The verdicts are arriving -- and the results
are looking good.
Time-lapse, or so-called 4-D, seismic technology
is proving its worth as a reservoir management tool -- not just
on new fields where the technique is applied from inception, but
at all stages of a field's lifecycle.
Its value, especially for older fields, is found
via pre-production baselines for future monitoring and in conjunction
with calibration to well data.
"The high cost of drilling in newer, deep-water assets
places much responsibility on seismic data for the characterization
of the reservoir, estimating its volumetrics and, through the reservoir
simulator, its potential flow performance," said Philip A. F. Christie,
scientific advisor for Schlumberger Cambridge Research. "Deep-water
reservoirs are often produced through flexible risers to floating
production systems, which make re-entry for surveillance logging
prohibitively expensive."
Under these conditions time lapse seismic is extremely
useful to complement data from permanently installed sensors --
and the good news is a high quality 3-D survey acquired for characterization
before first oil can become the reference survey for subsequent
reservoir monitoring from first oil through to abandonment.
Christie, who presented a paper on "Time-Lapse Seismic
From Exploration Through Abandonment" at the Petroleum Exploration
Society of Great Britain's reservoir geophysics seminar held earlier
this year, examined published results from four different fields
at four different stages from exploration to abandonment, defining
how time-lapse seismic was an important tool at all those stages.
"An exploration 3-D seismic survey represents an
opportunity to establish a baseline for future time-lapse analyses,"
he said in his paper, "if the seismic data quality is good enough
with measurements of source signature, navigation repeatability,
calibrated sensors, together with estimates of uncertainty in these
measurements.
"Such a calibrated exploration 3-D survey can be
termed 4-D ready."
First at Foinaven
Foinaven, operated by BP 190 kilometers west of the
Shetland Islands in 500 meters of water, is the first example of
a 4-D project launched before first oil and where the baseline survey
was acquired as a dedicated 4-D reference before production.
"Time-lapse seismic is the only practical way of
monitoring the field because it is produced through an FPSO," Christie
said, "so routine re-entry into the wells to obtain production log
information at reservoir level is not an easy option.
"The key reservoir property is that the oil is fully
saturated with gas, and that the reservoir pressure is close to
bubble point -- the pressure below which free gas will come out
of solution in the pore space."
Christie cited a paper written by M. Cooper, J. Bouska,
E. Thorogood, A. O'Donovan, P. Kristiansen and P. Christie, "Foinaven
Active Reservoir Management: The Time-Lapse Signal," presented at
the 1999 SEG annual meeting.
He found that on the 1995 baseline survey there was
virtually no free gas in the reservoir compartment under analysis.
By the 1998 repeat survey, however, the amplitudes were much brighter
with gas evolution indicated, even to the oil-water contact, due
to the pressure reduction following production.
When amplitudes increase, scientists can deduce that
this part of the reservoir is in pressure communication with the
producer wells. Also, the consistency of features such as minor
faulting in the two surveys adds confidence to a detailed interpretation.
Amplitude build-up in the 1998 data to the north of a fault separating
two compartments indicates gas evolution that has not breached the
fault, although there is reservoir rock on both sides of the fault.
This suggests the fault is sealing at the time of
the repeat survey, he said.
"Foinaven demonstrated that 4-D seismic works West
of Shetland and provided reservoir-related interpretation," Christie
said. "Because all the wells had been drilled by the time of the
repeat survey, there is little intervention that can be based upon
the results of the 4-D. But the success of the project in delivering
reservoir-related information motivated the extension of the technology
to other West of Shetland assets.
"Foinaven is a good example of designer 4-D giving
results early in a field, which can help to guide the development
of the field and manage its production."
The Draugen Field
The Draugen Field, operated by Shell in the North
Sea's Norwegian sector, is in the early maturity stage with full
plateau production from seven producers in the center of the field
and five water injectors at the northern and southern ends of the
field.
It includes the worlds most prolific oil well, producing
at 76,750 barrels of oil a day. The field has no primary gas cap,
but has uncertain connection to the aquifer.
Christie's talk referenced a paper published last
year by SEG, "Time-Lapse Seismic Surveys in the North Sea and Their
Business Impact," written by K. Koster, P. Gabriels, M. Hartung,
J. Verbeek, G. Deinum and R. Staples.
A 3-D survey acquired in 1990 prior to first oil
in 1993 was used as a legacy baseline for a repeat survey acquired
in 1998 to help determine the optimal location for a new well planned
for 1999.
"The Draugen story is one of uncertain support from
the aquifer and fault seal," Christie said. "Shell was carrying
three reservoir models with different fault transmissivities, all
of which matched the production history data equally well and indicate
the level of uncertainty in the original model.
"The time-lapse seismic difference maps were able
to image the impedance change due to water replacing oil in the
reservoir. The seismic best matched one of the reservoir models,
though it also indicated that a fault in the north was sealing,
since the seismic difference switched off to the south of the fault."
Shell discarded two of the models and manually updated
the third until it matched both the seismic and the production history.
At this point, Shell had reduced uncertainty by having a model consistent
with the seismic and the production.
The manually updated flow model was used to predict
the performance of the planned infill well at several alternative
locations. The optimal location was found to the north of the platform,
rather than the originally planned westerly location, and the resulting
delay in anticipated water cut extended the forecasted plateau production
by a year and increased predicted recovery by 12 million barrels.
"As a result, in 1999 the new well was successfully
drilled at the revised location and an additional two wells have
been drilled extending the field life to 2015," Christie said.
"Draugen represents an example of 4-D seismic helping
to prolong plateau production."
The Gullfaks Field
The Gullfaks Field, also in the North Sea's Norwegian
sector, was discovered in 1984 with first oil two years later. The
3-D used as a baseline was before first oil, but the first repeat
survey was in 1995, a year after peak production.
Christie cited a talk presented on the field at last
year's EAGE annual meeting, "Four-D Seismic Enhances Oil Recovery
and Improves the Reservoir Description," by L.K. Strønen and P.
Diagranes.
The geology in the "middle-aged" field is complex
with multiple, high porosity and permeability sandstone reservoir
units in tilted fault blocks beneath an erosional unconformity with
the sealing shales. The reservoir has saturated oil with no free
gas and pore pressure is maintained with water injectors.
Base reserves were estimated at 1.3 billion barrels.
"The strongest seismic difference is in the amplitude
of the top reservoir," Christie said. "The oil sand is 'softer'
than the overlying shale with a lower Poisson's Ratio, so there
is a good stack response at the top reservoir in the 1985 data."
By 1995, the reservoir sand had been drained and
water had replaced oil in the pore spaces. As a result, the impedance
and the Poisson's Ratio had both increased to become closer to the
overlying shale, resulting in a dimmer response on the seismic.
The two-way-time of a deeper horizon has also decreased
because of the increase in velocity of the produced reservoir rock
resulting in a seismic "pull-up" effect, according to Christie.
Seismic inversion was used to map the fluid distributions
in 1985 and 1995, using well data to constrain the inversion and
to distinguish between possible changes in saturation, temperature
and pore pressure. The results show several areas of unchanged oil
saturation in both 1985 and 1995, indicating by-passed oil.
The 4-D map was used to guide successful drilling
of in-fill wells, accessing an additional 9.5 million barrels of
oil and helping to increase recoverable reserves to 1.9 billion
barrels. Two more repeat surveys were conducted in 1996 and 1999.
The Sleipner Field
Time-lapse seismic data can even be useful in fields
that have reached retirement.
The Sleipner Field, operated by Statoil in the North
Sea's Norwegian sector, was discovered in 1974 but production did
not commence until 1996. Recoverable reserves are 128 billion cubic
meters of gas and 170 million barrels of condensate.
Christie cited another paper presented at the EAGE
annual meeting in Glasgow, Scotland, "Reservoir Geology of the Utsira
Sand in the Southern Viking Graben Area -- A Site for Potential
CO2 Storage," by I. Brevik, O. Eiken, R. Arts, E. Lindeberg
and E. Causse. He said carbon dioxide is found in association with
the gas in West Sleipner at an unsaleable concentration of 9 percent
and is separated offshore for re-injection in an underground storage
structure from the East Sleipner platform.
This presented an opportunity for exploring an environmentally
acceptable method of carbon dioxide sequestration, and so a European
Union-supported project called Saline Aquifer CO2 Storage
was established.
The CO2 repository is the massive, brine-filled
Utsira sand body, which is 200 meters thick with high porosity and
permeability. It is estimated that the sand could accommodate 400
years of Western Europe's CO2 production.
A legacy baseline seismic survey was acquired in
1994, and between the start of injection in 1996 and the repeat
survey in 1999, some two million tons of CO2 were injected.
"CO2 arriving at the Sleipner A platform
from Sleipner West is injected via an extended reach well into the
Utsira Formation," he said. "The 1994 survey, prior to injection,
shows the top Utsira Formation, intended to be the top of the storage
interval. In the 1999 repeat survey, the water sands have become
partially saturated with gas and have dramatically increased amplitudes
due to the gas. Its upward migration is indeed contained by the
shales above the top Utsira Formation. The progressive push-down
effect on deeper horizons due to the low velocity gas is also evident."
According to Christie, this is an example of how
4-D seismic may be used to demonstrate seal integrity in gas storage
structures, either for CO2 sequestration or for the storage
of associated gas, even after the original oil production has ceased.
"In this sense, monitoring the seal integrity is
a motivation for continuing to run monitor surveys to check that
the buoyancy forces created by displacing the brine with CO2,
together with the increased pressurization of the reservoir, do
not cause breakdown of the sealing caprock," he said.
"In the case of Sleipner, the large capacity of the
Utsira sand and the presence of the infrastructure may allow CO2
sequestration to continue long after the abandonment of the field
as a hydrocarbon producer."