h2 = w f 2 .
(7)
2
h1
h1 h1
(t, x)
K t x (t, x)
K t
T1 (t, x) =
2
K th1 hx1 (t, x)
K hx1
(t, x).
From (8) we have a condition when one the eigenvectors vanish. However, for the more general case we would need to
formulate a penalty function that can deal with the cases where
the components gradually change, and at places where the eigenvectors are equal in size, and equal amount of penalty should
be used for the two directions. One such choice is to define
S(Tm ) =
j=1
This property is clearly not always satisfied (although counterparts in higher dimension hold more frequently with increased
dimensionality), it is a property that can be used as a penalty
from which the simultaneously source data can be decoded.
Even if (8) is not satisfied, the relation can be used to minimize the energy of the decoded data in the directions carried
from h1 and h2 , respectively.
2016 SEG
SEG International Exposition and 86th Annual Meeting
2
X
j=1
T
j emj emj
2 0
and
s1 (1 , 1 ) = s2 (1 , 1 ),
implying that (8) will be satisfied in the case where there is locally only energy in one direction, and where an equal amount
of penalty will be applied in the case where there is the same
amount of energy in both directions.
This definition now allows for the generalization of (8) to penalty
functionals
ZZ
( f1 )T S(T1 ) f1 (t, x) dtdx,
and
ZZ
( f2 )T S(T2 ) f2 (t, x) dtdx,
respectively, for the two wavefields. The expressions above describe the energy in the undesirable direction, given the knowledge of the bandpassed versions h1 and h2 , respectively.
Before we use these expressions to define a minimization problem that describes the decoding procedure, we incorporate the
original cone condition (1) in the formulation. To this end, we
will now work with sampled representations of fb1 and fb2 . By
abuse of notation, we will use fb1 and fb2 to denote these sampled values. We let Fc denote the inverse Fourier operator
that is restricted to function supported on the cone defined by
(1), and recall the definition of the apparition operator T from
(2). The relationship (2) is then satisfied for (the non-unique)
solutions to
b
f1 , b
f2
(8)
Tm =
1 2
1 22
s1 ( ) = 2 exp
2
2 1
1
1 2
1 2
s2 ( ) = 1
exp
.
2 1
2 1
min
T
s j ( )emj emj
,
with
s12 (t, x) = 0,
f1 e12
2
X
with the additional constraint that fb1 and fb2 have support on
the cone defined by (1). To obtain a unique approximate solution, we now add the directionality penalties and consider
min
b
f1 , b
f1
ZZ
T
Fc fb1
S(T1 ) Fc fb1
(t, x) dtdx
ZZ
T
+
Fc fb2
S(T2 ) Fc fb2
(t, x) dtdx.
+
(9)
Page 58
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
and
Fc D1 Fc
AD =
.
Fc D1 Fc
and
CONCLUSIONS
The current main principle in seismic multishooting has been
to shoot with random dithering for one of the sources. We challenge this thinking by demonstrating that the methodology of
seismic apparition in combination with directional dealiasing
is a powerful combination to be able to rely on periodic shooting for accurate separation of the data. The proposed dealiasing method using directional regularization was applied to a
synthetic North Sea data set to appropriately precondition the
data to exploit the wavefield seismic apparition effects. The
numerical example shows a low-residual simultaneous sources
separation in a realistic acquisition scenario. This multishooting, dealias and decoding method can be generalized to more
than two sources.
b2 = Fc T d.
ACKNOWLEDGEMENTS
Fc T Fc
Fc Fc
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 59
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Andersson, F., and A. A. Duchkov, 2013, Extended structure tensors for multiple directionality
estimation: Geophysical Prospecting, 61, 11351149, http://dx.doi.org/10.1111/13652478.12067.
Andersson, F., Y. Morimoto, and J. Wittsten, 2015, A variational formulation for interpolation of seismic
traces with derivative information: Inverse Problems, 31, 055002, http://dx.doi.org/10.1088/02665611/31/5/055002.
Bouska, J., 2009, Distance separated simultaneous sweeping: Efficient 3D vibroseis acquisition in Oman:
79th Annual International Meeting, SEG, Expanded Abstracts, 15,
http://dx.doi.org/10.1190/1.3255248.
Ellis, D., 2013, Simultaneous source acquisition-achievements and challenges: Presented at the 75th
EAGE Annual International Conference & Exhibition, SPE EUROPEC 2013,
http://dx.doi.org/10.3997/2214-4609.20130080.
Hale, D., 2011, Structure-oriented bilateral filtering of seismic images: Presented at the 81st Annual
International Meeting, SEG, Expanded Abstracts, http://dx.doi.org/10.1190/1.3627947.
Howe, D., A. Allen, M. Foster, I. Jack, and B. Taylor, 2008, Independent simultaneous sweeping:
Presented at the 70th EAGE Annual International Conference and Exhibition, SPE EUROPEC
2008, http://dx.doi.org/10.3997/2214-4609.20147588.
Ji, Y., E. Kragh, P. Christie, 2012, A new simultaneous source separation algorithm using frequencydiverse filtering: Presented at the 82nd Annual International Meeting, S SEG, Expanded
Abstracts, http://dx.doi.org/10.1190/segam2012-0259.1.
Knutsson, H., 1989, Representing local structure using tensors: 6th Scandinavian Conference on Image
Analysis, Linkping University Electronic Press, 244251.
Langhammer, J., and P. Bennion, 2015, Triple-source simultaneous shooting (ts3), a future for higher
density seismic?: Presented at the 77th Annual International Conference and Exhibition, EAGE,
Extended Abstracts.
Moore, I., D. Monk, L. Hansen, and C. Beasley, 2012, Simultaneous sources: The inaugural full-field,
marine seismic case history from australia: ASEG Extended Abstracts, 2012, 14,
http://dx.doi.org/10.1071/ASEG2012ab181.
Ramirez, A., F. Andersson, T. Wiik, and P. Riste, 2015, Datadriven interpolation of multicomponent data
by directionality tensors: Presented at the 77th Annual International Conference and Exhibition,
EAGE, Extended Abstracts.
Robertsson, J. O. A., L. Amundsen, and A. S. Pedersen, 2016, Wavefield signal apparition, part I:
Theory: Presented at the 78th Annual International Conference and Exhibition, EAGE, Extended
Abstracts.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 60
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
Wide Azimuth (WAZ) geometries have become a standard
technique for acquiring seismic data in subsalt areas such
as Gulf of Mexico or Gulf of Guinea. Compared to Narrow
Azimuth (NAZ) geometries, WAZ geometries provide a
huge uplift in terms of illumination and imaging as they are
characterized by a higher fold and a richer azimuth
distribution. However, considering the investment that
represents such an acquisition, optimizing the design is
paramount and essential to ensure that the acquisition
meets the imaging objectives but also to minimize the
survey costs. Due to their limitations to properly handle
complex geology, classic ray-tracing modeling tools have
been found not sufficient to provide robust enough support
for optimization of advanced survey geometry. Looking for
more reliability, 3D acoustic Full Wave modeling has
therefore been used to design and optimize Wide Azimuth
marine acquisition in complex geological context. Because
of the always developing HPC capacity, several
cumbersome marine geometries have been evaluated within
a reasonable time frame.
Model building
The acoustic model is composed of a 3D velocity cube and
a 3D density cube. The velocity field contains all the
information related to the geological complexity of the area
whereas the density field is a layer cake model with 3D
structures inserted at target level to evaluate the
illumination capacity of different acquisition designs. The
velocity cube is typically built with two main elements: a
background PSDM model which brings the kinematic
information and perturbations aiming at representing the
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 61
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 2:: Schematic draw of Octopus type geometry (left) and Mad
Mad-Dog
Dog type (right) with four tiles. Tiles are built-up
built
by successive
passes of the streamer vessel away from the source sail line. Both source vessels sail on the same line as many times as the required
number of tiles.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 62
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 3: FD modeling results and model. Upper left: time slice for dense Mad-dog type acquisition in one direction.
Upper right, same layout for orthogonal direction. Lower left: density model. Lower right time slice for the Bi-Waz
configuration. Green circles highlight areas with better imaging compared to orange circles. The Bi-WAZ geometry
combines positive results from both directions.
As expected, FD modeling clearly demonstrates that the bidirectional acquisition combines the benefits from each
azimuth. It also shows that a sparse bi-directional Mad-Dog
acquisition with 3 Tiles returns an equivalent imaging
quality as our initial base case. Indeed, most of the seismic
information recorded by the 4th Tile of a mono-directional
WATS is captured by the orthogonal direction of a BiWATS design. Also, Bi-WATS with 3 and 4 Tiles return
similar target illumination, proving that the 4th Tile is
optional.
Even though bi-WATS improves the imaging quality in this
complex geological environment, its cost has to be
carefully evaluated and benchmarked against other designs.
Such challenging acquisition layout has to be justified not
only on a technical side, but also from the cost effective
aspect. Hence, simulations taken into account several
operational parameters such as the amount of stand-by, line
change duration, vessels speed, run-in and run-out length or
infill rate have been carried out. The results of those
simulations show that even with two directions of
acquisition, the Bi-WATS design reduces the acquisition
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 63
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 64
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Lencrerot, R., J. Colonge, and F. Studer, 2016, Optimising an acquisition design for sub-salt targets using
full wave modelling: 78th Annual International Conference and Exhibition, EAGE, Extended
Abstract, Tu STZ0 01.
Naranjo, J. C., E. J. Ekstrand, J. Etgen, K. L. Hargrove, J. G. Brewton, O. Garcia, G. Astvatsaturov, D.
Hite, B. Howieson, B. Bai, M. Bhargava, and C. Barousse, 2011, Survey design and
implementation of a multiple wide-azimuth towed streamer seismic acquisition strategy at the
tiber discovery; Deep Water Gulf of Mexico, USA: 81st Annual International Meeting, SEG,
Expanded Abstracts, 132135.
Regone, C., 2006, Using 3D finite-difference modeling to design wide azimuth surveys for improved
subsalt imaging: 76th Annual International Meeting, SEG, Expanded Abstracts, 28962899.
Sadeghi, E., A. Melois, and A. C. Bon, 2009, Wave equation based on 3D SRME impact on wide azimuth
towed streamer survey design: 79th Annual International Meeting, SEG, Expanded Abstracts, 75
78, http://dx.doi.org/10.1190/1.3054859.
Vermeer, G. J. O., 2009, Wide-azimuth towed-streamer data acquisition and simultaneous sources: The
Leading Edge, 28, 950958, http://dx.doi.org/10.1190/1.3192843.
VerWest, B. J., and D. Lin, 2007, Modeling the impact of wide-azimuth acquisition on subsalt imaging:
Geophysics, 72, no. 5, SM241SM250, http://dx.doi.org/10.1190/1.2736516.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 65
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Challenges to extending the usable seismic bandwidth at the seafloor in the deep water GoM
Joe Dellinger*, BP America
Summary
2016 SEG
SEG International Exposition and 86th Annual Meeting
The first thing to notice is that the spectra are all thick or
spikey except at the lowest frequencies. Close inspection
of the top two curves reveals that the thickness is the result
of a uniformly spaced spike comb with a spacing of 12
spikes per Hertz. This is because the Atlantis shot interval
was 12 seconds (= 12 / Hz). A repeating signal can be
modeled mathematically as single instance of the signal
convolved with a Dirac comb in the time domain. In the
frequency domain, the convolution becomes multiplication.
The effect of repeating the signal is thus to multiply its
spectrum by the Fourier transform of a Dirac comb, which
is also a Dirac comb. A thick, spikey spectrum thus
indicates a repeating signal in the time domain. Note above
20 Hz the airgun spectra are not as thick, because the
Atlantis airgun signature was less repeatable at these
frequencies.
Figure 2 demonstrates that there is another strong repeating
signal in the background noise above 35 Hz. The wide
interval between the prominent spectral spikes indicates a
repeat rate of about 7.5 per second. This is likely the
signature of a ships propeller or a pump. What about the
rest of the background noise? It rolls off below 7 Hz with
the same slope of about 16 dB / octave as the Atlantis
airgun signatures do, and has a thick, spikey spectra
indicating a repeating source. Is it another airgun? In this
case the spikes are not uniformly spaced. Figure 3
demonstrates that this noise is the combined spectrum of
three distant airguns, each with a different repeat interval.
With the giant ear of our 500+ ocean-bottom nodes we can
monitor all the seismic activity in the Gulf.
Page 66
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 1: Three power spectral density plots calculated over a one-minute time window, averaged over all the ocean-bottom nodes in
the Atlantis array on 10 February 2006. The black curve is for a minute when the Atlantis airgun array was centered over the array. The
red curve is for a minute when the airgun array was 7 km away, at the far end of a shot line. The blue curve is for a minute when the
Atlantis airgun array was not firing, and thus represents the baseline noise level to be overcome. The frequency axis has a log scale.
Figure 2: The background noise from Figure 1, this time plotted with a linear frequency scale to better show the prominent regularly
spaced spectral lines above 35 Hz, corresponding to a signal repeating about 7.5 times a second, most likely from a propellor or a pump.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 67
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
S
Figure 4: Beam-forming the low-frequency seismic
noise at frequencies below 2 Hz reveals that the
microseismic background travels with phase
velocities of about 2000 m/s and predominantly
arrives from within 90 degrees of a southeasterly
direction at Atlantis. The amplitude of this energy
varies with the sea state.
Figure 5: Smoothed power spectral densities recorded at a single node on the abyssal plane for fourteen different 3-hour windows
from a 2009 Atlantis dataset. Above 4 Hz, the acoustic power level primarily depends on the proximity and repeat interval of the
nearest operating airgun array. Below 2 Hz, the acoustic power level depends on the sea state.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 68
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 69
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Clarke, R., G. Xia, N. Kabir, L. Sirgue, and S. Michell, 2006, Case study: A large 3D wide-azimuth
ocean-bottom-node survey in deepwater GoM: 76th Annual International Meeting, SEG,
Expanded Abstracts, MC 1.2, 11281132, http://dx.doi.org/10.1190/1.2369717.
Crawford, W. C., and S. C. Webb, 2002, Variations in the distribution of magma in the lower crust and at
the Moho beneath the East Pacific Rise at 910N: Earth and Planetary Science Letters, 203,
117130, http://dx.doi.org/10.1016/S0012-821X(02)00831-2.
Dellinger, J., and J. Ehlers, 2007, Low frequencies with a dense OBS array: The Atlantis-Green Canyon
earthquake dataset: 77th Annual International Meeting, SEG, Expanded Abstracts, 3640,
http://dx.doi.org/10.1190/1.2792377.
Estabrook, B. J., D. W. Ponirakis, C. W. Clark, and A. N. Rice, 2016, Widespread spatial and temporal
extent of anthropogenic noise across the Gulf of Mexico: Implications of Ocean Noise and the
Marine Ecosystem (Endangered Species Research).
Hegna, S., and G. Parkes, 2011, The low frequency output of marine air-gun arrays: 81st Annual
International Meeting, SEG, Expanded Abstracts, 7781, http://dx.doi.org/10.1190/1.3628192.
Longuet-Higgins, M. S., 1950, A theory of the origin of microseisms: Philosophical Transactions of the
Royal Society of London A: Mathematical: Physical and Engineering Sciences, 243, 135,
http://dx.doi.org/10.1098/rsta.1950.0012.
Regone, C. J., 1997, Measurement and identification of 3-D coherent noise generated from irregular
surface carbonates, in Marfurt, K. J, ed., Carbonate seismology: SEG, 281306,
http://dx.doi.org/10.1190/1.9781560802099.ch11.
Reilly, J. M., 2016, Marine broadband technology: History and remaining challenges from an end-user
perspective: The Leading Edge, 936941, http://dx.doi.org/10.1190/tle35040316.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 70
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Source deghosting
Johan O. A. Robertsson*, Dirk-Jan van Manen, ETH-Zurich, Seismic Apparition GmbH; Fredrik Andersson,
Lund University; Kurt Eggenberger, Seismic Apparition GmbH; Lasse Amundsen, NTNU, Statoil
(B)!
(A)!
Summary
!(rad/s)!
!(rad/s)!
Signal!cone!
H
Zero!!
signal!
kN
Zero!
signal!
k!(1/m)!
+kN
Zero!
signal!
kN
H+
H
Zero!
signal!
k!(1/m)!
+kN
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Source deghosting
where = ! ! .
In the frequency-wavenumber
domain we therefore obtain the following expression for
the ghost wavefield using capital letters for the variables in
the frequency-wavenumber domain:
! , ! , ! =
!
!
!
!
1 + , ! , ! ! , ! , ! +
1 , ! , ! ! , ! ! , !
(4)
(1)
1+
!"!!
!
!
1 + 1
!
+ 1
!
!!
!"!!
+ 1 1
!
!!
=
(2)
, ! , ! = !!! !! =
(3)
(5)
1+
!"!!
!
!
!
1 + 1
+ 1
!
!!
!"!!
+ 1 1
!
!!
=
(6)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Source deghosting
, ! , ! = !! .
(7)
!
!
1 + !! ! , ! , ! +
, ! ! , !
!
!
1
(8)
!
!
!
1 + , ! , ! ! , ! , ! +
, ! , ! ,
!
!
1+
(9)
!
!
!
1 , ! , ! ! , ! , ! +
, ! , ! .
!
!
1
(10)
!!! ! ! ! !!! ! !
!!!!!
(11)
(12)
(13)
and
! ! = !! !!!!! /!! ! .
(14)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Source deghosting
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
Ziolkowski et al. (1982) proposed to place an array of nearfield hydrophones in close proximity to the air guns in an air
gun array (usually roughly 1 m above each air gun in an air
gun array), with the initial purpose to measure near-field
wavelets from all the air guns. By treating the air guns as
notional sources, a recursive inversion process was proposed
(Ziolkowski et al. 1982; Parkes et al. 1984) to separate the
NFH data into individual point sources located at the air gun
locations. With these point sources, a direct water arrival at
any location, including a far-field source signature, could be
obtained by various modeling methods. Lots of studies have
been carried out to improve the far-field source signature
estimation (Kragh et al. 2000; Hopperstad et al. 2006; Niang
et al 2013; Ni et al 2014; Ni et al. 2015).
2016 SEG
SEG International Exposition and 86th Annual Meeting
Method
() =
=1 () {
},
(1)
Page 77
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
+ ()
(2)
Ziolkowskis method enjoys some success in fullycalibrated and well-controlled experimental cases
(Ziolkowski 1987). However, real data exhibit uncertainties
in amplitude and airgun geometry, as well as angle- and
frequency-dependent sea-surface reflectivity. Therefore, we
improve the notional source inversion method by including
extra terms to tackle amplitude uncertainty and sea-surface
reflectivity, and an extra step to update the geometry. The
updated equation is as follows:
() = {
=1, () [
()
+ (, , )
} + (),
2(
]+
(3)
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 78
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 3. (a) True NFH data. (b) Modeled direct wave. (c)
Adaptive subtraction result of (a) and (b). Ideally (c) only contains
reflection data.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 79
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusion
We propose new algorithms and workflows to better take
advantage of near-field hydrophone data for signal
enhancements, including separating direct wave and
reflections on both active and passive NFH arrays, and
deriving zero and nearly-zero offset traces for shallow water
de-multiple processing. The synthetic results shown in this
paper cover the use of NFH data for the derivation of farfield source signature, mitigating bubble effects, and doing
shallow water de-multiple. The shallow water de-multiple
results show that with the help of NFH data, the missing
near-offset data reconstructions are much more accurate, and
hence the stacked results after the de-multiple processing are
2016 SEG
SEG International Exposition and 86th Annual Meeting
Figure 5. (a) True stack. (2) Demultiple result using the full offset
data. (3) Demultiple result using interpolated data with NFH data.
(4) Demultiple result using extrapolated data. All stacks use only
offset beyond 120 m.
Page 80
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Hopperstad, J. F., and R. O. B. E. R. T. Laws, 2006, Source signature estimation Attenuation of the
seafloor reflection error in shallow water: 68th Annual International Conference and Exhibition,
EAGE, Extended Abstracts, http://dx.doi.org/10.3997/2214-4609.201402115.
Kragh, E., R. Laws, and A. zbek, 2000, Source signature estimation Attenuation of the sea-bottom
reflection error from near-field measurements: 62nd Annual International Conference and
Exhibition, EAGE, Extended Abstracts.
Kragh, J. E., J. O. A. Robertsson, E. J. Muyzert, and C. Kostov, 2015, Zero-offset seismic trace
construction: U.S. Patent 8,958,266 B2.
Martin, G. S., K. J. Marfurt, and S. Larsen, 2002, Marmousi-2: An updated model for the investigation of
AVO in structurally complex areas: 72nd Annual International Meeting, SEG, Expanded
Abstracts, 21, 19791982, http://dx.doi.org/10.1190/1.1817083.
Martin, G. S., R. Wiley, and K. J. Marfurt, 2006, Marmousi-2: An elastic upgrade for Marmousi: The
Leading Edge, 25, 156166, http://dx.doi.org/10.1190/1.2172306.
Ni, Y., F. Haouam, and R. Siliqi, 2015, Source signature estimation in shallow water surveys: 85th
Annual International Meeting, SEG, Expanded Abstracts, 7175,
http://dx.doi.org/10.1190/segam2015-5844754.1.
Ni, Y., T. Payen, and A. Vesin, 2014, Joint inversion of near-field and far-field hydrophone data for
source signature estimation: 84th Annual International Meeting, SEG, Expanded Abstracts, 57
61, http://dx.doi.org/10.1190/segam2014-1193.1.
Niang, C., Y. Ni, and J. Svay, 2013, Monitoring of air-gun source signature directivity: 83rd Annual
International Meeting, SEG, Expanded Abstracts, 4145, http://dx.doi.org/10.1190/segam20130940.1.
Norris, M. W., R. E. Lory, and Y. A. Paisant, 2011, Zero-offset profile from near-field hydrophone data:
73rd Annual International Conference and Exhibition, EAGE, Extended Abstracts,
http://dx.doi.org/10.3997/2214-4609.20149255.
Parkes, G. E., A. Ziolkowski, L. Hatton, and T. Haugland, 1984, The signature of an airgun array:
Computation from near-field measurements including interactions Practical considerations:
Geophysics, 49, 105111, http://dx.doi.org/10.1190/1.1441640.
Ziolkowski, A., G. Parkes, L. Hatton, and T. Haugland, 1982, The signature of an air gun array:
Computation from near-field measurements including interactions: Geophysics, 47, 14131421,
http://dx.doi.org/10.1190/1.1441289.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 81
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Amundsen, L., 1993, Wavenumber-based filtering of marine point-source data: Geophysics, 58, 1335
1348, http://dx.doi.org/10.1190/1.1443516.
Amundsen, L., and J. O. A. Robertsson, 2014, Wave equation processing using finite-difference
propagators, Part 1: Wavefield dissection and imaging of marine multicomponent seismic data:
Geophysics, 79, no. 6, T287T300, http://dx.doi.org/10.1190/geo2014-0151.1.
Amundsen, L., and H. Zhou, 2013, Low-frequency seismic deghosting: Geophysics, 78, no. 2, WA15
WA20, http://dx.doi.org/10.1190/geo2012-0276.1.
Amundsen, L., T. Rsten, J. O. A. Robertsson, and E. Kragh, 2005, On rough-sea deghosting of streamer
seismic data using pressure gradient approximations: Geophysics, 70, no. 1, V1V9,
http://dx.doi.org/10.1190/1.1852892.
Amundsen, L., H. Zhou, A. Reitan, and A. B. Weglein, 2013, On seismic deghosting by spatial
deconvolution: Geophysics, 78, no. 6, V267V271, http://dx.doi.org/10.1190/geo2013-0198.1.
Beasley, C. J., R. T. Coates, Y. Ji, and J. Perdomo, 2013, Wave equation receiver deghosting: A
provocative example: 83rd Annual International Meeting, SEG, Expanded Abstracts, 42264230,
http://dx.doi.org/10.1190/segam2013-0693.1.
Berkhout, A. J., and G. Blacquire, 2016, Deghosting by echo-deblending: Geophysical Prospecting, 64,
406420, http://dx.doi.org/10.1111/1365-2478.12293.
Bracewell, R., 1999, The Fourier transform and its applications (3rd ed.): McGraw-Hill Science.
Fokkema, J. T., and P. M. van den Berg, 1993, Seismic applications of acoustic reciprocity: Elsevier
Science, http://dx.doi.org/10.3997/2214-4609.201411552.
Halliday, D., J. O. A. Robertsson, I. Vasconcelos, D. J. van Manen, R. Laws, K. Ozdemir, and H.
Groenaas, 2012, Full-wavefield, towed-marine seismic acquisition and applications: 82nd Annual
International Meeting, SEG, Expanded Abstracts, http://dx.doi.org/10.1190/segam2012-0446.1.
Moldoveanu, N., 2000, Vertical source array in marine seismic exploration: 70th Annual International
Meeting, SEG, Expanded Abstracts, 5356, http://dx.doi.org/10.1190/1.1816117.
Posthumus, B. J., 1993, Deghosting using a twin streamer configuration: Geophysical Prospecting, 41,
267286, http://dx.doi.org/10.1111/j.1365-2478.1993.tb00570.x.
Robertsson, J. O. A., and L. Amundsen, 2014, Wave equation processing using finite-difference
propagators, Part 2: Deghosting of marine hydrophone seismic data: Geophysics, 79, no. 6,
T301T312, http://dx.doi.org/10.1190/geo2014-0152.1.
Robertsson, J. O. A., and E. Kragh, 2002, Rough sea deghosting using a single streamer and a pressure
gradient approximation: Geophysics, 67, 20052011, http://dx.doi.org/10.1190/1.1527100.
Robertsson, J. O. A., L. Amundsen, and . Sjen Pedersen, 2016a, A seismic shift: Wavefield signal
apparition: Physical Review Letters, submitted.
Robertsson, J. O. A., L. Amundsen, and . Sjen Pedersen, 2016b, Wavefield signal apparition, Part I:
Theory: EAGE.
Robertsson, J. O. A., D. F. Halliday, D. J. van Manen, I. Vasconcelos, R. M. Laws, and H. Gronaas, 2012,
Full-wavefield, towed-marine seismic acquisition and applications: 74th Annual International
Conference and Exhibition, EAGE, Extended Abstracts, http://dx.doi.org/10.3997/22144609.20148849.
Sjen Pedersen, ., L. Amundsen, and J. O. A. Robertsson, 2016, Wavefield signal apparition, Part II:
Applications: EAGE.
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Tian, Y., G. Hampson, K. Davies, J. Cocker, and A. Strudley, 2015, GeoSource de-blending, de-ghosting,
and designature: EAGE.
van Manen, D. J., J. O. A. Robertsson, F. Andersson, and K. Eggenberger, 2016, Aperiodic wavefield
signal apparition: Dealiased simultaneous source separation: 86th Annual International Meeting,
SEG, Expanded Abstracts, submitted.
van Melle, F. A., and K. R. Weatherburn, 1953, Ghost reflections caused by energy initially reflected
above the level of the shot: Geophysics, 18, 793804, http://dx.doi.org/10.1190/1.1437929.
Weglein, A. B., S. A. Shaw, K. H. Matson, J. L. Sheiman, R. H. Stolt, T. H. Tan, A. Osen, G. P. Correa,
K. A. Innanen, Z. Guo, and J. Zhang, 2002, New approaches to deghosting towed-streamer and
ocean-bottom pressure measurements: 72nd Annual International Meeting, SEG, Expanded
Abstracts, 10161019, http://dx.doi.org/10.1190/1.1817121.
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
In the current economic climate cost and efficiency of a
new marine seismic experiment are more crucial than ever.
However, current economic conditions also lead to
increased risk adversity towards novel acquisition or
processing techniques. In consequence, seismic explorers
find themselves in a situation where they are tasked with
improving survey performance without jeopardizing the
survey target.
The last decade has seen the emergence of blended and
continuous acquisition techniques that promise to deliver
more data in less time by near-simultaneously triggering
multiple seismic sources (cf. Beasley, 2008; Berkhout,
2008; Hampson et al., 2008). However, deploying these
techniques in the field usually entails specialized
acquisition with multiple source vessels and randomized
time delays (cf. Borselen et al., 2012; Maraschini et al.,
2012). These specialized acquisition schemes inevitably
increase the cost and complexity of seismic operations.
While deblending techniques based on random noise
attenuation and dip separation achieve a good separation of
the blended sources, they cannot deliver the unperturbed
shot records in the target interval.
In this abstract, we will present a novel acquisition and
2016 SEG
SEG International Exposition and 86th Annual Meeting
Method
During blended acquisition seismic sources are triggered at
intervals shorter than the total record length. In
consequence, each seismic record contains signal from
multiple sources. To protect the primary survey target from
seismic interference, we define a minimum unperturbed
record length. The next blended source is triggered no
earlier than the resulting time delay. This allows us to
deliver unperturbed seismic data over the primary target
window, while still having access to longer records.
To separate the blended seismic sources, we combine two
complimentary processing approaches. In a first step, we
estimate and remove parts of the seismic wavefield
effectively eliminating the predictable part of the blended
seismic signal. This estimation and subtraction (SES)
approach is inspired by demultiple processing sequences. In
a second step, we attenuate residual seismic interference
using erratic noise attenuation and/or time-variant low-pass
filtering. We note that we treat linear events (direct wave)
and hyperbolic events (seafloor reflection) separately.
To estimate the blended direct wave, we carry out a median
stack of all blended shot records in the common channel
domain. This approach honors the linear moveout of the
direct wave. Some attention is required for near-offset
traces to avoid wavelet truncation effects and filter roll-on.
This method of direct wave estimation makes two
assumptions: stability of the source signature along the
profile and sufficient variations in shot time delay to
remove wrap-around signal. The estimated blended direct
Page 82
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Synthetic Example
The following synthetic example compares two deblending
methods: multi-directional vector median filtering (MDVMF) (Huo et al., 2012) and our SES method. This
example demonstrates that the SES method works at least
as well as other deblending methods.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 83
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 2: Synthetic shot gathers (a-c) and stacked seismic sections (d-f) before and after deblending. Only the near-offset traces (< 1 km) were
used in the stack. Flow 1 included a multi-directional vector median filter in the channel domain. Flow 2 applied our SES method.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 84
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 3: Observed shot gathers (a-c) and near-offset stacked seismic sections (d-f) before and after attenuation of the direct wave and near
seafloor reflections. Only the near-offset traces (< 1 km) were used in the stack. The direct wave was estimated using a median stack and the near
seafloor energy was estimated using a Radon transform.
Acknowledgements
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 85
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Beasley, C., 2008, A new look at marine simultaneous sources: The Leading Edge, 27, 914917,
http://dx.doi.org/10.1190/1.2954033.
Berkhout, A., 2008, Changing the mindset in seismic data acquisition: The Leading Edge, 27, 924938,
http://dx.doi.org/10.1190/1.2954035.
Hampson, G., J. Stefani, and F. Herkenhoff, 2008, Acquisition using simultaneous sources: The Leading
Edge, 27, 918923, http://dx.doi.org/10.1190/1.2954034.
Huo, S., Y. Luo, and P. Kelamis, 2012, Simultaneous Sources Separation via Multidirectional VectorMedian Filtering: Geophysics, 77, V123V131, http://dx.doi.org/10.1190/geo2011-0254.1.
Hemon, C., and D. Mace, 1978, Essai dune application de la transformation de Karhunen-Love au
traitement sismique: Geophysical Prospecting, 26, 600626, http://dx.doi.org/10.1111/j.13652478.1978.tb01620.x.
Maraschini, M., R. Dyer, K. Stevens, and D. Bird, 2012, Source Separation by Iterative Rank Reduction Theory and Applications, 74th EAGE Conference and Exhibition incorporating EUROPEC 2012,
A044, http://dx.doi.org/10.3997/2214-4609.20148370.
van Borselen, R., R. Baardman, T. Martin, B. Goswami, and E. Fromyr, 2012, An Inversion Approach to
Separating Sources in Marine Simultaneous Shooting Acquisition Application to a Gulf of
Mexico Data Set: Geophysical Prospecting, 60, 640647, http://dx.doi.org/10.1111/j.13652478.2012.01076.x.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 86
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
In a blended acquisition source encoding is needed for the
separation of the blended sources. In marine seismic
surveys, many approaches of temporal source encoding
have been employed (Abma et al., 2015; Mueller et al.,
2015; Wu et al., 2015; Vaage, 2002). In this work, we
consider the naturally blended source, i.e. source ghosts, as
part of the blending code (Berkhout and Blacquire, 2014).
With the help of this natural blending code in depth, it is
possible to use the source ghost for deblending.
Additionally, it is easy to combine with man-made source
codes and provides an interesting alternative to deal with
the current depth distributed broadband source.
In this abstract, we present three cases where source ghosts
are treated as signal and then separated from the source
response. In the first case, two sources are activated near
simultaneously at different lateral locations. They are
towed at different depths, and therefore these two sources
also have different source ghosts correspondingly. In the
second case, the blended source geometry is the same as in
the first case. However this time each physical source is
activated in a shot-repetition fashion, i.e. activated twice
with certain time delays (Wu et al., 2015). The third case
contains two sources situated at the same lateral position
but at different depths. This is an analog of the current
depth distributed source and a field data example will be
discussed.
Forward model
The forward model of blending with the source ghost is
formulated based on the matrix representation described in
2016 SEG
SEG International Exposition and 86th Annual Meeting
(1)
(3)
Page 87
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 1: Blending two sources of different depths: a) blended data, b) pseudo-deblended left source, c) deblended and deghosted left source;
blending two sources of different depths and shot repetition encoding: d) blended data, e) pseudo-deblended left source, f) deblended and
deghosted left source; blending multi-level sources: g) blended data, h) pseudo-deblended data, i) deblended and deghosted source.
2016 SEG
SEG International Exposition and 86th Annual Meeting
= ( H ) H (H ) H .
(4)
Page 88
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 89
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Acknowledgements
Figure 2: a) Blended field data and d) its f-k spectrum; b) deblended and deghosted data and e) its f-k spectrum.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 90
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Abma, R., D. Howe, M. Foster, I. Ahmed, M. Tanis, Q. Zhang, A. Arogunmati, and G. Alexander, 2015,
Independent simultaneous source acquisition and processing: Geophysics, 80, WD37WD44,
http://dx.doi.org/10.1190/geo2015-0078.1.
Berkhout, A. J., 1982, Seismic migration, imaging of acoustic energy by wave field extrapolation, A:
Theoretical aspects: Elsevier.
Berkhout, A. J., and G. Blacquire, 2014, Combining deblending with source deghosting: 76th EAGE
Meeting, Extended Abstracts.
Mahdad, A., P. Doulgeris, and G. Blacquiere, 2011, Separation of blended data by iterative estimation
and subtraction of blending interference noise: Geophysics, 76, Q9Q17,
http://dx.doi.org/10.1190/1.3556597.
Mueller, M. B., D. F. Halliday, D. J. van Manen, and J. O. A. Robertsson, 2015, The benefit of encoded
source sequences for simultaneous source separation: Geophysics, 80, V133V143,
http://dx.doi.org/10.1190/geo2015-0015.1.
Vaage, S. T., 2002, Method and system for acquiring marine seismic data by using multiple sources: US
patent 6,906,981.
Wu, S., G. Blacquire, and G. van Groenestijn, 2015, Shot repetition: an alternative approach to blending
in marine seismic: SEG Technical Program Expanded Abstracts,
http://dx.doi.org/10.1190/segam2015-5858788.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 91
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 92
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Acquisition Efficiency
Efficiently acquiring a survey of these dimensions depends
to a large extent on the node inventory available, the
capability of deploying a large receiver patch and reliably
recording data during the entire time the node is on the
seabed. This has been enabled by advances made in node
battery endurance helped by reduced power requirements
of modern atomic clocks. The 2015 surveys set a new
record for the largest live deep water OBN patch. Nodes
were on the seabed an average of 85 and 95 days on the
Atlantis and Thunder Horse surveys respectively, compared
to only 30 days during the 2005 survey. The low recording
endurance resulted in the 2005 acquisition being acquired
in two patches, introducing a zipper with a large percentage
of shot points repeated as a consequence. An additional
consequence of the limited battery life restricted the cross
line offset to 6Km. Node recording locations were
increased from 1628 on the Atlantis baseline survey up to
1912 stations on the 2014/5 monitor. The additional nodes
were included to extend the image. The introduction of two
ROVs and the HSL to the operation contributed to
significant improvement in efficiency compared to previous
operations.
We can visualize the Thunder Horse acquisition
productivity by extracting the daily production profiles
(Figure 2). The figure demonstrates the sensitivity of
extending
the
source
patch
on
the
node
deployment/retrieval. Extending the cross-line offset
requirement from 10km to 14km caused significant waiting
time for the node vessel, as it was waiting on the source
vessel to move a sufficient distance in order to clear nodes
for retrieval.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 93
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
With the availability of the aforementioned highproductivity source methods the focus should be on how to
improve the receiver efficiency of such surveys and where
the bottlenecks are. For instance the need for high accuracy
receiver positioning is fundamentally important for 4D
repeatability, however the desire to lay or shoot as close to
the pre-plot impacts productivity, resulting in slow
deployment speeds. With the gradual increase in receiver
inventory, dense efficient OBS patch geometry options
could be exploited. With high trace density 3D surveys
there is a case to relax the pre-plot to post-plot positioning
requirements to enable faster deployment.
Time and motion studies
We have actively been conducting time and motion studies
to assess current and developing acquisition technologies
using a GOM design area. Using various ocean bottom
survey (OBS) technologies and known operating conditions
with benchmarked vessel and equipment rates, it is possible
to estimate survey durations and costs.
Figure 5 graphed results of a time and motion study for a
deepwater (approx. 2,000m) OBS surveys using different
deployment methods and source effort. Using a triple
source array reduces the total number of sail lines to
acquire. For this particular survey extent and modelled
survey geometry, adding a second source vessel further
reduces survey duration. For certain shot constrained
scenarios although survey duration has been reduced, the
additional cost of the second source boat increases the
overall survey cost. Hence a single triple source vessel
could be more desirable. Faster receiver deployment will
reduce both survey duration and cost. This combined
application of several technologies optimizes the duration
of the survey and provides significant cost savings
compared to conventional ROV guided node deployment
operation with a single, dual-source vessel. These savings
Page 94
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Acknowledgments
The Authors would like to thank BP, BHP Billiton and
ExxonMobil for permission to publish this work. Also
thanks to FairfieldNodals deep-water node crew who
acquired the surveys discussed.
ISS is a registered trade mark of BP p.l.c.
Conclusions
Advancements in deep-water OBN surveys have led to
more operational efficiency and improved imaging over the
Atlantis and Thunder Horse fields. However there is more
that can be achieved. Through the combination of new
efficient node deployment methods for deep-water,
simultaneous source techniques, and larger node
inventories more routine application of higher density OBN
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 95
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Beaudoin, G., and A. A. Ross, 2007, Field design and operation of a novel deepwater, wide azimuth node
seismic survey: The Leading Edge, 26, 494, http://dx.doi.org/10.1190/1.2723213.
Ellis, D., 2013, Simultaneous source acquisition Achievements and challenges: 75th Annual
International Conference and Exhibition, EAGE, Extended Abstracts, Th-08-01.
Hager, E., M. Rocke, and P. Fontana, 2015, Efficient multi-source and multi-streamer configuration for
dense cross-line sampling: 85th Annual International Meeting, SEG, Expanded Abstracts,
100-104, http://dx.doi.org/10.1190/segam2015-5857262.1.
Keggin, J., and R. Abma, 2012, Simultaneous shooting, today and tomorrow: 74th Annual International
Conference and Exhibition, EAGE, Extended Abstracts, http://dx.doi.org/10.3997/22144609.20149761.
Ourabah, A., J. Keggin, C. Brooks, D. Ellis, and J. Etgen, 2015, Seismic acquisition, what really matters?:
85th Annual International Meeting, SEG, Expanded Abstracts, 6-11,
http://dx.doi.org/10.1190/segam2015-5844787.1.
Reasnor, M., G. Beaudoin, M. Pfister, I. Ahmed, S. Davis, M. Roberts, J. Howie, G. Openshaw, and A.
Longo, 2010, Atlantis time-lapse ocean bottom node survey: A project team's journey from
acquisition through processing: 80th Annual International Meeting, SEG, Expanded Abstracts,
41554159, http://dx.doi.org/10.1190/1.3513730.
Van Gestel, J., E. LHeureux, J. R. Sandschaper, P. O. Ariston, N. D. Bassett, and S. Dadi, 2015, Atlantis
beyond 4D ocean bottom nodes acquisition design: 85th Annual International Meeting, SEG,
Expanded Abstracts, 125-129, http://dx.doi.org/10.1190/segam2015-5847522.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 96
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 97
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 98
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 2. Hydrophone FK spectra: a) 2D-deghosted after deblending; b) 2D-deghosted single-source data; and, c) single-source hydrophone data
before deghosting.
Figure 3. Deblended CMP gather a) before and b) after Radon multiple suppression, showing how deblending has preserved the weak primary
reflectors below 3.5 seconds. The blue and green curves respectively indicate the water bottom and the start of its multiple train.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 99
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 4. PSTM of a 12-streamer swath: (a) simultaneous-source data after removing the undershoot vessel with deblending; and, (b) control
data acquired with only the streamer vessel shooting.
Conclusions
Advances in 3D acquisition and processing over the past
twenty-five years have produced high-quality data and
made a tremendous impact on our business: from derisking
exploration wells to reservoir management and reserves
estimates. With that change has come a natural reluctance
to tamper with success, particularly with a technology such
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 100
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Davies, D., and M. Ibram, 2015, Evaluating the impact of ISS HD-OBC acquisition on 4D data: 77th
Annual International Conference and Exhibition, EAGE, Extended Abstracts, N101 05.
Jenkerson, M., H. Clark, R. Houck, S. Seyb, and M. Walsh, 1999, Effects of seismic interference on 3D
data: 69th Annual International meeting, SEG, Expanded Abstracts, 1208-1211,
http://dx.doi.org/10.1190/1.1820722.
Liu, Z., B. Wang, J. Specht, J. Sposato, and Y. Zhai, 2014, Enhanced adaptive subtraction method for
simultaneous source separation: 84th Annual International meeting, SEG, Expanded Abstracts,
115-119, http://dx.doi.org/10.1190/segam2014-1510.1.
Mahdad, A., P. Doulgeris, and G. Blacquiere, 2011, Separation of blended data by iterative estimation
and subtraction of Blending interference noise: Geophysics, 76, no. 3, Q9Q17,
http://dx.doi.org/10.1190/1.3556597.
Sabel, P., F. Bker, C. Otterbein, B. Szydlik, I. Moore, and C. Beasley, 2014, An interpreters view on 4D
with simultaneous-source acquisition: 76th Annual International Conference and Exhibition,
EAGE, Extended Abstracts, ELI2 07.
Van den berg, E., and M. P. Friedlander, 2008, Probing the Pareto frontier for basis pursuit solutions:
SIAM Journal on Scientific Computing, 31, 890912, http://dx.doi.org/10.1137/080714488.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 101
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Dirk-Jan van Manen , Johan O. A. Robertsson, ETH Zurich, Seismic Apparition GmbH; Kurt Eggenberger,
Seismic Apparition GmbH; Fredrik Andersson, Lund University; Lasse Amundsen, NTNU Trondheim
SUMMARY
In wavefield signal apparition, acquisition with periodic variations in source activation parameters shifts all or part of a signal cone out to, e.g., the Nyquist wavenumber, enabling perfect
separation of simultaneous-source (sim-source) data below a
certain temporal frequency. Cyclic convolution in the spatial
frequency domain is used to generalise the signal apparition
concept to the case of aperiodic variations in the source activation parameters. The separation is formulated as a leastsquares reconstruction in the frequency-wavenumber ( f k) domain. The methodology is used in combination with normal
moveout correction to de-alias the sim-source separation and
enables shooting on position in sim-source acquisition.
Figure 1: Left (a): Conventional acquisition: all signal energy sits inside a signal cone (yellow) bounded by the propagation velocity of the recording medium. Right (b): Acquisition
following the principles of periodic signal apparition. After
Robertsson et al. (2016).
INTRODUCTION
In simultaneous source acquisition, two (or more) sources interfere during acquisition as for instance a second source is
triggered simultaneously or close in time with the first. The
wavefields interfere and at the receiver location the sum of the
wavefields is measured. The interference of signals is handled in data processing to decode the information generated
from each source. The literature on the subject is vast. Key
references can be found in Pedersen et al. (2016). Recently,
the concept of wavefield signal apparition was introduced to
enable perfect sim-source separation below a certain temporal
frequency by shifting all or part of a wavefield signal cone out
to the Nyquist wavenumber through acquiring the data with periodic variations in a source (activation) parameter (Robertsson
et al., 2016). In the following, we briefly review the concept
of signal apparition before extending it to the case of aperiodic variations in activation parameters and showing how the
corresponding reconstruction methods can be used to improve
sim-source separation in the presence of aliasing.
1
1
[1 + (1)n ] + A [1 (1)n ] ,
2
2
(1)
and an expression for the wavenumber spectrum of the resulting wavefield, H(k), in terms of the spectrum of the conven-
2016 SEG
SEG International Exposition and 86th Annual Meeting
N1
1X
m(n) f (n)ei2kn/N
N
1
1
[1 + A] F(k) + [1 A] F(k kN ).
2
2
n=0
(2)
Page 102
0.8
0.8
0.6
0.6
f [Hz]
f [Hz]
0.4
0.2
0.2
0
1
0.5
0
k [1/m]
(4)
.
..
..
CM =
.
M(N 2) M(N 3)
M(N 1) M(N 2)
..
.
M(2)
M(1)
M(3)
M(2)
..
..
.
.
.
M(0) M(N 1)
M(1)
M(0)
(5)
From the above description it is clear that the effect of a general non-periodic modulation function will be to introduce additional, scaled replications of the signal cones of the source(s)
along the wavenumber axis (or axes in 3D). Both the position
and the factor scaling the replications are determined by the
discrete Fourier transform of the aperiodic modulation function, M(k), which cyclically convolves the signal cones.
Cyclic convolution is illustrated in Figure 2. In panel (a), a signal cone in a normalized f k-domain is shown. A real and symmetric aperiodic amplitude modulation function and its discrete Fourier transform are shown in the bottom and top parts
of panel (d), respectively. The corresponding cyclic convolution matrix is shown in panel (c). The effect of applying the
2016 SEG
SEG International Exposition and 86th Annual Meeting
0.5
0.8
0.6
0.4
0.5
0.2
0.5
0
k [1/m]
0.5
amplitude [a.u.]
0
k [1/m]
(b)
0.5
(3)
0.5
(a)
Let s(n) represent some source wavefield acquired as a function of a spatial coordinate. Here we exploit the lesser-known
dual of the convolution theorem, which states that multiplication in the space domain of s(n) with a so-called modulation
function m(n), corresponds to cyclic convolution of the discrete Fourier transform of the source wavefield, S(k), with the
discrete Fourier transform of the modulation function M(k) =
F {m(n)}, followed by an inverse discrete Fourier transform:
0
1
0.5
0.4
1
0.5
amplitude [a.u.]
transforms yield the separate wavefields g and h in the timespace domain. A practical choice for A(), as pointed out by
Robertsson et al. (2016), is to choose the case where every
second source is excited a time T later compared to neighboring recordings: A() = eiT or a(t) = (t T ), as it does not
require the ability to flip polarity of the source signal. The apparition concept can be extended to simultaneous acquisition
of more than two source lines by choosing different modulation functions.
k [1/m]
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
0
1
x 10
0.5
0
k [1/m]
0.5
5
0
5
(c)
50
100
150
x [m]
200
250
(d)
(6)
(7)
The data in s2 times, d2 , can be obtained by applying the inverse of the applied time shifts, T (n), to the data in s1 times:
d2 (t, n) = F 1 {F {d1 (t, n)}eiT (n) },
(8)
Page 103
40
30
30
1.5
20
1.5
20
10
2.5
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
f [Hz]
0.5
t [s]
40
f [Hz]
t [s]
0
0.5
3
5000
0
x [m]
0
10
5000
10
2.5
(a)
0
k [1/m]
5
3
x 10
(b)
3
5000
0
x [m]
5000
(c)
0
10
0
k [1/m]
5
3
x 10
(d)
Figure 3: Sim-source data in the tx- and f k-domain (a and b) and after reversible NMO correction (c and d). See text for details.
G=
CM11
CM21
CM12
CM22
=
CM12
I
CM21
.
(9)
(10)
i 0, . . . , N 1
j 0, . . . , Kmax Kmin .
I
M
C
21
M
C
12
I
,
(12)
(13)
2016 SEG
SEG International Exposition and 86th Annual Meeting
of the data. Separation according to periodic apparition principles can only be applied in the diamond-shaped region below
22 Hz. Clearly some other methodology is needed to separate
the sources without suffering from aliasing effects. In Figure 3c, the sim-source data are shown with a reversible normal
moveout (NMO) correction applied. The f k-transform of the
NMO-corrected data is shown in Figure 3d. As can be seen, the
NMO correction de-aliases the data and the apparated s2 data is
now clearly visible at low frequencies around kN . However,
due to the space- and time-variant NMO stretch, the encoding time shift is now also space- and time-variant. We model
the effective time shift using an exact expression for NMO
stretch (Barnes, 1992). To deal with the time-variant time
shift, the NMO-corrected data are transformed to the timefrequency domain using the S-transform (Stockwell, 1996).
The S-transforms of the separated, NMO-corrected sources are
then obtained by applying the aperiodic reconstruction method.
In Figure 4 (central column), reconstructed s1 data are shown
in various domains and compared to the reference solution (left
column) by differencing (right column). In Figure 4a-4c, one
slice of the inverted S-transform, i.e., the f x-spectrum of s1
at t = 0.92 s, is evaluated. In Figure 4d-4f, the estimated NMOcorrected s1 wavefield after inverse S-transform is evaluated.
In Figures 4g-4i, the f k-spectrum of the NMO-corrected s1
wavefield is evaluated. In Figures 4j-4l, the reconstructed s1
data after removing the NMO correction are evaluated. Finally, in Figures 4m-4o, the f k-spectrum of the s1 wavefield
is evaluated. Note that the wavefield has been reconstructed
successfully, mostly without suffering from the aliasing.
CONCLUSIONS
Cyclic convolution in the spatial frequency domain was used
to generalise wavefield signal apparition to aperiodic encoding
functions. This makes it possible, for example, to deal with
perturbations or deviations from nominally periodic encoding,
or to acquire sim-source data on positions. Alternatively, it
can be used when processing makes the encoding effectively
aperiodic. It was shown how this can be exploited to perform
anti-aliased source separation when used in conjunction with
an NMO correction and a time-frequency decomposition.
ACKNOWLEDGEMENTS
We thank Seismic Apparition GmbH for permission to publish
proprietary, patent-pending work.
Page 104
40
30
30
30
20
f [Hz]
40
f [Hz]
40
10
20
10
0
5000
0
x [m]
0
5000
5000
0
x [m]
0.5
0.5
0.5
1
t [s]
t [s]
t [s]
1.5
1.5
1.5
2.5
2.5
2.5
0
x [m]
3
5000
5000
0
x [m]
3
5000
5000
30
30
30
f [Hz]
40
f [Hz]
40
10
20
10
0
k [1/m]
0.005
0
0.01
0.01
20
10
0.005
(g)
0
k [1/m]
0.005
0
0.01
0.01
0.5
0.5
1.5
2.5
2.5
3
5000
5000
0
x [m]
(j)
3
5000
5000
30
30
30
f [Hz]
40
f [Hz]
40
10
20
10
0
k [1/m]
0
10
5
3
x 10
(m)
5000
(l)
40
0
10
0
x [m]
(k)
20
0.01
1.5
2.5
0
x [m]
0.005
1
t [s]
1
t [s]
0
k [1/m]
(i)
0.5
3
5000
0.005
(h)
1.5
5000
(f)
40
0.005
0
x [m]
(e)
20
5000
(c)
0
0.01
0
x [m]
(b)
(d)
f [Hz]
0
5000
5000
3
5000
t [s]
20
10
(a)
f [Hz]
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
f [Hz]
20
10
0
k [1/m]
(n)
0
10
5
3
x 10
0
k [1/m]
5
3
x 10
(o)
Figure 4: Source 1 reconstructions in the (a-c) S-transform f x-, (d-f) NMO-corrected tx-, (g-i) NMO-corrected f k-, (j-l) tx-, and
(m-o) f k-domains. The reference, reconstructed, and difference data are shown in the left, centre, and right panels, respectively.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 105
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Barnes, A. E., 1992, Another look at NMO stretch: Geophysics, 57, no. 5, 749751.
http://dx.doi.org/10.1190/1.1443289.
Pedersen, . S., L. Amundsen, and J. O. A. Robertsson, 2016, Wavefield signal apparition, Part II:
Application to simultaneous sources and their separation: 78th Conference & Exhibition, EAGE,
Extended Abstracts.
Robertsson, J. O. A., L. Amundsen, and . S. Pedersen, 2016, Wavefield signal apparition, Part I:
Theory: 78th Conference & Exhibition, EAGE, Extended Abstracts.
Stockwell, R. G., L. Mansinha, and R. P. Lowe, 1996, Localization of the complex spectrum: the S
transform: Signal Processing: IEEE Transactions on, 44, no. 4, 9981001.
http://dx.doi.org/10.1109/78.492555.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 106
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
DEBLENDING METHOD
INTRODUCTION
Simultaneous-source acquisitions (Beasley et al., 1998) are already familiar to the majority of exploration geophysicists.
The decreasing oil costs on the market are even stronger forcing oil and gas companies to rethink their expenses, including the ones related to seismic data acquisitions. Furthermore,
we are now facing the necessity to speed up not only the exploration stage acquisitions, but also the ones having reservoir
characterization and monitoring purposes. Therefore, new efficient acquisition technologies are encouraged to be brought
up to an operational level of maturity. Simultaneous-source
acquisitions allow saving time and money but provide certain
complexity for data processing. Indeed, sources emitting simultaneously interfere with each other, which leads to crosstalk in the data (Berkhout, 2008). A certain number of solutions to this problem have already been proposed by several researchers in this area. There are two main approaches: direct
imaging with inversion of the blended data (Berkhout et al.,
2008; Tang and Biondi, 2009; Xue et al., 2016), and preliminary signal separation (deblending) followed by more or less
conventional processing (Spitz et al., 2008; Cheng and Sacchi,
2015; Chen, 2015). In this paper, we present results obtained
using a method based on the second approach. Several authors
have proven that the existing deblending methods allow acceptable data quality for imaging and structural interpretation
(Dai et al., 2011; Verschuur and Berkhout, 2011; Henin et al.,
2015). There are less publications on 4D and reservoir applications of simultaneous-source data (Ayeni et al., 2009; Haacke
et al., 2015). The main goal of this study is to confirm the
possibility to use blended data in reservoir characterization.
CASE STUDY
Initial data
A 3D land data set was chosen for the study. The data were
acquired in a non-simultaneous mode. The acquisition was
performed using high density single-vibrator (V1) technology
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 3: Non-blended (a and b), blended (c and d) and deblended (e and f) raw common receiver and common shot gathers.
Figure 4: PreSTM results for non-blended (a) and deblended (b) seismic data.
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 5: Horizon maps on major discontinuity at approximately 600 ms through the GCC (a) and the amplitude cube
(b) of the non-blended data set; same horizon maps through
the GCC (c) and the amplitude cube (d) of the deblended data
set. Hardly any differences seen between the two data sets.
CONCLUSIONS
The advantages of simultaneous-source seismic acquisitions
are evident: the same (or even bigger) amounts of data can be
acquired in shorter time delays, which turns out to be beneficial for marine acquisitions as well as for land ones. The only
possible drawback of such acquisition designs cross-talk
is imposing challenges for processing and imaging, especially,
when passing on to highly precise and delicate production surveys. We presented a study which confirms that simultaneous
sources are not only applicable to structural interpretation purpose exploration surveys, but can also be used for reservoir
characterization without any loss in data quality. The overall
conclusion of the amplitudes interpretation quality control test
may be formulated as follows: even though the non-blended
and the deblended data sets are not exactly identical after processing, their level of semblance allows making the same amplitudes interpretation for the zones of interest. Consequently,
we have proven that ISS blended acquisitions are totally compatible with inversion and reservoir characterization.
ACKNOWLEDGMENTS
We are particularly grateful to Sonatrach, Cepsa and Total for
permission to publish the field data example. We would like to
thank Total and CGG for permission to publish this work.
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Abma, R., D. Howe, M. Foster, I. Ahmed, M. Tanis, Q. Zhang, A. Arogunmati, and G. Alexander, 2015,
Independent simultaneous source acquisition and processing: Geophysics, 80, no. 6, WD37
WD44, http://dx.doi.org/10.1190/geo2015-0078.1.
Abma, R., and J. Yan, 2009, Separating simultaneous sources by inversion: Presented at the 71st
European Association of Geoscientists and Engineers Conference and Exhibition,
http://dx.doi.org/10.3997/2214-4609.201400403.
Ayeni, G., Y. Tang, and B. Biondi, 2009, Joint preconditioned least-squares inversion of simultaneous
source time-lapse seismic data sets: 79th Annual International Meeting, SEG, Expanded
Abstracts, 39143918, http://dx.doi.org/10.1190/1.3255685.
Beasley, C. J., R. E. Chambers, and Z. Jiang, 1998, A new look at simultaneous sources: 68th Annual
International Meeting, SEG, Expanded Abstracts, 133135, http://dx.doi.org/10.1190/1.1820149.
Berkhout, A. J., 2008, Changing the mindset in seismic data acquisition: The Leading Edge, 27, 924938,
http://dx.doi.org/10.1190/1.2954035.
Berkhout, A. J., G. Blaquiere, and D. J. Verschuur, 2008, From simultaneous shooting to blended
acquisition: 78th Annual International Meeting, SEG, Expanded Abstracts, 28312838,
http://dx.doi.org/10.1190/1.3063933.
Chen, Y., 2015, Deblending by iterative orthogonalization and seislet thresholding: 85th Annual
International Meeting, SEG, Expanded Abstracts, 5358, http://dx.doi.org/10.1190/segam20155835651.1.
Cheng, J., and M. Sacchi, 2015, Separation and reconstruction of simultaneous source data via iterative
rank reduction: Geophysics, 80, no. 4, V57V66, http://dx.doi.org/10.1190/geo2014-0385.1.
Dai, W., X. Wang, and G. T. Schuster, 2011, Least-squares migration of multisource data with deblurring
filter: Geophysics, 76, no. 5, R135R146, http://dx.doi.org/10.1190/geo2010-0159.1.
Haacke, R. R., G. Hampson, and B. Golebiowski, 2015, Simultaneous shooting for sparse OBN 4D
surveys and deblending using modified Radon operators: Presented at the 77th European
Association of Geoscientists and Engineers Conference and Exhibition,
http://dx.doi.org/10.3997/2214-4609.201412873.
Henin, G., D. Marin, S. Maitra, A. Rollet, S. K. Chandola, S. Kumar, N. E. Kady, and L. C. Foo, 2015,
Deblending 4-component simultaneous-source data A 2D OBC case study in Malaysia: 85th
International Annual Meeting, SEG, Expanded Abstracts, 4347,
http://dx.doi.org/10.1190/segam2015-5899893.1.
Hugonnet, P., and J.-L. Boelle, 2007, Beyond aliasing regularisation by plane event extraction: Presented
at the 69th European Association of Geoscientists and Engineers Conference and Exhibition,
http://dx.doi.org/10.3997/2214-4609.201401830.
Hugonnet, P., J.-L. Boelle, and F. Prat, 2012, Local linear events extraction and filtering in the presence
of time-shifts: Presented at the 74th European Association of Geoscientists and Engineers
Conference and Exhibition, http://dx.doi.org/10.3997/2214-4609.20148307.
Massala, A., N. Keskes, L. Goncalves-Ferreira, C. Onu, S. Cordier, and P. Cordier, 2013, An innovative
attribute for enhancing faults lineaments and sedimentary features during 2G&R interpretation:
Presented at the Society of Petroleum engineers Annual Technical Conference and Exhibition,
http://dx.doi.org/10.2118/166122-MS.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 111
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Meunier, J., T. Bianchi, and S. Ragab, 2007, Experimenting single vibrator seismic acquisition: Presented
at the 69th European Association of Geoscientists and Engineers Conference and Exhibition,
http://dx.doi.org/10.3997/2214-4609.201401981.
Spitz, S., G. Hampson, and A. Pica, 2008, Simultaneous source separation: A prediction-subtraction
approach: 78th Annual International Meeting, SEG, Expanded Abstracts, 28112815,
http://dx.doi.org/10.1190/1.3063929.
Tang, Y., and B. Biondi, 2009, Least-squares migration/inversion of blended data: 79th Annual
International Meeting, SEG, Expanded Abstracts, 28592863,
http://dx.doi.org/10.1190/1.3255444.
Verschuur, D. J., and A. J. Berkhout, 2011, Seismic migration of blended shot records with surfacerelated multiple scattering: Geophysics, 76, A7A13, http://dx.doi.org/10.1190/1.3521658.
Xue, Z., Y. Chen, S. Fomel, and J. Sun, 2016, Seismic imaging of incomplete data and simultaneoussource data using least-squares reverse time migration with shaping regularization: Geophysics,
81, no. 1, S11S20, http://dx.doi.org/10.1190/geo2014-0524.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 112
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Shaohuan Zu and Hui Zhou, China University of Petroleum Beijing; Yangkang Chen, The University of Texas at
Austin; Haolin Chen and Mingqiang Cao, Dagang of BGP; Chunlin Xie, E.&D. Research Institute, Daqing Oilfield
Company
SUMMARY
Simultaneous sources acquisition (continuous recording, significant overlap in time) has many advantages over the traditional seismic acquisition (discontinuous recording, zero overlap in time). When focusing on data quality, blended acquisition (simultaneous sources acquisition) allows significantly
denser spatial sources sampling and much wider range of azimuths. This can improve the quality of subsurface illumination. When focusing on economic aspect, the blended acquisition can greatly shorten the survey time. However, many
challenges such as the continuous recording equipment, the
availability of boats units and the implementation of speed cable vessels are emerging when using simultaneous sources acquisition. Dagang Geophysical Prospecting Branch of BGP,
CNPC has made two field trials to explore the advantages of
simultaneous sources and obtained some experience. The goal
of this paper is to give an overview of the latest field trial and
display the very successful deblending results, which may give
an inspiration to the depressing oil price.
INTRODUCTION
Currently, in the climate of oil price drop keeps the oil and gas
industry on their toes, the advantages of simultaneous sources,
which can improve the quality of illumination with denser shot
coverage or shorten the survey time with wider shot coverage,
have attracted much attention from industry (Berkhout, 1982;
Abma and Yan, 2009; Abma et al., 2012; Abma, 2014; Zhang
et al., 2013) and academia (Bagaini, 2006; Beasley et al., 1998;
Li et al., 2013; Chen et al., 2014; Chen, 2014; Chen et al.,
2015; Chen, 2015; Gan et al., 2015). In land seismics, the simultaneous sources acquisition is the well known vibroseis acquisition. The slip-sweep technique can tremendously reduce
the cycle time and significantly increase productivity. This
method is based on transmitting specially encoded source sweeps
such that the interfering sources responses can be separated in
a preprocessing step.
However, in marine seismics, the use of impulsive sources is
not easy for signal encoding methods. Beasley et al. (1998)
proposes a method that does not require sources signature encoding, but relies on spatial sources positioning to allow for
separation of the signals in subsequent data processing. Hampson et al. (2008) utilizes a technique that two or more shots
with a small random time delay are acquired during the survey time. They demonstrate on three-dimensional (3D) field
data that for deep water with modest water-bottom reflectivity,
no special processing is required, whereas in shallow water
with stronger water-bottom reflectivity the use of shot separation techniques is necessary. Berkhout (2008) introduces the
2016 SEG
SEG International Exposition and 86th Annual Meeting
(1)
Page 113
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
source (d1 ) is coherent, but the record from the second source
(d2 ) is incoherent due to the dithering operator . Then we
apply the inverse of to Equation 1 and get the second blended
equation:
1 d = 1 d1 + d2 ,
(3)
where 1 stands for the inverse of . In the record of 1 d,
the record from the second source (d2 ) is coherent, however,
the record from the first source (1 d1 ) is incoherent. We process the blended record in the common receiver domain because one shot record is coherent and others are incoherent
(Hampson et al., 2008). Combing Equations 1 and 3, we can
obtain
Fm = e
d,
(4)
d
I
d1
.
where e
d=
,F=
and m =
d2
1 d
1 I
Equation 4 is an under-determined system (the number of Equation is less the number of unknown items). One way to overcome the ill-posedness of Equation 4 is to solve the minimization problem of the cost function
= arg minkFm e
m
dk22 + kSmk1 ,
m
(5)
where k k22 stands for the square of L2 norm of the data fitting that measures the differences between the observed and
the modeled data. k k1 stands for the L1 norm of the model
shaping which is used to penalize nonsparse solutions, S is
a sparsity promoting transform and is a penalty coefficient
which keeps the balance between the data fitting and the mode
shaping. Equation 5 can be minimized using iterative shrinkage thresholding algorithm (ISTA) (Beck and Teboulle, 2009)
h
i
mn+1 = C 1 T C mn + FT e
d Fmn ,
(6)
where C and C 1 are a pair of forward and inverse curvelet
transforms. T denotes the threshold function with an threshold parameter , FT is the adjoint of F. The threshold function can be divided into two types: soft and hard thresholdings.
Soft thresholding aims to remove data whose value are smaller
than a certain level and subtract the other values by this level
(Donoho, 1995). Hard thresholding only removes data whose
values are smaller than a certain level. In this paper, we choose
the hard threshold function because according to many numerical tests, it can achieve a higher signal-to-noise ratio (S/N)
than the soft thresholding function. Although the hard thresholding tends to bigger oscillation due to the discontinuity of
the thresholding function.
EXAMPLE
The marine test is carried out in a shallow water and oceanbottom-nodes are used (Figure 1). There are four node lines
(the red dotted lines in Figure 1) at the bottom of water to
receive the blended seismic record and each node line has 400
nodes. The blue and green dotted lines represent the two shot
lines, respectively, and each source vessel shots 301 times. The
sailing direction of the first source boat is A B C D (the
blue dots in Figure 1) and the direction of the second source
2016 SEG
SEG International Exposition and 86th Annual Meeting
2
3
4
Page 114
40
90
Trace
140 190
240
290
15
Trace
25
90
35
Trace
190
290
390
Trace
190
90
290
390
45
-500
A
B
Time (s)
Time (s)
Time (s)
-250
6
6
250
500
(b)
(b)
Trace
140 190
240
290
40
Trace
140 190
240
290
Time (s)
90
In order to quantitatively analyze the damage to signal, we select two unblended sections from Figure 4 as red and blue rectangles shown to analyze the frequency spectrum. In Figure 6a,
the red line represents the amplitude spectrum of the red rectangle shown in Figure 4 and the blue line denotes that of the
same section shown in Figure 5a. The coincidence of the two
lines demonstrates that the damage to the useful signal is very
weak. Figure 6b is the comparison of frequency spectrum between the two blue rectangles shown in Figure 4 and Figure
5b, respectively. Similarly, the blue line is very close to the
red line. From Figure 6, it is very obvious that the deblending
performance is very good and hardly find the damage to the
useful signal.
60
(a)
(b)
10
20
30
40
Frequency
50
60
70
80
90
100
blended data
deblended record of A
50
Amplitude
90
(a)
(a)
40
Time (s)
40
30
20
10
90
Trace
190
290
390
(a)
A
B
60
10
20
30
40
Frequency
50
60
70
80
90
100
blended data
deblended record of B
50
Amplitude
Time (s)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
40
30
20
10
6
0
(b)
8
2016 SEG
SEG International Exposition and 86th Annual Meeting
Figure 6: The comparison of frequency between the two rectangles shown in Figure 4 and that of deblended results shown
in Figure 5a and 5b.
Page 115
200
400
CMP
600
800
CMP
250
350
450
700
450
700
2
Time (s)
(a)
CMP
250
350
1000
0.5
Time (s)
1.5
Time (s)
3
2.5
4
3.5
(b)
4.5
(a)
200
400
CMP
600
800
1000
0.5
1.5
Time (s)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2.5
3.5
4.5
the marine trial directly. Without any preprocessing before deblending, the results of the simultaneous sources field trial are
very encouraging. We have got some experiences of simultaneous shooting from this trial. A comparison of two marine trials that are successful and failed respectively shows that
shot scheduling of simultaneous sources is an important factor in the success of the survey. Good shot scheduling can get
high quality separation performance with no impact from the
blended acquisition. Further analysis, such as AVO and migration comparisons of this trial, will be discussed in the near
future.
ACKNOWLEDGEMENTS
(b)
CONCLUSIONS
The authors would like to thank Dagang Geophysical Prospecting Branch of BGP, CNPC permission to present this paper.
We would like to thank Shuwei Gan, Zhaoyu Jin, Xingye Liu
and Lin Zhou for the inspiring discussion about simultaneous sources. Specially Zu would like to thank Ms. Shan Qu
for teaching him the knowledge about simultaneous sources.
This research is partly supported by the 973 Program of China
(2013CB228603), National of Major Science and Technology
Program (2016ZX05010001-002).
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 116
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Abma, R., 2014, Shot scheduling in simultaneous shooting: 84th Annual International Meeting, SEG,
Expanded Abstracts, 9498.
Abma, R., and J. Yan, 2009, Separating simultaneous sources by inversion: 71st Annual International
Conference and Exhibition, EAGE, Extended Abstract.
Abma, R., Q. Zhang, A. Arogunmati, and G. Beaudoin, 2012, An overview of BPs marine independent
simultaneous source field trials: 82nd Annual International Meeting, SEG, Expanded Abstracts,
15.
Bagaini, C., 2006, Overview of simultaneous vibroseis acquisition methods: 76th Annual International
Meeting, SEG, Expanded Abstracts, 7074.
Beasley, C. J., R. E. Chambers, and Z. Jiang, 1998, A new look at simultaneous sources: 68th Annual
International Meeting, SEG, Expanded Abstracts, 133135.
Beck, A., and M. Teboulle, 2009, A fast iterative shrinkage thresholding algorithm for linear inverse
problems: SIAM Journal on Imaging Sciences, 2, 183202, http://dx.doi.org/10.1137/080716542.
Berkhout, A., 1982, Seismic migration: Imaging of acoustic energy by wave field extrapolation, part a:
Theoretical aspects: Elsevier Scientific Publishing Company.
Berkhout, A. J. G., 2008, Changing the mindset in seismic data acquisition: The Leading Edge, 27, 924
938, http://dx.doi.org/10.1190/1.2954035.
Blacquiere, G., G. Berkhout, and E. Verschuur, 2009, Survey design for blended acquisition: SEG
Technical Program Expanded Abstracts, 2009, 5660.
Chen, Y., 2014, Deblending using a space-varying median filter: Exploration Geophysics, 5183.
Chen, Y., 2015, Iterative deblending with multiple constraints based on shaping regularization:
Geoscience and Remote Sensing Letters IEEE, 12, 22472251,
http://dx.doi.org/10.1109/LGRS.2015.2463815.
Chen, Y., S. Fomel, and J. Hu, 2014, Iterative deblending of simultaneous-source seismic data using
seislet-domain shaping regularization: Geophysics, 79, V179V189,
http://dx.doi.org/10.1190/geo2013-0449.1.
Chen, Y., J. Yuan, S. Zu, S. Qu, and S. Gan, 2015, Seismic imaging of simultaneous-source data using
constrained least-squares reverse time migration: Journal of Applied Geophysics, 114, 3235,
http://dx.doi.org/10.1016/j.jappgeo.2015.01.004.
Donoho, D., 1995, De-noising by soft-thresholding: Information Theory: IEEE Transactions on
Information Theory, 41, 613627, http://dx.doi.org/10.1109/18.382009.
Gan, S., S. Wang, Y. Chen, S. Qu, and S. Zu, 2015, Velocity analysis of simultaneous-source data using
high-resolution semblancecoping with the strong noise: Geophysical Journal International, 204,
768779, http://dx.doi.org/10.1093/gji/ggv484.
Hampson, G., J. Stefani, and F. Herkenhoff, 2008, Acquisition using simultaneous sources: 78th Annual
International Meeting, SEG, Expanded Abstracts, 28162820.
Jiang, Z., and R. Abma, 2010, An analysis on the simultaneous imaging of simultaneous source data:
Annual International Meeting, SEG, Expanded Abstracts, 31153119.
Li, C., C. C. Mosher, L. C. Morley, Y. Ji, and J. D. Brewer, 2013, Joint source deblending and
reconstruction for seismic data: 83rd Annual International Meeting, SEG, Expanded Abstracts,
8287.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 117
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Zhang, Q., R. Abma, and I. Ahmed, 2013, A marine node simultaneous source acquisition trial at
Atlantis, Gulf of Mexico: Annual International Meeting, SEG, Expanded Abstracts, 2013, 5258.
Zu, S., H. Zhou, Y. Chen, S. Qu, X. Zou, H. Chen, and R. Liu, 2016, A periodically varying code for
improving deblending of simultaneous sources in marine acquisition: Geophysics, 81, 113,
http://dx.doi.org/10.1190/geo2015-0447.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 118
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
INTRODUCTION
Simultaneous source marine acquisition mitigates the challenges posed by conventional marine acquisition in terms of
sampling and survey efficiency, since more than one shot can
be fired at the same time (Beasley et al., 1998; de Kok and
Gillespie, 2002; Berkhout, 2008; Beasley, 2008; Hampson
et al., 2008). The final objective of source separation is to
get interference-free shot records. Wason and Herrmann (2013)
have shown that the challenge of separating simultaneous data
can be addressed through a combination of tailored single- (or
multiple-) source simultaneous acquisition design and curveletbased sparsity-promoting recovery. The idea is to design a
pragmatic time-jittered marine acquisition scheme where acquisition time is reduced and spatial sampling is improved by
separating overlapping shot records and interpolating jittered
coarse source locations to fine source sampling grid. While the
proposed sparsity-promoting approach recovers densely sampled conventional data reasonably well, it poses computational
challenges since curvelet-based sparsity-promoting methods
can become computationally intractablein terms of speed and
memory storageespecially for large-scale 5D seismic data
volumes.
Recently, nuclear-norm minimization based methods have
shown the potential to overcome the computational bottleneck
(Kumar et al., 2015a), hence, these methods are successfully
used for source separation (Maraschini et al., 2012; Cheng
and Sacchi, 2013; Kumar et al., 2015b). The general idea is
that conventional seismic data can be well approximated in
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 119
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
(a)
(b)
(c)
(d)
Figure 2: Schematic representation of the samplingtransformation operator A during the forward operation. The
adjoint of the operator A follows accordingly. (a, b, c) represent a monochromatic data slice from conventional data volume
and (d) represents a time slice from the continuous data volume.
Rank-minimization formulations require that the target data set
should exhibit a low-rank structure or fast decay of singular values. Consequently, the sampling-restriction (M R) operation
should increase the rank or slow the decay of singular values.
As we know, there is no unique notion of rank for tensors,
therefore, we can choose the rank of different matricizations
of D (Kreimer and Sacchi, 2012) where the idea is to create
the matrix D(i) by group the dimensions of D(i) specified by i
and vectorize them along the rows while vectorizing the other
dimensions along the columns. In this work, we consider the
matricization proposed by Silva and Herrmann (2013), where
i = (nsx , nsy )i.e., placing both source coordinates along the
columns (Figure 3a), or i = (nrx , nsx )i.e., placing receiver-x
and source-x coordinates along the columns (Figure 3b). As
we see in Figure 3e, the matricization i = (nsx , nsy ) has higher
rank or slow decay of the singular values (solid red curve) compared to the matricization i = (nrx , nsx ) (solid blue curve). The
sampling-restriction operator removes random columns in the
matricization i = (nsx , nsy ) (Figure 3c), as a result the overall
singular values decay faster (dotted red curve). This is because
missing columns put the singular values to zero, which is opposite to the requirement of rank-minimization algorithms. On the
other hand, the sampling-restriction operator removes random
blocks in the matricization i = (nrx , nsx ) (Figure 3d), hence,
slowing down the decay of the singular values (dotted blue
2016 SEG
SEG International Exposition and 86th Annual Meeting
(e)
Page 120
(1)
acquisition, the simultaneous data volume b is 4-times subsampled compared to conventional acquisition. Consequently, the
spatial sampling of recovered data is improved by a factor of 4
and the acquisition time is reduced by the same factor.
Pn
Pn
s j k1 and s j is the vector of singular
where j f k.k = j f ks
values for each monochromatic data matricization. One of the
main drawbacks of the nuclear-norm minimization problem is
that it involves computation of the singular-value decomposition (SVD) of the matrices, which is prohibitively expensive
for large-scale seismic data. Therefore, we avoid the direct
approach to nuclear-norm minimization problem and follow a
factorization-based approach (Rennie and Srebro, 2005; Lee
et al., 2010; Recht and R, 2011). The factorization-based
approach parametrizes each monochromatic data matrix D(i)
as a product of two low-rank factors L(i) 2 C(nrx nsx )k and
H
R(i) 2 C(nry nsy )k such that, D(i) = L(i) R(i) , where k represents the rank of the underlying matrix and H represents the Hermitian transpose. Note that tensors L, R can be formed by concatenating each matrix L(i) , R(i) , respectively. The optimization
scheme can then be carried out using the tensors L, R instead
of D, thereby significantly reducing the size of the decision
variable from nrx nry nsx nsy n f to 2k nrx nsx n f
when k nrx nsx . Following Rennie and Srebro (2005), the
sum of the nuclear norm obeys the relationship:
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
nf
X
j
(i)
kD j k
nf
X
j
(i)
subject to kA (D)
kD j k
nf
X
1
j
bk2 e,
kL j R j k2F ,
(i) (i)
2016 SEG
SEG International Exposition and 86th Annual Meeting
CONCLUSIONS
We propose a factorization based nuclear-norm minimization formulation for simultaneous source separation and
interpolation of 5D seismic data volume. Since the samplingtransformation operator is nonseparable in the simultaneous
time-jittered marine acquisition, we formulate the factorization based nuclear-norm minimization problem over the
entire temporal-frequency domain, contrary to solving each
monochromatic data matrix independently. We show that the
proposed methodology is able to separate and interpolate the
data to a fine underlying grid reasonably well. The proposed
approach is computationally memory efficient in comparison
to the curvelet-based sparsity-promoting approach.
ACKNOWLEDGEMENTS
We would like to thank the BG Group for permission to use the
synthetic Compass velocity model, Chevron for providing the
3D time-stepping finite-difference modelling code and Tristan
van Leeuwen for valuable discussions. The authors wish to
acknowledge the SENAI CIMATEC Supercomputing Center
for Industrial Innovation, with support from BG Brasil and
the Brazilian Authority for Oil, Gas and Biofuels (ANP), for
the provision and operation of computational facilities and
the commitment to invest in Research & Development. This
work was financially supported in part by the Natural Sciences
and Engineering Research Council of Canada Collaborative
Research and Development Grant DNOISE II (CDRP J 37514208). This research was carried out as part of the SINBAD II
project with the support of the member organizations of the
SINBAD Consortium.
Page 121
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
(a)
(b)
(c)
(d)
(e)
Figure 4: Source separation recovery. A shot gather from the (a) conventional data; (b) a section of 30 seconds from the continuous
time-domain simultaneous data (b); (c) recovered data by applying the adjoint of the sampling operator M ; (d) data recovered via
the proposed formulation (SNR = 20.8 dB); (e) difference of (a) and (d) where amplitudes are magnified by a factor of 8 to illustrate
a very small loss in coherent energy.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 122
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Beasley, C. J., 2008, A new look at marine simultaneous sources: The Leading Edge, 27, 914917,
http://dx.doi.org/10.1190/1.2954033.
Beasley, C. J., R. E. Chambers, and Z. Jiang, 1998, A new look at simultaneous sources: 68th Annual
International Meeting, SEG, Expanded Abstracts, 17, 133135.
Berkhout, A., 2008, Changing the mindset in seismic data acquisition: The Leading Edge, 27, 924938,
http://dx.doi.org/10.1190/1.2954035.
Cheng, J., and M. D. Sacchi, 2013, Separation of simultaneous source data via iterative rank reduction:
83rd Annual International Meeting, SEG, Expanded Abstracts, 8893.
de Kok, R., and D. Gillespie, 2002, A universal simultaneous shooting technique: 64th Annual
International Conference and Exhibition, EAGE, Extended Abstracts.
Hampson, G., J. Stefani, and F. Herkenhoff, 2008, Acquisition using simultaneous sources: The Leading
Edge, 27, 918923, http://dx.doi.org/10.1190/1.2954034.
Kreimer, N., and M. D. Sacchi, 2012, A tensor higher-order singular value decomposition for prestack
seismic data noise reductionand interpolation: Geophysics, 77, no. 3, V113V122,
http://dx.doi.org/10.1190/geo2011-0399.1.
Kumar, R., C. Da Silva, O. Akalin, A. Y. Aravkin, H. Mansour, B. Recht, and F. J. Herrmann, 2015a,
Efficient matrix completion for seismic data reconstruction: Geophysics, 80, no. 5, V97V114,
http://dx.doi.org/10.1190/geo2014-0369.1.
Kumar, R., H. Wason, and F. J. Herrmann, 2015b, Source separation for simultaneous towed-streamer
marine acquisition A compressed sensing approach: Geophysics, 80, no. 6, WD73WD88,
http://dx.doi.org/10.1190/geo2015-0108.1.
Lee, J., B. Recht, R. Salakhutdinov, N. Srebro, and J. Tropp, 2010, Practical large-scale optimization for
max-norm regularization: Presented at the Advances in Neural Information Processing Systems,
2010.
Maraschini, M., R. Dyer, K. Stevens, and D. Bird, 2012, Source separation by iterative rank reductiontheory and applications: Presented at the 74th EAGE Conference and Exhibition,
http://dx.doi.org/10.3997/2214-4609.20148370.
Recht, B., M. Fazel, and P. Parrilo, 2010, Guaranteed minimum rank solutions to linear matrix equations
via nuclear norm minimization: SIAM Review, 52, 471501,
http://dx.doi.org/10.1137/070697835.
Recht, B., and C. R, 2011, Parallel stochastic gradient algorithms for large-scale matrix completion:
Technical report, University of Wisconsin-Madison.
Rennie, J. D. M., and N. Srebro, 2005, Fast maximum margin matrix factorization for collaborative
prediction: Proceedings of the 22nd international conference on Machine learning, ACM, 713
719, http://dx.doi.org/10.1145/1102351.1102441.
Da Silva, C., and F. J. Herrmann, 2013, Hierarchical Tucker tensor optimization - applications to 4D
seismic data interpolation: Presented at the EAGE Annual Conference Proceedings.
10.3997/2214-4609.20130390.
Wason, H., and F. J. Herrmann, 2013, Time-jittered ocean bottom seismic acquisition: 75th Annual
International Meeting, SEG, Expanded Abstracts, 32, 16, http://dx.doi.org/10.1190/segam20131391.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 123
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Results from a large field test using 2D ring arrays to address back-scattered surface noise in
land seismic acquisition
Christof Stork*, Carolyn Dingus, Nick Bernitsas, Dave Flentge, ION Geophysical
Summary
Back-scattered surface noise is often a large problem with
surface seismic data. This noise is badly aliased with
conventional 3D orthogonal geometry, which makes it very
difficult to remove in processing.
We performed a large field test of a new acquisition
geometry using 2D ring arrays that properly sampled the
back-scattered noise so it can be removed in processing.
This test with 4300 shots and 3300 receivers is large
enough to produce a 3D image. Since this test was
performed concurrently with a conventional 3D acquisition
survey using the same sources and similar number of
receivers, we can directly compare the processed image
quality from the 3D ring array results with the conventional
results. Results show that the new acquisition geometry
using 2D ring arrays is very effective at addressing the
back-scattered noise at similar acquisition effort.
Introduction
Because the near surface is a very strong wave guide that
traps much energy and is very heterogeneous, it often
creates strong back-scattered surface noise. This surface
noise can be 2x-10x greater than the reflection signal.
This back scattered noise is often not uniform in offset or
azimuth. As a result, this noise will effect seismic
interpretation attributes based on offset or azimuth. It is
often difficult to identify when noise artifacts influence
attribute patterns.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 124
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
55,000 feet
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 125
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
The near random nature of the ring array geometry did not
add significant cost since GPS was used for survey and
layout.
Acknowledgements
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 126
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 4a: Stack of raw ring array data with no filtering. Signal quality is poor.
Figure 4b: Stack of ring array data after 3D Radon filtering. Signal quality is much improved.
Figure 4c: Stack of conventional acquisition data with no filtering. Signal quality is similar to the raw ring array stack.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 127
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Auger, E., E. Schissel-Rebel, and J. Jia, 2013, Suppressing noise while preserving signal for surface
microseismic monitoring: The case for the patch design: 83rd Annual International Meeting,
SEG, Expanded Abstracts, 20242028, http://dx.doi.org/10.1190/segam2013-1396.1.
Barr, F. J., 1993, Seismic data acquisition: Recent advances and the road ahead: 63rd Annual
International Meeting, SEG, Expanded Abstracts, 12011204,
http://dx.doi.org/10.1190/1.1822334
Biondi, B. L., 2006, 3D seismic imaging: SEG.
Petrochilos, N., and J. Drew, 2014, Noise reduction on microseismic data acquired using a patch
monitoring configuration: A fayetteville formation example: 84th Annual International Meeting,
SEG, Expanded Abstracts, 23142318, http://dx.doi.org/10.1190/segam2014-1269.1.
Regone, C. J., 1997, Measurement and identification of 3-D coherent noise generated from irregular
surface carbonates [Abstract], in I. Palaz and K. J. Marfurt, eds., Carbonate seismology: SEG,
281306.
Regone, C. J., 1998, Suppression of coherent noise in 3-D seismology [Abstract]: The Leading Edge, 17,
15841589, http://dx.doi.org/10.1190/1.1437900.
Regone, C. J., and G. L. Rethford, 1990, Identifying, quantifying, and suppressing backscattered seismic
noise: 60th Annual International Meeting, SEG, Expanded Abstracts, 748751,
http://dx.doi.org/10.1190/1.1890320.
Schissel-Rebel, E., and J. Meunier, 2013, Patch versus broadband networks for microseismic: A signalto-noise ratio analysis: 83rd Annual International Meeting, SEG, Expanded Abstracts, 2104
2108, http://dx.doi.org/10.1190/segam2013-1285.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 128
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Title: High Density 3D Nodal Seismic Acquisition in Sub Andean Mountains - An Operational
Challenge
Author: J. Uribe* (Repsol), L. Rodriguez (Repsol), P. A. Munoz (Repsol), R. Parrado (Repsol), N. Sanabria
(Repsol).
Summary
Geological Objectives
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 129
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 130
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Cruz, C., J. Rodriguez, J. Hechem, and H. Villar, 2008, Sistemas petroleros de cuencas andinas:
Presented at the 7th Congreso de Exploracin y Desarrollo de Hidrocarburos, IAPG,
Abstracts, 159187.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 131
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 132
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
(n , ,x , )
sin(nx cos / )
n sin(x cos / )
(1)
2016 SEG
SEG International Exposition and 86th Annual Meeting
Figure 4: Map view of the equivalent cavity generated by a threehole array shooting.
Page 133
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusion
Acknowledgement
The authors wish to thank Crew 2137 and Crew 249 for
their hard works spent in the test acquisition, and thanks
BGP, CNPC for permitting to publish the related results.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 134
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
The first row is raw test records. The second row is the band-pass filtered (BP30-60Hz) records corresponding with the first row. The receiving
spread direction of the first column is 0, the middle column is 45, and third column is 90. The receiver array orientation is 0.
Figure 5: Comparison between seismic records obtained from different source-array orientations.
The records in the upper row are raw data. The source-array orientation of (a) is parallel to the receiving spread, and (b) is perpendicular to the
receiving spread. The records (c) and (d) are the corresponding band-pass filtered BP (30-60Hz) records of (a) and (b).
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 135
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Li, Q. Z., and Z. J. Chen, 1984, The suppression and improvement of the array effect on the reflection
wave: Oil Geophysical Prospecting, 19, 120.
Li, Q. Z., and J. D. Wei, 2008, Talk about importance of cross-line array of geophone on spread: Oil
Geophysical Prospecting, 43, 375382.
Ning, H. X., Y. H. Zhang, D. R. Zhang, Y. X. Jiang, and W. L. Jiao, 2011, Seismic source optimization in
Qilian mountain area: Oil Geophysical Prospecting, 46, 370373.
Qian, R. J., 2003, Analysis of explosive effect of dynamite source: Oil Geophysical Prospecting, 38, 583
588.
Robert, E., 2002, Sheriff, encyclopedic dictionary of applied geophysics, 4th ed.: SEG.
Tang, D. L., X. W. Cai, Y. Q. He, and H. X. Ning, 2014, Source and receiver arrays for prestack
migration: Oil Geophysical Prospecting, 49, 10341038.
Wu, Y. G., W. H. Yin, Y. Q. He, G. S. Li, H. X. Ning, D. R. Zhang, and Q. S. Pu, 2012, A uniformity
quantitative method for 3D geometry attributes: Oil Geophysical Prospecting, 47, 361365.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 136
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
In a context of low oil prices, the whole geophysical
industry has entered a potentially long-term, shady period
during which every cent will count to ensure
competitiveness, and even project feasibility. Marine
seismic, penalized by strong operating expenses, has been
first to suffer from the current situation, but land seismic
projects also show a downturn and are subject to important
cost cuts. Thus, it is more than ever paramount to identify
all the project expenses and reduce the less necessary or
less effective costs. This concerns vibrators that enable
acquiring low-priced VPs wherever possible, but also
require maintenance, consumables and staff, with the
associated costs. This abstract proposes several existing and
prospective solutions to ensure the vibrator contribution to
the crew operating expenses reduction.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 137
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 138
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 139
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
Cutting costs is a major concern for land seismic crews in
difficult market conditions. The vibrator item has been
addressed for a long time by contractors (e.g., through
vibrator fleet standardization and logistic and shooting
rationalization) and manufacturers who developed
numerous solutions to ease operation and maintenance and
then help with operating expenses reduction. Other
solutions are being developed or are still at the state of a
concept. Depending on their boldness and the industry
support for developing and testing them, the technical
landscape we have been used to for vibroseis is likely to
evolve significantly in the coming years and allow even
cheaper operations and cost per VP unless too greatly
reduced exploration budgets dissuade manufacturers from
performing the necessary and cash-consuming
research and development.
Acknowledgments
The author would like to thank Pascal Buttin (Sercel) for
constructive exchanges on prospective vibroseis seismic
acquisition and the management of Sercel for permission to
publish this work.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 140
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Bagaini, C., 2007, Enhancing the low-frequency content of vibroseis data with maximum displacement
sweep: 69th conference and exhibition, EAGE, Extended Abstract,
http://dx.doi.org/10.1190/1.2370368.
Blacquire, G., and G. Berkhout, 2013, Robotization in Seismic Acquisition: 83rd Annual International
Meeting, SEG, Expanded Abstract, http://dx.doi.org/10.1190/segam2013-0838.1.
Mass, S., G. Baeten, K. Hornman, and G. Irwin, 2013, Seismic cable handling system and method: Patent
WO 2013/134196 A2.
Sacilotto, J., M. Rombaud, and P. Barbe, 2000, Fully automatic mobile storage, handling, deployment and
retrieval equipment for belts of seismic sensors, buries them individually over large area at
controlled speed, avoiding onerous, time-consuming manual methods: French Patent 2785684A1.
Sallas, J. J., 2010, How do hydraulic vibrators work? A look inside the black box: Geophysical
Prospecting, 58, 318, http://dx.doi.org/10.1111/j.1365-2478.2009.00837.x.
Tellier, N., G. Caradec, and G. Ollivrin, 2014, Impact of the use of low-frequency vibrators on crew
productivity: 84th Annual International Meeting, SEG, Expanded Abstract,
http://dx.doi.org/10.1190/segam2014-0343.1.
Wei, Z., M. Hall, and T. Phillips, 2012, Geophysical benefits from an improved seismic vibrator:
Geophysical Prospecting, 60, 466479, http://dx.doi.org/10.1111/j.1365-2478.2011.01008.x.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 141
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Effects of wind noise on high-density seismic data and its field operation methods
Wu Yongguo*, He Yongqing, Liu Fengzhi,Yin Wuhai, BGP, CNPC, Hu Jie, Qinghai Oilfield Company, CNPC
Summary
The high-density data acquisition technology improves the
quality of data, but it is also facing many challenges,
especially in many regions of western China, where it
belongs to the arid plateau climate being characterized as
cold, hypoxia, drought and windy. Such bad natural
conditions, windy weather and short available production
duration affect the acquisition efficiency heavily. Aiming at
this problem, a mathematical model for wind noises is
established firstly and hereby the wind noises are simulated;
then the wind noises with different wind scales are added
into the wind-noise free shot record to do imaging impact
analysis; finally, the actual field wind noise records are
recorded and added into the wind-noise free shot records to
do imaging impact analysis and thereby get the required
high-density field operation method. Actual test data shows
that this method has a good result in solving the acquistion
efficiency problem of high-density data at areas with strong
windy meather.
Introduction
The purpose of high-density 3D survey is to increase shot
and receiver density by significantly increasing the source
and receiver points so as to significantly improve the data
quality. However, it also faces many challenges, especially
in many regions in western China, with arid plateau climate
being characterized as cold, hypoxia, drought and windy.
Such bad natural conditions windy weather (the average
wind duration is 10 hours per day) shortens the available
operation time per day and therefore affects the acquisition
efficiency heavily. For example, we standed by
traditionally when the wind scale was above moderate
breeze (4 level wind) or when the environmental noise was
greater than 7 micro-volts (mv) and accordingly a highdensity 2D JB project in western China had shut down for
18 days due to gale in the three-month acquisition period.
This brought huge investment to the field crew. The
traditional wind noise field operation method seriously
affected the field acquisition efficiency, extended the
production cycle and increased the field exploration costs.
Analyzing the influences of wind noise at different wind
scales on high-density acquisition seismic data and reestablishing the wind-noise-limited high-efficiency
acquisition method according to the wind noises will
reduce the impact on the field operation.
Wind noise forward modeling and impact analysis
Forward modeling of wind noises
2016 SEG
SEG International Exposition and 86th Annual Meeting
Assuming the earth is a homogeneous, isotropic, semiinfinite elastic medium, due to the viscosity of air, wind
noises received by geophones are mainly the surface
deformation caused by the force of wind to ground surface
(Sorrells, 1971). Supposing the wind blows from left to
right, the wind force to the geophone surrounding area is
distributed in the rectangular area on the windward side in
the form of point source. Considering the working range of
wind, the record for one trace is mainly affected by the
layout range of the geophone. The expression of drag force
of wind to the ground surface is:
1
(1)
F C v 2 S
d
v ( z) z
(3)
v10
10
(1 2 ) 4 / 3
v10
Based on the Shinozaki theory, the fluctuating wind
velocity time-domain signal can be derived from the
fluctuating wind velocity spectrum (Shinozaka, 1972), as
shown in formula (6):
N
v f (t) 2S (i ) cos(it i )
(6)
i1
Page 142
0.5s
1.0s
0.02
0.01
Amplitude
-0.01
0
Time(s) 2
0. 04
0
-0. 04
-0. 08
0
6
5
4
3
0.5s
Amplitude
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Effects of wind noise on high-density seismic data and field operation methods
2016 SEG
SEG International Exposition and 86th Annual Meeting
Time(s) 2
60
80
2
1
0
1.0s
20
40
Frequency (Hz)
Page 143
1.0s
2.0s
2.0s
Figure 5: Actual single shot records with different wind scales (a: 2
level wind(light breeze), b: 3 level wind(gentle breeze), c: 4 level
wind(moderate breeze), d: 5 level wind(fresh breeze), e: 6 level
wind(strong breeze), f: 7 level wind(moderate gale)) .
40
0.5s
35
2
3
4
5
6
7
30
25
20
15
10
Wind scale
Average energy(mv)
1.0s
5
0
-2 -1 1 2
Offset(km)
-3
-4
-5
-7 -6
0.5s
100
90
70
60
50
40
30
Wind scale
2
3
4
5
6
7
80
20
10
0
-2 -1 1 2
Offset (km)
-3
-4
-5
-7 -6
Average_energyr_ratio (%)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Effects of wind noise on high-density seismic data and field operation methods
Figure 7: The average energy ratio between the pre- and post- first
arrivals values of shot records with different wind scale and wind
noise.
of the shot records with wind noise of 5-7 level wind (fresh
breeze to moderate gale) is affected more obviously. The
envrionmental noises are monitored by the noise limit on
the instrument site to see whether it is suitble to continue
the acquisition. According to the previous acquisition
standards, it is not allowed to do acquisition data when the
background recording noises are greater than 7 micro-volts
(mv) or when the number of bad traces is more than 1/24 of
the total number of traces. Figure 6 is the analysis result of
the before first arrival values picked from the the different
shot records in figure 5. It indicates that the wind noises of
light-gentle breeze (2-3 level wind) is less than 7 mv, that
of moderate fresh breeze (4-5 level wind) is 10-15 mv,
and that of strong breeze to moderate gale (6-7 level wind)
is 20-40 mv. Based on the previous acquisition standard, it
2016 SEG
SEG International Exposition and 86th Annual Meeting
1
N Expect _ sec tion (2S / N )
Shot _ record
(S / N )
(7)
where N means the number of folds, Expect_section(S/N)
means the expected S/N ratio of the section and the
Shot_record(S/N) means the shot records S/N ratio; the
expected section S/N ratio in the formula is a constant; and
the fold for the high-density 3D survey is also given in the
Page 144
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Effects of wind noise on high-density seismic data and field operation methods
early operation stage of the project, so the S/N ratio of shot
record can be calculated. As can be seen obviously from
figure 8 that the S/N ratio of the stacked section for 6-7
level wind is stronger; the event is in poorer continuity and
distorted waveform at shallow-middle layer; with the
increase of fold, the S/N ratio in the section is improved
significantly, the continuity of event at shallow-middle
layer is even better; and when the fold is up to above 1,440,
the improvement of the S/N ratio in the section is no longer
obvious. This indicates that the increase of fold suppresses
the wind noises effectively and improves the S/N ratio of
the section, but when it is increased to a certain value, the
fold will no longer work much in improving the S/N ratio.
Applications
Through the above analysis, with the increase of wind scale,
the impact of wind noises on shot records and imaging is
obviously increased; different geometries also have
different wind noise suppression abilities, therefore, the
high-density acquisition methods are different for different
seismic and geologic conditions in different regions.
Different high-density field operation methods are designed
for different areas based on their own seismic-geologic and
climatic conditions and the high-density acquisition
parameters. When the NZ 3D high-density acquisition
project is implemented in western China, the wind is often
4-7 level wind in the work area, which has seriously
affected the acquisition efficiency of the project and the
1s
2s
3s
a
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 145
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Davenport, A. G., 1961, The spectrum of horizontal gustiness near ground in high winds: Royal
Meteorological Society, 87, 194211, http://dx.doi.org/10.1002/qj.49708737208.
Ministry of Housing and Urban-Rural Development of the Peoples Republic of China, 2012, Load code
for the design of building structures (MOHURD) GB 50009-2012: China Building Industry Press.
Shinozaka, M., and C. M. Jan, 1972, Digital simulation of random processes and its applications: Sound
and Vibration, 25, 111128, http://dx.doi.org/10.1016/0022-460X(72)90600-1.
Sorrells, G. G., J. A. McDonald, Z. A. Der, and E. Herrin, 1971, Earth motion caused by local
atmospheric pressure changes: Geophysical Journal of the Royal Astronomical Society, 26, 83
98, http://dx.doi.org/10.1111/j.1365-246X.1971.tb03384.x.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 146
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Field setup
Introduction
The moving coil geophone is the most widely used sensor
for land seismic acquisition. It is sensitive to velocity and
its response is essentially flat above the natural resonant
frequency. Below resonance it behaves like a low-cut filter.
Furthermore, the particle velocity sensed by the geophone,
being the derivative of particle displacement, decays
toward the low frequencies like a low cut filter, at the rate
of -6db/octave. Both filters are detrimental to broadband
applications where low frequencies not only are necessary
to increase the resolution of the signal, but also to help
quantifying the inversion of reflection data into acoustic
impedance (Poggiagliolmi et al., 2015). The recovery of
low frequencies below the geophone natural resonant
frequency by inverse filtering has been reported (Bertram
and Margrave, 2010, 2011, Zhang et al., 2012, Margrave et
al., 2012) but its success is limited by the ever presence of
noise in the data.
Unlike particle velocity, particle displacement
exhibits a broader band response especially at the low
frequencies. Therefore, it would be desirable to acquire
seismic data with displacement sensors rather than
geophones or indeed accelerometers. To the best of the
authors knowledge no progress has been made toward the
development of displacement sensors until recently
(Poggiagliolmi et al., 2012, 2013).
The purpose of the field experiment described herein
is to compare the output of a displacement sensor with
those of two commercially available geophones. From the
comparisons it is clear that the broadband response of the
displacement sensor far exceeds that of the geophones.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Field tests
Figure 2 shows an example of a recording made with the
above setup. Shown in this figure are the time and
frequency domain responses of the first arrival pulse
simultaneously picked-up by the three sensors. The UAS
senses particle displacement and its response is broadband
down to the very low frequencies (Figure 2, red trace). The
Page 147
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 148
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 3: First arrivals for a 5Hz geophone: simulated with the UAS (1, red line) and field measurement (2, blue line).
Figure 4: First arrivals for a 1Hz geophone: simulated with the UAS (1, red line) and field measurement (2, blue line).
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 149
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
Since geophones sense particle velocity, they lack
sensitivity at the low frequencies relative to their
performance at the high frequencies. Instead, particle
displacement sensors have a broad band response
especially at the very low frequencies.
Comparisons using field data acquired with a
displacement sensor, a 5 Hz geophone and 1Hz geophone
demonstrated that the response of the displacement sensor
is essentially flat at low frequencies down to 0.1Hz, and
lower, while the geophones attenuate these low
frequencies.
The validity of the derivative relationship between
the displacement sensor and geophones has been
demonstrated by direct field measurements. This is an
important result, because it confirms that the UAS sensor
output is directly related to particle displacement.
Field-recorded first arrivals transformed to zerophase wavelets were used to demonstrate that the wavelets
side-lobes magnitude is smaller for the displacement sensor
compared to that produced by conventional geophones.
Acknowledgements
The authors would like thank Entec Integrated
Technologies Ltd., for permission to publish this work,
which was partially supported with the grant n. 14502 by
the Petroleum Institute Research Centre (Abu Dhabi,
United Arab Emirates) and by OGS Italian National
Institute of Oceanography and Applied Geophysics (Italy).
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 150
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Bertram, M. B., and G. F. Margrave, 2010, Recovery of low frequency data from 10Hz geophones:
CREWES Research Report 22.
Bertram, M. B., and G. F. Margrave, 2011, Recovery of low frequency data from 10 Hz geophones:
Recovery CSPG CSEG CWLS Convention, Calgary, Expanded Abstracts.
Margrave, G. F., M. B. Bertram, K. L. Bertram, K. W. Hall, K. A. H. Innanen, D. C. Lawton, L.E.
Mewhort, and T.M. Phillips, 2012, A field experiment recording low seismic frequencies: AAPG
Datapages/Search and Discovery, 90174, GeoConvention, Calgary, Canada.
Poggiagliolmi, E., F. Accaino, and A. Vesnaver, 2015, Broadband data acquisition A case for
displacement sensors: 77th Annual International Conference and Exhibition, EAGE, Extended
Abstracts, http://dx.doi.org/10.3997/2214-4609.201414421.
Poggiagliolmi, E., A. Vesnaver, D. Nieto, and L. Baradello, 2012, The application of uncoupled sensors
to seismic exploration: Presented at the SEG/KOC workshop on Single Sensor Acquisition,
Kuwait.
Poggiagliolmi, E., A. Vesnaver, D. Nieto, and L. Baradello, 2013, An uncoupled acoustic sensor for land
seismic acquisition: 75th Annual International Conference and Exhibition, EAGE, Extended
Abstracts, http://dx.doi.org/10.3997/2214-4609.20130745.
Zhang, Y., Z. Zhihui Zou, and H. Zhou, 2012, Estimating and recovering the low-frequency signals in
geophone data: 82nd Annual International Meeting, SEG, Expanded Abstracts, 15,
http://dx.doi.org/10.1190/segam2012-1178.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 151
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 152
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
(a)
(b)
(a)
(b)
Figure 1 The amplitude spectra of adaptive (red) and
linear (shadow) vibroseis from excitation signal and
received seismic data.
High frequency processing technologies we used in this
abstract mainly include well-constrained deconvolution and
structure-constrained Q compensation technology. In the
conventional methods, the deconvolution parameters is
determined by parameter sweeping, which compares the
effect of the parameters on single shot section with stack
section. In this abstract, we use well-constrained
deconvolution technology, which gets preferable
deconvolution parameters by analyzing the crosscorrelation of the well-nearby seismic traces and synthetic
record or VSP stack and generating a series of matching
attributes (matching reliability, predictability, deliver
function, etc). The result of well-constrained deconvolution
technology can reflect the features of reflection wave in the
layer around the well objectively. The aim of Q
compensation is to compensate amplitude attenuation and
phase distortion caused by subsurface attenuation effects,
correct the stretch effect of the wavelet phase, compensate
the attenuation of wave frequencies and amplitude,
2016 SEG
SEG International Exposition and 86th Annual Meeting
(c)
(d)
Figure 2 (a) Shot gather from linear sweeping; (b) shot
gather from adaptive sweeping; (c) the spectra of time
window 0-1500ms; (d) the spectra of time window 15004000ms.
Structure-constrained Q compensation technology are used
for our high resolution processing, which uses spectrumratio method to compute the layer Q model with the help of
sliding time window along the layer on the basis of the
structure of the target layer. Structure-constrained 3D Q
model can be obtained by interpolation and smoothing
process. After achieving 3D Q model, we can implement Q
compensation. As the attenuated energy of wavelet being
compensated, the energy at high frequencies can be
strengthened effectively and the resolution of seismic data
can be improved.
Application example in China
We use the proposed method to seismic exploration in
TKX area of China, which focuses on lithologic trap. The
burial depth is between 1800-2500 m, and the thickness of
low velocity zone is between 12-25m. The 3D acquisition
and observing system is 36L4S120R. The size of bin is
2525m, and the fold number is 360. We use lowfrequency vibroseis KZ28LF, and the sweeping frequency
range is 1.5-84Hz. We use conventional acquisition and
Page 153
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
(a)
(b)
Figure3 The amplitude spectrum and phase spectrum
before (red) and after (blue) low-frequency-preserving
processing. (a) Amplitude spectrum; (b) Phase spectrum.
We use well-constrained deconvolution technology, which
can get preferable deconvolution parameters by analyzing
the cross-correlation of synthetic record at well and the
nearby seismic traces and generating a series of matching
attributes (matching reliability, predictability, deliver
function). After quantifying, comparing and analyzing the
outcomes, the deconvolution parameter can be determined.
By applying well-constrained deconvolution, the bandwidth
of target layer is extended by 8-10Hz. We do the
interpretation at whole area with the help of geological
stratification divided by drilling. The deep layers are
divided into 5 parts and choose time window along the
2016 SEG
SEG International Exposition and 86th Annual Meeting
(a)
(b)
(c)
(d)
Figure 4 Seismic section and spectrum of conventional
exploration (a, c) and broadband exploration (b, d).
From the final seismic section of two methods, we compare
the main target special lithology covered on T4 and tail
sandstone at the lower part of T4 (Figures 4a and 4b). From
the final section of broadband exploration, the special
lithology layer can be recognized clearly and the lateral
variations of tail sandstone can also be recognized.
However, from the final section of regular exploration, it is
difficult to recognize the special lithology part and the tail
sandstone. By analyzing the spectra of target layer T4, we
can indicate that the energy of broadband exploration at
middle and high frequency band is strengthened and the
effective bandwidth is extended by 8-16Hz. The effective
bandwidth of target layer in the final section of broadband
exploration is 1.5-60Hz.
From P-wave impedance inversion sections (Figure 5), we
can note that the sand body is multi-periodic with rich
information and high resolution. However, we can only
recognize sands group from the inversion section of
conventional exploration. According to well drilling data,
there are large groups of sandstones at the special lithology
part and tail sandstone part of T4. The percentage of sandmud stone predicted by RMS amplitude attribute of
broadband exploration data matches well with well drilling
data, and while the prediction error of conventional
exploration data is relatively big. We use newly acquired
3D data and indicated a number of lithologic traps after
careful interpretation and reservoir prediction. We
determine well GB1 for drilling according to broadband
Page 154
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
(a)
Figure 5 P-wave impedance inversion of conventional exploration (a) and broadband exploration (b).
(b)
Figure 6 Well GB1 (left) and seismic section (right) of broadband exploration.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 155
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Cao, W., Li, X., and Guo, H., 2009, Shaping design of vibroseis sweeping signal: Geophysical
Prospecting for Petroleum, 48, 611614.
Lan, J., 2008. Application of vibroseis nonlinear sweeping in high resolution seismic data
acquisition: Geophysical Prospecting for Petroleum, 47, 208211.
Tao, Z., Su, Z., Zhao, Y., and Ma L., 2010. The latest development of low frequency vibrator for
seismic: Equipment for geophysical prospecting, 20, 15.
Zhukov, A., 2013. The adaptive vibroseis technology: hardware, software and outcomes: 83rd Annual
International Meeting, SEG, Expanded Abstracts, 249253,
http://dx.doi.org/10.1190/segam2013-0542.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 156
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Initial 3C-2D surface seismic and walkaway VSP results from the 2015 Brooks SuperCable
experiment
Kevin W. Hall*1, J. Helen Isaac1, Joe Wong1, Kevin L. Bertram1, Malcolm B. Bertram1, Donald C. Lawton1,3,
Xuewei Bao2, and David W. Eaton2 (1CREWES and 2Microseismic Industry Consortium, University of Calgary, and
3
Containment and Monitoring Institute)
Summary
A 3C walkaway VSP and surface seismic experiment was
conducted at the Containment and Monitoring Institute
(CaMI) Field Research Station (FRS) in May of 2015. The
FRS is located near the town of Brooks in southern Alberta,
Canada. Multiple objectives for the program included
student training, surface source and receiver comparisons,
multi-component walkaway VSP acquisition, and velocity
tomography for site characterization.
Two parallel NE-SW receiver lines were laid out with one
line centered on a well and the other offset 100 m to the
northwest. Both receiver lines had single-component
geophones at a 10 m receiver spacing. In addition, the line
centered on the well had three-component geophones at a
30 m receiver spacing. A tool with three-component
geophones was deployed within the well at three different
levels, giving receiver positions from 106 to 496 m depth at
a 15 m spacing.
Two source lines centered on the well location, one linear
with a vibe point (VP) every 10 m and the other semicircular with a VP every 5 degrees were acquired three
times, once for each tool position in the well. The source
was an IVI EnviroVibe using a variety of filtered and
unfiltered maximal length sequence pilots (m-sequences) as
well as a linear 10-200 Hz sweep. This abstract presents a
first look at the data and some early results.
Well
Well
Introduction
A 3C walkaway VSP and surface seismic experiment was
conducted at the Containment and Monitoring Institute
(CaMI) Field Research Station (FRS) in May of 2015. Two
parallel NE-SW receiver lines were laid out with one line
(Line 108) centered on well CMCRI COUNTESS 10-2217-16, and the other (Line 106) offset 100 m to the
northwest (Figure 1). Receiver lines 106 and 108 had
single-component SM-24 geophones at a 10 m receiver
spacing connected to an Inova (ARAM) Aries SPML
recorder. In addition, receiver line 108 had threecomponent SM-7 geophones in nail-type casings at a 30 m
receiver spacing recorded by Inova Hawk nodal systems. A
three-component ESG SuperCable was deployed in the
well at three different levels, giving receiver positions in
the well from 106 to 496 meters depth at a 15 m spacing.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Two source lines were acquired three times, once for each
tool position in the well (Figure 2). The source was an IVI
EnviroVibe sweeping from 10-200 Hz linearly over 16 s
with an additional 4 s listening time. Source line 208 (NESW) had a Vibe Point (VP) every 10 m for surface 2D
seismic and walkaway VSP. A semi-circular source line
(Line 204) with a radius of 400 m and a VP every five
degrees was acquired for a velocity tomography study.
Finally, source line 208 was re-acquired using a variety of
filtered and unfiltered maximal length sequence pilots
Page 157
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 158
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
transitions between two values -1 and +1. The puresequences were modified by an Ormsby bandpass filter
with corners at [5-10-200-250] Hz. The m-sequence pilots
were all 16.376 seconds long. For comparison purposes, we
also recorded data using a standard linear sweep pilot (10 to
120 Hz swept over 16 seconds with 500 ms end tapers). In
all cases, listen time was 4 seconds, and in all cases we
recorded both correlated and uncorrelated data as well as
the signals from accelerometers mounted on the base plate
and reaction mass of the vibrator. It is hoped that these
accelerometer signals will provide clues as to how the
hydraulically-powered vibrator reacts to the sharp/smoother
transitions that are characteristic of pure and filtered msequences.
Pure m-sequences are used to estimate the impulse
response of linear systems (among many other uses). The
fact that the EnviroVibe generates multiples (which are
considered to be artifacts; Figure 7) in seismograms when
driven by pure m-sequences means that the EnviroVibe is
not a perfectly linear system, especially at high frequencies
since it cannot respond accurately to the step-function-like
transitions characteristic of pure m-sequences. Bandpass
filtering the m-sequence before using it as a sweep
introduces side-lobes in the wavelet, although smaller ones
than seen for a linear sweep. It also reduces the prominence
of the source-generated multiples in the recorded data
(Figure 8). Figure 9 shows the same VP acquired with a
linear sweep for comparison. It is difficult to see at this
scale, but the amplitude spectra for the m-sequence gather
contains more energy above 250 Hz than the amplitude
spectra for the filtered m-sequence gather.
Examination of these preliminary results indicates that the
pure m-sequences and the particular filtering applied are
not well suited for the EnviroVibe and its Pelton controller.
It appears that the pure and currently filtered TREF pilots
probably should not contain any energy above 125 Hz. In
addition, we have not yet succeeded in ascertaining how the
many settings available in the Pelton controller should be
set in order for the hydraulics and position controls to best
allow the ground force signals to closely follow the msequence TREF signal (for example: Should the phase lock
be disabled? Can we prevent the controller from learning,
which causes the ground force to grow with repeated
sweeps?).
Surface seismic processing included refraction statics, air
blast attenuation, spike and noise burst edit, surface wave
noise attenuation, and Gabor deconvolution. In order to
compare Aries to Hawk data, we post-stack migrated
receiver stacks using a finite difference migration and
applied a bandpass filter of 10-15-80-90 Hz. The migrated
data are shown in Figure 10, which also shows for
comparison an arbitrary line extracted from a 2014 3D
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 159
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Acknowledgements
The authors thank the sponsors of CREWES and the
Microseismic Industry Consortium, and CMC Research
Institutes Inc. for access to the CaMI field research sites.
This work was funded by CREWES and Microseismic
Industry Consortium industrial sponsors, CMC, and
NSERC (Natural Science and Engineering Research
Council of Canada) through Collaborative Research and
Development grants. We would also like to thank ESG for
field support, as well as Halliburton (Landmark Graphics)
and Schlumberger for providing donated software.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 160
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
No references.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 161
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
0.8
0.8
0.6
0.6
0.4
0.4
0.2
0.2
0
10 -1
10 0
10 1
10 2
10 3
0
10 -1
10 0
10 1
frequency (Hz)
frequency (Hz)
wavelet
wavelet
10 2
10 3
0
10 -1
0.5
0.5
-0.5
-0.5
-0.5
-1
1.6
1.7
1.8
16.3
16.4
16.5
time (s)
integrated wavelet
16.2
0.5
0.5
0.5
-0.5
-0.5
-0.5
-1
1.6
time (s)
1.8
16.3
16.4
16.5
time (s)
1.4
integrated wavelet
1.2
10 3
Optimum depth
-1
16.2
time (s)
integrated wavelet
-1
10 2
wavelet
1.5
10 1
frequency (Hz)
0.5
-1
10 0
-1
16
16.2
16.4
time (s)
16.6
16.8
16
16.2
16.4
16.6
16.8
time (s)
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 162
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 163
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
for G
ceiling ,
1
for
> ceiling ,
(5b)
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 164
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Concluding remarks
To enhance the low frequencies in the seismic method, the
so-called ghost should be exploited. It means that sources
and detectors should be positioned at the optimum depth,
which is frequency dependent (0.25). In the DSA concept
dedicated narrow-band sources are deployed. This makes
the DSA concept very suited to enhance the low
frequencies at the source side: each source type can be
towed at its optimum depth.
Remaining imprints of the ghost should be mitigated by
deghosting. Deghosting can be carried out perfectly in
areas away from the notches. Inside the notch areas,
nonlinear processing should further improve the result.
Dedicated processing of the low frequencies such as
interpolation, regularization and lateral stacking further
contributes to an increased bandwidth of seismic data.
The trend information related to the very low frequencies is
retrieved from velocity information as well as gravity
gradiometry.
Acknowledgments
We acknowledge the members of the Delphi consortium for
the stimulating discussions and their financial support.
Page 165
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
ten Kroode, F., S. Bergler, C. Corsten, J. W. de Maag, F. Strijbos, and H. Tijhof, 2013, Broadband
seismic data The importance of low frequencies: Geophysics, 78, no. 2, WA3WA14,
http://dx.doi.org/10.1190/geo2012-0294.1.
Berkhout, A. J., 2012, Blended acquisition with dispersed source arrays: Geophysics, 77, no. 4, A19
A23, http://dx.doi.org/10.1190/geo2011-0480.1.
Dellinger, J. A., J. T. Etgen, and G. Openshaw, 2012, Seismic acquisition using narrowband seismic
sources, U.S. Patent 2012/0155217 A1.
Cambois, G., A. Long, G. Parkes, T. Lundsten, A. Mattsson, and E. Fromyr, 2009, Multi-level airgun
array: A simple and effective way to enhance the low frequency content of marine seismic data:
79th Annual International Meeting, SEG, Expanded Abstracts, 152156,
http://dx.doi.org/10.1190/1.3255140.
Soubaras, R., 2010, Deghosting by joint deconvolution of a migration and a mirror migration: 80th
Annual International Meeting, SEG, Expanded Abstracts, 34063410,
http://dx.doi.org/10.1190/1.3513556.
Brink, M., and M. Svendsen, 1987, Marine seismic exploration using vertical receiver arrays: A means
for reduction of weather downtime: 57th Annual International Meeting, SEG, Expanded
Abstracts, 184187, http://dx.doi.org/10.1190/1.1892129.
Amundsen, L., H. Zhou, A. Reitan, and A. B. Weglein, 2013, On seismic deghosting by spatial
deconvolution: Geophysics, 78, no. 6, V267V271, http://dx.doi.org/10.1190/geo2013-0198.1.
Beasley, C. J., R. Coates, Y. Ji, and J. Perdomo, 2013, Wave equation receiver deghosting: A provocative
example: 83rd Annual International Meeting, SEG, Expanded Abstracts, 42264230.
Robertsson, J. O. A., L. Amundsen, and . Pedersen, 2014, Deghosting of arbitrarily depth-varying
marine hydrophone streamer data by time-space domain modelling: 84th Annual International
Meeting, SEG, Expanded Abstracts, 42484252, http://dx.doi.org/10.1190/segam2014-0221.1.
Ferber, R., and C. J. Beasley, 2014, Simulating ultra-deep-tow marine seismic data for receiver
deghosting: 76th Annual International Conference and Exhibition, EAGE, Extended Abstracts,
http://dx.doi.org/10.3997/2214-4609.20141452.
Mayhan, J. D., and A. B. Weglein, 2013, First application of Greens theorem-derived source and receiver
deghosting on deep-water Gulf of Mexico synthetic (SEAM) and field data: Geophysics, 78, no.
2, WA77WA89, http://dx.doi.org/10.1190/geo2012-0295.1.
Weglein, A. B., and J. Zhang, 2005, Extinction theorem deghosting method using towed streamer
pressure data: Analysis of the receiver array effect on deghosting and subsequent free surface
multiple removal: 75th Annual International Meeting, SEG, Expanded Abstracts, 20952098,
http://dx.doi.org/10.1190/1.2148125.
Amundsen, L., and H. Zhou, 2013, Low-frequency seismic deghosting: Geophysics, 78, no. 2, WA15
WA20, http://dx.doi.org/10.1190/geo2012-0276.1.
Orji, O. C., W. Sollner, and L. J. Gelius, 2013, Sea surface reflection coefficient estimation: 83rd Annual
International Meeting, SEG, Expanded Abstracts, 5155, http://dx.doi.org/10.1190/segam20130944.1.
Leroy, C. C., S. P. Robinson, and M. J. Goldsmith, 2008, A new equation for the accurate calculation of
sound speed in all oceans: The Journal of the Acoustical Society of America, 124, 27742782,
http://dx.doi.org/10.1121/1.2988296.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 166
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Berkhout, A. J., 1985, Seismic migration: Theoretical aspects, 3rd ed.: Elsevier.
Berkhout, A. J., and G. Blacquire, 2016, Deghosting by echo-deblending: Geophysical Prospecting, 64,
406420, http://dx.doi.org/10.1111/1365-2478.12293.
Walker, C., D. Monk, and D. Hays, 2014, Blended source The future of ocean bottom seismic
acquisition: 76th Annual International Conference and Exhibition, EAGE, Extended Abstracts,
Th ELI2 16, http://dx.doi.org/10.3997/2214-4609.20141470.
Letki, L. P., and C. Spjuth, 2014, Quantification of wavefield reconstruction quality from multisensor
streamer data using a witness streamer experiment: 76th Annual International Conference and
Exhibition, EAGE, Extended Abstracts, Th ELI1 10, http://dx.doi.org/10.3997/22144609.20141448.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 167
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
In 2013 an experiment was conducted with short streamers
attached to wave gliders during an ocean-bottom node
(OBN) survey in the Gulf of Mexico (Moldoveanu et al.,
2014). The experiment proved that wave gliders can be
used as a platform for marine acquisition if an array of
sensors is attached to them. The limited amount of acquired
data was comparable, after processing, with hydrophone
OBN data in terms of signal-to-noise ratio and frequency
content. During this experiment we realized that towing a
31-m streamer is a challenge due to the high drag, reduced
maneuverability, inaccurate receiver positioning of along
the streamers, and contamination with typical streamer
noise, such as swell noise, current noise, and the noise
induced by the streamer movement. For these reasons we
decided to design a 3D sensor array to replace the streamer.
The 3D sensor array is attached to the wave glider sub by
a decoupling cable (Figure 1). A detailed view of the 3D
multimeasurement sensor array (3DSA) is presented in
Figure 2. It consists of a rigid frame that has five sensor
arms (green color) placed in vertical and horizontal planes.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 168
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
(a)
(b)
Figure 3. (a) OBC source patch (red) and receiver patch (blue);
(b) Deployment of three wave gliders on top of the OBC receiver
patch
2016 SEG
SEG International Exposition and 86th Annual Meeting
(a)
(b)
Figure 5. Comparison of OBC CRG (a) vs. 3DSA CRG (b), after
swell noise was removed from 3DSA data
Page 169
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 170
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
(a)
(b)
Figure 8. (a) Receiver patch (green dots) and source circle (yellow)
(b) centers of the source circles (yellow dots); The receiver patch
will move from center to center and record the data generated
along the source circles
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 171
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Moldoveanu, N., 2011, Attenuation of high energy towed streamer noise: 81st Annual International
Meeting, SEG, Expanded Abstracts, 35763580, http://dx.doi.org/10.1190/1.3627943.
Vasconcelos, I., 2013, Source-receiver, reverse-time imaging of dual-source, vector-acoustic seismic data:
Geophysics, 78, no. 2, WA123WA145, http://dx.doi.org/10.1190/geo2012-0300.1.
Moldoveanu, N., A. Salama, O. Lien, E. Muyzert, S. Pai, and D. Monk, 2014, Marine acquisition using
autonomous marine vehicle: A field experiment: 84th Annual International Meeting, SEG,
Expanded Abstracts, http://dx.doi.org/10.1190/segam2014-1498.1.
Muyzert, E., 2013, Design, modeling and imaging of marine seismic swarm surveys: Schlumberger,
Internal report.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 172
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
The five-source design was conceived in November 2014
and was followed by a field test in January 2015 (Hager et
al., 2015). The overall concept and test results led to the
first full commercial survey, Baxter, which was acquired in
November 2015. The very short time from concept to
commercial production bears testament to the overall
simplicity of the penta-source design.
The five sources provide the means for improving the
cross-line sampling, but come at the expense of in-line fold
of coverage. This compromise drives the need to reduce the
source firing interval to a distance that is as small as can be
tolerated technically and physically. In order to understand
the physical limits for the shotpoint interval a field trial was
run which showed that as the time between shots decreased,
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 173
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 2: Left, Baxter survey outline blue (approx 400 km2), Conventional survey outline green (45 km2), with conventional sail-lines plotted.
FOC area shown right in red. Right, unique fold-of-coverage for conventional and far right, penta-source shown for first 3 offsets (500-800 m).
The FOC area covers the race-track boundary where sail-line direction is reversed (blue arrows). All displays approx 2:1 aspect ratio
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 174
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Processing
Both the penta-source and the conventional data were
processed through an identical sequence for the fast track,
except for the source debubble and de-signature to account
for the different sources. Given the similarity in the spatial
sampling of sub-surface lines, this is considered reasonable.
The fast track was processed through to isotropic pre-stack
migration with no regularization so that the effect of the
geometry could be clearly understood and seen. The full
processing sequence will include source interference
removal, full source and receiver deghosting, regularization
and anisotropic depth migration. It is important to note that
no bespoke algorithms or processing sequences are needed
to process the penta-source data.
To image the entire subset of conventional data,
surrounding lines from the penta-source survey were used
to satisfy the migration aperture after matching with a
simple scalar to the conventional. For the fast-track, the
entire 45 km2 area was migrated (green outline Figure 2).
All subsequent displays have a single matching scalar of
0.64 that adjusts for the source volume difference.
Penta and Conventional Comparisons
Figure 3 illustrates what we might expect to happen when
migrating un-regularized data: migration artifacts are
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 175
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 5: Top near surface features 116 ms below seabed, conventional on left , penta-source on right. Bottom left timeslice 2700 ms through
mass transport complex; bottom right coherency slice 3100 ms through poloygnal faulting seen in Figure 4
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 176
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Hager, E., M. Rocke, and P. Fontana, 2015, Efficient multi-source and multi-streamer configuration for
dense cross-line sampling. 85th Annual International Meeting, SEG, Expanded Abstracts, 100
104, http://dx.doi.org/10.1190/segam2015-5857262.1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 177
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 178
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 179
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 180
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
The subsurface imaging results from this survey show very
good definition of post-salt and subsalt sediments. This
allows accurate mapping of the stratigraphic and fault traps,
delineation of the salt bodies, subsalt structure or salt
associated reservoirs, and amplitude anomalies potential
related to fluid presence.
Towed-streamer circular acquisition using a single receiver
vessel, towing short streamers, and a source vessel placed
behind the streamer vessel, proved to be an efficient
method to acquire full-azimuth and 10000 m maximum
offset in a much obstructed area of the Gulf of Mexico
Shelf. Reprocessing all underlying NAZ seismic data was
essential to complement the new data.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 181
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Bakulin, A., M. Woodward, Y. Liu, O. Zdraveva, D. Nichols, and K. Osypov, 2010, Application of
steering filters to localized anisotropic tomography with well data: 80th Annual International
Meeting, SEG, Expanded Abstracts, 42864290, http://dx.doi.org/10.1190/1.3513765.
Dragoset, B., E. Verschuur, I. Moore and R. Bisley, 2010, A perspective on 3D surface-related multiple
elimination: Geophysics, 75, no. 5, 75A24575A261, http://dx.doi.org/10.1190/1.3475413.
Jiao, K., D. Sun, X. Cheng, and D. Vigh, 2015 Adjustive full waveform inversion: 85th Annual
International Meeting, SEG, Expanded Abstracts, 10911095,
http://dx.doi.org/10.1190/segam2015-5901541.1.
Moldoveanu, N. J., Kapoor, and M. Egan, 2008, A single-vessel method for wide-azimuth towedstreamer acquisition: 78th Annual International Meeting, SEG, Expanded Abstracts, 6569,
http://dx.doi.org/10.1190/1.3054856.
Woodward, M., D. Nichols, O. Zdraveva, P. Whitfield, and T. Johns, 2008, A decade of tomography:
Geophysics, 73, no. 5, 511, http://dx.doi.org/10.1190/1.2969907.
Zdraveva, O., R. Hubbard, M. OBriain, D. Zhang, and C. Vito, 2012, Anisotropic model building in
complex media: Three successful strategies applied to wide azimuth data from Gulf of Mexico:
74th EAGE Conference and Exhibition, Extended Abstracts.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 182
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
() = ()(),
(1)
() = 0 + 2 (),
(2)
Page 183
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
()
()
()
.
(/)
() =
2 2
,
1 0
= .
2016 SEG
SEG International Exposition and 86th Annual Meeting
(5)
(6)
Page 184
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
5.
Discussion
For linear sweeps, dispersion corrections based on equation
(7) completely remove the Doppler dispersion and the
forward and inverse f-k transforms are stable and efficient.
Synthetic seismic data shown in Figure 3(a) are phase
corrected using the f-k method, and the data after correction
are shown in Figure 6(a). For validating the effectiveness of
the phase correction process, the same synthetic seismogram
is generated but with a stationary source. The ideal
seismogram does not suffer the phase dispersion because the
speed of the boat is 0m/s and no Doppler Effect exists. The
subtraction error of the corrected data and the ideal
seismogram [Figure 6(b)] is hardly observed except for the
end traces which are affected by the boundary. We have
compared our result against the Hampson and Jakubowicz
method and the subtraction errors are basically the same.
2 ( 0 )
1 0
(9)
2016 SEG
SEG International Exposition and 86th Annual Meeting
Figure 6. (a) Seismic data shown in Figure 3(a) is phasecorrected using f-k domain method. (b) Subtraction error
between phase-corrected data shown in (a) and the synthetic
data generated with stationary source ( = 0/).
Page 185
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
4.
5.
Figure 8. (a) Input common-receiver gather from noenlinear sweep. (b) Field data shown in (a) after Doppler Effect
and spatial aliasing are removed.
Conclusions
Field Data
Ocean-Bottom receiver field tests with different marine
vibrator sweep parameters were conducted by Geokinetics.
One experiment had sweep parameters similar to those in
Table 1 except the sweep was highly non-linear. Each sweep
was designed to accommodate a specific exploration goal.
Analytic expressions of the sweeps were not necessarily
known, but of course the pilot sweeps were saved. The data
were corrected for phase distortion in the f-k domain with the
following procedure.
1. Generate synthetics of the direct arrival (slope =
pDA) with and without boat movement.
2. Fourier transform both synthetics and subtract
phase spectra; phase distortion dist(f) at pDA.
3. Create f-k matrix for phase-correction filter.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 186
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Aldridge, D. F., 1992, Mathematics of linear sweeps: Canadian Journal of Exploration Geophysics, 28,
no. 1, 6268.
Bagaini, Claudio. 2006. "Overview of simultaneous Vibroseis acquisition methods." SEG Expanded
Abstract. 10.1190/1.2370358.
Dragoset, W. H., 1988, Marine vibrators and the Doppler effect: Geophysics, 53, no. 11, 13881398.
http://dx.doi.org/10.1190/1.1442418.
Hampson, G., and H. Jakubowicz, 1995, The effects of source and receiver motion on seismic data:
Geophysical Prospecting, 43, no. 2, 221244. http://dx.doi.org/10.1111/j.13652478.1995.tb00133.x.
Kak, A. C., and A. Rosenfeld, 1982, Digital picture processing: Academic Press Inc.
Pramik, Bill, Lee M. Bell, Adam Grier, and Allen Lindsay. 2015. "Field testing the AquaVib: an alternate
marine seismic source." SEG Expanded Abstract. 10.1190/segam2015-5925758.1.
Rietsch, E., 1977, Computerize analysis of vibroseis signal similarity: Geophysical Prospecting, 25, no. 3,
541552. http://dx.doi.org/10.1111/j.1365-2478.1977.tb01186.x.
Sheriff, R. E., 2002, Encyclopedic Dictionary of Applied Geophysics: Society of Exploration
Geophysicists. http://dx.doi.org/10.1190/1.9781560802969.
Yilmaz, O., 2001, Seismic data analysis: Society of Exploration Geophysics.
http://dx.doi.org/10.1190/1.9781560801580.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 187
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Optimization of sea surface reflection coefficient and source geometry in conventional dual
source flip/flop marine seismic acquisition
Maksym Kryvohuz*, Shell International Exploration and Production Inc.; Xander Campman, Shell Global
Solutions International
Summary
In the present paper we demonstrate how redundant nearfield data from dual airgun array marine seismic source can
be used for estimation of the sea surface reflection
coefficient, optimization of the geometry of airgun arrays
and inversion for source signatures of individual airguns.
The latter allows prediction of accurate 3D far field
signatures of the gun arrays. The optimized sea surface
reflection coefficient is found as a function of frequency
and angle of incidence at the sea surface, and agrees with
the Rayleigh model of reflection from rough surfaces.
Introduction
Estimation of signatures of marine seismic sources is
important for successful offshore exploration for
hydrocarbons. It serves as an input to many seismic
processing algorithms. Areas where knowledge of an
accurate source signature is of great value includes
deconvolution, multiple attenuation, modelling and
inversion, 4D, AVO analysis, reservoir monitoring, and
analysis of marine multicomponent recordings.
As noted by Ziolkowski (1991) in his paper Why dont we
measure seismic signatures, it is possible to use
measurements to obtain signatures of marine seismic
sources rather than getting their wavelets using statistical
methods or modelling. Several methods have been
proposed in the past for determination of signatures of
airgun arrays from measurements (Ziolkowski et al., 1982;
Landr et al., 1994; Laws et al., 1998; Amundsen 1993),
from which the method based on near-field measurements
(Ziolkowski et al., 1982) seems to be the most simple and
practical. Deriving signatures from measurements promises
to properly handle phase and amplidutes of the far-field
estimates better than modeled signatures do, in particular at
low frequencies. The Ziolkowski method is based on the
assumption that an array of N interacting airguns (air
bubbles) can be treated as N non-interacting point sources.
The total wavefield of the gun array is then a superposition
of the fields from N notional point sources. Near-field
recordings of airgun arrays acoustic field at N different
places (i.e. by N different hydrophones installed on the
airgun array) are sufficient to perform the inversion of such
near-field data to find N source signatures of notional point
sources.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 188
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 1: Dual source in flip/flop marine seismic acquisition. Gray cylinders represent airguns, green cones represent NFHs, blue spheres
represent air bubbles released by airguns of the firing array, and the blue plane represents the sea surface.
rij (t )
1
pi (t )
s j t
c
j 1 rij ( t )
rijg (t )
1
,
R * g s j t
c
r
(
t
)
j 1
ij
g
rij (t ) , rij (t ) are source-hydrophone
(1)
where
and ghost-
2016 SEG
SEG International Exposition and 86th Annual Meeting
(2)
p Zs ,
(3)
p Z( )s ,
(4)
min p Z( )s
,s
(5)
s( ) ( Z( )T Z( )) 1 Z( )T p
(6)
Page 189
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Examples
Synthetic test
min p Z( )s( )
(7)
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 190
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 191
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Amundsen, L., 1993, Estimation of source array signatures: Geophysics, 58, 18651869,
http://dx.doi.org/10.1190/1.1443402.
Brekhovskikh, L. M., and Yu. P. Lysanov, 2003, Fundamentals of ocean acoustics: Springer.
Golub, G. H., and V. Pereyra, 1973, The differentiation of pseudo-inverses and nonlinear least squares
problems whose variables separate: SIAM Journal on Numerical Analysis, 10, 413432,
http://dx.doi.org/10.1137/0710036.
Klver, T., 2015, Derivation of statistical sea-surface information from dual-sensor towed streamer data:
77th Annual International Conference and Exhibition, EAGE, Expanded Abstracts,
http://dx.doi.org/10.3997/2214-4609.201413187.
Landr, M., J. Langhammer, R. Sollie, L. Amundsen, and E. Berg, 1994, Source signature determination
from ministreamer data: Geophysics, 59, 12611269, http://dx.doi.org/10.1190/1.1443683.
Laws, R., M. Landr, and L. Amundsen, 1998, An experimental comparison of three direct methods of
marine source signature estimation: Geophysical Prospecting, 46, 353389,
http://dx.doi.org/10.1046/j.1365-2478.1998.980334.x.
Ni, Y., C. Niang, and R. Siliqi, 2012, Monitoring the stability of airgun source array signature: 82nd
Annual International Meeting, SEG, Expanded Abstracts, http://dx.doi.org/10.1190/segam20120875.1.
Orji, O. C., 2013, Sea surface reflection coefficient estimation: 83rd Annual International Meeting, SEG,
Expanded Abstracts, http://dx.doi.org/10.1190/segam2013-0944.1.
Ziolkowski, A., 1991, Why dont we measure seismic signatures? Geophysics, 56, 190201,
http://dx.doi.org/10.1190/1.1443031.
Ziolkowski, A., G. Parkes, L. Hatton, and T. Haugland, 1982, The signature of an air gun array:
Computation from near-field measurements including interactions: Geophysics, 47, 14131421,
http://dx.doi.org/10.1190/1.1441289.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 192
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
As exploration and new field development
projects move towards geologically complex regions (subsalt, sub-basalt, Arctic), new challenges in acquiring and
processing seismic data emerge. Often, in these complex
regions, current acquisition geometries have limited
azimuths and/or offsets and hence provide sub-optimal
images. Moreover, in-sea and sub-sea installations place
restrictions on streamer acquisition both because of standoff zones for streamer vessels as well as noise from the
drilling activity. In such cases, acquisition design should be
based on operational as well as technical constraints.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 193
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 194
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Courtesy of WesternGeco
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 195
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
WAZ B
OBN
Courtesy of FairfieldNodal
WAZ D
OBN
Courtesy of FairfieldNodal
Figure 3: (A) Shot ray hit density in the image area for 12
km nominal offset, ray hits are gradually diminishing away
from center. This optimized shot coverage is used to design
shot layout and shooting direction. (B) Receiver ray hit
density in the image area; red is high hit and blue is low hit
density. Receiver coverage was planned in such a way that
most of the high hit density area can be covered.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 196
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Olofsson, B., P. Mitchell, and R. Doychev, 2012, Decimation test on an ocean-bottom node survey:
Feasibility to acquire sparse but full-azimuth data: The Leading Edge, 31, 457464,
http://dx.doi.org/10.1190/tle31040457.1.
Wang, K., and P. Hatchell, 2013, Impact of receiver decimation and mispositioning on time-lapse OBN
data: 75th Annual International Conference and Exhibition, EAGE, Extended Abstracts,
http://dx.doi.org/10.3997/2214-4609.20130128.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 197
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Optimal towing depth for marine seismic data minimizing the noise from normal modes
Toan Dao* and Martin Landr, Norwegian University of Science and Technology
Summary
The sea layer is an effective medium to transmit sound
waves. The acoustic signal from noise sources can be
detected at hydrophones from tens to hundreds of kilometers
away at any depth. Noise level is not monotonically
distributed in the water column. Contrary to the common
belief that it is more quiet in deeper water which is generally
applied to low-frequency surface waves, we find that noise
levels spatial distribution is non-monotonic with respect to
depth and is closely related to normal modes. Normal modes
are waves that travel within the water layer; along the
seafloor, reflect back to the free surface and are trapped in
the water column. Based on the theory for normal modes, we
find that in a 135 m water column, the noise level at 8 m is
less than at 60 m and greatest in between. We test our
hypothesis using a unique dataset acquired offshore Norway
that employed three streamer configurations: constant at 8
m, varying between 15-35 m and constant at 60 m deep. The
finding in the field data is consistent with the theory. Based
on this, we suggest to determine an optimal towing depth for
a given area based on theory when the normal mode signal
is minimal.
Introduction
2016 SEG
SEG International Exposition and 86th Annual Meeting
)} ( ) sin
S(). exp { (
sin
(1)
where
2
=
( ) =
12
2 ,
(2)
2
1
sin2 tan
2
2
, (3)
sin cos
12
1=
2
1
2
2
1
(4)
2
2
2
Page 198
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 199
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 200
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Acknowledgement
We thank Lundin for providing and allowing us to use the
data in this study. The Norwegian Research Council is
acknowledged for financial support the ROSE consortium at
NTNU.
Figure 4: Frequency spectra of the noise along the shot line of from top to bottom: 8m, 30m and 60m depth.
Figure 6: RMS amplitude of the noise along the shot line of 8 m, 30 m and 60 m data after 5-80Hz bandpass filtering.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 201
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Dhelie, P.E., J. E. Lie, V. Danielsen, A. K. Evensen, and A. Myklebostad, 2014, Broadband seismic A
novel way to increase notch diversity: 84th Annual International Meeting, SEG, Expanded
Abstracts, 148152, http://dx.doi.org/10.1190/segam2014-1330.1.
Elboth, T., B. A. P. Reif, O. Andreassen and M. B. Martell, 2012, Flow noise reduction from
superhydrophobic surfaces: Geophysics, 77, no. 1, P1P13, http://dx.doi.org/10.1190/geo20110001.1.
Ewing, W. M., W. S. Jardetsky, and F. Press, 1957, Elastic waves in layered media: McGraw-Hill.
Kroode, F ten, S. Bergler, C. Corsten, J. W de Maag, F. Strijbos, H. Tijhof, 2013, Broadband seismic data
The importance of low frequencies: Geophysics, 78, no. 2, WA3WA14,
http://dx.doi.org/10.1190/geo2012-0294.1.
Landr, 2007, Attenuation of seismic water-column noise, tested on seismic data from the Grane field:
Geophysics, 72, no. 4, V87V95, http://dx.doi.org/10.1190/1.2740020.
Landr, M and L. Amundsen, 2014, Is it optimal to tow air guns shallow to enhance low frequencies?
Geophysics, 79, A13-A18.
Landr, M., and P. Hatchell, 2012, Normal modes in seismic data Revisited: Geophysics, 77, no. 4,
W27W40, http://dx.doi.org/10.1190/geo2011-0094.1.
Parkes, G. E., and S. Hegna, 2011, How to influence the low frequency output of marine air-gun arrays:
73rd Annual International Conference and Exhibition, EAGE, Extended Abstracts, H012.
Pekeris, C. L., 1948, Theory of propagation of explosive sound in shallow water: Geological Society of
America Memoir, 27, 1117.
Wenz, M. G., 1962, Acoustic ambient noise in the ocean: Spectra and sources: The Journal of the
Acoustical Society of America, 34, 19361951, http://dx.doi.org/10.1121/1.1909155
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 202
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 203
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 204
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2 + +
2016 SEG
SEG International Exposition and 86th Annual Meeting
(1)
Page 205
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 206
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Berkhout, A. J., 2012, Blended acquisition with dispersed source arrays: Geophysics, 77, no. 4, A19
A23. http://dx.doi.org/10.1190/geo2011-0480.1.
Meier, M. A., S. E. Heiney, J. Tomic, P. Ibanez, and C. N. Byrne, 2015, Low frequency seismic
acquisition using a counter rotating eccentric mass vibrator: US Patent application
US2015/0041242 A1.
Reust, D. K., O. A. Johnston, J. A. Giles, and S. Ballinger, 2015, Very low frequency seismic source:
85th Annual International Meeting, SEG, Expanded Abstracts, 221 225.
Tsingas, C., Y. Kim, and J. Yoo, 2015, Broadband Acquisition, Deblending and Imaging Employing
Dispersed Source Arrays: EAGE Workshop on Broadband Seismic - A Broader View for the
Middle East, EAGE, BS27.
Wei, Z., 2015a, A new generation low frequency seismic vibrator: 85th Annual International Meeting,
SEG, Expanded Abstracts, 211 215.
Wei, Z., 2015b, Extending the vibroseis acquisition bandwidth with a newly designed low frequency
seismic vibrator: EAGE Workshop on Broadband Seismic - A Broader View for the Middle East,
EAGE, BS06.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 207
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Improving acquisition efficiency with a situational awareness (SA) based SIMOPS management
solution
Gary Pemberton, Stuart Darling*, and Emma McDonald, ION
Summary
Situational awareness (SA) describes the accuracy of a
persons knowledge and understanding of a situation. It
directly impacts the quality of decisions made by
personnel. SA can be severely compromised when
personnel are overloaded with complex and dynamic
information, as experienced during simultaneous operations
(SIMOPS) management in many offshore fields. When
conducting seismic acquisition, poor SA can result in an
increase in risk exposure, inefficiency and expense.
This paper presents a SIMOPS management system which
works to improve the SA of all users (including, but not
limited to seismic operations personnel), therefore
improving the efficiency and reducing the risk exposure of
acquisition. This is primarily achieved through a linked
Gantt chart and map, and the ability to model forward in
time, showing the users where vessels and other infield
equipment (such as streamers, dive crews, drilling rigs etc.)
plan to be in the future. A knowledge-based system
automatically provides alarms when rules are breached,
reducing the chance of user error and increasing visibility.
Several case histories from commercial field operations
have demonstrated that downtime and risk exposure can be
reduced through effective SA.
The cost of simultaneous operations
Oilfields are complex environments, with multiple
operations occurring simultaneously. The variety and
intricacy of SIMOPS considerations have increased
markedly as oilfields and infrastructure have developed.
Managing SIMOPS has become a major task, and errors
due to poor management can have a massive cost in QHSE
and monetary terms.
Seismic acquisition is a challenge for both the operator and
contractor, as it introduces a substantial moving obstruction
into an environment already crowded with Exploration and
Production activity. This occurs alongside industries such
as fishing and shipping which operate across oilfields.
When a seismic crew is added, the SIMOPS complexity
increases and careful planning is required. Communication
is crucial to minimizing standby - operational plans are
exchanged infield daily, often verbally or via email (as text
or spreadsheet). The problems associated with this timing
and distribution are well recognized by the time an email
is sent the information may be out of date, and spreadsheets
cannot clearly represent the spatial nature of the operations.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 208
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 1: Levels of situational awareness and feedback loop (based on Endsley 1995)
Automated Processes
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 209
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 2: Interactive Gantt chart showing overlapping task schedules. These are some of the facts which rules are monitored against.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 210
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 3: Linked Gantt chart and map showing multiple activities. Note seismic streamer vessel in yellow at the bottom of the map.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 211
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Bonaceto, C., and K. Burns, 2006, Using cognitive engineering to improve systems engineering:
Manuscript submitted for Presentation at the 2006 International Council on Systems Engineering
Conference, http://dx.doi.org/10.1002/j.2334-5837.2006.tb02784.x.
Davidson, J. and M. J. Lochmann, 2012, Lessons and insights from unexpected places, SPE Intelligent
Energy International, SPE-149998-MS, http://dx.doi.org/10.2118/149998-MS.
Dictionary of Military and Associated Terms, 2005, accessed 23 March 2016 from
http://www.thefreedictionary.com/common operational picture.
Endsley, M. R., 1995, Toward a theory of situation awareness in dynamic systems: Human Factors, 37,
3264, http://dx.doi.org/10.1518/001872095779049543.
Endsley, M.R., 2000. Theoretical underpinnings of situation awareness: a critical review: Situation
Awareness Analysis and Measurement, 332.
Galloway, A., J. A. McDermid, J. Murdoch, and D. J. Pumfrey, 2002, Automation of system safety
analysis: possibilities and pitfalls: Proceedings of International Systems Safety Conference,
Denver, accessed 16 March 2016 http://wwwusers.cs.york.ac.uk/~djp/publications/Automation.pdf
Gersh, J. R., J. A. McKneely, and R. W. Remington, 2005, Cognitive engineering: understanding human
interaction with complex systems: Johns Hopkins APL Technical Digest, 26, 377382.
Glandrup, M., 2013. Improving situation awareness in the maritime domain: Situation Awareness with
Systems of Systems, 21-38: Springer, http://dx.doi.org/10.1007/978-1-4614-6230-9_2.
Graham, B., W.K Reilly, F. Beinecke, D.F.Boesch, T.D. Garcia, C.A. Murray, and F. Ulmer, 2011, Deep
water: the gulf oil disaster and the future of offshore drilling, report to the President: National
Commission on the BP Deepwater Horizon Oil Spill and Offshore Drilling.
Harrald, J., and T. Jefferson, 2007, Shared situational awareness in emergency management mitigation
and response: 40th Annual Hawaii International Conference on Systems Science, 23,
http://dx.doi.org/10.1109/HICSS.2007.481.
Hetherington, C., R. Flin, and K. Mearns, 2006, Safety in shipping: the human element: Journal of Safety
Research, 37, no. 4, 401411. http://dx.doi.org/10.1016/j.jsr.2006.04.007.
Jones, D. G., and M. R. Endsley, 1996, Sources of situational awareness errors in aviation: Aviation,
Space, and Environmental Medicine, 67, 507512.
Leveson, N. G., 2011, Risk management in the oil and gas industry: Testimony before the United States
Senate Committee on Energy and Natural Resources, http://mitei.mit.edu/news/risk-managementoil-and-gas-industry
Lochmann, M., and I. Brown, 2014, Intelligent energy: A strategic inflection point: SPE Annual
Technical Conference and Exhibition, SPE, http://dx.doi.org/10.2118/170630-MS.
Lochmann, M., S. Darling, G. Pemberton, R. Mills, and J. Altaie, 2015, Improving situational awareness
to prevent conflicts in simultaneous offshore operations: SPE Middle East Intelligent Oil and Gas
Conference and Exhibition, SPE-176796-MS
Miller, G., 1956, The magical number seven, plus or minus two: some limits on our capacity for
processing information: Psychological Review, 63, 8197, http://dx.doi.org/10.1037/h0043158.
Pemberton, G., S. Darling, C. Koheler, and E. McDonald, 2015, Managing simultaneous operations
during seismic acquisition: First Break, 33, 7581.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 212
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Roberts, R., R. Flin, and J. Cleland, 2013, Situation awareness in offshore drillers: Proceedings of the
11th International Conference on Naturalistic Decision Making (NDM 2013), Arpege Science
Publishing
Safahani, M. S., and L. S. Tuttle, 2013, Situation awareness and its practical application in maritime
domain: The International Maritime Human Element Bulletin HE01120, accessed 16 March 2016
http://www.he-alert.org/en/articles.cfm/page/37
Simons, D. J., 2000, Attentional capture and inattentional blindness: Trends in Cognitive Sciences, 4,
147155, http://dx.doi.org/10.1016/S1364-6613(00)01455-8.
Sneddon, A., K. Mearns, and R. Flin, 2006, Situation awareness and safety in offshore drill crews:
Cognition Technology and Work, 8, 255267. http://dx.doi.org/10.1007/s10111-006-0040-1.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 213
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Implementation of Vibrator Autonomous Shooting in Urban Area, Case Study from Kuwait
Bader Al-Ajmi *, Khaled SalahEldin , Meshari Al-Awadi , Ding Guandong , Li Wenjing , Nie Mingtao , Liu
Introduction
It is an extremely challengeable seismic acquisition
covering water bay areas, transition zone and crowded
urban areas over 700 km2. BGP allocated suitable and
sufficient equipment, including two types of recording
systems combining cable system for marine portion and
cable-less system for land portion. The cable-less nodal
recording system offered a golden opportunity for
application and popularization of autonomous shooting
technology since it could realize continuous recording. The
bottleneck of high production efficiency is not node
recording but vibrator shooting instead, especially in urban
areas. The Digital-Seis System (DSS) was equipped for the
purpose of autonomous shooting so that it was possible to
enhance operation productivity. DSS mainly contains two
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 214
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Methodology
Normally, vibrators shooting is under control of the
vibrator controller system mounted on recording room by
radio which is constrained with hardware, electromagnetic
interference, bad weather, and barriers like mountains or
urban architectures. In some cases, these communication
limitation cannot be avoided when using radios, the best
way is to minimize the usage of radios or shorten the
contact range. The autonomous shooting will reduce the
period of using radios whatever it is analog or digital and
the most time of using radios are communication within
one group or fleet. The most significant is that the
autonomous shooting can fulfill automatically sweeping
based on preset sequential time window without any other
control or triggering from outside of DSS system, which
could not only improve productivity but also lower HSE
risks. Regarding urban seismic vibration, there are a lot of
tall buildings and blind areas which will block the radio
communication, especially long range connection.
Furthermore, crowed vehicles and numerous human
activities will bring high potential HSE hazards for seismic
operations, especially big and unwieldy recording truck
moving. Autonomous shooting method can release seismic
operations from such inconvenience.
The methodology of autonomous shooting depending upon
DSS contains four (4) aspects. First, the real-time
information is exchanged by both of digital radios and
satellite data link or 4G network. Second, 4G network link
is the main channel for QC & status information exchange
and the digital radio data link is alternative. Third, the
satellite data link is for GPS & Differential signal. Fourth,
the DSS software for integrated information processing.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 215
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
The third step is sending SPS file from QC office in realtime. Once the sweep parameter changes or source code
switches, QC office need to update shot points in a block or
zone and spread to vibrators and recording truck. Both of
DSG and DSC can monitor shot point status that green
points mean already acquired, yellow points indicate lack
of vibrator performance files, and red color points represent
not being shot yet. Regarding shot points chosen to be
acquired, the vibration section can take point by point
method or line by line or block by block method under
Program mode. The common way is to choose default
method following appointed time window to shoot when
READY through matching coordinates.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 216
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
Figure 7: Time sequences with constant period
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 217
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
No references.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 218
INTRODUCTION
Seismic airguns are the predominant source used in marine
seismic surveys. They function by discharging highly pressurized air forming a bubble that expands and contracts in the water, exciting pressure waves over a wide range of frequencies.
The low frequency waves are used to image targets of interest.
Several studies have emphasized the need for improved low
frequency content (below 30 Hz) for sub-salt and sub-basalt
imaging (Ziolkowski et al., 2003). The high frequency energy
(above 150 Hz) is generally useless for seismic imaging as it is
attenuated before it reaches the target or scattered by the heterogeneous overburden. In addition, current seismic acquisition and processing techniques sample at 2 ms and only utilize
frequencies up to 220 Hz. Thus, reducing the proportion of
high frequency energy generated would improve the efficiency
of the airgun. Furthermore, ocean noise from marine seismic
surveys is thought to have a significant impact on marine life
(Weilgart, 2007; Nowacek et al., 2015). The specific impact
of marine seismic surveys on the plethora of different marine
species is complicated and understanding is hampered by limited data (Weilgart, 2013). However, it is likely that reducing
the high frequency noise that is not used for seismic imaging
will have environmental benefits without compromising survey quality.
Chelminski et al. (2016) proposed a low-pressure source (LPS)
with radically reduced pressure and increased volume. They
argue that the LPS will be more efficient and have lower high
frequency content, alleviating environmental concerns. To investigate this idea, Chelminski Technology and Dolphin Geophysical conducted field tests of a LPS prototype in June 2015.
2016 SEG
SEG International Exposition and 86th Annual Meeting
DATA
Data was collected over two days at Lake Seneca, a 200 m
deep lake in upstate New York. The LPS prototype was suspended at variable depth from a crane over the side of the boat.
Two airgun volumes, 598 in3 and 50 in3 , were tested at a range
of depths (5, 7.5, 10, 15, and 25 m measured depth) and pressures (135 psi to 1320 psi for the 598 in3 airgun and 510 psi
to 1850 psi for the 50 in3 airgun). Observations were made
with a 24 channel downhole array in the far-field, 75 m below
the airgun, with a spacing of 2 m between the channels. The
observations are recorded at 32 kHz, a much higher temporal
resolution than in industry seismic surveys, where 0.5 kHz is
the standard sampling rate.
210
810 psi, 598 in 3 , 5 m
610 psi, 598 in 3 , 10 m
410 psi, 598 in 3 , 15 m
200
power (dB)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
190
180
170
160
150
140
101
102
frequency (Hz)
f =k
(1 + D/10)5/6
,
(paVa )1/3
(1)
Page 219
MODEL
We solve the Euler equations governing the motion of a compressible fluid and evaluate the solution on the bubble wall to
give a nonlinear ordinary differential equation for the bubble
dynamics. Our work differs from previous studies (e.g., Ziolkowski, 1970; de Graaf et al., 2014) as we use the modified
Herring equation (Herring, 1941; Cole, 1948; Vokurka, 1986)
rather than the Gilmore (1952) equation to describe the bubble
motions. The modified Herring equation is
3
p p
R
RR + R 2 = b
+
p ,
2
c b
(2)
where R and R = dR/dt are the radius and velocity of the bubble wall, respectively, pb is the pressure inside the bubble, and
p , and c are the pressure, density, and speed of sound,
respectively, in the water infinitely far from the bubble. Without the pb term, equation 2 is the Rayleigh equation (Rayleigh,
1917) which is a statement of conservation of momentum for
an incompressible fluid. The pb term is a correction for compressibility that allows for energy loss through acoustic radiation. The Herring equation assumes a constant, rather than
pressure dependent, speed of sound, which is well justified as
1. The modified version of the Herring equation neR/c
) type correction factors (Vokurka, 1986).
glects the (1 R/c
The bubble is coupled to the airgun by mass conservation. We
solve for the exit velocity of the flow out of the airgun at each
time step rather than assuming choked flow. The airgun is assumed to discharge adiabatically. The temperature of the bubble is governed by the first law of thermodynamics for an open
system. This allows for heat conduction across the bubble wall
2016 SEG
SEG International Exposition and 86th Annual Meeting
p(r,t) =
V (t r/c ) V (t r/c )2
,
4r
32 2 r4
(3)
normalized R
1
0.8
0.6
0.4
0.2
Gilmore
Modified Herring
normalized p
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
0.1
0.2
0.3
0.4
0.5
1
Gilmore
Keller and Kolodner
0.5
0
-0.5
0.1
0.2
0.3
0.4
0.5
time (s)
Figure 2: Bubble radius (top) and near-field pressure perturbation in the water, p = pb p , (bottom) as computed by the
Gilmore (1952) equations and with the analogous equations
from Herring (1941) and Keller and Kolodner (1956), which
are used in this work. The bubble radius from the modified
Herring equation is used as an input to the Keller and Kolodner (1956) pressure equation. The bubble radius and pressure
perturbation are normalized by the maximum of the Gilmore
(1952) solutions. The initial conditions of Ziolkowski (1970)
are used where the initial volume of the bubble is equal to the
volume of the airgun. The discontinuity in the derivative of the
radius and pressure is due to the airgun port opening instantaneously.
The observed pressure perturbation in the water is a superposition of the direct arrival and the ghost, which is a wave that
is reflected from the surface of the water and arrives at the
receiver at a later time. In the near-field, the amplitude of the
ghost is much smaller than that of the direct arrival as the ghost
travels along a much longer path, reducing the amplitude by
geometrical spreading. In the far-field, the path length for the
Page 220
(4)
MODEL VALIDATION
In order to validate our model, we compare our simulation results to the lake data. The model has several tunable parameters. We tune these parameters so that the model fits the farfield data for one airgun firing configuration (Figure 3). We
can then match the measurements from the other firing configurations by varying the airgun properties (Figure 4). This is
done without any further tuning of the model parameters.
The magnitude of the pressure perturbation depends upon the
location of the receiver relative to the airgun. To remove this
dependency, we normalize all observations and simulations by
multiplying the pressure perturbation by r, the distance from
the airgun to the receiver, and state the result in bar m. The
port area of the airgun used in the lake was measured as 11
in2 . In our simulations, we use a reduced area of 4 in2 to best
fit the data. de Graaf et al. (2014) used a similar approach to
avoid over predicting the amplitude of the initial peak when
modeling conventional airguns.
p (bar m)
1
Data
Model
0.5
0
-0.5
0.1
0.2
0.3
0.4
0.5
time (s)
power (dB)
200
180
160
140
120
100
101
102
frequency (Hz)
p (bar m)
direct arrival and the ghost are almost the same. The ghost
must be accounted for in order to accurately simulate the observed pressure perturbations, especially in the far-field. The
pressure perturbation due to the ghost signal is calculated by
replacing the path length of the direct arrival, r, with the path
length of the ghost, r + 2D, in equation (3). The sea surface is
assumed to have a reflectivity of -1 (Ziolkowski, 1982). The
reflectivity can be frequency dependent, especially in rough
seas. The lake surface was relatively flat during data acquisition and we found that -1 was an appropriate choice for this
work.
Data
Model
1
0
-1
-2
0.1
0.2
0.3
0.4
0.5
time (s)
200
power (dB)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
180
160
140
120
100
101
102
frequency (Hz)
Figure 4: Comparison between far-field observations and simulations for an airgun fired at a depth of 15 m, pressure of 1030
psi, and volume of 598 in3 . The tunable model parameters are
the same as for Figure 3.
6000 in3 and pressure of 600 psi to 1000 psi. The LPS will
have a much larger port area than conventional airguns, 62 in2
compared to 16 in2 .
The simulation results are in agreement with the RayleighWillis equation (Figure 5) and display similar trends to the
data (see Figure 1). The fit to the data and agreement with
the Rayleigh-Willis equation validates our model and enables
us to use it to investigate airgun firing configurations not tested
in the lake, such as the proposed LPS.
Larger volume conventional airguns (2000 in3 ) have been proposed as a solution to improve the low frequency content (Ziolkowski et al., 2003). However, the larger volume airguns are
heavy and have maintenance issues because of the high pres-
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 221
210
810 psi, 598 ci, 5 m
610 psi, 598 ci, 10 m
410 psi, 598 ci, 15 m
180
160
150
140
101
102
frequency (Hz)
pressure (bar m)
4
2
0
-2
0.2
0.4
0.6
0.8
time (s)
210
200
190
180
pressure (bar m)
170
6
1000 in3 airgun at 2000 psi
4000 in3 LPS at 600 psi
4
2
0
-2
0.2
0.4
0.6
0.8
time (s)
220
power (dB)
power (dB)
190
power (dB)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
200
200
180
160
101
102
frequency (Hz)
CONCLUSION
There is significant interest in reducing the high frequency
noise that is produced by seismic airguns as this is thought
to adversely impact marine life. In addition, it is desirable to
improve their imaging capabilities and efficiency. The lowpressure source has been proposed as an improvement to conventional seismic airguns that will achieve these goals.
We present a numerical model for seismic airguns and lowpressure sources that we validate against high resolution farfield data from a lake. Numerical simulations show that the
proposed low pressure source can reduce the high frequency
noise without compromising the usable low frequency content
compared to a conventional airgun and is thus more efficient
and environmentally friendly. Furthermore, the low-pressure
source can be manufactured and operated at far larger volumes
than conventional airguns enabling the low frequency content
to be improved resulting in better sub-salt and sub-basalt imaging capabilities.
170
160
10
10
frequency (Hz)
ACKNOWLEDGMENTS
We thank Chelminski Technology and Dolphin Geophysical
for providing us access to the data from Lake Seneca. We acknowledge the Stanford Exploration Project and their sponsors
for financial support.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 222
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Chelminski, S., J. Chelminski, S. Denny, and S. Ronen, 2016, The low-pressure source: Harts E&P
Magazine, 14.
Cole, R. H., 1948, Underwater explosions: Princeton University Press,
http://dx.doi.org/10.5962/bhl.title.48411.
de Graaf, K. L., I. Penesis, and P. A. Brandner, 2014, Modelling of seismic airgun bubble dynamics and
pressure field using the Gilmore equation with additional damping factors: Ocean Engineering,
76, 3239, http://dx.doi.org/10.1016/j.oceaneng.2013.12.001.
Gilmore, F. R., 1952, The growth or collapse of a spherical bubble in a viscous compressible liquid:
Hydrodynamics Laboratory, California Institute of Technology, Technical report.
Herring, C., 1941, Theory of the pulsations of the gas bubble produced by an underwater explosion:
Technical report, OSRD Report 236.
Keller, J. B., and I. I. Kolodner, 1956, Damping of underwater explosion bubble oscillations: Journal of
Applied Physics, 27, 11521161, http://dx.doi.org/10.1063/1.1722221.
Landr, M., and R. Sollie, 1992, Source signature determination by inversion: Geophysics, 57, 1633
1640, http://dx.doi.org/10.1190/1.1443230.
Li, G., Z. Liu, J. Wang, and M. Cao, 2014, Air-gun signature modelling considering the influence of
mechanical structure factors: Journal of Geophysics and Engineering, 11, 025005,
http://dx.doi.org/10.1088/1742-2132/11/2/025005.
Li, G. F., M. Q. Cao, H. L. Chen, and C. Z. Ni, 2010, Modeling air gun signatures in marine seismic
exploration considering multiple physical factors: Applied Geophysics, 7, no. 2, 158165,
http://dx.doi.org/10.1007/s11770-010-0240-y.
Nowacek, D. P., C. W. Clark, D. Mann, P. J. O. Miller, H. C. Rosenbaum, J. S. Golden, M. Jasny, J.
Kraska, and B. L. Southall, 2015, Marine seismic surveys and ocean noise: Time for coordinated
and prudent planning: Frontiers in Ecology and the Environment, 13, 378386,
http://dx.doi.org/10.1890/130286.
Rayleigh, L., 1917, On the pressure developed in a liquid during the collapse of a spherical cavity:
Philosophical Magazine Series 6, 34, 9498, http://dx.doi.org/10.1080/14786440808635681.
Ronen, S., S. Denny, R. Telling, S. Chelminski, J. Young, D. Darling, and S. Murphy, 2015, Reducing
ocean noise in offshore seismic surveys using low-pressure sources and swarms of motorized
unmanned surface vessels Low- pressure sources and Swarms of motorized unmanned surface
vessels: 85th Annual International Meeting, SEG, Expanded Abstracts, 49564960,
http://dx.doi.org/10.1190/segam2015-5928795.1.
Safar, M. H., 1976, The radiation of acoustic waves from an airgun: Geophysical Prospecting, 24, 756
772, http://dx.doi.org/10.1111/j.1365-2478.1976.tb01571.x.
Schulze-Gattermann, R., 1972, Physical aspects of the air- pulser as a seismic energy source:
Geophysical Prospecting, 20, 155192, http://dx.doi.org/10.1111/j.1365-2478.1972.tb00628.x.
Vokurka, K., 1986, Comparison of Rayleighs, Herrings, and Gilmores models of gas bubbles: Acta
Acustica United with Acustica, 59, 214219(6).
Weilgart, L., 2007, The impacts of anthropogenic ocean noise on cetaceans and implications for
management: Canadian Journal of Zoology, 85, 10911116, http://dx.doi.org/10.1139/Z07-101.
Weilgart, L., 2013, A review of the impacts of seismic airgun surveys on marine life: CBD Expert
Workshop on Underwater Noise and its Impacts on Marine and Coastal Biodiversity, 19.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 223
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Willis, H. F., 1941, Underwater explosions: the time interval between successive explosions: Technical
report, British Report WA-47-21.
Ziolkowski, A., 1970, A method for calculating the output pressure waveform from an air gun:
Geophysical Journal of the Royal Astronomical Society, 21, 137161,
http://dx.doi.org/10.1111/j.1365-246X.1970.tb01773.x.
Ziolkowski, A., 1982, An airgun model which includes heat transfer and bubble interactions: 52th Annual
International Meeting, SEG, Expanded Abstracts, 187189.
Ziolkowski, A., P. Hanssen, R. Gatliff, H. Jakubowicz, A. Dobson, G. Hampson, X. Y. Li, and E. Liu,
2003, Use of low frequencies for sub-basalt imaging: Geophysical Prospecting, 51, 169182,
http://dx.doi.org/10.1046/j.1365-2478.2003.00363.x.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 224
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
There is an exciting new technology that has become popular
with recreational flyers and a growing cadre of geoscience
professionals (Cicoria, 2015). It is the unmanned aerial
vehicle (UAV) or airborne drone (Chamayou, 2015; Whittle,
2015) - a flying platform with propulsion, positioning, and
remote or self control. Most drones also have some kind of
sensors which capture and possibly transmit information.
There has been considerable coverage in the news and
technical press about the burgeoning promise, along with
concerns, of drones (Horgan, 2013; Adams and Bushwick,
2014). The promise of UAVs is legion - from remote rescue
to deliveries, farming, and forensics. Drones are being used
in humanitarian response efforts after disasters. For
example, after the April 2015 7.8 magnitude earthquake in
Nepal, drones were deployed and able to survey sites,
inspect buildings and roads, and create 3D maps of cultural
heritage sites such as temples (Team Rotordrone, 2015).
However, there are issues which include safety, reliability,
and privacy. Preliminary regulations governing the
recreational and commercial operation of drones have been
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 225
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Coupling
The first experiment compares the output from the
geophones, as deployed with variable coupling on
different surfaces (Figure 3), with the seismic
drone. This test was undertaken at the University
of Houston campus.
Recording a shot gather
The second experiment compared the drone and
the cabled system at our La Marque Geophysical
Observatory 15 miles north of Galveston, Texas.
The recording system and source are Geometrics
StrataVisor NZ and 40 kg propelled energy
generator (PEG.) We deployed a 24-channel 2D
line. The drone flew closely to each receiver,
landed, and recorded each shot.
Soil penetration
The third experiment compares soil penetration
and the angle of incidence in three different soil
types. This is important to ensure quality data for
coupling in various soils. We also need to test
whether the drone can takeoff, even when the
geophones are well planted in soil.
Experiments
We conducted three experiments to test the drones seismic
capabilities:
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 226
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Survey Results
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 227
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
Further capabilities
We noted previously that the vibration or seismic sensing
platform could be detached from the aircraft and left to
monitor In addition, the platform could also be an
autonomous rover in its own right (Figure 9). It could thus
move to various positions to achieve, for example, more
recording locations, better coupling, sunlight, shade, etc. It
could then recovered by the drone or by other means.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 228
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Adams, E., and S. Bushwick, 2014, Successful drone delivery: Popular Science, 285, 4350.
Chamayou, G., 2015, A theory of the drone: The New Press.
Cicoria, N., 2015, Unmanned aerial vehicles (UAVs) in seismic: How to evaluate drone technology and
operators effectively: The Source: Canadian Association Geophysical Contractors, 12, 1618.
Gifford, C. M., 2005, Robotic seismic sensors for polar environments: Masters thesis, University of
Kansas.
Horgan, J., 2013, The drones come home: National Geographic, 223, 122135.
Lorge, G., 2015, Come on in, the waters fine! Water-tight multirotors from QuadH2O, in J. Babler, ed.,
Make: Magazine: Maker Media Inc. 44, 4243.
Masunaga, S., 2016, Dynamic growth for drones ahead: Houston Chronicle, B3.
Oleson, T., 2013, Droning on for science: Unmanned aerial vehicles take off in geosciences research:
Earth, 58, 2837.
Postel, J. J., T. Bianchi, and J. Grimsdale, 2014, Drone seismic sensing method and apparatus: U. S.
Patent US 2014/0307525 A1, https://www.google.com/patents/US20140307525.
Sonner, S., 2016, Drone journey called a first for autonomous deliveries: Houston Chronicle, B3.
Team RotorDrone, 2015, Disaster in Nepal: Drones aid first responders in the aftermath of a devastating
earthquake: RotorDrone Magazine, 4247.
Whittle, R., 2015, Hellfire meets PredatorHow the drone was cleared to shoot: Air and Space
Magazine, 29, 2227.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 229
Summary
10
Introduction
1
1000000
100000
Number of channels per crew
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
10000
1000
100
10
1
1960
1970
1980
1990
2000
2010
2020
2030
2040
2050
Year
2016 SEG
SEG International Exposition and 86th Annual Meeting
1975
1980
1985
1990
1995
2000
2005
2010
2015
Year
Page 230
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 231
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 232
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
HSE benefits
The method contributes a positive impact on the
environment:
Conclusions
Generally speaking, the system can be seen as an evolution
combining the advantages of both streamer and OBC/OBN
techniques.
It may be considered as an evolution of multi-sensor towed
streamer technique in which the number of cables is
theoretically unlimited and the noise level is reduced thanks
to limited tension, larger depth and the stationary cable;
thus providing true 4C with low-frequency content.
It can be seen as a floating OBC technique in which cables
are easily moved and reconfigured, removing the need for
coupling to the sea bed, seabed interface related noise and
time-consuming operations (deployment, recovery, rollover); thus providing an efficient high end 3D method the
only downside being not to be able to record shear waves.
It may also be understood as performing the equivalent of
land acquisition at sea with a perfectly sampled geometry,
and without the complex statics from the surface weather
zone and without the terrestrial obstacles such as mountains
or forests, thus offering full-offset full-azimuth in a regular
square bin with high fold.
Page 233
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Elboth, T., B. A. Reif, and . Andreassen, 2009, Flow and swell noise in marine seismic data:
Geophysics, 74, no. 2, Q17Q25, http://dx.doi.org/10.1190/1.3078403.
Elboth, T., D. Lilja, B. A. Reif, and . Andreassen, 2010, Investigation of flow and flow noise around a
seismic streamer cable: Geophysics, 75, no. 1, Q1Q9, http://dx.doi.org/10.1190/1.3294639.
Schoenberger, M., and J.-F. Mifsud, 1974, Hydrophone streamer noise: Geophysics, 39, 781793,
http://dx.doi.org/10.1190/1.1440466.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 234
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Processing considerations for seismic data acquired with wave gliders towing a 3D Sensor Array
Philippe Caprioli*, Bent Andreas Kjellesvig, Nick Moldoveanu and Ed Kragh, Schlumberger
Summary
Field test
Introduction
The wave glider is an Autonomous
Marine Vehicle (AMV) with the
unique feature of using wave
motion for propulsion, which
enables it to stay offshore for long
periods of time without servicing.
The AMV consists of a surface
float and a submerged glider (sub)
connected by an electromechanical
Figure 1: Wave glider
umbilical (Figure 1). The wave
with 3DSA
glider can be equipped with a
range of metrological, oceanographic sensors, and
communication systems. It is well suited for a variety of
applications (Pai 2013), including the acquisition of seismic
data (Moldoveanu et al. 2014).
In the application described here, the wave glider is equipped
with a 3D Sensor Array (3DSA). The 3DSA consists of 15
hydrophones mounted on cubic frame of about 1 m in size,
as illustrated in Figure 1. The presence of a buoyancy engine
just below the top arm ensures that this arm remains on the
high side, and that the entire array is oriented and does not
rotate on its axis while moving through the water. The 3D
distribution of the hydrophones enables deriving the spatial
gradients (via Finite Difference) of the recorded pressure
field. This paper discusses acquisition geometry QC, and the
challenges and benefits when using (either explicitly or
implicitly) such pressure gradients in processing one line of
3DSA field data.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 235
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
= ( , , ) dependent phase
shift operators:
.
.
=
+
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 236
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 237
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusion
This study shows that useful seismic data can be acquired
with a wave glider towing a 3D sensor array of hydrophones.
The distribution of sensors in a 3D sensor array enables
estimating the spatial gradients of the recorded pressure
wavefield. Using a field data example, we found genuine
2016 SEG
SEG International Exposition and 86th Annual Meeting
Acknowledgments
To Abu Dhabi Marine Operating Company for allowing the
3DSA and OBC data comparison, to Liquid Robotics Oil &
Gas for running the field operations, and to Schlumberger
for permitting the publication. We also thank our colleagues
Leendert Combee and Simon Baker for valuable input.
Page 238
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Esmersoy, C., 1990, Inversion of P- and SV-waves from multicomponent offset vertical seismic profiles:
Geophysics, 55, 3950, http://dx.doi.org/10.1190/1.1442770.
Leaney, W. S., 1990, Parametric wavefield decomposition and applications: 60th Annual International
Meeting, SEG, Expanded Abstracts, 10971100, http://dx.doi.org/10.1190/1.1889919.
Moldoveanu, N., L. Combee, P. Caprioli, E. Muyzert, and S. Pai, 2016, Marine seismic acquisition with
3D sensor arrays towed by wave gliders: SEG.
Moldoveanu, N., A. Salama, O. Lien, E. Muyzert, and S. Pai, 2014, Marine acquisition using wave
gliders: A field feasibility test: 76th Annual International Conference and Exhibition, EAGE,
Extended Abstracts, http://dx.doi.org/10.3997/2214-4609.20141469.
Pai, S., 2013, Wave glider Introduction to an innovative autonomous remotely piloted ocean data
collection platform: Presented at the SPE Offshore Europe Oil and Gas Conference and
Exhibition, http://dx.doi.org/10.2118/166626-MS.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 239
In the last few years, the interest towards the utilization of Dispersed Source Arrays (DSAs) in seismic acquisition has considerably grown. The proposed approach offers a wide range
of practical advantages, while no physical constraint restrains
us from utilizing diverse sources with different spectral properties during seismic surveys. As a consequence, the use of
simple autonomous source boats with airgun arrays of different sizes or marine vibrators producing sweeps with different
frequency ranges (in marine) and simple autonomous source
trucks (on land) becomes a practical proposition in DSA acquisitions.
This concept could give to the system additional operational
flexibility and facilitate the automation of seismic data collection. Therefore, with this study we intend to investigate the advantages that DSAs and system decentralization would bring
to seismic data acquisition. Although the main focus of this
research is on the marine environment, a generalization to the
applications of the method on land is possible. Preliminary
examples show that it is possible to produce valid migration
outputs from 3D decentralized DSA data.
the concept. Moreover, the improved operational flexibility allowed by the DSA concept can be an important step towards
the robotization and the decentralization of the seismic acquisition process.
An interesting first attempt of DSA land data acquisition and
inversion (FWI-based) has been successfully carried on and
presented by Kim and Tsingas (2014). In the following, the
theoretical framework and the inversion scheme of reference
is presented, the benefits of seismic acquisition system decentralization are introduced, and the first preliminary migration
results of 3D decentralized DSA data are provided.
0.5
t(s)
SUMMARY
1.5
INTRODUCTION
Following the guidelines drawn by Berkhout (2012), we suggest to replace (or reinforce) traditional broadband units with
narrow(er)band devices, together representing a Dispersed Source Array (DSA). The whole inhomogeneous ensemble of
sources incorporated to the arrays is required to cover the entire temporal and spatial bandwidth of interest, according to
the users needs. Figure 1 illustrates the principle by showing a comparison between two blended shot records: the first
with three identical broadband sources, the second with three
different narrowband source units.
1000
2000
3000
4000
5000
3000
4000
5000
(a)
0.5
t(s)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Matteo Caporal , Gerrit Blacqui`ere and Mikhail Davydenko, Delft University of Technology
1.5
These measures may have extremely fruitful practical implications. Narrowband source units can be technically simpler to
produce and more effective from an acoustic energy transmission point of view than todays broadband alternatives. Modern multiple-driver loudspeaker systems are based on the same
key concept and their improved performances are demonstrated
(Davis and Patronis, 2006). Additionally, the method will allow us to adjust the source interval in a frequency dependent
manner and revolutionize the way we address the ghost problem in marine seismic acquisition, permitting us to tow different devices at different depth below the water surface, according to their specific bandwidth. In fact, to reduce the destructive effect of the ghost, each source type can be placed at the
optimum depth below the water surface, i. e. at zs = c /4, one
quarter of its central frequency wavelength c (ghost matching). As a result, the ghost wave field will enhance the signal instead of compromising it. The reader is referred to Caporal and Blacqui`ere (2015) for an illustrative explanation of
2016 SEG
SEG International Exposition and 86th Annual Meeting
1000
2000
x(m)
(b)
THEORETICAL FRAMEWORK
In this section, wave field extrapolation based modeling will
be briefly discussed by means of the W RW+ model proposed by Berkhout (1982) and extended to the description of
Page 240
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
M
X
(3)
Delphi
#
W (zd , zm )R (zm , zm ) W+ (zm , zs ) S (zs )
(1)
m=1
S (zs ) is the source matrix. It contains amplitude and phase information about the source wave field at the frequency under
consideration. Key information about the spectral properties
of the different DSA sources is therefore enclosed here. Each
column represents one source (array). Each row corresponds to
a different spatial coordinate. W+ (zm , zs ) and W (zd , zm ) are
the forward, down- and upgoing wave field propagation matrix, respectively. Each column contains a discretized Greens
function describing the wave propagation from one lateral location in space at depth level zs (zm ) to all grid points at depth
level zm (zd ). R (zm , zm ) is the reflectivity matrix, describing
the (angle - dependent) scattering occurring at depth level zm .
In other words, it specifies how the incident wave field is converted into the reflected wave field. D (zd ) is the detector matrix. It contains amplitude and phase information about the
detectors at the frequency under consideration. Each row represents one detector (array). Each column corresponds to a
different spatial coordinate.
Within this framework, it is possible to introduce the concept
of blending by defining the so called blending matrix (zs ).
All information about the linear combination of the different
sources of the array to be employed during the blended experiments is encoded here. The blended source matrix S0 (zs ) can
be formulated by
S0 (zs ) = S (zs ) (zs )
Residual
Imaging
incl. coda
and transmission
Simulated data
at every grid
point
Gradient
Velocity
FWMod
(non-linear)
Reflectivity
Source-side
wavefield
3
(2)
2016 SEG
SEG International Exposition and 86th Annual Meeting
Observed data
Page 241
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
SYSTEM DECENTRALIZATION
System decentralization represents a fundamental change in
the architecture of a very wide range of scientific studies. It
constitutes both an opportunity and a challenge in information technology and communications networks as well as in
the business arena (Bacon, 1990). With this current research
we intend to emphasize and investigate the fundamental improvements that decentralization would bring to seismic data
acquisition.
With DSAs, the use of relatively simple autonomous devices
becomes a practical proposition in seismic surveys. In marine
environment we can consider utilizing at the same time several
source boats with airgun arrays of different sizes or marine vibrators producing sweeps with different frequency ranges. On
land a combination of autonomous Vibroseis trucks of varied
dimensions is suggested.
Additionally, a higher level of decentralization could affect
the way receivers are deployed. Specifically, it is proposed
to move from the considerably complex traditional acquisition
systems, such as single seismic broadband source boats dragging a set of numerous streamer cables, to a swarm of many
simple source-detector subsystems sailing together (see Figure 3). Each subsystem consists of a DSA robot dragging a
single streamer cable and towing a single narrowband seismic
source at its optimum depth.
Figure 3: Comparison between a traditional centralized acquisition system (left) and a decentralized acquisition system
(right).
2007). Nevertheless, to our knowledge, no extensive and conclusive research has been conducted and published concerning
the applications of this concept to exploration geophysics.
EXAMPLE
In this section, we will demonstrate the feasibility of the 3D
decentralized DSA acquisition method with numerical examples of forward modeling and migration. To illustrate the concept, four different source unit types are used: ultralow- (2 - 6
Hz), low- (5 - 15 Hz), mid- (10 - 30 Hz) and high-frequency
sources (20 - 60 Hz). Note that each source spans a frequency
bandwidth corresponding to the same number of octaves. In
such situation, given bandwidths are partially overlapping.
The velocity and density models used as reference are shown
in Figures 4 and 5, respectively. The marine subsurface profile
is featured by two different reflectors: the first (ocean bottom)
at a depth of 500 meters, while the second is defined by the
function f (x, y) = zl acos(2x/X)sin(2y/Y ), where zl is
the central depth level (750 meters), a is the variation amplitude (25 meters), x and y the lateral variables and X = Y = 2000
meters the lateral sizes of the model. Two different datasets
corresponding respectively to a conventional broadband and a
decentralized DSA acquisition scheme have been simulated.
In both cases each source involved has been placed at the optimum depth below the water surface, i.e. at zs = c /4. No
deghosting has been performed on the source side.
Although some sort of central coordination can still set restrictions for the system components (such as the target area
coordinates and the source density requirements), each subsystem is expected to sail autonomously and to fulfill specific
individual objectives. In particular, every unit must be able to
take crucial decision on the spot and modify its own behavior
(e.g. the sailing speed and direction) to promptly adapt to environmental changes and accomplish a given task (e.g. obstacle
avoidance, pattern formation, subsystem unit failure compensation).
In the last few decades, promising advances in swarm robotics
have started making it feasible to deploy large numbers of
inexpensive robots for tasks such as surveillance and search
(Iocchi et al., 2001; Bahceci et al., 2003; Bayndr and Sahin,
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 242
For the first simulation, the data have been modeled considering a regular source spacing of 10 meters along the inline direction and 50 meters along the crossline direction (Figure 6).
Sources have been chosen to be broadband units with a bandwidth ranging from 5 to 60 Hz. For the second simulation,
a number of 10 DSA source boats were left free to sail with
the only constraints of not colliding on each others and to stay
within the target area. No offline path computation is involved.
The interval between consecutive shots was kept the same as
the inline spacing of the first simulation for the high-frequency
sources and decreased linearly with the maximum emitted frequency for the other sources. As we can see from Figure 7,
their paths are random-like and clearly non-symmetric. On the
receivers side, a single ocean bottom node (OBN) located in
the centre of the area is chosen.
(a)
2000
y(m)
1500
1000
500
(b)
500
1000
1500
2000
x(m)
2000
1500
y(m)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
1000
500
500
1000
1500
2000
x(m)
2016 SEG
SEG International Exposition and 86th Annual Meeting
CONCLUSIONS
The DSA concept offers a wide range of practical advantages,
while no physical constraint restrains us from employing diverse sources with different spectral properties during seismic
surveys. Narrowband frequency sources can be technically
simpler to produce and more effective from a energy transmission point of view. Their utilization will allow shot densities
to be chosen in a frequency dependent manner and offer the
possibility to tow the devices at depths (in marine) that are
optimum for their specific central frequency giving extra benefits if we look at the source ghost issue. Furthermore, surveys
may be carried out by acquisition systems that are less complex and considerably more flexible than the ones that are used
today. System decentralization becomes a practical proposition for DSA surveys: more information can be gathered with
less complexity by moving from a single, complex centralized
system to a network of several, simple decentralized subsystems. Promising preliminary migration examples of 3D decentralized DSA data encourage us to carry this research forward in the near future and to explore the significant potential
improvements for exploration geophysics.
ACKNOWLEDGEMENTS
The authors would like to thank the sponsors of the Delphi
consortium for the stimulating discussions during the Delphi
meetings and the continuing financial support.
Page 243
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 244
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 245
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Fold
Noise
Amplitude
Dynamic range of weak
signals
I ( dB ) = 20 log
max
max
Atotal
Anoise
20 log min
min
Asignal
Asignal
(2)
at a moment.
Figure 1 shows the weak signal dynamic range analysis
idea: replace the gather with the section to perform the
extraction of weak signals so as to improve the
identification accuracy; pick up the target weak layer and
the rms energy corresponding to the position of a strong
layer, and then calculate the energy ration between weakand strong-layers , the relationship between the gather and
section energy can be considered as only having the fold
factor involved. This relationship can be expressed as
N SW
F N GW
= w
P
NS
F P N GP
(1)
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 246
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 247
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 2: a) The weak- and strong-layer analysis window of the section, b) the RMS amplitude curves for the weak- and strong-layers in the
section, c) the energy ratio curves for the weak- and strong-layers, d) the energy variation curves for the strong- and weak-layer signales in the
gather.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 248
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Bekara, M., and M. van der Baan, 2009, Random and coherent noise attenuation by empirical mode
decomposition: Geophysics, 74, no. 5, V89V98, http://dx.doi.org/10.1190/1.3157244.
Bose, S., H.-P. Valero, Q. Liu, R. G. Shenoy, and A. Ounadjela, 2009, An automatic procedure to detect
microseismic events embedded in high noise: 79th Annual International Meeting, SEG, Expanded
Abstracts, 15371541.
Cooper, N. M., 2000, Dynamic range and instrumentation Myths, facts and current status: 70th Annual
International Meeting, SEG, Expanded Abstracts, 2932, http://dx.doi.org/10.1190/1.1816049.
Han, W., and J. Zhang, 2011, A theoretical study on the weak reflection seismic signal characteristics and
their identifications: Petroleum Geophysical Prospecting, 46, 232236.
Lao, Y., 2010, A study on the Curvelet transform based the high-density weak seismic signal detection
and de-noising method Qingdao: China University of Petroleum.
Li, Y., B.-J. Yang, X.-P. Zhao, M. Zhang, and H.-B. Lin, 2005, An algorithm of chaotic vibrator to detect
weak events in seismic prospecting records: Chinese Journal of Geophysics, 48, 15021508,
http://dx.doi.org/10.1002/cjg2.800.
Liu, Y., 1996, The time-frequency characteristic and instantaneous dynamic range of seismic Signal:
GPP, 35, 6670.
Liu, Z., Y. Sun, S. Yang, G. Qin, and L. Deng, 2009, Calculation of spectral attribute based on the weak
signal separation method: Presented at the Beijing 2009 International Geophysical Conference
and Exposition, 156, http://dx.doi.org/10.1190/1.3603681.
Ma, N., L. Chen, X. Wang, 2009, Application of wavelet transform in weak signal detection: Journal of
Harbin Institute of Technology, 41, 257258.
Ni, Y., 2011, Progress of vibroseis seismic acquisition technology: Petroleum Geophysical Prospecting,
46, 349356.
Ras, P., M. Daly, and G. Baeten, 1999, Harmonic distortion inslip sweep records: 69th Annual
International Meeting, SEG, Expanded Abstracts, 609612.
Sandip, B., P. Henri, and D. Alain, 2009, Semblance criterion modification to incorporate signal energy
threshold: 79th Annual International Meeting, SEG, Expanded Abstracts, 376380,
http://dx.doi.org/10.1190/1.3255652.
Simon, H., and X. Li, 1999, Detection of signals in chaos: Proceedings of the IEEE, 85, 95122,
http://dx.doi.org/10.1109/5.362751.
Zeng, G., and X. Hou, 2009, Implementation of weak signal detection using wavelet transform and the
bispectrum method: Digital Signal Processing, 33, 6062.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 249
Disributed Principal Component Analysis for Data Compression of Sequential Seismic Sensor Arrays
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Bo Liu 1 , Hilal Nuha1 , Mohamed Mohandes1 , Mohamed Deriche1 and Faramarz Fekri2
1: King Fahd University of Petroleum and Minerals, KSA; 2: Georgia Institute of Technology, USA
SUMMARY
This work considers the data compression of sequential seismic sensor arrays. First, the statistics of the seismic traces
collected by all the sensors are modeled by using the mixture
model. Hence, a distributed Principle Component Analysis
(PCA) compression scheme for sequential sensor arrays is designed. The proposed scheme does not require transmitting the
traces, leading to a more efficient computation and compression compared with the conventional local PCA compression.
Furthermore, an efficient communication scheme is developed
for the sequential sensor array for delivering the local statistics to the fusion center. In this communication scheme, the
sensors update and pass a data package consisting of cumulative variables. The size of the data package does not increase
throughout the process, which is more efficient than the direct communication scheme. Finally, the performance of the
proposed scheme is evaluated by using both real and synthetic
seismic data.
INTRODUCTION
The petroleum industry uses seismic reflections as one of the
principal tools to explore and evaluate oil fields. The impetuous increase of the number of sensors (geophones) and the
number of shots in the modern seismic data acquisition during
the last several decades has caused a swift increase in volume
of seismic data stored and transferred. With the emergence
of 3D technology, the data volume is particularly large and it
is usually more than 1012 bytes in 3D surveys (Wood, 1974;
Chen and Larner, 1995). It is therefore desirable to compress
the data, in order to reduce the costs of storage and transmission (Donoho et al., 1999; Zarantonello and Bevc, 2005).
Data compression techniques are usually classified into two
categories, i.e., lossless and lossy (Kass and Li, 2012). No
information loses during the lossless compression procedure,
and the original data can be perfectly reconstructed from the
compressed data. Lossy compression, on the other hand, uses
approximations to represent the original data. Unlike lossless compression, the original data can not be perfectly reconstructed after lossy compression. However, the amount of
data reduction possible using lossy compression is often much
higher than through lossless techniques. Therefore, we focus
here on lossy compression and specifically on PCA compression.
PCA is a standard tool in modern data analysis, which with
minimal effort provides simple way to reduce a complex data
set to a lower dimension. The target of PCA is to identify a set
of the most important coordinates, i.e., Principal Components
(PCs), in terms of the second order moment of the original
data, then to re-express the original data on this set of PCs
2016 SEG
SEG International Exposition and 86th Annual Meeting
(Huang, 1996; Liu et al., 2012; Shlens, 2014). The hope is that
by keeping the data projected on the most important PCs the
dimension is decreased, and the volume of the data shrinks.
In this work, we consider a PCA based seismic data compression for sequential seismic sensor arrays. The assumptions and
constrains of this work are listed as follows. First, the sensors
(geophones) have an ability to store amount of data, to undertake computation and to communicate with other sensors.
Secondly, the sensors are arranged in a sequential order and
only the last sensor has an access to the fusion center. Our target is to obtain a set of global PCs in the fusion center to be
used among all sensors for compression instead of local PCs,
subjects to a requirement that during the procedure of obtaining the global PCs, only statistics of the traces are transmitted
among the sensors instead of all traces.
MIXTURE MODELING
In this section, the concept of the mixture model is introduced,
and the seismic traces collected by multiple sensors are modeled as the realizations of a mixture model. Then a distributed
PCA scheme is proposed.
First, for an n-dimensional stochastic variable
x
[x1 , x2 , , xn ]T ,
(1)
K
X
i fi (x),
(2)
i=1
{i , i }.
(3)
One can easily obtain the first two moments of mixture model
(2) as
K
X
i i
(4)
i=1
K
X
i=1
i i +
M
X
i=1
i (i )(i )T .
(5)
Page 250
Disributed Principal Component Analysis for Data Compression of Sequential Seismic Sensor Arrays
One trace is a time series with a length, say n, recorded by a
sensor from one shot. Let it be defined as
(i)
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Xj
(i)
(i)
(6)
from the i-th sensor and the j-th shot. Trace X j is regarded as
a realization of a n-dimensional stochastic variable x, which is
assumed to follow an unknown local PDF fi (x). Once the i-th
(i) i
sensor collects a number of traces, say Ni traces {X j }Nj=1
, the
first two moments of PDF fi (x) can be estimated as
i
1
Ni
Ni
X
(i)
Xj
(7)
j=1
Ni
1 X (i)
(i)
(X j i )(X j i )T .
Ni 1
(8)
i
the original data. Finally, the compressed traces {Y j }Nj=1
, for
i = 1, 2, , K are transferred to the fusion center.
(i)
The volume of all traces is nN. By implementing the distributed PCA compression, it is required to store k global PCs,
k coefficients and K means. The compression ratio is
kn + kN + Kn
.
(13)
nN
By implementing the local PCA compression, it is required to
store k local PCs for each of the sensors, k coefficients and K
means. The compression ratio is
Rd
Kkn + kN + Kn
,
(14)
nN
which is larger than Rd by using same number of PCs. As for
the computation cost, the distributed PCA only needs matrix
decomposition once at the fusion center, while the local PCA
needs to do it K times.
Rl
j=1
Suppose that there are K sensors considered in the sensor array, applying the same modeling formulation mentioned above
to other sensors finally yields K local PDFs, i.e., fi (x), for
i = 1, 2, , K. Since the sensors are generally with different
geographical locations, the local PDFs fi (x) are with nuances
from each other. To capture the global statistic of the seismic
traces collected by the sensor array, we construct the global
PDF of the traces, x, as a convex sum of the local PDFs fi (x)
as in (2) with weights
i
PK
Ni
,
N
(9)
PPT ,
(10)
[p1 , p2 , , pk ], k n,
(11)
Yj
(i)
Pk X j , j = 1, 2, , Ni .
(12)
2016 SEG
SEG International Exposition and 86th Annual Meeting
K
1X
Ni i
N
(15)
i=1
K
K
1X
1X
Ni i +
Ni (i )(i )T
N
N
i=1
i=1
Page 251
Disributed Principal Component Analysis for Data Compression of Sequential Seismic Sensor Arrays
=
K
1X
Ni i + i iT i T iT + T
N
i=1
1
N
i=1
K
K
1X
1X
Ni i iT
Ni i T
N
N
i=1
K
1X
Ni iT + T
N
i=1
K
X
i=1
K
1X
Ni i +
Ni i iT
N
K
X
1
N
i=1
!T
Ni i
i=1
EXPERIMENTAL RESULTS
i=1
K
1X
Ni i
N
i=1
+ T .
(16)
(a) Sensor Si
Receiving from sensor Si1
Pi1
Pi1
j=1 N j ,
j=1 N j j ,
j=1 N j j ,
Pi1
T
j=1 N j j j }
{
Pi1
Pi
Pi1
j=1 N j =
j=1 N j + Ni
Pi
Pi1
j=1 N j j =
j=1 N j j + Ni i
Pi
j=1 N j j =
Pi
T
j=1 N j j j
Pi1
j=1 N j j + Ni i
T
T
=
j=1 N j j j + Ni i i
Pi1
(c) Sending the updated cumulations to the next sensor (or the Gateway, if i = K):
Pi
Pi
j=1 N j ,
Pi
j=1 N j j ,
T
j=1 N j j j }
Pi
j=1 N j j ,
2016 SEG
SEG International Exposition and 86th Annual Meeting
Real Data
We extract 18 traces for each sensor from East Texas USA data
base (Mousa and Al-Shuhail, 2011), in which the shots are
generated in 80 100 feet depth on 18 different locations. We
consider 10 sensors that are arranged in a sequential array with
an interval of 220 feet. The length of each trace is 1501 time
samples, which is approximately 3 seconds. Since the magnitude of the shot wave is attenuated as it propagates to the sensors over distance and the sensors are away from each other,
different sensors receive traces with different magnitude. See
Figure 2 as an example, for a sensor tha is far away from the
shots, its traces are with too small magnitude to be visually
observed. To mitigate this nuisance, we normalize the traces
with small wave magnitude, for example, see Figure 3.
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
trace no
Ni i +
200
400
600
800
time
1000
1200
1400
1600
trace no
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
K
1X
200
400
600
800
time
1000
1200
1400
1600
Pk
i
Pi=1
,
n
i=1 i
(17)
Page 252
To get a given NCE, the distributed PCA needs more PCs than
the local PCA, however, the distributed PCA compression still
has some overall benefits. First, it requires one matrix decomposition in the fusion center, instead of 10 matrix decompositions as in the local PCA compression. Secondly, it has lower
compression ratio than the local PCA. For example, to preserve approximately 99% energy, we need 12 PCs for the local
PCA or 34 PCs for the global PCA. According to (13)-(14),
the compression ratio of the distributed PCA is Rd = 0.26 and
the compression ratio of the local PCA is Rl = 0.73.
110
100
90
80
70
60
50
40
30
20
Global Dist PC
Local PC sensor i
1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39 41 43 45 47 49
number of PCs
Figure 4: Then NCEs of the local PCA and the global PCA by
using 10 sensors (real data).
90
80
70
NCE %
NCE %
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Disributed Principal Component Analysis for Data Compression of Sequential Seismic Sensor Arrays
60
50
40
30
Global Dist PC
Local PC sensor i
20
10
10
13
16
19
22 25 28 31
number of PCs
34
37
40
43
46
49
Figure 5: Then NCEs of the local PCA and the global PCA by
using 10 sensors (synthetic data).
CONCLUSIONS
In this work, we present a distributed PCA compression scheme
for sequential sensor arrays. The seismic traces collected by
multiple sensors are modeled as realizations of a mixture PDF,
whose mean and covariance can be estimated at a fusion center in a distributed fashion without accessing the traces. A set
of global PCs is determined by decomposing the global covariance and then used by all the sensors to project the traces.
The advantage of the proposed scheme compared with the local PCA scheme is twofold. First, it has a lower computation cost in terms of matrix decomposition, and secondly, it
achieves a lower compression ratio. Furthermore, an efficient
communication method is proposed to alleviate the communication burden of the sequential sensor array, in which the data
package transmitted among the sensors does not accumulate
in size. Finally, a number of experiments are implemented to
show the efficiency of the proposed scheme.
ACKNOWLEDGMENTS
The work presented in this paper is supported by Center for
Energy and Geo processing (CeGP) at King Fahd University
of Petroleum and Minerals (KFUPM) and Georgia Institute of
Technology.
Synthetic Data
In the rest of this section, we test the proposed scheme by using
synthetic seismic traces. First, Ricker wavelet is employed as
shot source to simulate 18 shots, which are recorded by 10
sensors. Each trace is with 1501 samples and a synthetic noise
with 40 dB SNR is added.
The NCEs are plotted in Figure 5. One can notice that all
the phenomena observed in the experiment by using the real
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 253
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 254
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
caused by
bandwidth.
different
offsets
and
source
frequency
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 255
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 256
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 257
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 258
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
El-Emam, A., and S. Khalil, 2012, Maximizing the value of single-sensor measurements, Kuwait
experience: SEG, http://dx.doi.org/10.1190/segam2012-1000.1.
Racked, G., A. El-Emam, K. S. Al-Deen, J. Al-Jenai, and B. Al-Ajmi, 2010, Designing 3D seismic
surveys to image and characterize multi-objective targets, study from Kuwait: SEG,
http://dx.doi.org/10.1190/1.3513819.
Fletcher, R. P., S. Archer, D. Nichols, and W. Mao, 2012, Inversion after depth imaging: SEG,
http://dx.doi.org/10.1190/segam2012-0427.1.
Moldoveanu, N., R. Fletcher, A. Lichnewsky, D. Coles, and H. Djikpesse, 2013, New aspects in seismic
survey design: SEG, http://dx.doi.org/10.1190/segam2013-0668.1.
Vigh, D., K. Jiao, W. Huang, N. Moldoveanu, and J. Kapoor, 2013, Long-offset-aided Full-waveform
Inversion. 75th Annual International Conference and Exhibition, EAGE, Extended Abstracts.
http://dx.doi.org/10.3997/2214-4609.20130825.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 259
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Introduction
The conventional 2D single-line data from the area shows
that the target layer was clearer in the plain region, but
subsurface structures can hardly be seen once entering
mountains and loess. The quite low S/N ratio results from
the extreme terrains in Central Asia, including mountains,
loess, rivers and hills. The rugged surfaces and great
elevation variations of mountainous area as well as the
complex underground structures lead to the imaging
difficulties of the data from mountainous area
comprehensively. The distribution of inhomogeneous nearsurface bodies results in large numbers of secondary
disturbances and the loess of hundreds of meters thick has a
significant attenuation effect to the shooting energy,
leading to the quite low energy at middle-deep target layers.
To solve the above problems, the crooked wide-line
acquisition scheme was widely employed in the early stage
survey. The wide-line survey has also been broadly used in
recent years, for example in the complex mountains in
Western China (Zhang et al, 2012) and in Pakistan (Shi et
al, 2008). Some scholar made a study on the spacing of
receiver lines (Lv in 2013) and concluded that the
reasonable range is from 40-80 m. In the crooked wide-line
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 260
1.6
1.8
2
t/s
2.2
wide line
fitted
2.4
2.6
2.8
3
-1
-0.5
0.5
offset/m
1
x 10
60
40
20
amp
0
-20
-40
-60
0
50
150
100
200
t/ms
receiver point
shot point
200
receiver point
shot point
200
crossline/m
crossline/m
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
-200
-400
-600 -400 -200
200
400
0
-200
-400
inline/m
Figure 1: Wide line geometry
-500
500
inline/m
Figure 4: Geometry with shot- point offsets
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 261
1200
600
60
30
150
no offset 200
offset
180
210
crossline/m
1000
400
800
600
400
200
330
5500
5000
4500
inline/m
300
240
270
300
Figure 5: The poststack waveform amplitudes of the side
interference in different directions
2016 SEG
SEG International Exposition and 86th Annual Meeting
200
100
amp
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
90
120
-100
-200
0
100
50
150
t/ms
Figure 7: Poststack waveform of the effective signal at the
inflection
Results
Figure 8 shows a single shot record with AGC of a line and
its pre-stack time migration profile. The convex portion in
the middle is a high mountain with an elevation difference
over 1,000 m. It is loess on the right side of the mountain.
The green arrow points to a inflexion of the wide line,
where the bending angle is 7 approximately. The target
layers as deep as 5-6s, the faults and the geological
structures just below the inflection point are clearly visible
in the figure.
Page 262
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
Based on the above analysis, the wide line geometry used
in an area of Central Asia has a significant effect in
suppressing side interferences and thereby the S/N ratio is
improved obviously. The narrow azimuth of the wide line
geometry enables the side interferences to be canceled out
mutually after NMO correction, while the effective signals
from the target layer right below the seismic line stack inphase hardly with any influence, thus the side influences
are suppressed and the S/N ratio is enhanced greatly. The
wide line geometry makes it possible to acquire higher S/N
ratio data without adding much cost. The crooked line
scheme has avoided the extreme terrains in the operation,
improved the efficiency of field operation and has
preserved the effective signals to a great extent.
Consequently, the crooked wide-line geometry is relatively
cost-effective acquisition scheme in the complex area of
Central Asia.
inflexion
mountain
loess
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 263
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Gonghe, L., 2013, The relationship between wide line geometry parameters and S/N ratio: Geophysical
Prospecting for Petroleum, 52, 495501.
Shi, J., G. He, Z. Zhang, 2008, Complex surface seismic acquisition technique in Pakistan: Oil
Geophysical Prospecting, 43, 1114.
Zhang, C., D. Qiao, S. Li, M. Zhang, G. Gan, Y. Xu, and Y. Tang, 2012, The wide line seismic
exploration technique of complex mountain in west Qaidam Basin: Oil Geophysical Prospecting,
7, 189193.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 264
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
The imaging resolution analysis for complex models applied in seismic survey design
Baoqing He12*, Donglei Tang1, Yongqing He1, Xiaobi Xie2, Bo Long1( 1BGP CNPC, 2Modeling and Imaging
Laboratory, Earth & Planetary Science, University of California, Santa Cruz, CA)
Summary
In this paper, based on a complex seismic-geolocial model,
the Time-Spatial Domain Acoustic Equation FiniteDifference Forward Modeling with given seismic
geometries and wavelets are used to get the Point
Spreading Function (PSF) at different analysis positions, on
the basis of which, we further calculate the illumination
spectrum and the envelope of a point scatter. In order to
provide an useful tool for seismic survey design and
optimization, in the numerical experiment, with the
maximum offset unchanged, we use variable source
interval or receiver interval to analyze the influences of
different geometry parameters to the PSDM resolution.
Introduction
Nowadays more and more seismic survey design
technologies have been used before seismic acquisition for
giving a balance between the cost and the quality of seismic
data. We all hope that we can get good enough seismic data
with minimum cost. The wave equation modeling based
and the wave equation illumination analysis based seismic
survey designs have been used nearly throughout each large
seismic survey design project. The seismic imaging
resolution has been the highest goal of the seismologist, but
the methods are mostly based on the ray tracing theory
(Gelius et al., 2002; Lecomte, 2008;). Such high frequency
asymptotic approximation based ray tracing theory cannot
deal with the model for complex structures, but it has its
unique characterisitcs, i.e. it can get the ray-path from the
source through the subsurface interface to the receiver,
such that it can easily get the local open angle of incident
wave and reflect wave at the reflection position. This open
angle is necessary for either imaging resolution or PSDM
amplitude of the target. In other words, the angle domain
illumination is closely related with the imaging resolution.
Xie et al., (2005, 2006); Wu et al., (2006) expanded the
above methods to the one way wave equation, which
required to calculate a large number of frequencies due to
the application of the wave equation operators in the F-K
domain. In this way, it is very difficult to meet the require
for a broadband seismic wavelet. In addition, the angle
limit of one way propagator is also a flaw. Although the
local incident angles at different scatter points can be
calcualted, the calcualtion amount is enormous, especially
when the broadband wavelet is used as the input source.
There are several ways to get seismic illumination based on
full-wave equation (Xie and Yang, 2008; Yang et al., 2008;
Cao and Wu, 2009; Yan, et al., 2014), but the computation
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 265
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Numerical example
2
k x
2
(3)
Rz =
k z
where Rx and Rz are the horizontal and the vertical size of
imaging resolutions respectively at the corresponding
interested point, i.e. the range of the PSF reflects the size of
resolution, and its amplitude reflects the intensity of
illumination, which is closely relative with the migration
imaging resolution.
Rx=
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 266
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 267
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 4: PSF of different Ricker wavelets. Left: 16 Hz; middle: 24 Hz; and right: 32 Hz.
Figure 5: Zoom in of the PSFs in Figure 4. Left: 16 Hz; middle: 24 Hz; and right: 32 Hz.
Figure 6: The wavenumber spectrums achieved from the PSFs in Figure 4 by Fourier transform. Left: 16 Hz; middle: 24 Hz; and right: 32 Hz.
Figure 7: The zoom in of the wavenumber spectrums in Figure 5. Left: 16 Hz; middle: 24 Hz; and right: 32 Hz.
(a)
(b)
(c)
(d)
(e)
Figure 8: The zoomed in PSFs of different receiver intervals: (a) 5m; (b) 10m; (c) 20m; (d) 40m; and (e) 50m.
(a)
(b)
(c)
(d)
(e)
Figure 9: The zoomed in PSFs of different source intervals: (a) 20m; (b) 40m; (c) 50m; (d) 100m; and (e) 200m.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 268
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Cao, J., and R. S. Wu, 2009, Full-wave directional illumination analysis in the frequency domain:
Geophysics, 74, no. 4, S85S93, http://dx.doi.org/10.1190/1.3131383.
Cao, J., 2013. Resolution/illumination analysis and imaging compensation in 3D dip-azimuth domain.
83rd Annual International Meeting, SEG, Expanded Abstracts, 39313936,
http://dx.doi.org/10.1190/segam2013-0380.1.
Chen, B., and X. B. Xie, 2015. An efficient method for broadband seismic illumination and resolution
analyses. 85th Annual International Meeting, SEG, Expanded Abstracts, 42274231,
http://dx.doi.org/10.1190/segam2015-5926976.1.
Gelius, L. J., I. Lecomte, and H. Tabti, 2002, Analysis of the resolution function in seismic prestack depth
imaging: Geophysical Prospecting, 50, 505515, http://dx.doi.org/10.1046/j.13652478.2002.00331.x.
Lecomte, I., 2008, Resolution and illumination analyses in PSDM: A ray-based approach: The Leading
Edge, 27, 650663, http://dx.doi.org/10.1190/1.2919584.
Mao, J., and R. S. Wu, 2011, Fast image decomposition in dip angle domain and its application for
illumination compensation: 81st Annual International Meeting, SEG, Expanded Abstracts, 3201
3204, http://dx.doi.org/10.1190/1.3627860.
Wu, R. S., X. B. Xie, M. Fehler, Xie, X.B., and Huang, L.J., 2006. Resolution analysis of seismic
imaging. 68th Annual International Conference and Exhibition, EAGE, Extended Abstracts,
G048.
Xie, X. B., R. S. Wu, M. Fehler, and L. Huang, 2005. Seismic resolution and illumination. A waveequation based analysis. 75th Annual International Meeting, SEG, Expanded Abstracts, 1862
1865, http://dx.doi.org/10.1190/1.2148066.
Xie, X. B., S. W. Jin, and R. S. Wu, 2006, Wave-equation based seismic illumination analysis:
Geophysics, 71, no. 5, S169S177, http://dx.doi.org/10.1190/1.2227619.
Yan, R. H., H. Guan, X.-B. Xie, and R.-S. Wu, 2014, Acquisition aperture correction in the angle domain
toward true-reflection reverse time migration: Geophysics, 79, no. 6, S241S250,
http://dx.doi.org/10.1190/geo2013-0324.1.
Yang, H., X. B. Xie, S. Jin, and M. Luo, 2008, Target oriented full-wave equation based illumination
analysis, 78th Annual International Meeting, SEG, Expanded Abstracts, 22162220,
http://dx.doi.org/10.1190/1.3059326.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 269
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
120
100
60
sensitivity
7
V/g
0.28(V/m/
s)
Alias (Hz)
400
180
distortion
0.18
0.2%
0.002
Page 270
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure1 The record comparison among different geophones (MEMS, LPS, 36 20DX array from left to right)
Figure2 The stack section comparison among different geophones (MEMS, LPS, 36 20DX array from left to right)
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 271
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 272
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure5 The migration comparison between previous conventional 20DX geophone and single-point LPS data (left: Section of
20DX geophone data reprocessed in 2010; right: Section of single-point LPS data acquired and processed in 2016)
Conclusions
Compared with conventional geophone array, the singlepoint LPS keeps better event consistency, increased
resolution and accurate imaging substantially, while its
SNR defect could be compensated by high density
geometry. The field production proves single-point LPS
technique to be an effective way for high resolution
prospecting on land.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 273
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Lv, G. H., 2009, Analysis on principles and performance of seismic geophone and relevant issues:
Geophysical Prospecting for Petroleum, 48, 531543.
Yuan, X., 1986, Sensor technology Handbook: National Defense Industry Press.
Cyril M.Harris,AIIan G.PiersoI., 2008, Shock and vibration handbook: China Petrochemical Press.
Zhao, D., W. Han, and D. Wei, 2012, Practice and exploration of high precision prospecting: Geological publishing
house.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 274
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
It was found that the direct arrival has the great effect on
the signal-to-blending noise ratio (S/N) and the
convergence rate (see Figure 2). Figure 2 shows the
contrast of residual between the data containing direct
arrival and the data which does not, and both kinds of data
was the result processed through the same number of
iterations.
At the same number of iterations, the residual energy of the
data containing direct arrival (Figure 2a) is larger than that
of data which does not contain the direct arrival (Figure 2b).
Therefore, the convergence rate of the data containing
direct arrival will be lower than that of the data which does
not.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 275
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Simultaneous
sources data
Next blending
Pseudodeblending
Subtracted the
direct arrival
Figure 2: The residual between the data containing direct arrival (a)
and the data which was not (b) . And these pictures are conducted
through the same number of iterations.
Build the
model of direct
arrival
Coherent
filtering
Method
Based on the conventional separation methods of
simultaneous source data, where the separation process was
addressed as de-noising one, we proposed a new separation
method. The process flows of conventional method were
illustrated in Figure 3. And the process flows of our method
are illustrated in Figure 4.
During the process flows of our method, the model of
direct arrival was built from the near-surface velocity
(onshore exploration) or sea-water velocity (offshore
exploration) and then subtracted adaptively from the
simultaneous source dataset.
And in the Figure 4, the two blue boxes show the additional
processing flow which was used to attenuate the direct
arrival.
Simultaneous
sources data
Next blending
Subtracted by
first Pseudodeblending
results
Added with
last separation
result
Separation
results
Example
To evaluate the effectiveness of our method, two tests on
synthetically blended dataset were performed. The SDR
(shot density ration) was 2 in these two dataset and the
models were presented in Figure 5. There are 100 receivers
with 10m receiver separation.
Pseudodeblending
Coherent
filtering
Subtracted by
first Pseudodeblending
results
Added with
last separation
result
Separation
results
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 276
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
(1)
Where, the subscript rms stands for root mean square; the
Q is the unblended data and QN is the separation results.
method
conventional filtering
method
our method
Acknowledgments
Model
Simple
model
Hess
model
16.0
5.3
39.5
10.5
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 277
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Figure 6: The comparison of deblending results and the residual map (especially indicated by the green oval) between
conventional iteration de-noising method and our method.Where, Figure A shows the blended data. Figure B shows the
deblended data separated through the conventional iteration filtering deblending method. Figure C shows the deblended results
separated by our method. Figure D shows the unblended shots. Figures E show the residuals of the data in Figure B and D. Figure
F show the residuals of the data in Figure C and D.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 278
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Abma, R., D. Howe, M. Foster, I. Ahmed, M. Tanis, Q. Zhang, A. Arogunmati, and G. Alexander, 2015,
Independent simultaneous source acquisition and processing: Geophysics, 80, no. 6, WD37
WD44, http://dx.doi.org/10.1190/geo2015-0078.1.
Beasley, C. J., 2008, A new look at marine simultaneous sources: The Leading Edge, 27, 914917,
http://dx.doi.org/10.1190/1.2954033.
Berkhout, A. J., 2008, Changing the mindset in seismic data acquisition: The Leading Edge, 27, 924938,
http://dx.doi.org/10.1190/1.2954035.
Berkhout, A. J., 2012, Blended acquisition with dispersed source arrays: Geophysics, 77, no. 4, A19
A23, http://dx.doi.org/10.1190/geo2011-0480.1.
Berkhout, A. J., G. Blacquire, and D. J. Verschuur, 2009, The concept of double blending: Combining
incoherent shooting with incoherent sensing: Geophysics, 74, no. 4, A59A62,
http://dx.doi.org/10.1190/1.3141895.
Berkhout, A. J., G. Blacquire, and D. J. Verschuur, 2012, Multiscattering illumination in blended
acquisition: Geophysics, 77, no. 2, P23P31, http://dx.doi.org/10.1190/geo2011-0121.1.
Hampson, G., J. Stefani, and F. Herkenhoff, 2008, Acquisition using simultaneous sources: 78th Annual
International Meeting, SEG, Expanded Abstracts, 28162820.
Ibrahim, A., and M. D. Sacchi, 2014, Simultaneous source separation using a robust Radon transform:
Geophysics, 79, no. 1, V1V11, http://dx.doi.org/10.1190/geo2013-0168.1.
Long, A., E. von Abendorff, M. Purves, and J. Norris, 2013, Simultaneous long offset (SLO) towed
streamer seismic acquisition: 75th Annual International Conference and Exhibition, EAGE,
Extended Abstracts, Tu 04 02.
Mahdad, A., P. Doulgeris, and G. Blacquire, 2011, Separation of blended data by iterative estimation
and subtraction of blending interference noise: Geophysics, 76, no. 3, Q9Q17,
http://dx.doi.org/10.1190/1.3556597.
Moore, I., D. Monk, L. Hansen, and C. J. Beasley, 2012, Simultaneous sources: The inaugural full-field,
marine seismic case history from Australia: 74th Annual International Conference and Exhibition,
EAGE, Extended Abstracts, paper no. 160.
Poole, G., K. Stevens, M. Maraschini, T. Mensch, and R. Siliqi, 2014, Blended dual-source acquisition
and processing of broadband data: 76th Annual International Conference and Exhibition, EAGE,
Extended Abstracts, Th ELI2 05.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 279
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Milos Cvetkovic*, Laurie Geiger, Tim Seher, Richard Clarke, Cesar Arias, Oscar Ramirez and Steve Ward,
Spectrum Geo Inc.
Summary
Modern 2D seismic programs in frontier areas need to be
effective and flexible in order to provide high quality data
for block definition, petroleum system identification and
prospect interpretation. This can be achieved with an
integrated workflow which starts from basin evaluation and
proceeds through modeling and finally data acquisition.
This process must be aligned with environmental and
economic constraints while still meeting the requirements
of data processing and effective seismic imaging. We
present a case study from the southern Gulf of Mexico,
offshore the Yucatan Peninsula in an unexplored basin.
Introduction
While the Bay of Campeche has been a location for
significant exploration activity since the discovery of the
Cantarell Field in 1976, the basin offshore the Yucatan
Peninsula has yet to see any hydrocarbon exploration. The
presence of a working petroleum system adjacent to the
Yucatan Peninsula would indicate there should be many
opportunities for potential hydrocarbon discoveries north
and northwest of the Yucatan. A practical and effective
seismic survey design for the deepwater Yucatan Basin
must take into account imaging objectives ranging from
basement structures to shallow sediments. These include a
significant Tertiary and Cretaceous sedimentary section as
well as a Jurassic salt layer overlying rift basins which
contain potential source rocks.
Modeling and acquisition
There are only a few seismic datasets available in this area,
however there are a sufficient number of detailed crosssections in the public domain to enable us to construct a
reasonably detailed subsurface model. Figure 1a shows a
Pre-Stack Time Migration (PSTM) stack with regional
interpretation extending from offshore the Yucatan
Peninsula into abyssal the plane.
The goal of this modeling exercise is to optimize the
proposed 2D seismic survey in order to achieve good
imaging of multiple identified target sections. These
include 1) the shallow Tertiary post-salt section, 2) the presalt section, 3) the steeply dipping escarpment and 4) the
deep basement and Moho structures. Our objective is to
capture all possible target types with a combination of
structural and stratigraphic synthetic models. We use an
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 280
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
2016 SEG
SEG International Exposition and 86th Annual Meeting
Figure 4: Acquisition parameter refinement via modeling. Top RTM stack with 15s record length and 12km of offset, with single
synthetic shot gather; bottom - RTM stack with continuously
recorded 15s record length, 5s overlap and 12km of offset, with
single shot gather. The yellow arrows indicate areas with blending
noise. The green arrows indicate areas that are improved versus the
image using 12s record length.
Page 281
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
We stress that the first processing step does not require shot
time randomization. The moveout differences between the
primary arrivals and the blended secondary arrivals are
sufficiently large for source separation in simple geological
settings. The second processing step profits from shot delay
variations due to variations in vessel speed caused by ocean
currents. If these variations are small, random noise
attenuation only allows for limited uplift. Last, time-variant
low-pass filtering is rarely used in de-blending, but allows
further attenuation of blended energy at late times.
For deghosting we use the method proposed by Yilmaz and
Orhan (2014) where the problem of obtaining the unknown
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 282
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
Conclusions
We have presented a robust integrated workflow for
developing and delivering a frontier area acquisition
program that offers an uncompromised dataset for modern
day exploration needs.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 283
Downloaded 12/16/16 to 79.62.242.250. Redistribution subject to SEG license or copyright; see Terms of Use at http://library.seg.org/
EDITED REFERENCES
Note: This reference list is a copyedited version of the reference list submitted by the author. Reference lists for the 2016
SEG Technical Program Expanded Abstracts have been copyedited so that references provided with the online
metadata for each paper will achieve a high degree of linking to cited sources that appear on the Web.
REFERENCES
Benson, C. E., P. Esestime, and M. Cvetkovic, 2015, Using geological constraints to improve velocity
model building for depth imaging in Frontier areas Case studies: 77th Annual International
Conference and Exhibition, EAGE, Extended Abstracts.
Cogan, M., and S. Boochoon, 2013, Sub-salt seismic interpretation using RTM vector image partitions:
83rd Annual International Conference and Exhibition, EAGE, Extended Abstracts, 13381342.
Cvetkovic, M., C. Caldern-Macas, P. Farmer, and G. Watts, 2014, Efficient numerical modelling and
imaging practices for aiding marine acquisition design and interpretation, First Break, 32, 99
105.
Rodriguez, A. B., 2011, Regional structure, stratigraphy, and hydrocarbon potential of the Mexican sector
of the Gulf of Mexico: M.S. thesis, The University of Texas.
Yilmaz, O., and E. Baysal, 2015, An effective ghost removal method for marine broadband seismic data
processing: 77th Annual International Meeting, SEG, Expanded Abstracts.
2016 SEG
SEG International Exposition and 86th Annual Meeting
Page 284