1998 SPE49130 Day Griffin Martins

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


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


3/13

SPE 49130 REDEVELOPMENT AND MANAGEMENT OF THE MAGNUS FIELD FOR POST-PLATEAU PRODUCTION 3

rapidly from ~15 mstbd to


4/13


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


5/13


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


6/13


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.


7/13


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


8/13


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


9/13


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


10/13


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


11/13


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.


12/13


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.


13/13


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.

Date post:	06-Apr-2018
Category:	Documents
Upload:	juan-m-villalobos-ii
View:	213 times
Download:	0 times

1998 SPE49130 Day Griffin Martins

Documents