The harsh but prolific West of Shetlands
— among the world's first deepwater producing regions — has been
a proving ground for time lapse seismic.
Today oil is flowing from several fields West of
Shetlands, thanks in large part to technological achievements in
floating production systems and time-lapse, or 4-D, seismic — and
the successes achieved there are furthering the use of the technique
around the world.
"Four-D seismic is rapidly becoming business as usual
for BP within the United Kingdom Continental Shelf," report BP officials
in a recent paper.
Multiple data sets covering multiple fields are being
used in a range of different ways, the officials said, varying between
simplistic qualitative comparisons to sophisticated models — and
they may be used to:
- Improve reservoir understanding.
- Reduce uncertainty in production forecasting.
- Design effective well interventions.
- Locate possible infill targets.
Graeme Bagley, Ian Saxby, John McGarrity and Chris
Pearse, all with BP in Aberdeen, Scotland, presented a paper titled
"4-D Time Lapse Seismic — Examples from the Foinaven, Schiehallion
and Loyal Fields, UKCS, West of Shetland" at the Petroleum Exploration
Society of Great Britain's reservoir geophysics seminar last year.
As part of the Foinaven Active Reservoir Management
(FARM) project, 4-D seismic data was acquired across part of the
field, with a baseline survey shot in 1995 prior to field start-up
and a repeat survey in 1998 after 10 months of production. The project's
results led to the recommendation to acquire a full field survey
in association with the Schiehallion and Loyal fields in 1999.
A subsequent survey was acquired in 2000 covering
the same area — and BP's intent is to acquire a 4-D survey every
year.
"So far, the data is being used in a qualitative
to semi-quantitative manner," the authors wrote, "but advances in
technology will very soon allow BP to use the data in a more rigorous,
quantitative and predictive manner."
The availability of low cost, high quality data on
demand and the ability to acquire data around complex infrastructure
will allow further step changes in our ability to manage reservoirs
effectively.
Objectives at Foinaven
Foinaven was the first significant test for 4-D,
or time-lapse, seismic technology.
The field, operated by BP, is 170 kilometers north
of the Scottish mainland in water depths between 400 and 600 meters.
The Paleocene turbidite reservoirs lie between at about 2,100 meters
and are cut by faults and sealed by up-dip sand pinchouts.
Sand unit thickness varies from zero to 90 meters
and is generally of good quality, with 24 to 28 percent porosity
and 250 to 500 millidarcies permeability.
Original oil in place is over one billion barrels,
and 250 million barrels of that total is currently being developed
through the Petrojarl Foinaven FPSO. The phase II program will develop
an additional 52 million barrels of reserves.
The oil is fully saturated and the reservoir is at
bubble point pressure, which is critical for interpreting the 4-D
seismic response of the field.
"The objectives of ongoing 4-D seismic acquisition
at Foinaven are two-fold," Bagley said:
- To understand fluid movement within the reservoir, which is
critical to off take management and successful production forecasting.
- To assess connectivity between future reservoir targets and
the existing development.
"Pressure support is vital if the field is to continue
producing oil and manageable gas rates," he added.
The FARM experiment acquired 4-D seismic data using
both surface tow and sea floor hydrophones of one area of the field.
The results from these experiments gave BP and its co-venturers
the confidence to acquire full field 4-D seismic in addition to
the baseline exploration 3-D survey acquired in 1993, which were
acquired in conjunction with the adjacent Schiehallion and Loyal
fields.
Valuable Insights
The rock and fluid properties at Foinaven are such
that oil and gas bearing reservoirs show as high amplitude reflectors.
Because the reservoir is at bubble point, any drop
in reservoir pressure due to production results in the evolution
of free gas and a brightening of the seismic reflectors.
"This effect is the dominant one that may be observed
on Foinaven 4-D seismic data — indeed the response appears to predict
high gas oil ratio before a producing well," Bagley said.
"Conversely, water injection should result in a reduction
of seismic amplitude due to a drop in oil saturation assuming no
pressure increase," he added. "This effect is less obvious."
The need for water injection is well illustrated
at a production well, where producing gas-oil ratio increased early
in the well's life due to a lack of effective water injection, resulting
in a brightening of the seismic data over the reservoir around the
well. As water injection was brought on, the seismic amplitude dropped
back to something close to that of the baseline survey, suggesting
a drop in reservoir gas-oil ratio or an increase in water saturation.
"Qualitative comparison between the 4-D seismic and
results from the reservoir simulation model have confirmed the understanding
of communication within the field, the importance of gas management
and the validity of the ongoing reservoir management program," Bagley
said.
"The 4-D continues to offer insights into communication
between existing and planned production wells, which is used in
the ranking of subsurface target opportunities."
Development at Schiehallion
The objectives of the 1999 4-D seismic at the Schiehallion
Field were to gather additional data on which to interpret dynamic
performance and to influence development drilling plans. Another
survey with the same objectives was acquired in 2000.
Schiehallion, which is operated by BP for partners
Shell, Amerada Hess, Statoil, Murphy Oil and OMV, is 170 miles north
of the Scottish mainland in about 350 meters of water.
The Paleocene turbidite reservoir sands lie at a
depth of around 2,000 meters. Horizontal wells are necessary to
contact sufficient net pay in the 10 to 50 meter thick sands and
produce at economic rates.
Original oil in place was 1.36 billion barrels of
oil, of which 40 million had been recovered by 2000. First oil from
Schiehallion flowed in 1998.
Three-D seismic was acquired over Schiehallion in
1993 and 1996 prior to the field coming on line. The 1999 and 2000
surveys were acquired using different vessels, source and streamer
configuration, but a common acquisition direction was maintained.
The cost of each 4-D survey was about 0.6 million pounds, including
processing.
In the original pre-development surveys, the oil-bearing
channel sands are evident as high amplitude areas. In response to
production and injection, a combined response to pressure and saturation
is observed:
- At injector locations, increased saturation results in reflectivity
reduction.
- In areas where pressure is elevated but saturation remains
unchanged, the amplitude brightens.
- At producers, amplitudes in response to pressure decline. However,
since the initial pressure is close to bubble point, when pressure
declines below this, the evolving gas leads to an increase in
reflectivity and significant brightening.
"By providing additional data from which to interpret
pressure and saturation changes, the presence and effect of transmissibility
barriers and connected hydrocarbon volumes, 4-D has impacted the
drilling and planning of six wells," he said. "More than four million
barrels net to BP of incremental production that these wells are
predicted to achieve is attributed to
4-D.
"Used in conjunction with reservoir simulation, 4-D
is proving to be a very useful interpretation tool," he continued.
"Because the observed effects are a combination of both pressure
and saturation, there may be ambiguity in the interpretation. Deconvolution
of these effects and calibration of the seismic response is the
next step in removing this potential problem.
"More stringent demands will be made on the data
quality and repeatability as increased sensitivity is sought."
Improved Understanding at Loyal
The Loyal Field, owned by BP and Shell, lies north
of Schiehallion in water depths of 450 meters. It was discovered
in 1993, with initial oil in place estimated at 200 million barrels.
Development drilling began at Loyal in 1998 as part
of the Schiehallion project.
The reservoir is a Paleocene marine turbidite sandstone
with 25 percent porosity.
"The aim of the 4-D is to improve understanding of
the connectivity within the field leading to better reservoir management
and optimal locations for the remaining development wells," Bagley
said.
Three-D seismic was acquired in 1991, 1999 and 2000.
All the production facilities are sub-sea and tied back to the Schiehallion
vessel, which allows complete coverage of the field with each repeat
surface tow survey.
The 1999 data were acquired after only eight months
of production, with just four months of water injection from a single
well. A second water injector was brought on line in 1999. No water
breakthrough had been seen at any of the producers at the time of
the 2000 survey.
In general, the strength of the 4-D seismic signal
was bigger than expected.
At the time of the 1999 survey, the amount of free
gas had increased as a result of reservoir pressure dropping below
bubblepoint, causing amplitudes to brighten. By the 2000 survey,
significant volumes of water had been injected and the increase
in free gas had slowed.
For instance, significant dimming was observed between
two wells from the 1999 to the 2000 surveys and this is interpreted
as evidence of water flood front movement.
"The timing of the water cut in one of the wells
has given us confidence in the ability of the 4-D seismic to image
fluid changes," Bagely wrote. "The data are now being used in the
history match process to calibrate simulated reservoir responses
away from wells."
Changes to the relative water injection rates at
the two injectors have been justified by the data, he said.
"The locations of the two remaining base development
wells to the north are being planned using the 4-D seismic," he
said. "Applying the 4-D analysis directly to the area of 'dim pay'
on the seismic remains a challenge and improvements in data resolution
and repeatability are required to map a coherent signal over the
full extent of the field."