Perturbations in 4D marine seismic
`
`Tim Brice
`WesternGeco Australia
`
`Tim.brice@westerngeco.com
`
`Leif Larsen
`WesternGeco UK
`Iarsenl@gatwick.geco-prakla.slb.com
`
`Steve Morice
`WesternGeco UK
`morices @gatwick.geco-prakla.slb.com
`
`Morten Svendsun
`WestemGeco Norway
`svendsem @oslo.geco-prakla.slb.com
`
`SUMMARY
`
`c-f
`
`new concept
`
`for acquiring calibrated towed streamer
`new acquisition and
`seismic data is introduced through
`processing system The specification of the new system
`has been defined by rigorous analysis of the factors that
`the sensitivity of seismic data in 4D studies
`imaging New sensor
`and
`streamer
`source
`technology
`
`limit
`
`technology
`
`and
`
`new
`
`and
`
`advances
`
`in
`
`positioning
`
`techniques
`
`and
`
`data processing have
`
`addressed
`
`these
`
`limitations
`
`Sensitivity
`
`the most significant
`perturbations to the seismic signal are swell noise and
`
`analysis revealed that
`
`sensor sensitivity variations Conventional
`of hydrophones
`are designed
`to suppress
`however
`new technique for data-adaptive coherent
`
`analog groups
`
`swell noise
`
`accurate
`
`depth
`
`recent
`
`development
`
`compensates
`source array geometry by monitoring
`and
`the position
`azimuth of each subarray using GPS receivers mounted
`on the floats
`
`for variation
`
`in
`
`New calibrated positioning and streamer control systems
`are part of the new acquisition system Active vertical and
`lateral streamer control
`is achieved
`using steerable birds
`and positioning uncertalnty is reduced through an in-built
`fully braced acoustic ranging system
`Calibrated marine seismic data
`are achieved
`
`through
`
`the sensor
`quantifying the source output
`The consequential
`
`positioning uncertainty
`
`in seismic fidelity
`reliable 4D analysis
`
`result
`
`in better
`
`responses
`
`and
`
`improvements
`imaging and more
`
`Keywords Calibrated seismic data perturbations
`4D seismic
`Q-Marine
`recording
`single sensor
`Notional Source Method steerable streamers
`
`noise attenuation delivers even greater noise suppression
`for densely spatially sampled single-sensor data
`Although modem source
`controllers provide
`the signature of an alrgun array may
`alrgun firing control
`to shot This can be due to factors such as
`vary from shot
`in the array geometry
`wave
`variations
`and
`
`changes
`
`air pressure variations
`method
`action
`for
`source array is the
`estimating the far-field signature of
`Notional Source Method proprietary to Schlumberger
`which has been steadily refined since its first disclosure
`
`INTRODUCTION
`
`Reservoir monitoring
`
`using time lapse or 4D seismic has
`for understanding the dynamic
`demonstrated its effectiveness
`behaviour of
`the
`reservoir
`and
`value
`for
`reservoir
`management 4D seismic studies can
`
`its
`
`Map fluid movements pore fluid saturation changes and
`
`pressure changes
`Identify unswept oil and mull drilling opportunities
`
`Identify flow units and flow barriers
`Monitor the performance of enhanced recovery programs
`and constraln the
`the
`spatial properties of
`Quantify
`reservoir model
`
`Most
`
`within
`
`data
`
`have been made for
`time lapse survey interpretations
`reservuirs with highly purous sands where fluid replacement
`has the greatest effect on seismic response The largest
`time
`lapse signals are induced by the replacement of either oil or
`water by gas the replacement of oil by water generally leads
`smaller time lapse signal Altering the pressure regime
`to
`reservoir will
`induce time lapse anomalies within the
`and the period between
`surveys will also affect
`magnitude of the time lapse signal
`
`the
`
`The method has
`that will open for
`yet unrealized potential
`wider use if higher repeatability and greater sensitivity can be
`obtalned Wider
`be more quantitative
`could
`over shorter
`time intervals and
`
`applications
`
`interpretations monitoring
`application of the technology
`complex reservoirs However
`is only detectable
`signals
`
`if
`
`the time lapse signal
`is not masked
`by noise
`between different phases
`Perturbations errors and differences
`4D seismic study create noise when datasets acquired at
`of
`for changes This noise can mask
`different
`times are analysed
`the subtie variations in the seismic response of the reservoir
`in pore fluids The
`indicate changes
`sources
`of
`the
`that
`can be internal or external
`to the acquisition
`perturbations
`system Examples of internal perturbations
`calibration
`variations
`and
`positioning
`
`to smallcr tightcr
`
`and more
`
`like all
`
`it
`
`are hydrophone
`errors External
`
`perturbations include environmental
`such as rough sea
`and tidal variations We can divide the perturbations internal
`and external into
`
`effects
`
`Those
`
`that
`
`affect
`
`such
`
`sensor
`
`the received signal
`as
`sensitivity variations and ambient noise like swell
`Those
`
`that affect
`
`signature such as
`shot-to-shot variations in the source output and the array
`
`the emitted source
`
`directivity
`Those that affect positioning accuracy
`as sea currents and positioning network
`
`and reliability such
`
`ASEG
`
`Geophysical Conference and Exhibition August2001 Brisbane
`
`Extended Abstracts
`
`PGS Exhibit 1071, pg. 1
`PGS v. WesternGeco (IPR2014-00688)
`
`

`

`Perturbations in 4D marine seismic
`
`Brice at al
`
`then defined to attenuate
`signal The result is
`
`the noise while leaving the reflected
`improvement of the signal-
`significant
`to-noise ratio in the region of 6-12dB of the seismic data
`
`and increased
`
`agalnst noise during data acquisition
`in marginal weather see Figure
`
`robustness
`
`have
`
`The effects of these perturbations
`been studied using
`both modelled and real data examples and the impact on
`It was with these factors in
`seismic data quality quantified
`mind that Schlumberger developed
`new marine acquisition
`system called Q-Marine The design features of system that
`from conventional marine acquisition systems
`may be summarised as follows
`
`differentiate it
`
`en
`
`Calibrated positioning with streamer steering capability
`
`Calibrated sensors individually recorded
`
`Calibrated sources
`
`These
`
`features
`
`are
`
`intended
`
`to
`
`address
`
`the
`
`sources
`
`of
`
`the quality of seismic data both for
`impact
`perturbation that
`high resolution imaging and repeatability in 4D studies
`
`CONTROL OF PERTURBATIONS
`
`Ca
`
`5-
`
`Calibrated receivers
`
`The two most significant perturbations to the signal received
`towed
`by
`streamer are swell noise and sensor sensitivity
`streamers contaln hard wired spatial arrays of
`to suppress noise these arrays also act
`for non-vertical
`seismic
`
`Conventional
`
`hydrophones designed
`
`as
`
`filter
`
`ray paths
`
`spatial
`receiver elements forms
`Connecting the outputs of individual
`receiver groups however there are drawbacks
`associated with
`the nature of the analog array The outputs of these elements
`the
`are simply susnmed without
`any processing applied as
`group forming is hard-wired in the acquisition system The
`ideal would be to record each individual
`receiver element into
`its own seismic channel single sensor
`receivers
`the wavefield
`dense grid of
`
`Figure
`
`and
`
`Comparison of shot domain
`on
`conventional
`simultaneously
`analog
`streamer Nessie
`left
`point-receiver
`right The point-receiver
`data are displayed after digital
`group forming using the data adaptive
`filter
`
`data recorded
`
`group-forming
`streamer
`
`coherent noise
`
`Calibrated positioning
`
`is regarded as the
`Repeatability of the acquisition footprint
`for 4D surveys Some of the most successful
`same
`have
`been
`those
`conducted
`using the
`
`key requirement
`4D studies
`
`successive
`
`between
`surveys This
`acquisition configurations
`and depths The
`means identical streamer lengths separations
`source arrays should be identical with the same tow depths
`sailines should have identical directions and shot
`locations to
`the baseline survey With
`towed
`to exactly replicate receiver
`streamer system it
`is very difficult
`positions due to the effects of current on streamer feathering
`
`those of
`
`conventional
`
`Variations in the acquisition footprint
`
`also depend
`accuracies
`
`across
`
`from survey to survey
`on positioning accuracies Relative positioning
`combination of GPS
`on
`spread depend
`range finders
`compasses
`network
`solver Crossline accuracies
`
`and
`
`the
`
`are
`
`acoustic
`
`and
`
`nodes
`performance of
`than inline and conventional
`systems have crossline
`poorer
`errors which may be as high as 12m Archer et al 1999
`errors and Morice et al
`studied the effects of positioning
`2000 analysed the impact on 3D imaging as functions of dip
`and frequency
`In broad terms they suggest
`target accuracy
`4D
`of 2-3m is required to reduce non-repeatable noise in
`signal below other sources of error
`
`recording
`
`so that
`
`samples
`
`alias free
`
`however
`and equipment
`this has been prohibited by cost
`limitations The Q-Marmne system is able to record data from
`in the streamer and can record more than
`each single sensor
`
`hard-wired
`The
`
`4000 individual
`hydrophones on each of up to 20 streamers
`total of up to 80000 channels Onboard computer
`perform sophisticated noise attenuation
`processing systems
`much more
`
`giving
`
`effectively
`
`than
`
`arrays while
`
`preserving
`
`signal
`
`compensates
`
`and
`
`dropouts
`
`amplitude
`processing
`system
`hydrophone sensitivity variations
`for individual
`New proprietary
`tubular
`high
`fidelity
`each delivered with its own aging
`are used
`hydrophones
`These values
`profile and sensitivity certificate
`are stored
`the streamer electronics for automatic data calibration
`
`in
`
`sensors allows more effective removal of
`Recording individual
`noise caused by the movement of the equipment
`through the
`The principal
`affecting marine towed
`water
`noise source
`streamer data is the flow of water across the streamer Vertical
`
`cross-flow can be introduced by wave
`
`action
`
`and is often
`
`termed swell noise Horizontal cross-flow is introduced by the
`
`towing of the equipment and local water currents All sources
`number
`of cross-flow generates vibrations in the streamer in
`of characteristic modes
`chaotic
`appearing completely
`to handle in data from conventional
`streamer
`impossible
`The same noise
`spatial sampling
`groups due to inadequate
`from single sensor
`will however
`and
`records appear coherent
`removable utilizing adaptive noise attenuation methods
`
`and
`
`new technique for data adaptive
`coherent
`noise attenuation
`has been developed Ozbek 2000 which delivers greater
`noise suppression for densely sampled point receiver data than
`The method works by searching
`for
`conventional arrays
`low frequency
`low velocity
`coherent
`noise fields across
`groups of adjacent
`suitable multi-channel
`
`filter
`
`is
`
`traces
`
`see Figure
`and constant
`ASEG lfl Geophysical Conference and Exhibition August 2001 Brisbane
`
`The new system provides lateral steering of the streamers
`is now possible to
`addition to depth control Utilising this it
`steer several degrees laterally from the natural streamer feather
`new positioning system
`angle This ability is coupled with
`fully braced acoustic network from the front
`based on
`to the
`tall of the streamers providing greatly improved relative and
`absolute accuracy Using the new steering system streamers
`can be towed with crossline separations as littie as 25
`
`in
`
`Streamer steering devices are mounted inline with the streamer
`to avoid noise and provide the desired mix of horizontal and
`two independentiy operating wings
`The ability to steer horizontally provides close
`streamer separation giving improved crossline
`
`vertical
`
`force with their
`
`Extended Abstracts
`
`PGS Exhibit 1071, pg. 2
`PGS v. WesternGeco (IPR2014-00688)
`
`

`

`Perturbations in 4D marine seismic
`
`Brice at
`
`the ability
`sampling and
`for optimal
`and repeatability from survey to survey
`
`coverage
`
`to steer
`
`sub-surface
`
`-c
`
`Ca
`
`Ct
`
`5-
`
`Figure
`
`Inline steerable bird Tbe two wings operate
`and lateral steering Tests
`to allow vertical
`independenfly
`show that
`
`these
`
`devices
`
`generate
`
`less
`
`noise
`
`than
`
`conventional depth control birds
`
`The new accurate positioning system provides
`
`full
`
`streamer
`
`acoustic
`
`network
`
`independent of
`
`streamer
`
`length
`
`The
`
`acoustic
`
`positioning
`
`robust broadband
`
`sources
`generate
`source having its own signal code
`signal with each acoustic
`enabling simultaneous shooting The positioning signal can be
`by any seismic hydrophone in the streamer and
`received
`super fast acoustic
`network solver performs correlation and
`The
`calculations
`
`range
`positioning
`accuracy better than meters anywhere along the streamers
`
`is an
`
`absolute
`
`result
`
`Streamer steering and accurate positioning improves the safety
`
`of seismic operation in busy oil
`steered around potential hazards
`
`fields as the streamers
`
`can be
`
`such as surface installations
`
`Calibrated sources
`
`The signature of an airgun source array is known to vary from
`to the next depending on variations in the individual
`one shot
`airgun firing times alrgun chamber pressure sea conditions
`and array geometry
`These shot
`to-shot variations reduce the
`and resolution of
`
`the seismic data and can mask
`accuracy
`subtle effects due to reservoir pore fluid changes
`in time-lapse
`shows
`how modelled changes
`studies
`Figure
`in fluid
`can be masked by source and receiver variations
`
`contract
`
`three plots show modelled data with
`
`The upper
`Figure
`Sm change
`the time lapse change
`in the oil/water
`contact
`right is as much as 25dB lower than the seismic response
`three plots illustrate how the
`of the reservoir The lower
`noise behind
`from random shot amplitude
`the signal
`and random receiver
`can
`variations
`variations
`sensitivity
`mask the seismic differences The variations are 5% in each
`case
`realistic ranges for conventional
`
`technology
`
`Removal of these shot-to shot variations may be accomplished
`by signature deconvolution for which the far field signature is
`that would be recorded by
`below the
`The
`
`required This is the
`hydrophone
`
`placed
`
`signal
`
`source
`
`directly
`array
`distance from the source array needs to be sufficient such that
`the array output appears
`to the hydrophone as
`single pulse
`number of pulses
`from the individual
`rather
`than
`is in excess of lOOm it
`this distance
`
`Since
`
`alrguns
`
`is impractical
`
`to
`
`record the far field signature during production
`
`The far-field signature can be estimated from the combination
`of array components One approach
`gun
`to use
`signature from library of previously recorded gun signatures
`and combine these using special
`source modelling software to
`far field signature The downside of this method is
`calculate
`
`static
`
`is
`
`that
`
`the estimate does
`
`not
`
`include
`
`shot
`
`to shot
`
`variations
`
`caused by timing air pressure changes and varying depths etc
`
`is to use
`
`gun recorded
`
`better approach
`real signature for
`during the shot
`this is recorded by
`field hydrophonc
`near
`NFH which is placed
`to each gun However
`directly next
`each NFH not only receives
`pulse from its own
`the acoustic
`gun but also its ghost arrival bounced
`off the sea surface
`signals from all uther guns
`in the array and
`the NFH is not
`elements Thus
`pure measure of the gun
`signature The new system takes the recorded NFH signals and
`an uncontaminated
`near field signature in real time
`computes
`This is done using the Notional Source Method of Ziolowski
`et al 1982 which takes
`into account
`the various pressure
`
`their ghost
`
`ASEG
`
`Geophysical Conference and Exhibition August2001 Brisbane
`
`Extended Abstracts
`
`PGS Exhibit 1071, pg. 3
`PGS v. WesternGeco (IPR2014-00688)
`
`

`

`Perturbations in 4D marine seismic
`
`Briceatal
`
`the relative movement of
`source signature The notional sources
`by subtracting scaled and time delayed signatures of
`the adjacent airgun hydrophones from each respective
`The
`reference
`
`geometry
`and time delays of the
`
`is used
`
`wavefields
`
`and
`
`generate
`
`calculated
`
`notional
`
`calculate
`
`hydrophone
`the travel path
`
`source
`
`distance
`
`the bubbles to
`
`are
`
`airgun
`
`to
`
`CONCLUSIONS
`
`Data recorded with the new generation of acquisition system
`will minimise the impact of the acquisition footprint
`on time
`The amount of ambient noise in the data
`
`lapse interpretations
`
`will
`
`through the use of single sensor
`be greatly reduced
`recording and digital group forming The quality of the final
`image and frequency content of data will be and the accuracy
`of
`improve to allow the detection of
`interpretations will
`weaker
`time-lapse signals This will allow periods between
`
`to be reduced giving greater control of the
`monitoring surveys
`reservoir management
`
`ACKNOWLEDGMENTS
`
`This paper is the result of contributions from numerous people
`which includes but is not limited to the following persons
`G.J.M Baeten
`Belougne
`Curtis
`Christie
`Combee
`Laake J.E Martin
`Orban
`Ozbek
`Kragh
`Peters P.L Vermeer
`
`REFERENCES
`
`and Cross
`Archer
`Ridyard
`Gilkas
`Pinel
`1999 Spatially and temporally correlated errors how do they
`17 11
`themselves in seismic data First Break
`manifest
`355-362
`
`and Canter
`Welker
`Ronen
`Morice
`2000-
`Expanded abstracts 62nd Conf European Association of
`Geoscientists and Engineers
`
`Ozbek
`
`2000-Expanded abstracts 70th Ann Mtg SEQ
`
`Parkes G.E Hatton
`Ziolkowski
`and Haugland
`1982 The signature of an airgun array Computation from
`near-field measurements including interactions Geophysics
`47 10 1413
`1421
`
`interfering energy and accurate measurements of the
`adjacent
`source geometry are derived from dGPS units on the subarray
`floats The far-field signature is then derived from the notional
`source signatures raw NIFH signals notional
`source signals
`and the computed far-field signature estimate are recorded on
`The
`
`shot
`
`far-field
`
`auxiliary channels
`then
`
`signature is
`
`on
`
`used
`
`for each
`
`tape
`for accurate
`
`source
`
`deconvolution
`
`is that by combining
`during processing
`the notional source signatures offiine far-field estimates with
`azimuth
`selected
`surface
`reflection
`
`operator
`
`angle
`
`or
`
`further
`
`refinement
`
`coefficients can be generated
`
`Experiments using
`showed
`
`signal
`
`hydrophone towed deep below the source
`between
`the measured
`excellent agreement
`and the estimate arrived at by the notional
`algorithm Compensation
`for
`the
`shot-to-shot
`variations
`improves the bandwidth of the seismic data and enhances
`analysis such as AVO Measurement
`accuracy of advanced
`in the
`and
`compensation for variations
`source
`signature
`4D study removes major perturbation
`between
`phases of
`from the time-lapse signal
`
`far-field
`
`source
`
`the
`
`a0
`
`00
`
`cl
`
`11
`
`ASEG
`
`Geophysical Conference and Exhibition August2001 Brisbane
`
`Extended Abstracts
`
`PGS Exhibit 1071, pg. 4
`PGS v. WesternGeco (IPR2014-00688)
`
`