Field Appraisal Basics PDF

BHP Petroleum
Fundamentals of Field Appraisal

and
Methods for Improving
Appraisal Decisions
by
C G McKinley
Contributions from:
J R Allen
M Bhatia
P Behrenbruch
R A Hogarth
February 1998
DRAFT VERSION 2.0
Field Appraisal: Technical Guidelines February 1998
DRAFT VERSION 2.0
Contents
1 Introduction 3
2 Fundamentals of Field Appraisal 4
2.1 Development Concept 4
2.2 Development Concept Maturity 7
2.3 Economics 8
2.4 Economic Maturity 10
2.5 Appraisal Requirements 12
2.6 Appraisal Costs 15
3 Methods for Improving Appraisal Decisions 17
3.1 Appraisal Strategy 17
Maturity Crossplot 17
Developing an Appraisal Strategy 18
Scale and Confidence Appraisal 23
3.2 Appraisal Measures 24
Reserves Accounting 24
Segmentation 25
Risked Reserves 25
Definitions 27
3.3 Sensitivity Analysis 30
3.4 Forward Modelling 30
Monte Carlo Forward Modelling 30

Appraisal by Objectives 35
Value of Information 36
4 Conclusions 41
2
DRAFT VERSION 2.0
1. Introduction
Field appraisal is an activity which has received growing attention in the industry over
the last few years, as operators strive to improve project economics by reducing cycle
times and pre-development costs.
Within BHPP, a number of key exploration areas have entered the appraisal phase,
and our ability to efficiently bring these projects to development could have
considerable impact on the performance of the company over the short to medium
term.
This paper discusses the fundamentals of field appraisal, challenging some of our
traditional views, and describes a number of methods which may be used to make
more objective judgements and improve the decision making process.
Its purpose is to generate discussion across the company and provide some guidelines
as to how we might improve our approach to field appraisal, recognising the
differences which can exist in appraising discoveries in different geographic areas due
to geological, cultural or political constraints.
3
DRAFT VERSION 2.0
2. Fundamentals of Field Appraisal
Field appraisal is driven by our perception of the economic risk of developing a

resource with our current knowledge.
In the ideal world, if we were certain that a discovery could be developed

economically, there would be no need to gather additional information and a project
could proceed immediately. Any remaining risks and uncertainty would be addressed
during field development and production.
However, in most cases, there is considerable risk and uncertainty in the size and
value of a hydrocarbon discovery, which prevents us from moving immediately to
develop the field. Field appraisal involves the acquisition of additional technical
information, typically seismic or well data. These data, when integrated with our
existing knowledge of the field, are expected to reduce the level of risk and/or
uncertainty towards an acceptable level.
The decision to proceed with a development project involves two fundamental

considerations.
We must select a development concept, and we must ensure that the economics of the
preferred development are robust. That is, we must satisfy ourselves that the selected
development has a high likelihood of making a satisfactory return on investment, or in
the worst case is unlikely to lose money - all based on an imperfect view of the size of
the resource and the cost of development.
2.1 Development Concept
“Where and when will what volumes of which fluids be produced?”
Only when we have a reasonable answer to this question can we consider development
options, facilities design and costs.
Hence, the choice of development concept is influenced by four key factors.
Location The proximity of existing infrastructure, including production and

processing facilities and export routes, play an important role. Concept selection is
also influenced by local factors such as bathymetry or topography, climate and the
areal extent of the field. Legislative, environmental or political requirements may also
exist.
Timor Sea, Australia -a floating production, storage and offloading (FPSO)

concept was developed for a number of small oilfields offshore northern
Australia due to their remote location and lack of existing infrastructure
4
DRAFT VERSION 2.0
Wytch Farm, UK - the onshore area of the Wytch Farm oilfield was
developed in the mid 1980’s from several onshore drill sites tied back to a
central processing facility. The field extends offshore into shallow waters
and the original development concept for this area was a manmade gravel
island with drill site, full processing facilities and pipeline to shore. The
proposed island was located only 3km from the major tourist resort of
Bournemouth and within the Poole Harbour Marine Park. Due to these
environmental constraints, this development concept was rejected. The
offshore reserves have subsequently been developed using ultra extended
reach drilling from an existing onshore site - with stepouts of up to 10km.
Volumes The size of the resource generally influences the scale of the development. A
well defined estimate of the reserves to be developed is often critical to concept
selection and maximising the value of the project. A poorly constrained estimate of
reserves and production can result in facilities which are either too large - spare
capacity, wasted capital - or too small - constrained production, modification costs.
Ula, Norway - the Ula oilfield is trapped in massive Jurassic sands in a

simple dome structure, and appraised by three crestal and one downdip
well, which encountered a clear oil-water contact on the western flank.
Development of 150 MMstb reserves was sanctioned using three linked steel
platforms for drilling/processing/accommodation, with an export pipeline
via the nearby Ekofisk field. The facilities were designed to handle up to 100
Mbd oil production.
The first three downdip wells on the eastern flank , which were planned as
peripheral water injectors, all encountered a full oil column. Further
drilling confirmed a deeper oil-water contact, and upgraded reserves to 450
Mmstb. If this volume had been expected during facilities design, processing
capacity would have been doubled to optimise offtake rates. Despite
expensive modifications which eventually took production up to 140 Mbd,
the field remained facilities constrained throughout its five year plateau
period and lost in excess of $100 million in potential project value.
Fluids The type(s) of fluids to be produced, including variations in amount or

composition over time, impact process design, material selection, production handling
requirements, injection or compression capacity and export options.
Gyda, Norway - the 200 Mmstb Gyda oilfield development, like others in
the area, was designed to recover low GOR oil trapped in a faulted anticline
structure. During production, a large southern extension of the field was
discovered. This highly productive reservoir contained high GOR oil which
could not be optimally developed due to gas handling facilities constraint.
During main field decline, the area was eventually brought onstream - six
years after discovery.
5
DRAFT VERSION 2.0
Natuna, Indonesia - the giant Natuna gas field is ideally located for Asian
LNG export markets and contains over 200 Tscf gas-in-place. However, the
gas contains over 30% CO2, which requires additional processing and gas
disposal facilities, and the use of special alloys in construction -
significantly increasing capital and operating costs. As a result, Natuna
remains uncompetitive and undeveloped.
Productivity In addition to understanding the size of the resource, we must understand

how and when reserves will be produced. The number and size of facilities will be
influenced by productivity - well numbers, flow rates and gas/water cut development
over time, pressure depletion or maintenance requirements - and may constrain how
the field is produced. The economic viability of a discovery depends on the field
having sufficient productivity to minimise well numbers and optimise the offtake rate.
Offshore and Onshore Fields, UK, US and Australia - Figure 1 plots

reserves versus well rate for a number of fields in the UK, US and Australia.
Most of the offshore UK fields are economically viable, despite the high
development threshold, because they are typically large and highly
productive. Although Field A and Field B are similar in size, Field B - an
Eocene heavy oil field - has not been developed due to low productivity.
The offshore US fields are generally smaller, with lower well rates, but most
are economically viable due to a lower development threshold. Even
extremely small fields, such as Field C, may be developed if productivity is
high.
In general, onshore fields have a lower development cost and even small,
low productivity reservoirs may be successfully developed. In Australia,
onshore fields with reserves of only 2-3 Mmstb and well rates less than 1
Mbd are economic. However, even for these small fields it is important to
have an accurate estimate of productivity prior to the development decision.
Field D was developed on the assumption that well rates would be 2 Mbd.
Early production wells were only able to produce 0.5 Mbd, and the field
failed to make an economic return.
Whilst many fields in a given play or area will have similar productivity
characteristics, we must remember that anomalous fields exist and can
provide the opportunity for both outstanding success and failure. Field E,
located only 30km from Field D, has unusually large reserves and high
productivity - making it one of the most profitable oilfields in Europe.
In some cases, the productivity of a reservoir may be improved by the application of

horizontal well technology (Figure 2). If a reservoir has good effective vertical
permeability, a horizontal well may increase productivity by a factor of 3-5+, and
delay water coning and/or gas cusping. However, in reservoirs with poor effective
vertical permeability, a horizontal well may give little or no increase in productivity
6

Field Appraisal: Technical Guidelines
Well Productivity (1): Development Thresholds

100
50
Offshore UK
Field E Offshore US
Field C
20
Onshore UK
Onshore Australia
10
Field A
Well Rate
(Mbd or MMcfd)
Field D
Development Threshold
Offshore UK
1
Field B
Offshore US
Onshore UK
Onshore Australia
0.1
1 10 100 1000
Reserves / Technically Recoverable Volume (MMboe)
6 August, 1999C:/appraisal.ppt PERT - Australia/Asia Region

December, 1997 Figure 1
BHP Petroleum
DRAFT VERSION 2.0

Well Productivity (2): Horizontal Wells
Kv/Kh=1.0
40
30
Kv/Kh=0.1
Productivity
Index
20
(stb/d/psi)
Kv/Kh=0.001
10 Vertical Well PI
0 400 800 1200 1600 2000 2400 2800
Horizontal Well Length (m)

BHP Petroleum
8
as the pressure support required to maintain well rates may be insufficient. A
substantial productivity benefit is required to offset the incremental cost of a
horizontal well - typically 30 - 100% more than an equivalent vertical well.
Horizontal wells may also be successfully employed where access and sweep, rather
than productivity, is the critical factor. A single horizontal may penetrate and drain
several isolated fault blocks or, combined with extended reach drilling technology,
recover hydrocarbons from peripheral areas of a field which would not otherwise be
developed.
Griffin Field, Australia - the Birdrong reservoir in the Griffin field is a

heterogeneous, low productivity reservoir. Development using vertical wells
(PI = 3) is uneconomic. However, the reservoir has been successfully been
developed using three high angle to horizontal wells which have improved
well productivity by a factor of 3-5.
2.2 Development Concept Maturity
The stages which lead towards the selection of a single development concept can be
described in terms of the development concept maturity of a discovery (Figure 3).
Concept Uncertain
The development concept for a discovery is considered to be uncertain when we are

unable to identify, with a reasonable degree of certainty, a realistic development
option. This may be due to lack of key information - the size of the resource, fluid
types and reservoir productivity may be unknown or ill defined - or technological
barriers, such as development in ultradeep water or ultra high pressure areas.
Appraisal to identify development concepts may be justified, with acquisition of new

information aimed at removing uncertainties.
Multiple Options
In many cases, where we have more information, or technology is not a barrier, there
may be multiple options - two or more identified, realistic concepts for development
of a discovery. Typically, we carry these options because the range or possible
reserves is wide, and each option is often best suited to a particular field size and
offtake.
Appraisal to eliminate development options may be justified, with the acquisition of

additional information aimed at narrowing the reserves range.
In some cases, where the field reserves and potential production are well defined, we
may still have alternatives due to several tie-in options or export routes. In the latter,
more refined development cost estimates and economics evaluation may be required.

Development Concept Maturity
Concept
Concept 1 Concept 2 Concept 3 Justification
Maturity
" " "

Concept Constrain
Identify development concepts
Uncertain development Reduce
concept risk and/or
Multiple uncertainty
Options ✖ ✔ ✔ Eliminate development option
Single
Selection ✖ ✔ ✖ Optimise development plan
" Unidentified
✔ Identified
✖ Excluded

BHP Petroleum
Single Selection
In order to make a development decision, we must make a single selection for our
development concept. This may be done by choosing our preferred option from a
number of options, in the light of additional information which refines our
understanding of the field. In some circumstances, there may always be only one
development option available due to specific infrastructure, geographic or legislative
considerations.
Appraisal to optimise the selected development concept may be justified, although it

is often difficult to justify the additional appraisal expenditure against the
improvement in economic value - which may result from better definition of
throughput capacity or process design etc.
2.3 Economics
“What volumes will be produced when, at what price, and for what costs including
taxes?”
Only when we have a reasonable answer to this question can we evaluate the
economics of a development option.
Hence, the economics of field development is influenced by five key factors, some of
which may be controlled or influenced by an operator, and some of which are
generally outside our control.
Reserves Whilst the size of the in-place resource is predetermined and outside our
control, an operator can maximise both the volume and value of reserves recovered
from a field. This requires a sound understanding of in-place volumes, reservoir
heterogeneities, productivity and recovery mechanisms. Prudent reservoir
management and the successful implementation of secondary or tertiary recovery
methods can both accelerate and increase field reserves.
Prudhoe Bay, Alaska - the giant Prudhoe Bay oil and gasfield was
discovered in 1967, and developed in the 1970’s with estimated reserves of
9600 Mmstb. The initial field development plan was based on gravity
drainage, with pressure support from gas re-injection into the large gas cap.
In the last ten years, a number of pilot and full scale enhanced oil recovery
(EOR) projects have been implemented including a downdip pattern
waterflood and water-alternating-gas (WAG) injection. New technologies
such as horizontal wells and coiled tubing sidetracks have also been
employed to economically access bypassed oil zones. As a result reserves
have been upgraded to over 12000 Mmstb, despite little change in estimated
in-place volumes.
The use of the term reserves refers only to volumes which are, or have the potential to
be, economically recoverable. Volumes which could technically be recovered, but are
considered uneconomic, are referred to as technically recoverable volumes.
DRAFT VERSION 2.0
Product Price In general, product prices are controlled by global energy demand,
economic cycles and political events. However, an operator can have some influence
over price by developing hydrocarbon types which are in demand, attracting a price
premium, sound marketing and active hedging against unfavourable price and/or forex
movements.
Kutubu, PNG - the Kutubu oilfield development in the remote highlands of

Papua New Guinea, produces light, sweet crude from the prolific Toro
reservoir. Oil is exported via pipeline to a marine terminal in the Gulf of
Papua and offloaded directly to tankers for transport to refineries. The
crude is ideally suited to most refining processes, being easy to breakdown
with no significant residues, and is in demand from refineries on the east
coast of Australia and west coast US. As a result, Kutubu crude consistently
attracts a premium over the regional marker, Indonesian Tapis crude.
Capital and Operating Costs The cost of appraising, developing and producing a
discovery is often a major factor in field economics, particularly in remote or offshore
environments. The cost of appraisal can vary considerably, depending on geographic
location and the level of confidence required to reach a development decision.
Development costs may be minimised through innovative design, synergy with
existing facilities and efficient project management and execution. During production,
efficient operation and the tie-in of satellite developments to extend field life and
reduce operating costs plays an increasingly important role.
Elang/Kakatua, Australia - the Elang/Kakatua development, offshore

northern Australia has minimised development costs by leasing an FPSO
facility which BHPP used previously on the Skua field.
Griffin Area, Australia - the main Zeepaard reservoirs in the Griffin and
Chinook Scindian fields, offshore western Australia, are prolific producers
which are now entering decline. Rising unit operating costs may be curtailed
by further development of the overlying Birdrong reservoir. Two new
horizontal wells and workover of an existing well to provide pressure
support is expected to boost production and extend the economic life of the
field.
Fiscal Regime Although there is often little room to alter the fiscal regime within
which a development occurs, there can be opportunities to amend fiscal terms to take
account of changing levels of risk or economic return. Entry into unexplored or high
risk areas, or the development of sub-economic satellite fields in a mature basin, may
attract tax or other financial concessions.
12
DRAFT VERSION 2.0
Marginal Field Development, UK - the giant North Sea oilfields developed
in the 1970’s and early 1980’s have been very profitable under a benign UK
fiscal regime. However, in the late 1980’s the number of smaller field
developments began declining as operators faced the relatively high costs
and risks of these marginal projects. The petroleum industry actively lobbied
the British government, highlighting the changing risk/reward balance and
requesting lower taxes and other incentives to develop these resources. In
the early 1990’s the government announced sweeping changes in the
treatment of marginal fields, encouraging many new small field
developments which have taken the UK to a second peak in oil production in
the late 1990’s.
Cycle Time Reducing the time between discovery and first production, and
accelerating production without loss of reserves, can significantly increase the
economic value of a project (Figure 4), provided that the main risks and uncertainties
have been correctly identified and managed. In some cases, the commercial pressure
to reduce cycle time may compromise genuine appraisal requirements, and may then
result in loss of value through inappropriate development or production. In practical
terms, reduced cycle times require an early, clear direction and fully aligned
partnerships between joint venturers, contractors and government.
Foinaven, UK - Foinaven, discovered in 1990, is the first deep water

Atlantic development, off the west coast of Scotland. Following encouraging
results from early appraisal wells, the field was sanctioned for fast track
development using an FPSO facility. It was recognised from the outset that
this approach may not lead to the optimal development, but a rapid return
on the substantial investment was essential to project economics. Both
appraisal and detailed facilities design were undertaken simultaneously,
with modifications as a result of new information being incorporated where
possible. As a result, this frontier project has gone from discovery to first oil
in seven years. Whilst this in itself is a remarkable achievement and success,
more production history will be needed to determine whether the project will
succeed.
Given the lack of control we have over product prices or fiscal regime, an operator
may improve the economics of a development by increasing reserves, reducing
development costs and shortening cycle times (Figure 5).
2.4 Economic Maturity
The degree of economic risk, based on the viability of a development across a range of
possible scenarios, can be described as the economic maturity of a discovery (Figure
6).
13

Cycle Time
80 80
Cum Cash Flow ($m)

Cum Cash Flow ($m)
40 40
0 0
6.0 years 13.5 years 3.5 years 11.0 years
-40 -40
4.0 years 2.0 years
-80 -80
0 4 8 12 0 4 8 12
Time (yr) Time (yr)
NPV10 = $16.9 m NPV10 = $28.0 m

BHP Petroleum
Uneconomic
A discovery is considered to be uneconomic if it fails to make an acceptable return

from any scenario, including upside cases.
No further appraisal is justified, as additional information is unlikely to make the

project economic at this time. A substantial change in economic conditions e.g. oil
price or fiscal terms, or a technological breakthrough which lowers development cost
e.g. compact LNG, is required.
Feasible
A discovery is considered to be economically feasible if it makes an acceptable return

from upside scenarios, even though the most likely view is that development is
uneconomic.
Appraisal to establish an economic project may be justified, with acquisition of new

information aimed at proving the upside potential of the discovery.
Viable
A discovery is considered to be economically viable if the most likely scenario makes

an acceptable return, even though downside scenarios may be uneconomic.
Appraisal to provide downside protection may be justified, with acquisition of new

information aimed at removing risks and adding proven reserves.
Robust
A discovery is considered to be economically robust if it makes an acceptable return

from any scenario, including downside cases.
A discovery can be economically robust for at least one development option, even
though other development options have not been eliminated. In these cases, further
appraisal to constrain the development concept may be justified.
Even when a discovery is fully mature - robust, single selection - further appraisal to
optimise the development plan may be justified if the improved value of the project is
greater than the cost of continued appraisal. In practical terms, however, this type of
appraisal is often difficult to evaluate and support.
The economic maturity of a discovery can change without any further appraisal, due to
breakthrough of technological barriers or improved economic assumptions e.g. higher
oil price or discount rate.
Key changes in the maturity of a discovery are also linked to the BHPP tollgate
process.

Shifting the Balance

Product
Developable Price
Reserves
Negotiation Skills
Facilities Design Government Influence
Efficient Process Well Numbers
Clear Direction Efficient Operation
Partnerships Satellite Tie-In
Product Quality
In Place Volumes Marketing
Reservoir Productivity
Recovery Factor
Near Field Exploration
Fiscal
Capital and Regime
Cycle Operating
Time Costs

BHP Petroleum
DRAFT VERSION 2.0

Economic Maturity
Economic
Downside Most Likely Upside
Maturity
Uneconomic ✖ ✖ ✖ No appraisal justified
Feasible ✖ ✖ ✔ Establish economic project Constrain

development Reduce
concept risk and/or
Viable ✖ ✔ ✔ Provide downside protection uncertainty
Robust ✔ ✔ ✔ Optimise development plan
✔ Economic
✖ Uneconomic

BHP Petroleum
17
DRAFT VERSION 2.0

Development Concept and Economic Maturity
Project Identification Project Definition

Blackback
SE Mananda Moran Lambert BSFN-ROD

Single
Selection
Multiple
Options Scott Reef Sunrise Bayu- Bayu
Development Area Undan Undan
Concept Gas Liquids
Maturity
Concept
Uncertain
Ultra Deepwater Neptune
GOM
Uneconomic Feasible Viable Robust

Economic Maturity

BHP Petroleum
18
New discoveries which are potentially economic or existing, uneconomic discoveries
which become economically feasible, may be considered to have entered the project
identification phase.
Discoveries which are highly mature - those with a high probability of being economic
based on one or two clearly identified, workable development concept(s) - may be
selected for further study and progression towards development - this represents the
start of the project definition phase.
2.5 Appraisal Requirements
The appraisal requirements for two discoveries of similar size, with similar risks and
uncertainties may differ markedly due to a number of external factors. In considering
an appropriate level and cost of appraisal we must also consider the following:
Market - Oil or Gas? Oil is an easily transported product which is freely traded on
the world market at spot prices. In contrast, gas is usually sold by long term contract to
consumers - power generators, heavy industry or domestic users - who demand
security of supply. Gas supply may be via an onshore gas pipeline system or via LNG
export, involving expensive, complex processing facilities and shipping fleet. For
many gas developments, appraisal must be more thorough to ensure that contract
volumes are met, and high facilities costs are justified.
NW Shelf Project, Australia - the NW Shelf Project exports over 7 mtpa

LNG to Japanese energy companies. These buyers demand security of
supply over the 20 year period of the sales contract, and a high degree of
confidence in gas reserves was required prior to signing a sales contract in
the mid 1980’s. During early appraisal, reserves from two giant gasfields,
North Rankin (8 Tscf) and Goodwyn (4 Tscf), were thought to be sufficient to
meet a gas sales contract. However, in order to prove P90 reserves of 10-12
Tscf, both fields were extensively appraised. The nearby Angel discovery,
which is still not developed, was also appraised to meet the proven reserves
target and provide some form of contingency.
Location - Onshore or Offshore? The cost of development offshore is typically an

order of magnitude more expensive, and less flexible, than onshore development -
unless an onshore location is particularly remote or inaccessible. As a result, appraisal
in offshore situations generally needs to be more thorough before a development
decision can be taken. For offshore subsea developments, the option to use appraisal
wells for later production or injection may influence or limit well location and data
acquisition.
Onshore Projects - whilst many onshore developments are relatively low

cost, some major onshore projects have proved to be as expensive, or more
expensive, than a similar sized offshore development. These include:
Cusiana/Cupiagua, Colombia - high well costs, steep terrain

Kutubu, PNG - remote location, mountainous terrain
Prudhoe Bay, Alaska - remote location, arctic climate
Infrastructure - None or Existing? The cost of the first development in an area may
DRAFT VERSION 2.0
be high if export pipeline and terminal facilities are required. With no guarantee of
subsequent developments, appraisal requirements may be greater to reduce risks on
the higher cost of greenfield facilities. In contrast, second stage developments may
have the benefit of existing pipeline networks or facilities, and even small satellite
fields may be tied-in economically. In these circumstances, little or no appraisal of a
discovery may be required e.g. tie-in of discovery well for immediate production to
nearby field.
However, it is often the owners of the first stage developments and infrastructure who
make better economic returns, as they often benefit from lucrative pipeline tariffs and
extended field life from the early fields.
Kutubu, PNG - the 300 MMstb Kutubu oil development, in the Papuan
highlands, was Papua New Guinea’s first petroleum project. At the time of
project sanction, no other commercial discoveries had been made and the
project therefore had to bear the full cost of infield facilities, 400km
overland pipeline to the Gulf of Papua, and marine export terminal. The
development cost of $1 billion was supported by a high level of confidence in
reserves from over 30 exploration and appraisal wells on the structure.
Cooper/Eromanga Basin, Australia - the 40 MMstb Jackson oilfield in

southwest Queensland is one of Australia’s largest onshore oilfields
supplying oil via pipeline to a refinery in Brisbane. Following the
development of Jackson, many small 1-5 MMstb discoveries in the area were
successfully developed, with production from the discovery well transported
via road to Jackson for processing and export.
Timing - Single Stage or Phased Development? Many development schemes, usually

larger ones, require the building and installation of all facilities prior to first
production. In some circumstance, particularly onshore, a field may be developed in
phases, with limited initial expenditure, and the benefit of production history before
committing further capital. In these cases, full field appraisal may not be complete
before the initial development decision is made.
Don/Deveron, UK - Don and Deveron are small, complex oilfields located

close to the giant Thistle oilfield in the northern North Sea. Early appraisal
wells had shown the fields to be highly faulted, with a number of isolated
reservoirs containing a variety of fluid types. Extensive appraisal to resolve
these uncertainties would erode the value of these marginal fields. Instead,
they were developed in several phases to mitigate the outstanding concerns.
The initial development comprised an extended reach platform well on
Deveron and two subsea wells on Don. The Deveron well was a success, and
was followed up by two further wells to boost production. Although early
production from Don was encouraging, economic rates could not be
sustained and the second phase of the development was cancelled.
Economy - High or Low Price Forecasts? The current economic climate, and
forecasts of future market demand and price forecasts can have significant influence
on appraisal requirements. These may influence the assumptions we make in our
economic analysis, and the amount of capital available for both appraisal and
20
DRAFT VERSION 2.0
development activities. Although technical appraisal requirements do not change with
economic climate, the pressure to limit appraisal spending in times of low prices or
demand may be considerable. The irony of the economic cycle is that fields which
were appraised during high oil prices, often realise low prices during peak production
and vice versa.
Hutton, UK - the Hutton oilfield in northern North Sea was appraised and
developed in the early to mid 1980’s, during a period of high oil prices. The
development used the world’s first tension leg platform (TLP) and capital
costs were very high. The field came onstream in 1986, one year before a
major and sustained oil price fall. Full life cycle economics for the field are
marginal to subeconomic and had the field been considered for development
a few years later, the project would not have gone ahead.
Ettrick, UK - the Ettrick oilfield is a small, marginally economic oilfield in

the central North Sea, which was extensively appraised during the early
1980’s. The project was considered for development during the late 1980’s
- when oil prices ranged from $10-14/bbl - and was rejected, requiring
higher to make an acceptable return on investment. Had the project gone
ahead, the field would have delivered most of its production during the mid
1990’s, when average oil prices were well above the economic threshold.
These differences are highlighted by comparing the hypothetical appraisal of two

similar discoveries, in terms of size, risks and uncertainties.
Discovery A is an onshore oil field in an established producing area, 15 km from the

nearest oil processing and export facilities. Current reserves estimates range from 100
to 1000 MMstb. The operator has opted for a phased development option, with an
initial 50 Mbd production facility with the potential to add further capacity up to a
maximum of 300 Mbd. Recent and forecast oil prices are high due to constrained
supply.
The appraisal requirements for Discovery A are minimal, and the initial development
may be economically robust after only two or three wells. Further appraisal will be
required before or during the development of later phases.
Discovery B is an offshore gas field in a undeveloped basin, 500 km from shore.

Current reserves estimates range from 600 to 6000 Bscf. The operator has opted for a
single stage LNG project and secured a 15 year, 3 mtpa export contract. Recent and
forecast gas prices are low due to gas market oversupply.
In the case of Discovery B, appraisal requirements are signficant as the operator must
prove up sufficient reserves to met the gas sales contract, and ensure that downside
scenarios remain economic despite the high capital cost of the development. It may
take 10+ wells to reach this level of confidence. This may be difficult to justify in the
prevailing economic climate.
21
DRAFT VERSION 2.0
2.6 Appraisal Costs
The cost of appraising a discovery can vary considerably, depending on the types of
data required, geographic location and the level of confidence needed to reach a
development decision.
Technical information obtained during appraisal can be broadly divided into two
categories - remotely sensed and direct information.
Basic remotely sensed information, such as satellite images or airborne gravity and
magnetics data may be acquired cheaply - typically less than $0.2 million - and may be
useful in determining the structural size and complexity of a discovery in onshore
environments.
Reflection seismic data is the most common form of remotely sensed data used to
appraise hydrocarbon discoveries. The scale of seismic operations can range from a
few infill 2D lines to closely spaced 3D surveys covering the entire area. The cost of
seismic acquisition depends not only on the size of the survey, but the environment in
which the data is acquired. Small onshore surveys in easily accessible terrain may cost
only a few hundred thousand dollars. Offshore, acquisition is generally easy and costs
are generally related to the size of the survey and the level of activity and
competitiveness in the local area - costs can vary from less than $1 million to more
than $10 million. In rugged or inhospitable onshore areas, the acquisition of 3D
seismic data can be time consuming and expensive, costing $20+ million in some
instances.
Reprocessed seismic data may also be considered an appraisal tool, as the enhanced
imagery may significantly reduce risk or uncertainty. For example, the number of
appraisal wells required may be less, if a predictive reservoir model based on seismic
attributes is confirmed by later well results.
Direct information is obtained from wells, and includes information which can be
used to refine our understanding or both the static, or in-place, resource and the
dynamic, or production, potential.
Well costs, like seismic, vary greatly depending on the environment, well length and
amount of data acquired. Onshore appraisal wells may be drilled to shallow objectives
for less than $0.5 million. In more remote and rugged areas, where logistic support of
drilling operations is a large proportion of costs, and in areas of complex geology,
onshore wells can cost in excess of $30 million. In offshore environments, drilling of
simple shallow wells as part of an ongoing programme may cost $1-2 million.
Contracting a specialist rig to drill a deep high pressure and/or temperature appraisal
well in a remote offshore area can cost upwards of $30 million.
The amount of data acquired in a well also influences appraisal costs, although in
most instances the cost of data acquisition is a fraction of the cost of drilling to the
target.
Whilst it is important to consider ways of minimising data costs - only acquiring
22
DRAFT VERSION 2.0
information which will significantly impact the development decision - it is important
to recognise the prohibitive expense of returning to acquire information which was not
collected in earlier wells.
Well data types which are commonly used to improve the static reservoir description
include well logs, core and biostratigraphic data, fluid samples and static pressure
data. The dynamic reservoir description can be improved with laboratory core
measurements and production data from MDT and well testing.
However, well test data can be expensive to acquire - a short term drill stem testing
(DST) can add $1-5 million to the cost of a well. In complex reservoirs, a more
accurate estimate of long term productivity may be gained from extended well testing
(EWT). The ability to perform an EWT is influenced by the location of the discovery
and local environmental regulations.
In onshore oil and gas fields, temporary connection of the well to nearby infrastructure
may permit extended testing. Offshore, EWT has been limited to oil reservoirs, with
production via a contracted testing, storage and offlaoding tanker. Although this can
be a very expensive operation, the cost of an EWT is generally offset by the sale of
production.
In the early stages of appraisal it is relatively easy to justify appraisal expenditure

given the large reduction in risk or uncertainty and, if successful, large improvement
in project value which additional information will provide. In later stages of appraisal,
as each new piece of data has less influence on the development concept and
economics, it becomes important to recognise the point at which further appraisal
expenditure may erode the value of a project.
If a development does not proceed, usually due to unacceptable technical, commercial

or political risks, the cost of appraising the field may have to be written off - although
many fiscal regimes provide tax or other financial concessions for expenditure
writedowns.
23
DRAFT VERSION 2.0
3. Methods for Improving Appraisal Decisions
Making appraisal decisions - which well(s) to drill, what data types to acquire - is a
subjective process. It involves judgements based on experience, knowledge and
technical analysis.
Traditionally, we have used our judgement to assess the impact of new technical data
on our geological models, reservoir description, estimates of resource size etc. Often,
gaps in technical knowledge are filled simply to improve our understanding, without
considering whether this significantly impacts the choice or design of development
concept, or project economics.
In many cases, our decisions have become too subjective, focussed on technical
information and knowledge, and considering only the immediate future - not the entire
appraisal process.
There are a number of simple methods, both qualitative and quantitative, which may
be used to improve the quality of appraisal decisions we make. These methods
emphasise the need for a clear appraisal strategy, simple quantitative analyses to
support our decision making and measures to evaluate the success of appraisal
activities.
3.1 Appraisal Strategy
At the end of the appraisal process, we aim to have an economically robust

development ready for project sanction and execution. To develop an effective
appraisal strategy we must clearly understand where we stand today, what is required
to achieve our goal, and some measure of our progress along the way.
Maturity Crossplot
Traditionally, we categorise the current status of field appraisal with a few generic
terms - post-discovery, early appraisal, late appraisal - and describe the level of risk
and uncertainty by estimating a reserves range using deterministic or probabilistic
volumetrics. Appraisal requirements are generally determined by assessing what
technical information is needed to address the major risks and uncertainties. We tend
to look one step ahead at a time, and use experience to judge which well location will
have the biggest impact on the in-place or reserves range.
A simple method of portraying the current status of a discovery is in terms of its

development concept maturity and economic maturity, as described previously. By
plotting each discovery on a simple crossplot of the two measures (Figure 7), we
create a clear understanding of where we stand today and the direction we must move
in to reach a development decision.
In most cases, the preferred development concept for a discovery is selected before the
project is economically robust. On the maturity crossplot, most fields tend to move
vertically, then to the right. Of course, in acquiring the information needed to select a
development concept, we may also improve the economics of the project by finding
larger volumes or adding proven reserves.
24
DRAFT VERSION 2.0
The illustrated example plots the maturity of a number of undeveloped discoveries in

BHPP portfolio. In some cases, further appraisal must focus on both selecting a
development concept and improving economics e.g. Neptune. In other cases, such as
Moran, the development concept is well defined and the project requires additional
volumes or greater confidence in the discovered volume to proceed.
Developing an Appraisal Strategy
The strategy we adopt in appraising a discovery often depends as much on external,

non-technical factors as it does on technical considerations and experience. Influences
such as economic climate, corporate priorities - both our own and our partners - and
political factors must be considered.
In general, appraisal strategy consists of two elements - direction and pace.
The direction, or focus, of our appraisal efforts depends on the business goal we are
trying to achieve, and may change over time. A clear vision of the desired outcome at
the beginning of the appraisal process is essential - standalone oil rim development
within five years, low cost tieback to existing facilities within two years, prove
significant gas potential within three years, or exit area.
The pace of appraisal, whilst inherently linked to our business goals, is more often
driven by budgetary considerations, partnership issues, rig availability and legislative
requirements. In some areas of the world, operators must appraise a discovery and
declare commerciality within a relatively short period - or lose the acreage.
Colombia - fiscal terms in recently awarded licences to explore foldbelt

plays in the Colombian foothills include a limited five year period to declare
field commerciality. Only proven volumes within a given radius of successful
wells are considered - inferred volumes updip from a dry hole are not
include in the commercial area. This regime has required back-to-back
appraisal drilling with several rigs - to ensure sufficient areal well coverage
- and few downdip delineation wells - to ensure few dry holes.
In these circumstances, appraisal activity may be fast paced and not always optimal. In
other cases, lack of strategic direction and alignment between partners may delay
appraisal, eroding the potential value of any project.
Some common appraisal strategies are briefly described below.
A Conventional appraisal strategy is not driven by commercial or legislative pressures

to reduce cycle time. Acquisition of additional information is carefully planned, with
sufficient time between activities to fully evaluate data. 3D seismic data is generally
acquired after one or two wells. A discovery well on the crest of the structure is
usually followed by a downdip well to delineate fluid contacts, and lateral stepouts to
evaluate the areal extent of the field. Appraisal tends to be thorough, with all major
technical uncertainties addressed before a development decision is made.
A Fast Track appraisal strategy is driven by commercial or legislative pressures to
25
DRAFT VERSION 2.0
reduce cycle time. This strategy is often adopted in proven areas, with a good
knowledge of local geology and existing, nearby infrastructure providing the
opportunity for low cost development. In some cases, fast track developments may not
be appraised at all, with simple tie-back of the discovery well to adjacent facilities.
Major risks and uncertainties may not be fully understood or addressed, although the
potential rewards from early development may outweigh these risks.
If fast track appraisal is forced by licence terms, an operator may speed up appraisal
by drilling wells simultaneously, and acquiring 3D seismic early. This approach may
be inefficient as the results from one well may have influenced the location or
objectives or another.
Recently, there have been a number of fast track developments which have proven to
be sub-optimal or unsuccessful because key data e.g. 3D seismic was not acquired
prior to the development decision.
Cossack, Australia - the 200 MMstb Wanaea oilfield, offshore northwest

Australia, was discovered in 1989. The smaller Cossack oilfield, located just
5km to the northeast, was discovered in 1990. In early 1991, with just one
well and recently acquired 3D seismic data still being processed, the field
was sanctioned for fast track development using a converted tanker FPSO
facility to deliver early production prior to development of Wanaea.
During project execution, interpretation of the 3D dataset revealed a much

smaller, more faulted structure. Two appraisal / development wells were
hurriedly drilled on the flanks of the field, both of which encountered the
reservoir deep to prognosis. Reserves were downgraded from 80 MMstb to
30 MMstb, and fast track development eventually cancelled. Cossack has
been subsequently developed as an integral part of the Wanaea project.
A Sales Contract appraisal is driven by the need to prove a specific volume of

reserves, usually gas, to enter into a long term sales contract. In many cases, a large
volume of gas is required and the early stages of appraisal are focussed on determining
whether the discovery is sufficiently large to be attractive. If early signs are positive,
the later stages of appraisal are focussed on proving up P90 reserves as efficiently as
possible. If the discovery is too small, exploration and appraisal of other structures in
the area may be needed, and gas gathering infrastructure developed.
NW Shelf Project, Australia - the NW Shelf Project exports over 7 mtpa

LNG to Japanese energy companies. These buyers demand security of
supply over the 20 year period of the sales contract, and a high degree of
confidence in gas reserves was required prior to signing a sales contract in
the mid 1980’s. During early appraisal, reserves from two giant gasfields,
North Rankin (8 Tscf) and Goodwyn (4 Tscf), were thought to be sufficient to
meet a gas sales contract. However, in order to prove P90 reserves of 10-12
Tscf, both fields were extensively appraised.
The nearby Angel discovery, which is still not developed, was also appraised
to meet the proven reserves target and provide some form of contingency.
26
DRAFT VERSION 2.0
An Equity appraisal strategy is driven by the need to establish and maximise equity
position in a discovery which is located in one or more licence areas. Generally, the
initial discovery in one licence will be rapidly followed by a well in the other licence
aimed at confirming the mapped extension, and giving the second operator a
negotiating position. Although early appraisal may be done independently, later stages
are often planned and executed jointly, as part of the unitisation process. The location
of, and data acquired in, appraisal wells may have an impact on equity split and both
operators will aim to acquire information in areas which they perceive will strengthen
their position, and not drill to reduce uncertainty in areas which may weaken equity.
These factors must be recognised when making a development decision, because key
uncertainties may not have been addressed during appraisal due to equity sensitivity.
There is also a risk that excessive equity driven appraisal might erode overall project
value.
Bayu-Undan, ZOCA - the Bayu-Undan gas condensate field is located in

the Zone of Cooperation (A) between Indonesia and Australia. The field was
discovered by well Bayu-1 in licence PSC 91-13, which was rapidly followed
by well Undan-1 in adjacent licence PSC 91-12. Despite eight further
appraisal wells, aimed largely at resolving major technical uncertainties
and risks, large uncertainties in in-place volumes and areal extent of the
field remain. On the flanks of the field, further appraisal to resolve these
issues, either prior to or during development, is unlikely as negative results
may provide conclusive data which impacts the equity position of either
PSC.
A Long Term appraisal strategy is driven by market demands and current technology
limits. In some cases, a discovery may be economically feasible or viable, but cannot
be developed due to lack of market or immature technology - development may not be
possible for five to ten years or more. Ideally, no appraisal should be undertaken until
a clear market opportunity or technological breakthrough occurs. However, licence
terms may require a specific level of expenditure or activity, and appraisal may occur
in a number of phases over many years. This often leads to frequent review of data,
without any real gains, and loss of corporate knowledge as the team working the field
changes.
Sunrise-Troubadour, Australia - the Sunrise-Troubadour area, offshore

northern Australia, contains significant gas condensate resources which
have been intermittently explored and appraised since the 1970’s. The fields
lies in a remote area, with no existing infrastructure, and development has
been hindered by technical risks and high costs in a competitive Asian gas
market. However, the combination of proximity to market, low retention
costs and low political / fiscal risk make these fields an attractive long term
option as demand for energy in the region is predicted to grow.
27
DRAFT VERSION 2.0
A Development Utility appraisal strategy is driven by a desire to minimise
development costs and re-use existing exploration and appraisal wells in field
development. This strategy is often adopted in remote or inaccessible areas with high
well costs. Whilst this approach may lead to a cheaper development, it can also
compromise the appraisal programme - in terms of well locations and data acquisition.
For example, downdip wells to test the hydrocarbon-water contact may not be drilled
as these wells often have little development utility. In wells which are targetted as
future producers, well test data may not be acquired as this could affect the
completion and/or suspension design.
Kutubu, PNG - the Kutubu oilfield complex in the remote highlands of

Papua New Guinea was appraised by over 30 wells, costing $5-15 million
each. In early appraisal, no conscious effort to locate wells for future
development use was made. However, in the later stages of appraisal a
number of wells were relocated or sidetracked to maximise their utility as
producers. Over 40% of appraisal wells have been re-used in development.
A Near Field Exploitation appraisal strategy is driven by a desire to improve or

extend production from an existing field by developing additional volumes from
nearby discoveries. It is important to undertake the exploration and appraisal of
adjacent structures early enough to ensure their development before end of mainfield
life is reached. This is not always easy to achieve as the focus and impetus to develop
satellite resources may not come until production is in steep decline. Potential tie-ins
may also be owned by other parties with different priorities
Perseus, Australia - the giant North Rankin and Goodwyn gas condensate
fields, which supply gas from Triassic reservoirs to the NW Shelf LNG
Project, were discovered and appraised during the 1970’s and 1980’s.
North Rankin began production in 1984, and in 1991 an extended reach
platform well was drilled to appraise a Jurassic gas accumulation in the
graben between the two fields which was discovered during early appraisal.
Estimated reserves were 0.5 to 1.0 Tscf. The well was successful, and
brought immediately on stream, providing valuable production data over the
next four years. Over this time, reservoir pressure declined slowly and the
potential for significant upside volumes was recognised. Four appraisal
wells were drilled in the mid 1990’s, proving up a resource of 5.0+Tscf
which could now drive future NW Shelf Project expansion.
1/3-3 Discovery, Norway - the 10-20 Mmstb 1/3-3 discovery lies 8km north
of the 200 MMstb Gyda field, central North Sea. It was discovered in the
1980’s, and no further appraisal was undertaken due to its small size. Gyda
started production in 1990, and after an erratic two year plateau entered
steep decline in the mid 1990’s. Only then, did attention focus on the
possibility of further appraisal and subsea tie-back of 1/3-3. Unfortunately,
the opportunity to capture these reserves may have been missed due to the
long lead time between a development decision and peak production, and
the impending end of Gyda field life.
28
DRAFT VERSION 2.0
A Field Cluster , or Area Wide Development, appraisal strategy may be applied to
undeveloped resources where there is insufficient volumes in any one structure to
justify economic development. In these cases, appraisal of two or more structures may
occur in parallel, reducing the cycle time from discovery to development, but also
compromising the ability of one appraisal programme to learn from, or influence, the
other. The scale of field clusters can range from joint development of two or three
small, adjacent fields to regional hydrocarbon gathering projects.
Elang-Kakatua, Australia - the 10 MMstb Elang oilfield, offshore northern

Australia, was discovered and appraised in the mid 1990’s. The field has
insufficient reserves for standalone development, but with an additional 5
MMstb volume discovered at nearby Kakatua, the cluster is to be developed
using a leased FPSO vessel. It is expected that the 1 MMstb will also be
developed using these facilities.
ETAP, UK - the Eastern Trough Area Project (ETAP) is a $2.6 billion

dollar development of oil, gas condensate and gas fields in the central North
Sea. It involves 3 operators and 8 fields, none of which was economically
developable on its own. Many of the larger gas fields were discovered and
appraised in the 1970’s and early 1980’s, but it was only with the addition
of significant liquid volumes from a number of smaller discoveries in the late
1980’s that the project became economically viable.
A Production Led appraisal strategy is driven by the need to understand the

production history and future potential of a producing asset. Although we usually
think of appraisal as an activity which precedes development, extensive appraisal may
continue throughout the life of a field. In most cases, production led appraisal is
linked to improving or extending production from an asset which we have owned for
many years. However, it may also apply to producing or abandoned fields which have
been acquired for field redevelopment. Appraisal may involve the deepening or
sidetrack of existing wells to evaluate the potential of long ignored reservoirs, or
extensive well intervention, workover and testing to understand the habitat of
bypassed hydrocarbons.
Pedernales, Venezuela - Pedernales is a giant oilfield in eastern Venezuala,

containing over 1000 Mmstb of heavy oil, which has been inefficiently
developed - producing only 60 MMstb from over 60 vertical wells in 60
years. Redevelopment of the field is underway, involving workover and long
term testing of existing wells, and drilling of horizontal appraisal /
production wells to evaluate productivity. If successful, the field will be
redeveloped using modern well construction and completion technologies.
29
DRAFT VERSION 2.0
Beatrice, UK - Beatrice was one of the first North Sea oilfields to be developed,
located just a few kilometres offshore northeast Scotland. Oil is produced from
Middle Jurassic reservoirs, although an early appraisal well had also encountered
hydrocarbon shows in deeper Devonian sands. Late in field life, and following a
number of successful Devonian discoveries in adjacent licences, an abandoned
production well was deepened to appraise Devonian potential. Unfortunately, the well
failed to encounter commercial volumes in the Devonian.
Ula, Norway - the Ula oilfield contains 450 Mmstb trapped in Jurassic
sands in a simple dome structure. The discovery well encountered an oil
column in deeper Triassic sands on the crest of the field, although downdip
appraisal wells proved the accumulation to be of limited size and extent. The
development of coiled tubing drilling technology has dramatically lowered
the cost of infield appraisal and development, and in 1996 a production well
was deepened to evaluate Triassic reservoir continuity and productivity.
Scale and Confidence Appraisal
For most appraisal strategies, there are two distinct phases of appraisal - an early stage
aimed at establishing the scale of the resource and, if successful, a later stage designed
to increase confidence in the proven resource leading to a development decision
(Figure 8).
Scale appraisal is typified by an incomplete understanding of the major factors which

dictate the resource size and hence, the choice of development concept. Many
segments remain untested and fluid types or contacts may be unknown or ill defined.
Success or failure in any one well can dramatically alter our perception of reserves,.
As a result, this period of appraisal is characterised by a fluctuating reserves range and
unstable mean volumes.
The primary aim of scale appraisal is remove those major uncertainties which control
the gross size of resource, and therefore constrain or stabilise mean volumes. It is not
necessarily aimed at increasing proven volumes or reducing the reserves range.
In mature basins, where the local geology and expected field sizes are well
understood, scale appraisal may require only one or two wells to confirm the expected
model. In new or poorly explored areas, many wells may be required before the scale
of the resource is defined - particularly in areas of structural or reservoir complexity
Confidence appraisal is characterised by a generally narrowing reserves range and

more stable mean volume.
The primary aim of confidence appraisal is to establish the economic viability of the
resource, increasing proven reserves in the most efficient manner - leading to a
reduction in appraisal well numbers and an earlier development decision.
As in any appraisal campaign, surprises will occur and our perception of mean
reserves and associated range alter. However, the magnitude of these fluctuations
30

Scale and Confidence Appraisal
Scale
Scale Confidence
Confidence Development
Development
Large
Risked
Reserves /
Value
Medium
Small Decision Decision
No. of Wells / Time / Cost
C:/appraisal.ppt PERT - Australia/Asia Region

BHP Petroleum
should be significantly less than in the early stages. Larger changes during confidence
appraisal imply that scale appraisal issues were not fully addressed.
3.2 Appraisal Measures
Being able to measure, both predictively and retrospectively, the impact of appraisal
activities is a critical part of the decision making process. These measurements can be
based on:
In-place Volumes - provide a simple way of estimating the impact of appraisal,

especially if there is limited information or there are manpower or time constraints.
This approach may be used for screening studies, but has limited use in decision
making as it takes no account of recovery from the field, or production rates.
Reserves - is the most commonly used basis for measuring the impact of appraisal. It
represents a reasonable compromise between the need to have some understanding of
the recoverable volumes, to consider development concepts, and the additional time
and effort required to generate production or value
Production Profiles - are rarely used to measure the impact of appraisal. If production
profiles are generated, there is little additional effort required to perform an economic
evaluation. Hence, most studies will either use reserves or value as the basis of
measurement.
Economic Value - allows us to assess the full impact of appraisal - on facilities costs,
cycle time etc. - as well as on reserves or production. Although this approach requires
additional manpower and time, all major appraisal decisions should be supported by
and reviewed using key economic measures - changes to net present value, rate of
return, capital efficiency etc..
Reserves Accounting
Sound reserves accounting has become increasingly important as many of our

corporate performance measures are based on volume additions through exploration
and appraisal. Within the industry, many different ways of calculating and classifying
reserves have developed. Some operators use only deterministic calculation methods
and adhere strictly to US SEC or SPE reserves classifications. Others use a variety of
probabilistic calculation methods, and have developed unique in-house reserves
classifications and reporting guidelines.
Within BHPP, we increasingly use common probabilistic methods to estimate the

reserves range, but have no standard approach to defining the volumes attributable to
each exploration or appraisal well. A recurring problem has been the “overbooking”
of reserves attributable to early appraisal wells, which can require a decrease in
booked volumes, if the results of later wells are below expectation.
By adopting a standard step-wise approach to reserves accounting during appraisal we

can facilitate comparison between discoveries, and ensure progressive growth in
booked reserves.
DRAFT VERSION 2.0
Segmentation
Segmentation , or the division of a field into a number of geographic areas, can be

used as a basis for reserves accounting and to plan the appraisal programme.
The basis for subdivision of a discovery into segments (regions or areas) is usually
related to structure or anticipated well spacing, although other criteria such as simple
geometry, areas of common risk or cultural features are equally valid (Figure 9).
Segmentation may be both areal and vertical, with over- or underlying reservoirs
which have common features sharing the same segments (Figure 10).
There are no hard and fast rules regarding the definition or number of segments
(Figure 11), although too many segments - more than eight to ten - can result in
complicated calculations and a tendency towards a narrower reserves range when
added.
The goal is to divide the field into a workable number of geographic areas which can
be systematically appraised, and reserves volumes added.
Each segment carries a risk - the chance of success - in addition to an uncertainty in

the range of possible volumes. The definition of success can vary, but is usually
defined as the discovery of hydrocarbons which are likely to have commercial
significance in the foreseeable future.
Risk is expressed as a value between 1.0 (proven success) and 0.0 (proven failure).
Risked Reserves
Reserves estimates for a segment or discovery may be made using deterministic or

probabilistic methodologies.
Deterministic volumes are based on one or more discrete scenarios in which one value
is assigned for each input parameter, producing a single reserves volume for each
scenario. The inputs which we consider to be our best estimates are combined to
calculate deterministic most likely reserves. One or more of these most likely input
parameters may be systematically varied to create a number of deterministic upside
and deterministic downside cases.
Probabilistic volumes are based on statistical, or Monte Carlo, modelling in which a

range of values is assigned for each input parameter, producing a range of reserves
volumes. Each input parameter is described by a number of discrete values and
associated probabilities. A parameter distribution or probability density function is
statistically fitted to these data to create a continuous range of values. Different
distributions may be used to fit the data, including normal, lognormal and beta
functions. The choice of function can significantly impact the reserves range, and care
should be taken to ensure the fitted distribution is appropriate. The parameter
distributions are combined in a Monte Carlo model, which samples each distribution
and calculates volume many times - typically 1000 to 10000 - to generate a range of
possible reserves. Deterministic estimates should be used to constrain and sense check
the results of Monte Carlo modelling.
33

Basis of Segmentation
2
1 2 7 5
1 3
2 1 3
4 5 6
4 4
3
Structural Well Spacing
Poor Seismic Imaging
+ +
2 Backlimb Good Seismic Imaging
+ + + +
No Seismic Imaging
Plunge
1 3 Plunge + + + + +
5 Crest + + 1
+ 1+ 2 3
+ + + + Possible
4 Forelimb
+ +
Reservoir
Pinchout
Geometric Common Risk

PSC - A PL 42
m
200
11 1 3
2 2
0m
3 PSC - A
20
PSC - B
PL 42 PSC - B PL 42
Offshore
Onshore
Economic Regulatory

BHP Petroleum
DRAFT VERSION 2.0

Areal and Vertical Segmentation
Map View 3-Dimensional View

6
1 6 3
2 2 5
1 2 3 1
4 3
5
4

BHP Petroleum
35
DRAFT VERSION 2.0

Reality... and if all else fails
Good Seismic Imaging

Poor Seismic Imaging
+ +
+ + + +
No Seismic Imaging
+ + + + + 6
2 4
1 7
+ + + 1 +
+ + + +
+ + 5
33
Offshore
Onshore PSC - A PL 42
1 2 3
4
Combination
5 7
6
Arbitrary

BHP Petroleum
36
The reserves range is described by an expectation or exceedence curve which plots
volume (V) against the probability of exceeding volume V- from the minimum value
(P100) to the maximum (P0).
Unrisked volumes do not include any risk associated with an undrilled segment, and
are used to describe the potential reserves for each segment.
Risked volumes are calculated by combining the unrisked reserves with the segment
risk. This is, in effect, a discounting process to take account of the possibility that the
segment may contain no hydrocarbons.
In deterministic calculations, the risked volume is simply the unrisked value

multiplied by the risk:
Unrisked Reserves = 100 Mmstb

Risk = 0.50
Risked Reserves = 50 Mmstb
In probabilistic calculations, the expectation curve is adjusted to reflect the risk of

failure i.e. V = 0, and the probability distribution of the entire reserves range (P100 to
P0) factored, or compressed, to fit the remaining probability (Figure 12).
Volumes at specific confidence levels, or percentiles, are usually quoted to describe

the reserves range.
Unrisked reserves percentiles are identified by the prefix “P” and are usually quoted at
P90, P50 and P10 levels, although P85 and P15 volumes are quoted in preference to P90
and P10 in some companies. Mean (average), median (middle or P50) and mode (most
likely or ML) values are also used.
Risked reserves percentiles are identified by the prefix “RP” and are usually quoted at
RP90, RP50 and RP10 levels. Risked mean Rmean, risked median Rmedian or RP50, and
risked mode RML, values are also used.
Both unrisked and risked expectation curves for two or more segments may be
combined using probabilistic addition to determine the reserves range for part or all
of a discovery. In probabilistic addition, the level of dependency between segments
must be included - the degree to which a higher or lower outcome in one segment will
result in a similar outcome in another. Dependency can vary from total positive (+1.0)
to neutral (0.0) to total negative (-1.0).
If no dependency is applied, outcomes will be added together randomly (Figure 13).
In total positive dependency, high outcomes will always be combined with high
outcomes, and vice versa - the resulting reserves range will be wider, and equivalent

Unrisked and Risked Expectation Curves

100 100
Failure
UNRISKED RISKED
Risk 1.00 Risk 0.70
Success
50 P90 250 50 RP90 0
Success
P50 450 RP50 350
P10 800 RP10 750
0 0
0 500 1000 1500 0 500 1000 1500
100 100
RISKED RISKED
Failure
Risk 0.30 Risk 0.05
Failure
50 RP90 0 50 RP90 0
RP50 0 RP50 0
RP10 550 RP10 0
Success
Success
0 0
0 500 1000 1500 0 500 1000 1500

BHP Petroleum
DRAFT VERSION 2.0

Independent Addition (0.0)

100 100
RISKED RISKED
P73 Risk 1.00 Risk 0.70
P64
50
P45
RP90
RP50
RP10
250
450
800
+ 50
P42
RP90
RP50
RP10
0
250
500
P13
P10
0 0
0 500 1000 1500 0 500 1000 1500
100
RISKED
Risk 1.00
= 50 RP90
RP50
RP10
400
700
1100
0
0 500 1000 1500

BHP Petroleum
39
DRAFT VERSION 2.0

Total Positive Dependent Addition (+1.0)

100 100
RISKED RISKED
P73 P73
Risk 1.00 Risk 0.70
50
P45
RP90
RP50
RP10
250
450
800
+ 50
P45
RP90
RP50
RP10
0
250
500
P10 P10
0 0
0 500 1000 1500 0 500 1000 1500
100
RISKED
Risk 1.00
= 50 RP90
RP50
RP10
250
700
1300
0
0 500 1000 1500

BHP Petroleum
40
DRAFT VERSION 2.0

Total Negative Dependent Addition (-1.0)

100 100
RISKED P90 RISKED

P73 Risk 1.00 Risk 0.70
50
P45
RP90
RP50
RP10
250
450
800
+ 50
P55
RP90
RP50
RP10
0
250
500
P27
P10
0 0
0 500 1000 1500 0 500 1000 1500
100
RISKED
Risk 1.00
= 50 RP90
RP50
RP10
600
700
950
0
0 500 1000 1500

BHP Petroleum
41
to arithmetic addition (Figure 14).
With negative dependency, high outcomes will be combined with low outcomes, and
vice versa - the resulting reserves range will be narrower (Figure 15).
Discoveries may also be combined in this way to estimate the reserves range for a
complex of fields or an entire basin.
Definitions
The potential volume is the unrisked deterministic most likely reserves for a segment,
estimated prior to drilling.
The discovered volume is the deterministic most likely reserves for a segment,
estimated after drilling. As each segment is drilled out, discovered volumes are added
together arithmetically to build up the discovered volume for the field. When all
segments are drilled out, the discovered volume will equal the deterministic most
likely reserves for the field.
The proven volume is the P90 reserves for a segment, estimated after drilling. As each
segment is drilled out, P90 volumes are added together probabilistically to build up the
proven volume for the field. When all segments are drilled out, the proven volume
will equal P90 reserves for the field.
The risked proven volume is the risked RP90 reserves for a segment. RP90 volumes for
all segments, drilled and undrilled, are added together probabilistically to estimate the
risked proven volume for the field. This takes into account the probability that some
volumes are present in undrilled segments. When all segments are drilled out, the
risked proven volume will equal P90 reserves for the field.
The definitions used in our reserves databases differ from the above, and are based on
US Securities Exchange Commission (SEC) definitions of proven (1P), probable (2P)
and possible (3P) reserves. However, probabilistic volumetric results may be broadly
equated to these definitions as follows:
DRAFT VERSION 2.0
US SEC Reserves BHPP Probabilistic Reserves
Proven 1P Risked Proven RP90
Probable 2P Risked Mean RMean
Possible 3P Risked Upside RP10
These definitions ensure that both discovered and proven volumes will be
progressively added, and that adverse drilling results will not require a major
downward revision in volumes. This is illustrated by using worked examples for a
hypothetical oil field.
Example 1 - Mid Case Outcome (Figures 16-18)
This example shows a 100 MMstb oil prospect, with a predrill reserves range of 25 -
100 - 400 MMstb. The prospect is divided into four structural segments, based on
seismic mapping, each of which is assessed to have a chance of success (risk) of 0.25.
The pre-drill risked reserves range is 6 - 25 - 100 MMstb. Obviously, no volumes

have been discovered or proven.
Well 1 is drilled in the central segment, and discovers oil. The post-drill reserves for
this segment is estimated to be 15 - 40 - 130 MMstb, although the oil-water contact
has not been confirmed in the well. The segment now has a risk of 1.0, and the risks in
other segments have been adjusted due to the results of well 1. The discovered volume
is 40 MMstb - the mean reserves for the drilled segment, although the proven volume
is only 15 MMstb.
Well 2 is drilled downdip of the first well, in the same segment, to delineate the oil-
water contact. Following the well, the discovered volume remains unchanged, but the
proven volume has increased to 25 MMstb.
Wells 3 to 5 are each drilled in a new segment, adding discovered and proven volumes
on each occasion. The total discovered and proven volumes at the end of appraisal are
100 MMstb and 65 MMstb respectively. At this time the P90, RP90 and proven
volumes are the same.
Example 2 - Upside Outcome (Figures 19-21)
This example uses the same prospect, with the same discovery well results as
Example 1.
43
DRAFT VERSION 2.0
Well 2, however, finds a deeper than expected oil-water contact, increasing the
discovered volume to 60 MMstb, and proven volume to 35 MMstb.
Wells 3 to 5 are each drilled in a new segment and confirm the existence of a deeper
oil-water contact across the field. The total discovered and proven volumes at the end
of appraisal are 150 MMstb and 90 MMstb respectively. Again, the final P90, RP90
and proven volumes are the same.
Example 3 - Downside Outcome (Figures 22-24)
This example uses the same prospect, with the same results as Example 1 in the first
three wells.
Well 4, however, fails to encounter hydrocarbons. The discovered volume remains

unaltered at 65 MMstb. Note how, in this example, the proven volume actually
increases slightly due to better control in depth conversion and reservoir distribution
the well provides despite its failure. The risk on the undrilled segment has also
changed.
Well 5 also fails to encounter hydrocarbons, and the discovered and proven volumes
remain unchanged at 65 MMstb and 45 MMstb respectively. Even though the
unrisked and risked mean volumes for the field have fallen significantly with the
results of later wells, the discovered and proven volumes have been unaffected by the
adverse results.
Example 4 - Mixed Outcome (Figures 25-27)
This example uses the same prospect, with the same results as Example 3 in the first
three wells, with upside outcomes..
Well 4, however, fails to encounter hydrocarbons. Even though the unrisked mean
volumes fall from 150 to 120 MMstb, the discovered volume remains unaltered at 95
MMstb. Note how, in this example, the proven volume actually increases slightly due
to better control in depth conversion and reservoir distribution the well provides
despite its failure. The risk on the undrilled segment has also changed.
Well 5 finds only a small volume of hydrocarbons, further reducing the unrisked mean
to 100 MMstb. However, the discovered volumes increases slightly to 100 MMstb,
adding the small volume which has been found. Proven volume also increases to 60
MMstb, the same as the total P90 and RP90 volume.
In all cases, whatever has happened to our perception of mean field volumes, the
discovered, proven and risked proven volumes have increased or held steady.
44

Example 1: Mid Case Outcome (1) - 100 Mmstb
No Wells Well 1
0.25 0.40
5 - 20 - 50 (15) 5 - 20 - 50 (15)
0.25 1.00
10 - 50 - 150 (40) 15 - 45 - 130 (40)
5 - 25 - 100 (20) 5 - 25 - 100 (20)

5 - 30 - 100 (25) 0.25 5 - 30 - 100 (25) 0.80
0.25 0.80
Unrisked 25 - 125 - 400 Unrisked 30 - 120 - 380
Risked 0 - 0 - 100 Risked 15 - 89 - 310
Discovered 0 Discovered 40
Proven 0 Proven 15
Well 2
0.40
5 - 20 - 40 (15)
1.00
25 - 45 - 100 (40)
Key
5 - 25 - 80 (20)
5 - 30 - 80 (25) 0.80
P90 - Mean - P10 (Det ML)
0.80 RP90 - Rmean - RP10

Unrisked 40 - 120 - 300
Risked 25 - 89 - 244
Discovered 40
Proven 25

BHP Petroleum
DRAFT VERSION 2.0

Example 1: Mid Case Outcome (2) - 100 Mmstb
Well 3 Well 4
0.40 0.50
5 - 20 - 40 (15) 5 - 20 - 35 (15)
1.00 1.00
30 - 45 - 100 (40) 30 - 45 - 100 (40)
5 - 25 - 80 (20) 10 - 20 - 50 (20)
10 - 25 - 60 (25) 0.90 15 - 25 - 50 (25) 1.00
1.00
1.00
Risked 45 - 93 - 248 Risked 55 - 100 - 218
Proven 40 Proven 55
Well 5
1.00
10 - 15 - 25 (15)
1.00
30 - 40 - 60 (40)
Key
10 - 20 - 40 (20)
15 - 25 - 50 (25) 1.00
1.00 RP90 - Rmean - RP10

Unrisked 65 - 100 - 175
Risked 65 - 100 - 175
Discovered 100
Proven 65

BHP Petroleum
46
DRAFT VERSION 2.0

Example 1: Mid Case Outcome (3)

400
Proven
Discovered
350 Mean
P10
P90
RP90
300
Rmean
RP10
Risked Reserves / Value
250
200
150
100
50
0
0 1 2 3 4 5
Number of Wells / Time

BHP Petroleum
47
DRAFT VERSION 2.0

Example 2: Upside Outcome (1) - 100 to 150 Mmstb
No Wells Well 1
0.25 0.40
5 - 20 - 50 (15) 5 - 20 - 50 (15)
0.25 1.00
10 - 50 - 150 (40) 15 - 45 - 130 (40)
5 - 25 - 100 (20) 5 - 25 - 100 (20)

5 - 30 - 100 (25) 0.25 5 - 30 - 100 (25) 0.80
0.25 0.80
Risked 0 - 0 - 100 Risked 15 - 89 - 310
Proven 0 Proven 15
Well 2
0.40
5 - 25 - 40 (25)
1.00
30 - 70 - 120 (60)
Key
5 - 35 - 100 (30)
5 - 40 - 100 (35) 0.80
0.80 RP90 - Rmean - RP10

Unrisked 45 - 170 - 370
Risked 30 - 130 - 300
Discovered 60
Proven 30

BHP Petroleum
48
DRAFT VERSION 2.0

Example 2: Upside Outcome (2) - 100 to 150 Mmstb

Well 3 Well 4
0.40 0.50
5 - 25 - 50 (25) 5 - 25 - 40 (25)
1.00 1.00
35 - 65 - 100 (60) 35 - 60 - 80 (60)
5 - 35 - 80 (30) 15 - 35 - 60 (30)
15 - 40 - 80 (35) 0.90 20 - 35 - 60 (35)
1.00
1.00 1.00
Unrisked 60 - 165 - 310
Risked 55 - 137 - 272 Unrisked 75 - 155 - 240
Discovered 95 Risked 70 - 143 - 220
Proven 50 Discovered 125
Proven 70
Well 5 1.00
15 - 25 - 35 (25)
1.00
35 - 60 - 80 (60)
20 - 30 - 50 (30) Key
20 - 35 - 60 (35)
1.00
1.00
RP90 - Rmean - RP10
Unrisked 90 - 150 - 225
Risked 90 - 150 - 225
Discovered 150
Proven 90

BHP Petroleum
49
DRAFT VERSION 2.0

Example 2: Upside Outcome (3)

400
Proven
Discovered
350 Mean
P10
P90
RP90
300
Rmean
RP10
250
200
150
100
50
0
0 1 2 3 4 5

BHP Petroleum
50
DRAFT VERSION 2.0

Example 3: Downside Outcome (1) - 100 to 65 MMstb
No Wells Well 1
0.25 0.40
5 - 20 - 50 (15) 5 - 20 - 50 (15)
0.25 1.00
10 - 50 - 150 (40) 15 - 45 - 130 (40)
5 - 25 - 100 (20) 5 - 25 - 100 (20)

5 - 30 - 100 (25) 0.25 5 - 30 - 100 (25) 0.80
0.25 0.80
Risked 0 - 0 - 100 Risked 15 - 89 - 310
Proven 0 Proven 15
Well 2
0.40
5 - 20 - 40 (15)
1.00
25 - 45 - 100 (40)
Key
5 - 25 - 80 (20)
5 - 30 - 80 (25) 0.80
0.80 RP90 - Rmean - RP10

Unrisked 40 - 120 - 300
Risked 25 - 89 - 244
Discovered 40
Proven 25

BHP Petroleum
51
DRAFT VERSION 2.0

Example 3: Downside Outcome (2) - 100 to 65 Mmstb
Well 3 Well 4
0.40 0.20
5 - 20 - 40 (15) 5 - 20 - 35 (15)
1.00 1.00
30 - 45 - 100 (40) 30 - 40 - 80 (40)
5 - 25 - 80 (20) DRY
10 - 25 - 60 (25) 0.90 15 - 25 - 50 (25) 1.00
1.00 1.00
Risked 45 - 93 - 248 Risked 45 - 65 - 137
Proven 40 Proven 45
Well 5
1.00
DRY
1.00
30 - 40 - 60
Key
DRY
15 - 25 - 50 1.00
1.00
RP90 - Rmean - RP10
Unrisked 45 - 65 - 110
Risked 45 - 65 - 110
Discovered 65
Proven 45

BHP Petroleum
52
DRAFT VERSION 2.0

Example 3: Downside Outcome (3)

400
Proven
Discovered
350 Mean
P10
P90
300 RP90
Rmean
RP10
250
200
150
100
50
0
0 1 2 3 4 5

BHP Petroleum
53
DRAFT VERSION 2.0

Example 4: Mixed Outcome (1) - 100 to 150 to 100 MMstb
No Wells Well 1
0.25 0.40
5 - 20 - 50 (15) 5 - 20 - 50 (15)
0.25 1.00
10 - 50 - 150 (40) 15 - 45 - 130 (40)
5 - 25 - 100 (20) 5 - 25 - 100 (20)

5 - 30 - 100 (25) 0.25 5 - 30 - 100 (25) 0.80
0.25 0.80
Risked 0 - 0 - 100 Risked 15 - 89 - 310
Proven 0 Proven 15
Well 2
0.40
5 - 25 - 40 (25)
1.00
30 - 70 - 120 (60)
Key
5 - 35 - 100 (30)
5 - 40 - 100 (35) 0.80
0.80 RP90 - Rmean - RP10

Unrisked 45 - 170 - 370
Risked 30 - 130 - 300
Discovered 60
Proven 30

BHP Petroleum
54
DRAFT VERSION 2.0

Example 4: Mixed Outcome (2) - 100 to 150 to 100 MMstb

Well 3 Well 4
0.40 0.20
5 - 25 - 50 (25) 5 - 25 - 50 (25)
1.00 1.00
35 - 65 - 100 (60) 35 - 60 - 80 (60)
5 - 35 - 80 (30) DRY
15 - 40 - 80 (35) 0.90 20 - 35 - 60 (35)
1.00
1.00 1.00
Risked 55 - 137 - 272 Risked 55 - 95 - 150
Proven 50 Proven 55
Well 5
1.00
5 - 10 - 15 (10)
1.00
35 - 55 - 75 (55)
Key
DRY
20 - 35 - 55 (35)
1.00 P90 - Mean - P10 (Det ML)
1.00 RP90 - Rmean - RP10

Unrisked 60 - 100 - 145
Risked 60 - 100 - 145
Discovered 100
Proven 60

BHP Petroleum
55
DRAFT VERSION 2.0

Example 4: Mixed Outcome (3)

400
Proven
Discovered
350 Mean
P10
P90
RP90
300
Rmean
RP10
250
200
150
100
50
0
0 1 2 3 4 5

BHP Petroleum
56
3.3 Sensitivity Analysis
Sensitivity analysis has been used for some time to identify which of the input
parameters to volumetric calculation has the greatest effect on the volume range.
In sophisticated probabilistic modelling software such as Crystal Ball or @Risk,

sensitivity analysis can be automatically performed and the results displayed on a
tornado plot - a horizontal bar chart showing the negative and positive contributions
of each parameter in order of significance (Figure 28). This may be expressed as an
absolute or percentage deviation from the mean, or statistical variance.
If sensitivity analysis cannot be performed automatically, it can be calculated by

systematically modifying the input data to the probabilistic model. The probability
distribution for gross rock volume is combined with constant most likely values for all
other parameters, and the reserves range noted. This process is repeated for each
parameter in turn, and a summary plot of reserves range versus parameter produced.
Economic sensitivities can also be performed, and the results displayed on a spider
plot (Figure 29). By systematically increasing or decreasing inputs such as reserves,
capital cost or oil price, the sensitivity of project value to each parameter can be
established. Typically, the value of a project is most sensitive to changes in reserves
and oil price.
3.4 Forward Modelling
Traditionally, as in the above examples, we measure the impact of appraisal

retrospectively - assessing the impact on reserves after a well has been drilled. More
recently, a number of methods which attempt to predict the impact of appraisal have
been developed. These include Monte Carlo forward modelling, appraisal by
objectives and value of information (VOI) techniques. These approaches allow us to
use quantitative analysis as part of the decision making process - allowing comparison
of options, providing greater insight and strengthening our decisions.
Monte Carlo Forward Modelling
Generally, the justification for further appraisal of a discovery is driven by the need to
obtain additional information - more control on structure, more core or fluid samples
etc. When deciding where to position the next well(s), we consider the ability of each
potential location to provide technical information. We also make subjective
judgements as to the likely impact of each location may have on reserves - is it small
or large?
Whilst the need for more information and a "feel" for the worth of a well remain a key
part of the appraisal decision making process, the process can be strengthened by
attempting to quantify the impact of well(s) on the reserves range or economic value
of the project. We place considerable time and effort into determining these based on
current data; rarely do we consider how that range may change following appraisal
wells. Although we make every effort to ensure that our predictions are accurate,
experience shows that on many occasions the post-drill reserves expectation and range

Sensitivity Analysis: Tornado Plot
GRV
Recovery Factor
Net:Gross
Porosity
FVF
Oil Saturation
-80 -60 -40 -20 0 20 40 60 80

Reserves / Percent /Variance

BHP Petroleum
DRAFT VERSION 2.0

Sensitivity Analysis: Spider Plot

100
NPV ($m)
Capex 80
60
Opex
40
20
0
-100% -80% -60% -40% -20% 0% 20% 40% 60% 80% 100%
-20 Parameter Change
-40
Oil Price
-60
Reserves -80
-100

BHP Petroleum
59
is significantly different from pre-drill-volumes.
By determining what these differences could be for a number of future well locations
and outcomes we can produce a series of comparative statements which can aid the
decision making process.
For example,
Comparison of different outcomes at the same well location:
" If we drill well A and it comes in on prognosis, then the revised reserves range
will be a - b - c."
" If we drill well A and it comes in 50m high, with the expected fluid contact,
then the revised reserves range will be x - y - z."
" If we drill well A and it comes in on prognosis, but dry, then the revised
reserves range will be p - q - r."
Comparison of the same outcome at different well locations:
" If we drill well B instead of well A and it comes in on prognosis, but dry, then
the revised reserves range will be j - k - l."
Comparison of sequences of wells:
" If we drill wells A, B and C and they all come in on prognosis the RP90
reserves will not be economic.”
" We need wells A or B to come in 20m high, or have a deeper fluid contact, to
make the RP90 reserves economic.”
“ If we drill wells C and D the RP90 reserves may increase more.”
This technique of modelling future wells to assess their impact is referred to as Monte
Carlo forward modelling. It has been used to quantify the impact of a number of well
locations, each of which is a candidate for the next well, and for sequences of wells.
This has been used to prioritise drilling order and estimate appraisal well numbers and
costs.
Most studies which have used this method to date have been based on risked reserves.
However, the same methodology can be applied to evaluations based on in-place
volumes, or value.
DRAFT VERSION 2.0
It must be emphasised that the aim of this technique is not to reproduce every possible
outcome or combination of outcomes. Nor is it necessary to encompass the absolute
minimum or maximum outcomes which may occur. The aim is simply to develop a
framework of realistic scenarios around which further discussion can take place.
Key questions which forward modelling can help to answer include:
“ Do we need another appraisal well ?”
“ How could this well change our expectation or proven reserves ?”
“ How do we get from current proven reserve of x to an economic level of y ?”
An example of the quantitative output from forward modelling is presented (Figure

30). The horizontal axis shows success and failure outcomes for a hypothetical well
location, referenced to the current base case. The vertical axis is the quantitative
measure of full field risked reserves. The vertical bars represent the RP90 to RP10
reserves range and the circles represent the Rmean volume - red = base, green = success,
blue = failure.
Measures
Using Monte Carlo forward modelling three separate parameters can be measured and
compared for various well locations and outcomes (Figures 30, 31):
Model Impact - The impact of an appraisal well is defined as the difference in Rmean
reserves between success and failure cases. The greater the difference, the more
impact that well has. In the example, Well A clearly has more effect on mean reserves
than the other wells. This may be related to the ability of this well to differentiate
models or remove risk.
Proven Growth - Proven reserves growth is defined as the ability of a successful well
to increase the RP90 value relative to the base case. In the example, Well C is the best
well to achieve this objective.
Upside Constraint - During appraisal knowing when to stop is an important

consideration. Removal of upside potential is defined as the ability of a failed well to
reduce the RP10 value relative to the base case. This is also referred to as a "drill-to-
kill" measure. In the example, Well B is the best well to achieve this objective.
The emphasis placed on each of these measures tends to change with time
(Figure 32). In scale appraisal, the ability of a well to impact expectation reserves
and/or remove upside is generally more important than the need to increase proven
volumes. During confidence appraisal, the emphasis often shifts to proven reserves
61

Appraisal Drilling: Measures
RP10
Upside
Constraint
Risked Model
Reserves / RMean Impact
Value
Proven
Growth
RP90
Base Well X Well X

Success Failure

BHP Petroleum
DRAFT VERSION 2.0

Appraisal Drilling: Comparison of Measures
Model Well A
Impact
Proven Well C
Growth
Upside Well B
Risked
Constraint
Reserves /
Value
Success
Failure
Base Well A Well B Well C Well D

Scenario

BHP Petroleum
63
DRAFT VERSION 2.0

Appraisal Emphasis
Project Identification Project Definition

Blackback
SE Mananda Moran Lambert BSFN-ROD

Single
Selection
Appraisal
Emphasis
Model
Multiple Impact
Options Scott Reef Sunrise Bayu- Bayu
Development Area Undan Undan Upside
Concept Gas Liquids Constraint
Maturity
Proven
Concept Growth
Uncertain
Ultra Deepwater Neptune
GOM
Uneconomic Feasible Viable Robust

Economic Maturity

BHP Petroleum
64
growth as we aim to improve the robustness of the project.
Method
For each model, a detailed summary of the hypothetical well results for success and
failure outcomes is prepared.
This includes information regarding the location, depth and relevant results of the well
including thickness, reservoir property, pressure and fluid data. These summaries are
used to ensure that the new Monte Carlo model honoured all available data as if the
well had actually been drilled.
For each model, outcome and well location(s), revised full field risked reserves are
calculated.
Depth structure maps are adjusted to tie the new well(s), if it is modelled to have come
in high or low to prognosis.
The minimum and maximum depth structure maps are also altered to take account of
the new depth control the well(s) provides. If these maps are derived from the most
likely structure map using "Low" and "High" depth error surfaces, new grids can be
easily generated by posting zero error at the new well(s) and flexing the error surfaces
to tie the additional data. Other approaches may require hand editing of the minimum
and maximum depth structure maps to honour the new well(s).
A range of new gross pay isochores can be generated, taking into account any new
information the well(s) has provided regarding reservoir thickness and fluid contacts.
The most likely reservoir property maps and property distributions for input to the
Monte Carlo model must also be updated.
The risks applied to each segment should also be adjusted, if appropriate. For
example, a well may be modelled to encounter reduced pay in the eastern part of the
field - requiring revision of the most likely N:G map, predicted N:G range, and an
increased risk in the eastern segments.
Having adjusted input parameters and risks, the volume range for each segment is re-
calculated and segments combined using probabilistic addition. The results can then
be compared against the pre-drill base case, and other appraisal options and outcomes.
Using Monte Carlo forward modelling it is important to limit the number of wells,
sequences and models evaluated, as the process can become both manpower and time
intensive. One or two alternative models, and three to five well locations usually
provides a reasonable number of comparisons.
Whilst this approach has most use during scale appraisal, providing quantitative
comparisons of the impact of success or failure in a few key wells, it can also be
applied in a simplified form to confidence appraisal.
In projects where a high degree of confidence is required, e.g. large offshore oil or gas
discovery in remote basin, it is important to estimate the number of wells and
DRAFT VERSION 2.0
appraisal costs at an early stage.
By evaluating the impact of a complete sequence of appraisal wells on RP90 and RP10
volumes for the current model i.e. all future wells come in on prognosis, we can at
least get an estimate of how many wells are required to prove a specific volume, or
remove upside potential, if all goes to plan. From this benchmark, we can assume that
better than expected results will require less wells, and failure or poor results will
require more.
If manpower and time permits, confidence appraisal requirements may also be

assessed for upside and downside models.
For each of these scenarios, a conceptual appraisal programme is developed. The

programme addresses in logical order the remaining uncertainties including areal
extent, reservoir distribution etc. The later stages of the programme includes
"optional" appraisal wells which may be required e.g. to confirm structural elevation
between wells and increase RP90 reserves to a level where a sales contract can be met.
These conceptual programmes do not imply that every well must be drilled. The
precise number of confidence appraisal wells also depends on many external factors.
For example, appraisal well numbers for a standalone LNG development will depend
on scheme size, contract terms, number of other pools, marketing initiatives etc.
As for scale appraisal, the Monte Carlo model is updated to reflect the results of
confidence appraisal wells. Given the large numbers of wells in most cases, not every
well is modelled separately. Wells may be grouped together and error maps, parameter
ranges amended after Well 5, Well 8, Well 12 etc.
At each time step, the GRV range is updated to include the effect of greater structural
control from additional wells. Minimum and maximum error maps are amended to
include zero error at the new wells and new depth maps produced.
In addition, the range of other volumetric parameters is assumed to narrow as time

proceeds reflecting our improved understanding of reservoir property distribution,
recovery mechanisms, sweep efficiencies etc as more data is gathered in numerous
appraisal wells.
The results may be plotted as risked reserves range versus well number or time
(Figure 33), and used to estimate appraisal requirements for a given threshold or range
of uncertainty.
In this example, the analysis shows that eight wells are required to prove up 400
mmboe reserves. Of course, this approach is only an approximation and the actual
number of wells required in this scenario may be more or less - perhaps it will take
six to ten wells to meet our objective. If we needed to prove up 450 mmboe, the
number of wells required increases to twelve to sixteen.
66

Confidence Appraisal
500
RP10
Rmean
RP90
400
300
Risked
Reserves /
Value
200
100
0
0 2 4 6 8 10 12 14 16
No. of Wells / Time / Cost

BHP Petroleum
Appraisal by Objectives
This semi-quantitative method uses a simple decision matrix to weight the importance
of each key objective, and rate the ability of each appraisal well location to address
these objectives.
The importance of each parameter in the reserves calculation, and development

concept selection, is assessed using sensitivity analysis and the major controlling
factors identified.
In many cases, gross rock volume (GRV) will be the greatest uncertainty, but the
factors which control this uncertainty may vary greatly. Rock volume may be poorly
defined because of seismic data quality, velocity control, or ill-defined fluid contacts.
Net:gross uncertainty may be introduced because of lack of information on areal
reservoir distribution, or insufficient core data to define net pay cut-off, or both.
By defining the objectives which must be met to remove the major uncertainties, we
can then rate the suitability of each well location. For example, reducing the
uncertainty in GRV may require discrimination of an ambiguous seismic time pick in
part of a field, and definition of an oil-water contact. Some wells will be better located
to achieve the first objective, and others may be more likely to achieve the second.
Each objective is assigned an importance, or weighting, ranging from 1 (unimportant)

to 5 (very important). The ability of each well location to meet each objective is also
given a rating from 1 (unlikely to meet objective) to 5 (highly likely to meet
objective).
For each objective and each well location the two numbers are multiplied together,
and the values added to determine a total value for each well location. The location
with the highest value is considered to be most likely to provide maximum
information and have the greatest reduction in uncertainty.
Using the above example:

Well A Well B Well C
Objective Weighting Rating Value Rating Value Rating Value
SeismicPick 4 4 16 3 12 4 16
Delineate OWC 3 2 6 3 9 4 12
Total 22 21 28

Appraisal by Objectives: Well 2 Decision Matrix
Location A Location B Location C

Objective Weight Rating Value Rating Value Rating Value
Prove up gross rock volume 3 3 9 4 12 5 15
Provide opportunity to test sands 2 3 6 4 8 4 8
Discriminate seismic pick in west 2 0 0 2 4 4 8
Determine lateral continuity of sands 1 3 3 2 2 5 5
Total Value (Weight x Rating) 18 26 36

BHP Petroleum
DRAFT VERSION 2.0

Appraisal by Objective: Well 3 Decision Matrix
Location D Location E Location F

Objective Weight Rating Value Rating Value Rating Value
Prove up gross rock volume 3 5 15 4 12 1 3
Investigate distal sand quality 2 1 2 3 6 5 10
Investigate subcrop on flank 1 0 0 3 3 5 5
Investigate amplitude changes in 3D 2 0 0 4 8 5 10
Total Value (Weight x Rating) 17 29 28

BHP Petroleum
70
In this case, Well C has the highest value and should be prioritised.
Two examples from a field case study (Figures 34,35) show the decision matrices that
were constructed when deciding the location of the second and third wells on the
structure.
In this case, the decision to drill location C as the second well is clear. However,
choosing between locations E and F for the third well is more difficult - as the total
values are so close. Other factors such as cost, sidetrack potential, partner issues must
be also considered.
This emphasises the need to use several approaches when evaluating appraisal
options, using our experience and judgments to integrate all available information and
analyses, and make the final decision.
Value of Information
In the early stages of appraisal, measuring the impact of appraisal activities by

changes in the reserves range is a reasonable approximation - it can be generally
assumed that any increase or decrease in reserves will be reflected by a similar
movement in value. However, in later appraisal improving the value of a project may
not be so readily linked to reserves, and the possibility of eroding value through
additional appraisal expenditure may arise.
Assessing the value of information (VOI) is a method of evaluating the economic

benefit of acquiring information against its cost, and can also be a useful way of
justifying appraisal expenditure.
Some VOI methods use full economic analysis, with probabilistic distributions for all
input parameters including capex, opex and economic criteria. Other methods adopt a
simplified decision tree approach, with a few specific scenarios and associated
probabilities.
Whatever the complexity of the method, the principle is to calculate an expected

monetary value (EMV) for the project - taking account of the probability and value of
each success outcome and the probability and cost of each failure.
VOI compares the current EMV of a project, against predictions of EMV which take
into account changes in value or risk which will occur with the acquisition of
additional data. A full range of scenarios, both success and failure, should be included
to capture the range of possible outcomes.
For example, consider the appraisal of a small oilfield.
Although four wells have already been drilled, a number of significant risks and
uncertainties have been identified - the chance of having a successful development
based on our current knowledge is estimated to be only 50%.
DRAFT VERSION 2.0
The operator is contemplating another appraisal well prior to committing to
development of the field - to reduce uncertainty in the reserves range, and increase the
chance of a successful project.
For clarity, the example is based on the following assumptions and a definitive
outcome - the project results in either total success or total failure. Of course, in reality
there could be many outcomes - we may achieve only partial success, and steps would
be taken to mitigate against total failure.
Well Cost $5 million

Development Cost $40 million
Revenue $100 million
A simple decision tree may be used to estimate the value of information (Figure 36).
If we decide not to drill the well, we must also decide whether or not to develop the
field.
The EMV of not developing the field is zero - $0.0 million - we neither spend money
or make money.
No Well, No Development
EMV ( 0.0 x 1.0 ) = $0.0 million
If we decide to develop the field, there are two possible outcomes - a successful
development or failure. The net present value (NPV) of the success outcome is $60
million - revenue minus development cost. The failure outcome has an NPV of $-40
million - no revenue minus development cost.
No Well, Development
EMV = ( 60 x 0.50 ) + ( -40 * 0.50 ) = $10.0 million
This higher EMV indicates that, if we do not drill a well, we are more likely to
maximise value by proceeding with the development than doing nothing - the benefits
of success outweigh the cost of failure.
If we decide to drill the well, the well results will have an impact on the project. If the
well is successful, the chance of a successful project is estimated to increase to 70%.
However, if the well has poor results, the chance of a successful project falls to 30%.
72

Value of Information: Decision Tree (1)

Drill
Well X?
16.0 10.0
Decision Node
Yes 6.0 No
Decision Path
Well
Outcome Node
25.0 Result -5.0
0.70 0.30
Success Failure Positive Outcome
Develop Develop Develop Negative Outcome

Field? Field? Field?
25.0 -5.0 -15.0 -5.0 10.0 0.0
Yes No Yes No Yes No
0.70 Probability
Project Project Project 75 Net Present Value

Result Result Result
25.0 Expected Monetary Value
0.70 0.30 0.30 0.70 0.50 0.50
Success Failure Success Failure Success Failure
6.0 Value of Information
Final Prob 0.49 0.21 0.09 0.21 0.50 0.50
Well Cost 5 5 5 5 5 5 0 0 0
Dev Cost 40 40 0 40 40 0 40 40 0
Revenue 100 0 0 100 0 0 100 0 0
NPV 55 -45 -5 55 -45 -5 60 -40 0

BHP Petroleum
To estimate the overall value of the decision, we must also assess the chance of the
well itself being successful - in this example it is estimated to be 70%.
Whatever the well results, we still have the option to develop or not develop the field.
This creates six branches to this side of the decision tree:
1 Successful Well Develop Field Successful Development
2 Successful Well Develop Field Failed Development
3 Successful Well No Development
4 Failed Well Develop Field Successful Development
5 Failed Well Develop Field Failed Development
6 Failed Well No Development
Consider first the value of not proceeding with a development, having drilled the well.
Regardless of the outcome of the well the EMV of not developing the field is $-5.0
million - the cost of the well.
Well, No Development
EMV ( -5.0 x 1.0 ) = $-5.0 million
If the well has positive results, increasing the likelihood of a successful development,
the EMV of the decision to proceed with development is $25 million. The reduced
NPV of the success outcome, and the higher cost of failure reflects the cost of the
well.
Successful Well, Development
EMV = ( 55 x 0.70 ) + ( -45 * 0.30 ) = $25.0 million
This higher EMV indicates that, if we have a successful well, we are more likely to
DRAFT VERSION 2.0
maximise value by proceeding with the development than doing nothing - the benefits
of success outweigh the cost of failure.
If the well has negative results, decreasing the likelihood of a successful development,
the EMV of the decision to proceed with development is $-15 million.
Failed Well, Development
EMV = ( 55 x 0.30 ) + ( -45 * 0.70 ) = $-15.0 million
This lower EMV indicates that, if we have a failed well, we are more likely to
maximise value by stopping the project than proceeding with the development - the
cost of failure outweighs the benefits of success.
The EMV from drilling the well can be calculated by combining the higher EMVs
from each of the two main branches - well success and failure.
EMV = ( 25 x 0.70 ) + ( -5 * 0.30 ) = $16 million
The difference in EMV between the two main branches of the decision tree, in this
case Well versus No Well, represents the value of information. A positive VOI
supports the decision to acquire further information.
In this example, the EMV from drilling the well is $6 million higher than the EMV
from no well. This supports a decision to drill the well.
VOI = ( 16 - 6) = $6 million
EMV Well EMV No Well
However, this type of analysis is very sensitive to probabilities we assign to each

outcome model, and sensitivities should always be run to test the robustness of our
decisions.
For example, another person may estimate the chance of a successful project
following positive well results to be only 60%, rather than 70%. This would result in a
decrease in EMV on the left side of the decision tree, and a negative VOI - leading to
a different appraisal decision (Figure 37).
Similarly, an increase in well cost from $5 million to $11 million results in a decrease
in EMV on the left side of the decision tree, and a neutral VOI - leading to an unclear
appraisal decision (Figure 38).
75


Drill
Well X?
9.0 10.0
Decision Node
Yes -1.0 No
Decision Path
Well
Outcome Node
15.0 Result -5.0
0.70 0.30
Probability of a Success Failure Positive Outcome
successful project
reduced from
0.70 to 0.60 Develop Develop Develop Negative Outcome
15.0 -5.0 -15.0 -5.0 10.0 0.0
0.70 Probability

0.60 0.30 0.30 0.70 0.50 0.50
Final Prob 0.49 0.21 0.09 0.21 0.50 0.50
Well Cost 5 5 5 5 5 5 0 0 0
Dev Cost 40 40 0 40 40 0 40 40 0
Revenue 100 0 0 100 0 0 100 0 0
NPV 55 -45 -5 55 -45 -5 60 -40 0

BHP Petroleum
DRAFT VERSION 2.0


Drill
Well X?
10.0 10.0
Decision Node
Yes 0.0 No
Decision Path
Well
Outcome Node
19.0 Result -11.0
0.70 0.30
Well cost Success Failure Positive Outcome
increased from
5 to 11 million
Develop Develop Develop Negative Outcome
19.0 -11.0 -21.0 -11.0 10.0 0.0
0.70 Probability

0.60 0.30 0.30 0.70 0.50 0.50
Final Prob 0.49 0.21 0.09 0.21 0.50 0.50
Well Cost 11 11 11 11 11 11 0 0 0
Dev Cost 40 40 0 40 40 0 40 40 0
Revenue 100 0 0 100 0 0 100 0 0
NPV 49 -51 -11 49 -51 -11 60 -40 0

BHP Petroleum
77
Recalculation of predicted EMV changes for a range of probabilities, rather than a
single value allows us to understand and communicate the effect differing perceptions
of risk may have on our decisions.
Using the above example, we can determine predicted EMV changes for variations in
both well cost and chance of success (Figure 39). In this case we assume that well cost
could vary from $2 million to $18 million, and the chance of well success may be
between 50% and 90%. These ranges can be used to define a risk envelope.
Based on our most likely assumptions, the appraisal well can slightly increase project
value. However, if the well cost was higher - or chance of success lower - value would
be eroded. Within the risk envelope, only 55% of the outcomes are positive, providing
a measure of the risk cover, or economic robustness of our decision.
If we repeat the analysis for other combinations, such as well success and project
success (Figure 40), we can assess the risk cover for other aspects of the decision. In
this case, 75% of outcomes are positive.

Value of Information: Sensitivity Analysis (2)

30
25
20
Positive Outcomes
15 55%
Predicted 10 Most Likely

VOI Case
($ million) 5
0
90%
-5 Well X -
70% Chance of
Negative Outcomes
-10 45% Success
50%
-15
2 6 10 14 18
Well Cost ($ million)

BHP Petroleum
DRAFT VERSION 2.0

Value of Information: Sensitivity Analysis (1)

30 90%
Positive Outcomes
25 75%
Well X -
20 70% Chance of
Success
15
Predicted 10 Most Likely

50%
Case
VOI
($ million) 5
-5
-10 Negative Outcomes

25%
-15
50% 60% 70% 80% 90%
Development Project - Chance of Success

BHP Petroleum
80
4. Conclusions
Understanding the fundamentals of appraisal and the many factors which can control
or influence our appraisal decisions is an important part of BHPP’s business.
The pressure to reduce cycle time, whilst increasing the value of projects, requires us
to make faster, better decisions. The various guidelines and methods described in this
paper can be used to improve our decision making processes.
It is recommended that:
a clearly defined and documented appraisal strategy is developed for each

economically feasible discovery. This should include a description of the
business goals and drivers, development concept maturity and economic
maturity, and the key factors controlling the direction and pace of appraisal,
the concept of field segmentation is used for all discoveries in the calculation of
unrisked and risked reserves ranges,
the proposed definitions of discovered, proven volumes and risked proven

volumes are adopted and used consistently in reserves accounting and external
reserves reporting,
all future appraisal decisions are supported by quantitative analyses

demonstrating the potential impact of appraisal on reserves, or project value.
Wherever possible, alternative options and well locations should be evaluated
and compared, and the rationale for choosing the preferred option(s) clearly
documented. Objectives should be clearly defined and prioritised.
C G McKinley
February 1998

Field Appraisal Basics PDF

Uploaded by

Copyright:

Available Formats

Field Appraisal Basics PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Field Appraisal Basics PDF

Uploaded by

Copyright:

Available Formats

BHP Petroleum

Fundamentals of Field Appraisal

Methods for Improving

2 Fundamentals of Field Appraisal 4

2.1 Development Concept 4

2.2 Development Concept Maturity 7

2.4 Economic Maturity 10

2.5 Appraisal Requirements 12

2.6 Appraisal Costs 15

3 Methods for Improving Appraisal Decisions 17

3.1 Appraisal Strategy 17

3.2 Appraisal Measures 24

3.3 Sensitivity Analysis 30

3.4 Forward Modelling 30

Monte Carlo Forward Modelling 30

Field appraisal is driven by our perception of the economic risk of developing a

In the ideal world, if we were certain that a discovery could be developed

The decision to proceed with a development project involves two fundamental

2.1 Development Concept

“Where and when will what volumes of which fluids be produced?”

Hence, the choice of development concept is influenced by four key factors.

Location The proximity of existing infrastructure, including production and

Timor Sea, Australia -a floating production, storage and offloading (FPSO)

Ula, Norway - the Ula oilfield is trapped in massive Jurassic sands in a

Fluids The type(s) of fluids to be produced, including variations in amount or

Productivity In addition to understanding the size of the resource, we must understand

Offshore and Onshore Fields, UK, US and Australia - Figure 1 plots

In some cases, the productivity of a reservoir may be improved by the application of

Well Productivity (1): Development Thresholds

6 August, 1999C:/appraisal.ppt PERT - Australia/Asia Region

Well Productivity (2): Horizontal Wells

0 400 800 1200 1600 2000 2400 2800

Horizontal Well Length (m)

6 August, 1999C:/appraisal.ppt PERT - Australia/Asia Region

Griffin Field, Australia - the Birdrong reservoir in the Griffin field is a

2.2 Development Concept Maturity

The development concept for a discovery is considered to be uncertain when we are

Appraisal to identify development concepts may be justified, with acquisition of new

Appraisal to eliminate development options may be justified, with the acquisition of

Development Concept Maturity

" " "

6 August, 1999C:/appraisal.ppt PERT - Australia/Asia Region

Appraisal to optimise the selected development concept may be justified, although it

Kutubu, PNG - the Kutubu oilfield development in the remote highlands of

Elang/Kakatua, Australia - the Elang/Kakatua development, offshore

Foinaven, UK - Foinaven, discovered in 1990, is the first deep water

2.4 Economic Maturity

Cum Cash Flow ($m)

NPV10 = $16.9 m NPV10 = $28.0 m

5 August, 1999C:/appraisal.ppt PERT - Australia/Asia Region

A discovery is considered to be uneconomic if it fails to make an acceptable return

No further appraisal is justified, as additional information is unlikely to make the

A discovery is considered to be economically feasible if it makes an acceptable return

Appraisal to establish an economic project may be justified, with acquisition of new

A discovery is considered to be economically viable if the most likely scenario makes

Appraisal to provide downside protection may be justified, with acquisition of new

A discovery is considered to be economically robust if it makes an acceptable return

Shifting the Balance

5 August, 1999C:/appraisal.ppt PERT - Australia/Asia Region