Date post: | 06-Apr-2018 |
Category: |
Documents |
Upload: | juan-m-villalobos-ii |
View: | 213 times |
Download: | 0 times |
of 13
8/3/2019 1998 SPE49130 Day Griffin Martins
1/13
Copyright 1998, Society of Petroleum Engineers, Inc.This paper was prepared for presentation at the 1998 SPE Annual Technical Conference andExhibition held in New Orleans, Louisiana, 2730 September 1998.This paper was selected for presentation by an SPE Program Committee following review ofinformation contained in an abstract submitted by the author(s). Contents of the paper, as
presented, have not been reviewed by the Society of Petroleum Engineers and are subject tocorrection by the author(s). The material, as presented, does not necessarily reflect anyposition of the Society of Petroleum Engineers, its officers, or members. Papers presented atSPE meetings are subject to publication review by Editorial Committees of the Society ofPetroleum Engineers. Electronic reproduction, distribution, or storage of any part of this paper
for commercial purposes without the written consent of the Society of Petroleum Engineers isprohibited. Permission to reproduce in print is restricted to an abstract of not more than 300words; illustrations may not be copied. The abstract must contain conspicuous
acknowledgment of where and by whom the paper was presented. Write Librarian, SPE, P.O.Box 833836, Richardson, TX 75083-3836, U.S.A., fax 01-972-952-9435.
AbstractMagnus is the most northerly field in the UK North Sea, with
14 platform and two subsea producers, and 13 injectors of
which 8 are subsea. After 12 years of plateau production at
140,000 bopd, the field went on precipitous decline in 1995,
with an underlying decline rate of 60% per annum. Although
there had been significant preparation and investment for
decline, the severity of the well-related problems had not been
anticipated. Wells producing over 20,000 bopd fell over in a
matter of hours as a result of severe barium sulphate scaling,
amongst other factors.
This paper discusses the causes of the severe declines, theincreasing evidence of reservoir complexity, and the re-
engineering of the reservoir development that equipped the
field for post-plateau production. During the first three years
of decline, all production wells were either side-tracked or had
major interventions to maintain production, as we moved from
multi-zone to single zone completions. Crucial to the success
of the performance recovery has been the drilling of several
injectors, targeted to support specific zones/areas. Facilities
upgrades were also made with increased water-handling
capacity, gas lift capability, upgraded water injection pumps
and a subsea injection manifold all being added.
The subsequent reservoir and well management strategies
used are discussed, together with a summary of lessons from
the new wells. The main keys to success of the programme are
presented, including base production focus and scale
management operations. As a result of the re-engineering, the
base decline rate has been substantially reduced and
manageability has been restored to the field. The lessons
learned on Magnus may have relevance for future
developments reliant on a small number of wells, particularly
if they are subsea.
Field BackgroundMagnus is the most northerly of the presently producing
fields on the UKCS. Discovered in 1974, the reservoir is a
Upper Jurassic turbidite reservoir, with a high net to gross
upper reservoir (Magnus Sandstone Member - MSM) and a
low net to gross lower reservoir (Lower Kimmeridge Clay
Formation - LKCF), both averaging approximately 100mTVT
(Figures 1&2). The reservoir is a South-East dipping tilted
fault block of 14km by 3.5km, overlain by Cretaceous
mudstones. Reservoir quality improves towards the crest of
the structure, with only limited permeability remaining at the
oil water contact. Some faults can be seen on the seismic but
these are of limited throw. Only the upper MSM sands were
recognised when the development was sanctioned in 1978.
The field was developed with a single platform with
subsea wells on the flanks of the field to access areas that the
drilling technology of the day could not reach. A two platform
development was considered but rejected on the grounds of
excessive cost in the deep waters (186m) and harsh
environment of the Northern North Sea. Due to the limited
permeability in the water leg, the development plan was for pre-produced injectors, completed in the oil leg then later
converted, to sweep oil in a line drive towards the main
producers. The first line of producers would in turn be
converted to injection as the flood front advanced up dip. The
pressure support from the water aquifer was inadequate due to
its limited extent and poor permeability. STOIIP is presently
estimated to be ~1650 MMSTB with recoverable reserves
797 MMSTB of sweet light 39o
API oil (GOR 750scf/stb).
The MSM has the majority of the reserves and STOIIP with
an expected final recovery factor of ~65%; LKCF recovery
factor ~30%. Initial reservoir pressure was 6653psi at datum
(3050 mTVDSS) with a bubble point of 2750psi. The Magnus
field is operated by BP on behalf of the Magnus LicensePartners (BP 85%; AGIP UK 5%; Nipon Oil E&P UK Ltd
5%; Petrobras (UK) Ltd 2.5%; Talisman Energy (UK) Ltd
2.5%)
On plateau, production was processed through two
identical trains, with first a high pressure (HP) separator, then
a low pressure (LP) separator and finally a flotation unit,
cleaning the produced water prior to overboard discharge. Oil
was exported through a pipeline via Ninian to Sullom Voe.
Gas was exported via the Northern Leg Pipeline System to
Brent and finally St. Fergus. Gas compression was the main
SPE 49130
Redevelopment and Management of the Magnus Field for Post-Plateau ProductionSimon Day, Tim Griffin and Paul Martins, BP Exploration
8/3/2019 1998 SPE49130 Day Griffin Martins
2/13
2 SIMON DAY, TIM GRIFFIN AND PAUL MARTINS SPE 49130
facilities limitation during plateau production.
The highly productive nature of the Magnus sands (PIs of
100-200 stb/day/psi were common in conventional wells)
allowed the field to be developed with relatively few wells.
The platform was installed with only 20 well slots and space
for 9 subsea wells. The field was brought onto production in
August 1983 with water injection start-up following in August
1984 (delay due to the pre-produced injectors). Initial plateaurate was an annualised average 125 mstbd (daily peak
145 mstbd) but this was steadily increased to an annualised
average 140 mstbd (daily peak 163mstbd) in 1988 as the
facilities were debottlenecked. Water injection initially
averaged ~150 mbwd but a major upgrade in 1991 increased
this to ~240 mbwd (Figure 3).
Due to the initial lag in starting water injection and the
subsequent facilities upgrades, the field was under-voided
throughout most of the 12 years of plateau production
(Figure 3). In an attempt to keep a straight line drive, two of
the highest injectors (M17:C5, M13:C7 on Figure 1) were
choked back which also contributed to the under-voidage. Asa result, the average reservoir pressure steadily dropped until,
in the early 1990s, the crestal pressure was close to the oil
bubble point. In 1993, all the injectors were opened to
maximise injection and increase the reservoir pressure,
regardless of whether the flood-front was skewed.
Despite the pressure drop in the crest, the high
productivity of the wells meant plateau production could
easily be maintained. Plans were considered for dedicated
crestal injectors but not carried out due to the difficulty of
justifying the benefit with the existing reservoir description
and a field limited by production facilities.
Production and injection were generally perforated in all
reservoir intervals encountered. These large multi-zone wellswere PLT logged throughout their lives to track zonal pressure
differences. Some wells were also perforated in stages to
monitor the productivity of different zones. Although pressure
differences were found in wells, these were never extensive
and it appeared that most reservoir zones were in
communication. Compartmentalisation was found in two
places: The Northern area had some pressure differences from
the North Central area, although compartmentalisation was not
total; The Brent High held back significant pressure support
from the Crestal area (Figure 1). The latter was the most
significant from a reservoir development viewpoint - M6:B3
was planned to be one of the first wells in the field to water
out, when it would be converted to injection, in fact it was oneof the last wells to suffer significant water breakthrough in
January 1995. The lack of support from M5:C4 to the crest
contributed to the pressure decline in that area. In fact, the
Crestal area was only being significantly supported from
M17:C5 and M13:C7. The Brent High is the most significant
pressure barrier in the field, holding back up to 4000 psi
pressure difference. Other barriers marked on Figure 1 are at
best leaky, generating some pressure difference but also
allowing some sweep.
Water analysis and the behaviour of the first wells to
water-out under waterflood had shown Magnus was very
susceptible to barium sulphate scale formation. This scale
forms when barium from the formation water mixes with
sulphate from the injected seawater. The resultant compound
is very insoluble and precipitates out in the well liner, limiting
production and well access (scale can also form in the topsides
pipe-work if the conditions are appropriate). Calcium
carbonate scale formation was not expected (nor has proved)to be a significant problem.
Performance Coming Off Plateau ProductionWater breakthrough had occurred in several flank producers
during the plateau period, but none of the crestal producers
(Figure 1) had significant water breakthrough until late 1994.
The field well potential averaged very close to the plant
capacity during most of 1994, but for approximately six
months the well capacity was slightly lower than plant
capacity. However, the field annualised average was still
above the average plateau rate due to strongly increased plant
efficiency - in fact the 1994 average of 144 mstbd was thehighest in field history despite the marginal well stock.
However, by late 1994 increasing signs of reservoir decline
were visible and the field came off plateau production at the
end of 1994 / beginning of 1995. The decline was in several
wells but was driven in particular by three wells:
A5 The first crestal well to break significant water in
October 1994. The water-cut rose very rapidly, reaching 55%
by the end of January 1995. A significant point is that the
average reservoir pressure measured in this well only rose
slowly and lagged the water-cut rise. As a result the lift
performance and oil production from the well declined sharply
- by the end of January the well was producing ~5 mstbd,
down from ~28 mstbd in November. By June the well wasessentially dead.
A4 Another crestal well that broke water at a low level in
April 1994. The water-cut remained low, usually averaging 2-
3%. This well was not scale squeezed as field strategy had
been to scale squeeze wells when they reached 5% to avoid
any formation damage. A PLT in August 1994 showed a
blockage just below the highest perforations, due to scale
formation. From January 1995, production declined rapidly
from ~20 mstbd as scale plugged off the well-bore. By March
production was
8/3/2019 1998 SPE49130 Day Griffin Martins
3/13
SPE 49130 REDEVELOPMENT AND MANAGEMENT OF THE MAGNUS FIELD FOR POST-PLATEAU PRODUCTION 3
rapidly from ~15 mstbd to
8/3/2019 1998 SPE49130 Day Griffin Martins
4/13
4 SIMON DAY, TIM GRIFFIN AND PAUL MARTINS SPE 49130
the wells as it was well performance that was driving the
production decline.
Production Wells: The most significant change in the field
development strategy was a move from large multi-zone
producers to dedicated single zone producers. Though not
initially seen as a strategy for the entire field, this process was
actually started in July 1994 when M22:B7 was completed in
one zone shortly before the field came off plateau. Single zonecompletions improved the lift performance of the wells by
reducing water-cuts and stopping cross-flow between zones.
Production could now be optimised and interventions
performed with confidence that the well would not die due to
cross-flow.
Single zone completions also enabled more effective scale
squeezes as the reservoir pressures within the well were more
uniform. Combined with more effective inhibition chemicals1,2
scale squeeze life was approximately tripled. It is likely that
once pressures in the field are more uniform and the scaling
tendency has reduced as the sea-water cut of the field
increases, a return will be made to multi-zone wells.In a field with such a large well spacing, it was often not
possible to predict the order in which layers watered out nor
their individual water-cuts. As stated above, scale usually
formed rapidly after water breakthrough, precluding the use of
PLTs to measure layer water-cuts after breakthrough. In
addition, due to the poor lift performance of the wells, if a
scale mill-out was performed to regain access then the wells
would likely cross-flow and die, making any PLT virtually
worthless (although the drier zones were usually lower
pressure this could not be guaranteed). The solution was to
perform a series of liner sidetracks (8.5 sidetrack, usually
~50m from the original wellbore) and install gas lift
completions instead of the originally intended gas lift work-overs. Liner sidetracks offered numerous advantages
including: assured shut-off of water intervals behind new liner
and cement; open-hole logs, including accurate pressure for
reservoir monitoring and perforation decisions; clean well
bore without residual scale problems; only minimal increase in
time required over a simple work-over and much quicker and
more reliable if a water shut-off was planned as part of the
work-over. Three wells didhave a successful zonal shut-off to
change them to single zone production - two with plugs
immediately below the topmost zone and one with a more
complicated cement squeeze shut-off. Two other wells were
also directly worked over to gas lift rather than sidetracked.
One was successful from a reservoir viewpoint but later faileddue to mechanical problems. The other well struggled from
difficulties in shutting off one zone and from much higher
than expected zonal water-cuts (95% water under steady flow
conditions 11months after first water breakthrough - zone
~50m thick).
Obviously not all wells could be converted to single zone
production immediately and interventions were required to
maintain flow from existing wells. One of the most intractable
problems was controlling scale in multi-layered wells with
cross-flow - inhibitor had to be injected into the higher
pressure layers in the well. One solution was to inject
inhibition squeezes at higher rates - Magnus now injects
chemicals using the water injection system and can inject at up
to 21bpm, depending on the fracture gradient in the well.
The other solution was to reduce the injectivity of the
lower pressure zone to divert chemicals to the higher pressure
wet zones. First, mechanical diversion was considered but
rejected due to lack of reliability in wells where scale hadalready formed in the liner (assuming access was even
possible in the first place). The chosen method was to inject
the inhibition chemicals in a series of stages, separated by wax
diverters to progressively block the low pressure injectivity3.
The reservoir interval was first cooled with sea-water and then
the first stage of chemicals pumped followed by wax divertor.
The wax was designed to solidify on the cooled rock, reducing
the injectivity. Successive stages injected inhibitor and
blocked the injectivity until all the open intervals had been
inhibited. During the soak period the well would heat up, the
wax melted and was subsequently back produced and
processed normally with the oil stream. This method has beensuccessfully used on several Magnus wells and has helped to
extend squeeze lives. Wax diverted squeezes have been
combined with PLTs to confirm diversion was successful.
Injection: As Magnus has a limited aquifer, injection is of
crucial importance to maintain production. The under-voidage
had been a concern during plateau and efforts were made to
improve injection volumes during the last 4 years of plateau.
However, this focus has increased sharply post-plateau when
the benefits of increasing individual well production rates
through injection are more clear if the field production is
limited by the well potential rates. Three routes were pursued
to improve injection support in the field:
Drilling Dedicated Zonal Water Injectors: As theunderstanding of field complexity improved, it was possible to
target injectors to particular zones or compartments and
locally boost production. In addition, the opportunity has been
taken to convert one producer to injection duty where this was
higher value. The programme has been very successful, with
several wells doubling in oil production rate.
Water Injection Upgrades: Several major injection
upgrade options have been investigated, from additional main
injection, boosting SWIFT manifold wells or boosting a single
well (with an ESP). The highest value option was to add a
sixth injection pump (cost ~6MM) and this may be
implemented during late 1998 / early 1999. However, these
investigations also highlighted the possibility of impellerupgrades for the present main injection pumps. The new
impellers were optimised for the expected range in pressure
and injection rates forecast for the field (short term, high
pressure lower rates, longer term lower pressure higher rates).
The pumps were upgraded on a rolling basis to minimise the
injection loss. The performance improvement was equivalent
to the addition of a new pump but at a cost of ~0.5MM.
Water Injection Optimisation: As field oil rate dropped
below plateau rates, the drop in reservoir barrel offtake was
sufficient to take the field into positive voidage on an annual
8/3/2019 1998 SPE49130 Day Griffin Martins
5/13
SPE 49130 REDEVELOPMENT AND MANAGEMENT OF THE MAGNUS FIELD FOR POST-PLATEAU PRODUCTION 5
basis for the first time (1995). This, combined with zonal
dedicated injection wells and upgrades outlined above allowed
the water injection system to be more closely optimised than
the few years previously. The overall aim was to maintain
field voidage balance while targeting water at the highest
value (in terms of oil production) areas. The reservoir model
was used to value each well in terms of barrels of oil
production supported. A model of the water injection systemusing Petroleum Experts Prosper / Gap software was then used
to optimise the injection to each well. The injection system
was now optimised for maximum oil production rather than
water injection.
Production Result Post-PlateauMagnus production base decline rate reached 60% per annum
for the first year post-plateau. Only a rapid succession of new
wells and liner side-tracks sustained production - in January
1996, 58% of production was from wells drilled in the
previous year. The focus on base well management, water
injection optimisation and the move to single zonecompletions gradually stabilised the annual decline rates
(Figure 5). Indeed, in recent months the decline rate has
actually become negative as the field production rate has
steadily risen due to improved injection support for two key
zones in the crestal area of the field (G and A sands). This
production is a combination of new oil and acceleration and
base decline rates are expected to grow again during the latter
half of 1998. However, manageability has be restored to the
field and it is not expected that the high decline rates of 1995
will return. Wells are now more predictable and can in general
be choked back or turned off if required for field management
reasons without the fear that the well will die. As stated above,
scale squeeze life has approximately tripled due to acombination of new chemicals and the single zone
completions.
Figure 6 shows the source of production since the end of
plateau by year and type of operation. In this case
interventions have been separated out from the base
production rate used to calculate decline rates above. The high
level of activity, spread almost evenly between new
production wells, new injection wells and interventions can
easily be seen. The pre-1995 base (i.e. base before end of
plateau without new wells or interventions) drops close to zero
in mid 1996 (~5 mstbd, 0 mstbd by March 1997). No wells in
Magnus were side-tracked or had major interventions unless
they had died, so this gives a good indication of the likely production rate without investment (although the lack of
competition as wells died would probably have allowed the
remaining wells to last longer). The present Magnus
production of ~90 mstbd is entirely due to the investment
carried out 1995-1998.
The change in strategy to recomplete wells as single zone
producers has essentially been completed within three years.
Figure 7 shows the dramatic change from multi-zone
producers to a pattern of single zone wells. The field now has
several oil bearing zones behind pipe awaiting perforation
when the well performance allows it. The next stage of
development where the present zones are shut-off and other
zones opened up is presently just beginning. It is likely the
field will gradually move back to multi-zone completions
were pressure differences are not great and the scale can be
controlled. This will be required in late life in order to cycle
water and improve recovery.
With such a high level of new wells and liner side-trackactivity, only one producer now remains that produced before
the end of plateau. Figure 8 shows the cumulative percentage
of production by the age of the well for the last months
production. Even the one well still existing from before the
end of plateau has had a major intervention and is actively
supported by two new injectors, drilled or converted since the
end of plateau.
Water injection optimisation is estimated to have increased
production by 1-2 mstbd over the last few years. However the
drilling of new injectors (and conversion of one producer to
injection) has produced more significant levels of new
production as can be seen in Figure 6. Of the 19 platform andsubsea wells drilled on Magnus since coming off plateau, 6
were injectors (not including the producer converted to
injection). Injectors were targeted at particular zones and
compartments - two deserve further description as they also
show the increasing complexity of the reservoir description:
Crestal A Sands This zone is a high net to gross sand of
limited extent in the crest of the field. M26:A4 was a
dedicated A sand producer completed in July 1995
(production rate ~16 mstb/d). The reservoir pressure was low,
close to bubble point, pressure support coming from other
zones and areas of the field. M5:C4 was originally planned to
be the injector supporting the Crestal A sand, but the Brent
High prevented adequate water flux. M5:C4 was highly pressured (~6500 psi) compared to M26:A4 (~2800 psi). It
was decided to sidetrack M5:C4 to a location in better
communication with the low pressure reservoir around
M26:A4. The Brent High is not a single fault, but a line of
thinned reservoir with small faults relating to a high in the
underlying sediments. It was therefore extremely difficult to
pick the exact structure that was causing the pressure barrier.
The A sand thins and narrows updip, so the associated
reserves were reduced rapidly the further updip the well was
targeted. This tension between maximising reserves and
achieving communication led to a stepwise approach, where
the well was sidetracked a short distance first and tested. If the
first location was a failure then the well would be sidetrackedagain over the final barrier on the seismic before M26:A4.
M31:C4 completed drilling in October 1996 and encountered
a high pressure but water swept A sand. As this was not in
communication with M26:A4, the well was immediately
sidetracked as M31z:C4. This well encountered low pressure
unswept A sand, the same pressure as M26:C4. The well was
completed and brought on to injection in early November
1996. Initial injection rates were >40 mbwd, but rapidly
declined to ~4 mbwd over two days. The well was shut-in for
a PFO and later returned to injection when it followed a
8/3/2019 1998 SPE49130 Day Griffin Martins
6/13
6 SIMON DAY, TIM GRIFFIN AND PAUL MARTINS SPE 49130
similar pattern (40 mbwd dropping to 4 mbwd). The pattern
was repeated several times in the following weeks and the
sustainable injection rate remained at 4 mbwd. The PFO
showed no problems with the well and the reservoir was
interpreted as being connected to M26:A4 by long thin
channels, thereby limiting the steady state injection rate.
Injection of 4 mbwd was not sufficient to actively support
M26:A4 production rate, so a further sidetrack was required.No further features were detectable on the seismic so defining
the well target was problematic. In the intervening time,
advantage was taken of the upper sands encountered by
M31z:C4 as outlined below, therefore a different slot was
used for the next updip sidetrack. M34:C3 was completed in
September 1997, again encountering low pressure dry oil
bearing A sands. The well was brought on to injection at a
steady rate of 50 mbwd with the well choked back (rates of up
to 70 mbwd were achievable). No problems were found with
decreasing injection rates. M26:A4 production rate started
rising within 2 days and eventually more than doubled to
~35 mstbd (Figure 9). Water breakthrough at low levelsoccurred in January 1998 and the main water breakthrough is
expected June / July 1998. A second producer, further updip
than M26:A4 is planned for late 1998 / early 1999 to take
advantage of the good injection support from M34:C3.
Crestal G Sands As the main sands in the crest watered
out during 1995/6, three wells (M10:A3, M6:B3 and M24:B4
- Figure 1) produced the majority or all of their production
from the topmost sand in the MSM (G sands). This sand
remained at lower pressure than the main crestal sands and
one of the new SWIFT injectors (F2) was targeted as a crestal
injector to support these wells. F2 was completed in the upper
MSM on January 1996 And brought on to injection at a rate of
~30 mstbd. M24:B4 showed a steady improvement inproduction rate, increasing by 50% in 3 months (Figure 10).
However, neither M10:A3 nor M6:B3 showed any response,
despite the fact that M6:B3 was the closest well to F2 and that
M10:A3 was only 300m from M24:B4 - two of the closest
wells in the field. The second sidetrack of C4 (M31z:C4), to
support the A sands as outlined above, had a top hole section
that passed within 50m of F2. The majority of the upper MSM
showed high pressures (~5500psi), as would be expected so
close to an active injector. However, one sand section within
the G (it was split into two sands) showed much lower
pressures (~3500psi). After the failure of the A sand injection,
it was decided to perforate this lower pressure G sand to
confirm whether it was in communication with M10:A3(M6:B3 had died through raised water-cut and low reservoir
pressure). M10:A3 showed a steady rise in production,
confirming the connection (Figure 11). Although shut-in,
M6:B3 had also showed a reservoir pressure rise. M6:B3 was
lifted back onto production on August 1997 but produced at
very high water-cut. It was decided to convert M6:B3 to
injection duty to more actively support M10:A3. This was
completed in late December 1997 and the well began injection
at ~30 mbwd. M10:A3 production rate again showed a
significant rise (Figure 11). In addition, M10:A3 water-cut
dropped from ~15% to ~8% as the sweep patterns in the field
were changed.
The patterns of connectivity between the above wells
introduced a significant problem for sand correlation. Figure
12 shows the sand bodies as described by well picks and
confirmed reservoir communication. For F2 and M24:B4 to be
directly connected, the two sand bodies between the wells
would have to cross - not geologically possible. No definitivesolution to this has yet been determined - the present reservoir
description has the sands between F2 and M24:B4 thinning
and injection communication occurring via other lower sand
bodies (Figure 13).
The increasing complexity of the G sand can easily be seen
on a plot of crestal RFT and PLT reservoir pressures taken in
the zone during field life (Figure 14). During plateau, the
pressures were relatively uniform. Post-plateau, the
combination of dedicated crestal injection and the move to
single zone producers has given rise to a wide range in zone G
pressures, sometimes within the same well, as discussed
above. The offtake from this zone has increased post-plateau,so the differences are probably the result of reservoir
constraints rather than complete pressure barriers. However
the plot shows how the real reservoir complexity may not be
apparent while on plateau.
Conclusions and Lessons LearntThe Magnus field has obviously required a substantial
investment, both in plant and well facilities, to maintain
production. Indeed, without the investment since the end of
plateau, production would be close to zero instead of the
present daily average of ~90mstb/d. All investments have been
justified on the basis of the production / reserves that they
accessed or protected. However, such a degree of late lifeinvestment had not been anticipated in the original
development plan. The risk of requiring late life investment
(and well access) to maintain production should be included in
the planning of new developments.
The relatively few highly productive wells means the field
is very similar in concept to a number of developments taking
place today, albeit that non-conventional well technology is
now often used to achieve the high well productivity.
Reservoir simulation and extensive PLT data gathering had
not shown the extent of the problems the Magnus field would
face as it came off plateau. Zonal pressure differences within
wells were relatively modest during plateau production with
no indication of the large changes that would rapidly occur.Although many of the main reservoir features could be
identified, the full complexity of the reservoir was not
apparent until the field came off plateau (and the degree of
complexity is continually increasing). With relatively few
wells and large inter-well spacing, reservoir description and
performance uncertainty is substantially increased - prediction
of the performance during decline is especially problematic.
New developments that rely on small numbers of wells should
include this uncertainty in their predictions. Developments
that assume limited well access may especially be at risk.
8/3/2019 1998 SPE49130 Day Griffin Martins
7/13
SPE 49130 REDEVELOPMENT AND MANAGEMENT OF THE MAGNUS FIELD FOR POST-PLATEAU PRODUCTION 7
AcknowledgmentsThis paper is written by three reservoir engineers but
obviously the successful redevelopment of Magnus has
involved many hundreds of people, both onshore and offshore,
past and present. We extend our thanks and congratulations to
them. In particular, we would like to thank David Richards for
the use of his illustrations.The authors would like to thank the Magnus License
partners for their permission to publish this paper.
The views expressed in this paper are those of the authors
and do not necessarily represent those of BP Exploration.
References1. Collins, I.R., et al.: Extending Scale Squeeze Lifetimes Using a
Chemical Additive: From the Laboratory to the Field, paper
presented at the "Solving Oilfield Scale" Conference, Aberdeen
Jan. 22nd - 23rd 1997
2. Bourne, H.M. et al.: Combining Innovative Technologies to
Maximise Scale Squeeze Cost Reduction: A Laboratory to Field
Study, paper presented at 9th International Oilfield ChemicalSymposium, 22 - 25 March 1998 Geilo, Norway.
3. Ravenscroft, P.D., Cowie, L.G. and Smith, P.S.: Magnus Scale
Inhibitor Squeeze Treatments - A Case History, paper SPE
36612 presented at the 1996 Annual Technical Conference and
Exhibition, Denver, Oct. 6-9
8/3/2019 1998 SPE49130 Day Griffin Martins
8/13
8 SIMON DAY, TIM GRIFFIN AND PAUL MARTINS SPE 49130
M32:A2
M36:B7
Brent High
OWC
M29:A7
12A-21:D9
12A-17:D4
M25:A1
M37:B1
M4:C2
M30Z:A5
M10:A3
M24:B4
M27:B2
M6:B3M26:A4
M28:A6F2
M31Z:C4
M23:B6 M17:C5
12A-16:D3
F1
M35z:B5
12A-11Z:D5
PLATFORM
SWIFTF3
M33:C7
Production
Injection
M32Z:A2
M34:C3
Figure 1: Magnus field map with areas
F4z
F4
M35:B5
Southern
Crest
North Central
Northern
8/3/2019 1998 SPE49130 Day Griffin Martins
9/13
SPE 49130 REDEVELOPMENT AND MANAGEMENT OF THE MAGNUS FIELD FOR POST-PLATEAU PRODUCTION 9
GR0 250
NET20 0
RHOBG/CC1.95 2.95
NPHIV/V0.45 -0.15
ILM0.2 200
ILD0.2 200
3250
3300
3350
3400
3450
3500
3550
Depth
METRES
2900
2950
3000
3050
3100
3150
3200
CKHL0.5 5000
KHA980.5 5000
PHITV/V0.3 0
PERF15 0
CPORPU30 0
VSHV/V0 1
PHITV/V1 0
FOIL1 0
Figure 2: Typical log showing MSM and LKCF reservoirs
8/3/2019 1998 SPE49130 Day Griffin Martins
10/13
10 SIMON DAY, TIM GRIFFIN AND PAUL MARTINS SPE 49130
0
25000
50000
75000
100000
125000
150000
175000
200000
225000
250000
275000
300000
A
ug-83
J
an-84
J
un-84
N
ov-84
A
pr-85
S
ep-85
F
eb-86
Jul-86
D
ec-86
M
ay-87
Oct-87
M
ar-88
A
ug-88
J
an-89
J
un-89
N
ov-89
A
pr-90
S
ep-90
F
eb-91
Jul-91
D
ec-91
M
ay-92
Oct-92
M
ar-93
A
ug-93
J
an-94
J
un-94
N
ov-94
A
pr-95
S
ep-95
F
eb-96
Jul-96
D
ec-96
M
ay-97
Oct-97
M
ar-98
ProductionandInjection(Barrels
/day)
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
1
2
VoidageReplacementRatio(V
RR)
Water Injection
Oil Production
Water Production
Monthly VRR
Cumulative VRR
Figure 3: Magnus production, injection and voidage history from start of field life
0
5
10
15
20
25
0:00 2:00 4:00 6:00 8:00 10:00 12:00 14:00
Time in Hours
Gr
oss
Fluid
Rate
(mbbl/d)
0
20
40
60
80
100
120
WH
FP
(barg)
and
Choke
(%)
M10:A3 Gross Rate
M10:A3 Choke
M10:A3 WHFP
8/3/2019 1998 SPE49130 Day Griffin Martins
11/13
SPE 49130 REDEVELOPMENT AND MANAGEMENT OF THE MAGNUS FIELD FOR POST-PLATEAU PRODUCTION 11
Figure 6: Magnus recent production by source
0
20000
40000
60000
80000
100000
120000
140000
160000
180000
Sep-94
Nov-94
Jan-95
Mar-95
May-95
Jul-95
Sep-95
Nov-95
Jan-96
Mar-96
May-96
Jul-96
Sep-96
Nov-96
Jan-97
Mar-97
May-97
Jul-97
Sep-97
Nov-97
Jan-98
Mar-98
Production(stb/d)
Base pre 1994
Interventions
New Producers New Injectors
JAN. 1994 A1 A3 A4 A5 A6 A7 B1 B3 B5 B7 D6G G G G G G E G G G G
11 Wells E E E E E E D E E E E36 Open Zones D C C A L C C C C D3.3 zones/well C A A A
L L
JUN. 1998 A1 A2 A3 A4 A5 A6 A7 B1 B2 B3 B4 B5 B6 D9G E G G G G G E G G G G G G
14 Wells E D E E L E E E E E E E E17 Open Zones D C C C L C C C C C D1.2 zones/well C A A L A L C
L L
Figure 7: Showing wells with zones present and open perforations in black (D9 is a subsea well)
0%
25%
50%
75%
100%
125%
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Number of Years Well on Production
CumulativeP
ercentageof
OilPo
tential
Figure 8: Age of well versus cumulative percentage of present (June 1998) production. The remaining well
existing from before the end of plateau has had one major intervention to restore production and is supported
by two new injectors.
8/3/2019 1998 SPE49130 Day Griffin Martins
12/13
12 SIMON DAY, TIM GRIFFIN AND PAUL MARTINS SPE 49130
0
10000
20000
30000
40000
50000
60000
Aug-95
Sep-95
Oct-95
Nov-95
Dec-95
Jan-96
Feb-96
Mar-96
Apr-96
May-96
Jun-96
Jul-96
Aug-96
Sep-96
Oct-96
Nov-96
Dec-96
Jan-97
Feb-97
Mar-97
Apr-97
May-97
Jun-97
Jul-97
Aug-97
Sep-97
Oct-97
Nov-97
Dec-97
Jan-98
Feb-98
Mar-98
Apr-98
Producti
on/Injectionbbl/d
M31z:C4 INJECTION
M34:C3 INJECTION
M26:A4 GROSS PRODUCTION RATE
0
5000
10000
15000
20000
25000
30000
35000
40000
45000
50000
Apr-95
Jun-95
Aug-95
Oct-95
Dec-95
Feb-96
Apr-96
Jun-96
Aug-96
Oct-96
Dec-96
Feb-97
Apr-97
Jun-97
Aug-97
Oct-97
Dec-97
Feb-98
Apr-98
Production/Injectionbbl/d
F2 INJECTION
M31z:C4 INJECTION
COMPETITION FROM
M30z:A5
M24:B4 GROSS
PRODUCTION RATE
0
5000
10000
15000
20000
25000
30000
35000
40000
Jul-96
Aug-96
Sep-96
Oct-96
Nov-96
Dec-96
Jan-97
Feb-97
Mar-97
Apr-97
May-97
Jun-97
Jul-97
Aug-97
Sep-97
Oct-97
Nov-97
Dec-97
Jan-98
Feb-98
Mar-98
Apr-98
Production/Injectionbbl/d
M31z:C4 INJECTION
M6:B3 INJECTION
FLUSH PRODUCTION FROM
INJECTION DURING
PRODUCTION SHUTDOWN
M10:A3 GROSS PRODUCTION RATE
M10:A3
M2 7:B2
M26:A4 M6:B3
M24:B4
M31z:C4
12A-F2M34:C3
MSMESupport
M10 :A 3
M2 7:B2
M26:A4
M 24:B4
12A-F2
M6:B3
M31z:C4
M33:C7
Figure 12: Shows the correlation,
evolved through time, of the crestal G
sands using well picks and connectivity
information from production data. Wells
M24:B4 and F2 apparently have the same
sand bodies in different depth order - a
geological problem.
Figure 13: Problem resolved by
assuming reservoir thinning (possible
from seismic) and that connectivity
between M24:B4 and F2 is via lower
sands. Given the changes in G sand
description to date, this is probably an
interim resolution.
8/3/2019 1998 SPE49130 Day Griffin Martins
13/13
SPE 49130 REDEVELOPMENT AND MANAGEMENT OF THE MAGNUS FIELD FOR POST-PLATEAU PRODUCTION 13
Initial Reservoir Pressure
Plateau Decline
0
1000
2000
3000
4000
5000
6000
7000
83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98
Year (start)
D
atu
m
Pr
essu
re(psia
)
RFT Data
Surveillance D ata
Figure 14: Shows RFT and PLT pressure data from the crestal G sands. The
increasing range in pressures due to dedicated injection and zonal completions
illustrates how the level of complexity of a field can be masked during plateau.