RESERVOIR CHARACTERIZATION:

A M ULTI -D ISCIPLINARY
T EAM A PPROACH

Dr. Ton Grimberg

PetroSkills, LLC.
2930 South Yale Avenue
Tulsa, Oklahoma 74114-6252
United States
1.918.828.2500--Office
1.800.821.5933--USA
1.918.828.2580--Fax
www.petroskills.com

© PetroSkills, 2013. All rights reserved. Version 6

NOTICE AND DISCLAIMER

The information contained herein and/or these workshop/seminar proceedings (WORK) was prepared by
or contributed to by various parties in support of professional continuing education.

For purposes of this Disclaimer, “Company Group” is defined as PetroSkills, LLC.; OGCI Training, Inc.;
John M. Campbell and Company; its and their parent, subsidiaries and affiliated companies; and, its and
their co-lessees, partners, joint ventures, co-owners, shareholders, agents, officers, directors, employees,
representatives, instructors, and contractors.

Company Group takes no position as to whether any method, apparatus or product mentioned herein is
or will be covered by a patent or other intellectual property. Furthermore, the information contained
herein does not grant the right, by implication or otherwise, to manufacture, sell, offer for sale or use any
method, apparatus or product covered by a patent or other intellectual property right; nor does it insure
anyone against liability for infringement of same.

Except as stated herein, COMPANY GROUP MAKES NO WARRANTIES, EXPRESS, IMPLIED, OR

STATUTORY, WITH RESPECT TO THE WORK, INCLUDING, WITHOUT LIMITATION, ANY
WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.

Company Group does not guarantee results. All interpretations using the WORK, and all
recommendations based upon such interpretations, are opinion based on inferences from measurements
and empirical relationships, and on assumptions, which inferences and assumptions are not infallible, and
with respect to which competent specialists may differ. In addition, such interpretations, recommendations
and descriptions may involve the opinion and judgment of the USER. USER has full responsibility for all
interpretations, recommendations and descriptions utilizing the WORK. Company Group cannot and does
not warrant the accuracy, correctness or completeness of any interpretation, recommendation or
description. Under no circumstances should any interpretation, recommendation or description be relied
upon as the basis for any drilling, completion, well treatment, production or other financial decision, or any
procedure involving any risk to the safety of any drilling venture, drilling rig or its crew or any other
individual. USER has full responsibility for all such decisions concerning other procedures relating to the
drilling or production operations. Except as expressly otherwise stated herein, USER agrees that
COMPANY GROUP SHALL HAVE NO LIABILITY TO USER OR TO ANY THIRD PARTY FOR ANY
ORDINARY, SPECIAL, OR CONSEQUENTIAL DAMAGES OR LOSSES WHICH MIGHT ARISE
DIRECTLY OR INDIRECTLY BY REASON OF USER’S USE OF WORK. USER shall protect, indemnify,
hold harmless and defend Company Group of and from any loss, cost, damage, or expense, including
attorneys’ fees, arising from any claim asserted against Company Group that is in any way associated
with the matters set forth in this Disclaimer.

No part of the text or original figures of the WORK may be reproduced in any form without the express
written consent of PetroSkills, LLC. Use of the WORK as a reference or manual for adult training
programs is specifically reserved for PetroSkills, LLC. All rights to the WORK, including translation rights,
are reserved.

© COPYRIGHT PETROSKILLS, LLC., 2013

THIS WORK IS COPYRIGHTED BY PETROSKILLS, LLC. AND PRINTED UNDER EXCLUSIVE LICENSE BY
PETROSKILLS, LLC.

© PetroSkills, 2013. All rights reserved.

 VITA
DR. TON GRIMBERG
DR. TON GRIMBERG is currently an Instructor/Coach for Shell’s introductory course for
graduates. He teaches Reservoir engineering, Petrophysics and some aspects of
Geology and Geophysics. His Shell career spanned 25 years in such fields as
Petrophysics, Reservoir Engineering, Rock Mechanics, etc. He developed a range of
qualities such as strategy development, planning, negotiating, personnel management,
organizational development, change management, commercial techniques and
economic planning. He received his certification and PhD in Physics from the University
of Amsterdam.

PROFESSIONAL EXPERIENCE
Teaching Experience
 Instructor/coach at Shell’s introductory course for graduates teaching Geology,
Geophysics, Petrophysics, Reservoir Engineering and economics (Rijswijk, 2006-
present day)
 Mathematics teacher at Stanislas college (Delft, 1999-2005)
 Physics teacher at Technical High School (Amsterdam, 1967)

Shell Experience
 Liaison Officer between Shell in the Netherlands and Shell in Nigeria (1996-
1999)
 Chief petrophysicist Shell Group (The Hague, 1994-1996)
 Procurement and Logistics Manager (Nigeria, 1993-1994)
 Chief petrophysicist (Nigeria, 1989-1993)
 Manager petrophysical research and development (Rijswijk, 1987-1989)
 Section head rock mechanics (Rijswijk, 1983-1987)
 Section head reservoir engineering (Brunei, 1981-1983)
 Reservoir Engineer (Den Haag, 1979-1981)
 Research petrophysicist (Rijswijk, 1978-1979)
 Operational petrophysicist (Aberdeen, 1976-1978)
 Research petrophysicist (Rijswijk, 1974-1976)

At Shell, I have developed a range of qualities such as strategy development, planning,

negotiating, personnel management, organisational development, change management,
commercial techniques and economic planning.

EDUCATION
 Gymnasium-β (1965)
 Physics (University of Amsterdam, 1969)
 PhD in Physics (University of Amsterdam, 1972)

© PetroSkills, 2013. All rights reserved.

© PetroSkills, 2013. All rights reserved.
TABLE OF CONTENTS
CHAPTER 1 INTRODUCTION .......................................................................................... 1-1
WHAT IS RESERVOIR CHARACTERIZATION?
(FROM SPE RCHA TECHNICAL INTEREST GROUP) ....................................... 1-3
RC PRINCIPLES ........................................................................................ 1-4
MODELING OBJECTIVES ............................................................................. 1-5
STAGES OF RESERVOIR CHARACTERIZATION ............................................... 1-6
RESERVES ETIMATION WITH TIME ................................................................ 1-7
HYDROCARBON LIFE CYCLE ...................................................................... 1-10
RESERVOIR CHARACTERIZATION PROCESS ............................................... 1-15
DEFINE THE PROBLEM ............................................................................. 1-17
EVALUATE DATA AVAILABLE ..................................................................... 1-17
SET THE BASELINE PLAN ......................................................................... 1-17
BUILD THE STATIC MODEL ........................................................................ 1-18
BUILD THE DYNAMIC MODEL WITH DYNAMIC DATA ..................................... 1-19
DEVELOP PROJECTIONS AND RECOMMENDATIONS .................................... 1-20
CRITICAL COMPONENTS TO RC PROCESSES ............................................. 1-20
INTEGRATION .......................................................................................... 1-21
SUMMARY OF CHAPTER 1 ......................................................................... 1-24

C H A P T E R 2 D A T A F O R C H A R A C T E R I Z I N G F I E L D S ................................................. 2-1
OBJECTIVES ............................................................................................. 2-1
DEVELOP DETAILED CHARACTERIZATION PLAN ............................................ 2-2
PROBABILITY DISTRIBUTIONS ..................................................................... 2-3
MULTIPLE REALIZATIONS OF PERMEABILITY ............................................... 2-10
DEPTH OF INVESTIGATION AND RESOLUTION OF LOGGING TOOLS ................ 2-16
SCALES OF MEASUREMENT ...................................................................... 2-17
QUALITY/ERROR MINIMIZATION ................................................................ 2-23
DATA MANAGEMENT ................................................................................ 2-24
EXTRAPOLATION TO OTHER AREAS........................................................... 2-29
HOW CAN WE INTEGRATE DATA ................................................................. 2-31
DATA - SUMMARY .................................................................................... 2-31
SUMMARY OF CHAPTER 2 ......................................................................... 2-32

CHAPTER 3 RESERVOIR MODELS ................................................................................. 3-1

MODEL TYPES .......................................................................................... 3-4
ANOTHER VIEW OF THE RESERVOIR CHARACTERIZATION WORKFLOW ............ 3-5

© PetroSkills, 2013. All rights reserved. i

Reservoir Characterization
Table of Contents

THE INTEGRATED MODEL ............................................................................ 3-9

SUMMARY OF CHAPTER 3 ........................................................................... 3-9

C H A P T E R 4 G E O S T A T I S T I C A L M O D EL I N G .............................................................. 4-1
INTEGRATED MODELING ............................................................................. 4-2
MODELING CONCEPTS ............................................................................... 4-3
GEOSTATISTICAL APPLICATIONS ................................................................. 4-4
OBSERVATIONS ABOUT RESERVOIRS .......................................................... 4-4
GEOSTATISTICAL OVERVIEW ...................................................................... 4-5
METHODS TO ASSIGN VALUES .................................................................... 4-6
UNIVARIATE ANALYSIS ............................................................................... 4-8
BIVARIATE ANALYSIS ................................................................................. 4-8
GEOSTATISTICAL METHODS ..................................................................... 4-10
GEOSTATISTICAL TOOLS .......................................................................... 4-11
MAPS AND MAP MAKING (CHAMBERS ET AL) .............................................. 4-13
BUILDING A VARIOGRAM........................................................................... 4-13
BOOLEAN OR OBJECT BASED SIMULATION (SRIVASTAVA 1994) .................. 4-29
SIMULATED ANEALING (SRIVASTAVA 1994) ............................................... 4-36
GEOSTATISTICS CASE STUDIES ................................................................ 4-40
THE TRUTH ............................................................................................. 4-44
SUMMARY OF CHAPTER 4 ......................................................................... 4-49
GLOSSARY OF VARIOGRAM TERMS, KUPFERSBERG AND DEUTSCH, 1999 .... 4-50

C H A P T E R 5 S E Q U E N C E S T R A T I G R A P H Y ................................................................. 5-1
SCENARIO TREES ...................................................................................... 5-2
RESERVOIR CHARACTERISATION WORKFLOWS ............................................ 5-3
STATIC RESERVOIR MODELS ...................................................................... 5-4
FACIES ..................................................................................................... 5-5
STRATIGRAPHY ......................................................................................... 5-5
BASIC IDEAS & DEFINITIONS ....................................................................... 5-6
POWER OF SEQUENCE STRATIGRAPHY ....................................................... 5-8
SEQUENCE STRATIGRAPHIC CYCLES ......................................................... 5-11
BIOSTRATIGRAPHY .................................................................................. 5-21
LITHO- AND CHRONOSTRATIGRAPHY ......................................................... 5-22
FLOW CHART FOR MODELING ................................................................... 5-26
SUMMARY OF CHAPTER .......................................................................... 5-26

C H A P T E R 6 L I T H O L O G I C M O D E L S ........................................................................... 6-1
MONTE CARLO SIMULATION ........................................................................ 6-3
TRAPS ...................................................................................................... 6-4

ii © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Table of Contents

LITHOLOGIC COMPONENTS......................................................................... 6-4

LITHOLOGIC MODELS - ELEMENTS .............................................................. 6-4
ROLE OF THE GEOSCIENTIST IN RESERVOIR STUDIES ................................... 6-6
DEPOSITIONAL SETTINGS ........................................................................... 6-6
OUTCROPS ............................................................................................... 6-7
ALLUVIAL FANS ....................................................................................... 6-10
RIVERS................................................................................................... 6-11
DELTAIC DEPOSITIONAL ENVIRONMENTS ................................................... 6-12
EOLIAN DEPOSITIONAL ENVIRONMENT ...................................................... 6-14
TURBIDITES ............................................................................................ 6-14
GEOMETRIC COMPLEXITY OF RESERVOIRS ................................................ 6-15
MODELING FLUVIAL SANDSTONES (BRIDGE AND TYE, 2000) ....................... 6-21
TYPICAL GAMMA RAY RESPONSES ........................................................... 6-28
PREDICTING DYNAMIC RESERVOIR RESPONSE .......................................... 6-34
WATER BREAKTHROUGH SIMULATION ....................................................... 6-39
DIAGENESIS ............................................................................................ 6-41
CARBONATE RESERVOIRS ........................................................................ 6-50
CLASSIFICATION OF CARBONATE ROCKS ................................................... 6-55
USING DYNAMIC DATA TO CONFIRM STATIC MODELS ................................. 6-66
SUMMARY OF CHAPTER 6 ........................................................................ 6-67

CHAPTER 7 PETROPHYSICAL DATA .............................................................................. 7-1

WHAT IS PETROPHYSICS ............................................................................ 7-1
QUICK LOO EVALUATION ............................................................................. 7-3
AMOCO APPROACH TO PETROPHYSICAL MODEL .......................................... 7-4
THE LOGGING ENVIRONMENT ...................................................................... 7-5
PETROPHYSICAL PERSPECTIVES ................................................................ 7-6
ANOTHER CHALLENGE: RESOLUTION ........................................................... 7-7
HOW TO INTEGRATE PETROPHYSICAL DATA .................................................. 7-7
FLUID AND FACIES DETERMINATION ............................................................. 7-8
CLUSTER ANALYSIS ................................................................................. 7-10
PRESSURE AND FLUID SAMPLING.............................................................. 7-13
THE QUEST FOR PERMEABILITIES FROM LOGS ............................................ 7-14
THE PORO-PERM SHOW ........................................................................... 7-14
THE GHAWAR FIELD IN SAUDI ARABIA ....................................................... 7-17
ACTIVATION LOGGING .............................................................................. 7-24
FORMATION MICRO-SCANNING .................................................................. 7-29
FRACTURES ............................................................................................ 7-30
CASED HOLE LOGGING ............................................................................. 7-33
CORES ................................................................................................... 7-33

© PetroSkills, 2013. All rights reserved. iii

Reservoir Characterization
Table of Contents

SUMMARY OF CHAPTER 7 ......................................................................... 7-38

RAPID EVALUATION OF KEY WELLS BY A MULTI-DISCIPLINARY TEAM .............. 7-39
DATA REQUIREMENTS AND SOURCES FOR ANALYSIS.................................. 7-43
ANSCHUTZ RANCH WORKSHOP ................................................................ 7-44

CHAPTER 8 GEOPHYSICAL MODELS ............................................................................. 8-1

WAVE PRINCIPLES ..................................................................................... 8-2
ACQUISITION ............................................................................................. 8-2
PROCESSING ............................................................................................ 8-7
SEISMIC PROCESSING - SUMMARY ............................................................... 8-9
GEOPHYSICAL APPLICATIONS ................................................................... 8-10
INTERPRETATION ..................................................................................... 8-11
STRUCTURAL INTERPRETATION ................................................................. 8-13
SEISMIC COHERENCE .............................................................................. 8-17
SEISMIC ATTRIBUTES ............................................................................... 8-19
MONITORING FLUID FLOW - 4D SEISMIC .................................................... 8-27
CROSS-DISCIPLINARY GEOPHYSICAL RESERVOIR MONITORING (GRM) ON
VISUND NORD - MODEL UPDATE AS A BASIS FOR NEW WELLS ...................... 8-30
ADDITIONAL EXAMPLES OF 4D SEISMIC ...................................................... 8-36
WELL SEISMIC ........................................................................................ 8-39
WELL PLANNING WITH SEISMIC .................................................................. 8-43
SEISMIC WHILE DRILLING ......................................................................... 8-45
GAS POOL EXERCISE - DEBRIEF ............................................................... 8-47
INTEGRATION CASE HISTORY ................................................................... 8-49
SUMMARY OF CHAPTER 8 ......................................................................... 8-57

CHAPTER 9 RESERVOIR ENGINEERING MODELS ........................................................... 9-2

FORCES CONTROLLING FLOW ..................................................................... 9-4
DRIVE MECHANISMS .................................................................................. 9-5
GRAVITY DRAINAGE................................................................................. 9-18
GAS RESERVOIRS ................................................................................... 9-20
DYNAMIC PERFORMANCE OF GAS WELLS ................................................... 9-19
RESERVOIR FLUID TYPES ......................................................................... 9-22
SECONDARY RECOVERY MECHANISM ....................................................... 9-23
ENHANCED OIL RECOVERY ...................................................................... 9-26
UNCONVENTIONALRESOURCES………………………………………………. 9-28
VOLUMES OF HYDROCARBONS IN PLACE ................................................... 9-29
RESERVES .............................................................................................. 9-31
FLOW UNITS ........................................................................................... 9-36
HETEROGENEITY ..................................................................................... 9-39

iv © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Table of Contents

RESERVOIR MANAGEMENT ....................................................................... 9-47

FIELD SURVEILLANCE ............................................................................... 9-48
HORIZONTAL WELLS …………………………………………………………..9-49
UPSCALING ............................................................................................. 9-52
RESERVOIR SIMULATIONS ........................................................................ 9-57
3D RESERVOIR IMAGING - EXAMPLES......................................................... 9-65
VERIFYING STATIC MODELS WITH DYNAMIC DATA ...................................... 9-71
INTEGRATION CASE STUDY ...................................................................... 9-73
RESERVOIR ENGINEERING DUTIES ............................................................. 9-77
SUMMARY OF CHAPTER 9 ......................................................................... 9-81

C H A P T E R 10 E C O N O M I C S A N D R I S K I N G ................................................................ 10-1
RISKING / ECONOMICS CHALLENGE ........................................................... 10-1
OBJECTIVES FOR ECONOMICS .................................................................. 10-2
RISKED BUSINESS VENTURES ................................................................... 10-8
NET PRESENT VALUE (NPV) ................................................................... 10-10
CASH SURPLUS AND ACCOUNTING PROFIT ............................................... 10-11
VALUING RESERVOIR STUDIES / DATA (VOI) ............................................. 10-12
RC EXAMPLE ........................................................................................ 10-14
SUMMARY OF CHAPTER 10 ..................................................................... 10-15

C H A P T E R 11 O R G A N I Z A T I O N A L S T R U C T U R E ........................................................ 11-1
TEAM STYLES ......................................................................................... 11-1
ORGANIZATIONAL MANAGEMENT .............................................................. 11-3
WHITNEY CANYON FIELD TEAM - AMOCO - HISTORY ................................. 11-11
TEAM LEADERSHIP SKILLS ...................................................................... 11-12
TEAM BUILDING ELEMENTS .................................................................... 11-18
TEAM COMMUNICATION.......................................................................... 11-20
SUMMARY OF CHAPTER 11 ..................................................................... 11-24

C H A P T E R 12 P R O J E C T M A N A G E M E N T .................................................................... 12-1
PROJECT PLANNING AND IMPLEMENTATION ............................................... 12-3
SETTING RESERVOIR CHARACTERIZATION TEAM GOALS ........................... 12-12
SUMMARY OF CHAPTER 12 ..................................................................... 12-15

C H A P T E R 13 P R O B L E M S , P I T F A L L S , A N D T H E F U T U R E O F R E S E R V O I R ........ 13-1
C H A R A C T E R I Z AT I O N ............................................................................ 13-1
PROBLEMS, PITFALLS .............................................................................. 13-1
THEMES FOR SUCCESSFUL RESERVOIR CHARACTERISATION ....................... 13-3
THE FUTURE OF RESERVOIR CHARACTERIZATION ...................................... 13-3

© PetroSkills, 2013. All rights reserved. v

 Reservoir Characterization
Table of Contents

SUMMARY OF CHAPTER 13 ....................................................................... 13-6

C H A P T E R 14 R E F E R E N C E S ....................................................................................... 14-1

C H A P T E R 15 E X E R C I S E M A T E R I A L ....................................................... 15-1
G A S P O O L E X E R C I S E - P A R T 1 ........................................... 15-1
WELL CORRELATION ................................................................................ 15-3
VARIOGRAM EXERCISE KINGFISH FIELD ..................................................... 15-5
LITHO- VERSUS CHRONO-STRATIGRAPHIC CORRELATION ............................ 15-8
G A S P O O L E X E R C I S E - P A R T 1 ........................................................... 15-9
PRESSURE VERSUS DEPTH FOR THE GRASSHOPPER FIELD ........................ 15-11
SEISMIC INTERPRETATION ...................................................................... 15-17
MATERIAL BALANCE EXERCISE ................................................................ 15-18
WHAT IS THE RIGHT PERMEABILITY? ........................................................ 15-20
G U I D E L I N E S F O R D EL I V E R I N G A N E F F E C T I V E P R E S E N T A T I O N ...... 15-24

vi © PetroSkills, 2013. All rights reserved.

Chapter 1 I NTRODUCTION
Reservoir Characterization is an act of creativity!
If aimed at underpinning business decisions, its impact is immeasurable.
Jeroen van der Veer
Former CEO of Shell

Reservoir characterization embodies the integration of the technical disciplines of the

exploration and production. The intent of reservoir characterization is to define oil and gas
reservoirs as accurately as possible. This is usually accomplished following the exploratory
and delineation phases of the property and prior to or contemporaneously with the
development phase. The key objectives in a reservoir characterization process are:

 It must be economic

 Model each reservoir to provide evidence

 Explanation of past reservoir performance

 Prediction of future reservoir performance of field

 Minimization of development expenditures

 Formulation of a plan for the development of

the field throughout its life

 Provide evidence that underpins

management decisions

 All activities - to produce the evidence - in line with the corporate vision

The models developed during the course will be based on the application of state-of-
the-art technical applications within the framework of a multi-disciplinary team approach.
The integration of data required for modern reservoir characterization encourages
changes in the organizational structure of companies. Modern multi-disciplinary team
approaches relieve many of the conflicts that result from traditional reporting
relationships.

There are several key elements in successfully characterizing reservoirs. These can be
categorized into related areas of learning; the course will treat these categories in depth
in the following sections; case examples will be included in the discussions to clarify the
concepts and to demonstrate practical methods of solving these problems.

The areas we will investigate further are:

 Data for characterizing fields

© PetroSkills, 2013. All rights reserved. 1-1

Reservoir Characterization
Chapter 1

 Sources
 Scale of the data/extrapolation to other areas
 Cross-disciplinary applications/integration
 Quality/error minimization and measurement
 Data management/planning
 Understanding what is represented by data
 Geostatistical Applications
 Reservoir models
 Sequence stratigraphic
 Lithologic
 Geophysical
 Reservoir engineering
 Economics and risking
 Volumetrics
 Probability of success
 Financial returns of project
 Organizational structure
 Team styles
 Organizational management
 Team communications
 Assessment and evaluation
 Problems, pitfalls and the future of reservoir characterization

1-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

WHAT IS RESERVOIR CHARACTERIZATION? (FROM SPE RCHA TECHNICAL INTEREST GROUP)

Reservoir Characterization (RC) is the process of developing an understanding of a

hydrocarbon reservoir: including the rock and fluid properties, the flow units, and the
reservoir production mechanisms so as to understand and unlock the full reserves
potential wherever it might be in its life cycle.

1. RC is frequently an iterative process, which often involves development of a three

dimensional description of a hydrocarbon trap.

2. The RC process is most successful building on the synergy of ideas developed

within a multi-disciplinary team composed of geologists, geophysicists,
petrophysicists, sedimentologist, reservoir engineers, production engineers,
operation engineers and others.

3. Data gathering to describe the reservoir normally begins with seismic and continues
with well information and then actual production results.

4. An initial model of the reservoir may be built and refined on a continual basis as
further information is obtained so as to understand the movement of fluids and to
develop a plan of production with the ultimate goal of maximizing the hydrocarbon
recovery while achieving a positive cash flow.

Simulated Facies - Rotliegend Gas Field

(Shrestha et al, 2006)

© PetroSkills, 2013. All rights reserved. 1-3

Reservoir Characterization
Chapter 1

Simulated Porosity - Rotliegend Gas Field

(Shrestha et al, 2006)

Simulated Permeability - Rotliegend Gas Field

(Shrestha et al, 2006)

RC PRINCIPLES

 Each field is unique

 Identify critical elements and data (often takes time to select)

1-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

 Requires multi-disciplinary input and studies

 It is business

MODELING OBJECTIVES

 Assessment of reserves
 Production of the field development program (FDP)
 Identification of additional exploitation opportunities

Particularly, the third objective can have a significant impact on the economics of
projects. If an expensive pipeline is to be laid, the economics will improve if additional
reservoirs can be found along the track of the pipeline that may feed into it. If satellite
fields can be found in the vicinity of an offshore platform, the lifetime of the platform may
be extended improving the economics significantly.

© PetroSkills, 2013. All rights reserved. 1-5

Reservoir Characterization
Chapter 1

STAGES OF RESERVOIR CHARACTERIZATION

The hydrocarbon life cycle shows that RC plays a role from delineation op to tertiary
production. Throughout, creativity in RC is required. In this creativity, divergent and
convergent thinking is important. For any project, different stages are identified. The
early stage is a phase with divergent thinking, during which all possible options for the
project are being considered. At the end of this stage a feasibility report is presented.

In the second and third stages, the thinking is convergent in order to select and define
the project. The second stage produces the FDP (field development plan) and the final
stage a project execution plan including a specific analysis of required investments.

After the third stage, management can take the decision to go ahead with the project
and should make the necessary funds available.

Examples of projects are a green field development, an appraisal project, acquisition of

3D or 4D seismic, a water injection project, an EOR (enhance oil recovery) project, etc.
For each project, the three-staged approach should be applied.

1-6 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

Depending on the nature of the study, typical elements that may be included in the
study report, are:

 Uncertainty analysis

 Volumetrics

 Flow modeling

 Hardware requirements (facilities, wells)

 Field / reservoir surveillance programme

 Operational strategies (manpower, maintenance, production, HSE)

Clearly, a seismic project will have different elements from thos of a water injection
project. However, the three staged approach with divergent and convergent thinking is
applicable to all projects.

RESERVE ESTIMATION WITH TIME

An important product of RC is an estimate of the reserves that will be confirmed by a

project. Two phases are identified, namely when no production data is as yet available

© PetroSkills, 2013. All rights reserved. 1-7

Reservoir Characterization
Chapter 1

and when a field or reservoir has been on production for some time. Different study
methods are applied in these phases, namely volumetric methods and methods based
on production performance. The range of uncertainties will reduce over time, but even
at the time of decommissioning uncertainties remain. Yes, the actual recovery may be
well known, but STOIIP or GIIP may still be uncertain. With time, when more data and
knowledge become available, the relative risk will reduce.

Each project is special depending on the nature of the project and the techniques and
technologies that may be applied to provide the evidence for the business decisions.

The slide on the next page shows a range of techniques and technologies such as
using the MDT (modular dynamic tool), PVT (pressure, volume and temperature
analysis), SCAL (special core analysis) and core-calibrated FAL (formation attribute
log). When applying these tools, the purpose is to reduce the uncertainties and increase
the reserves.

Uncertainties cannot be reduced to zero, but are reduced from an initial level to an
acceptable level. Acceptable means that there are contingency plans available that can
deal with the uncertainties. This implies that any decisions being taken, are taken under
risk.

1-8 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

In this context, the concept of reservoir management must be highlighted. The purpose
of reservoir management is to manage the reservoir in line with corporate objectives.
There may be different corporate strategies such as optimizing production or optimizing
ultimate recovery. These are often conflicting strategies and cannot always be pursued
at the same time.

When should reservoir management start?

 Ideally, at discovery of reservoir

– Provide early monitoring, evaluation
– Save costs of more expensive methods

 Usually, not early enough

– Reservoirs, wells, systems typically ignored long before ‘management’ starts
– Secondary, tertiary programs usually identify need for management

© PetroSkills, 2013. All rights reserved. 1-9

Reservoir Characterization
Chapter 1

HYDROCARBON LIFE CYCLE

The purpose of any hydrocarbon project is to realize an ultimate cash surplus upon
completion:

Cumulative cash flow

Money

Cumulative Cash Flow

Decommissioning

Ultimate
Cash Surplus

Time
Payout
Maximum Time
Exposure
First
Oil

During the life cycle of a project, different stages are followed, namely:

 Exploration

 Delineation

 Development

 Maturity

In each of these phases, there are opportunities to optimize the project, which may lead
to acceleration of the project. In the following pages these acceleration opportunities are
being discussed.

1-10 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

STAGE 1: EXPLORATION
The first stage in the life of a reservoir may span a period of two years. At this stage, the
reservoir and its contents are just beginning to be defined and understood. Early
interpretation of geologic and seismic data is necessary to confirm the shape and extent
of the reservoir and its contents. Such interpretations and analyses lead to reduction of
uncertainty and risk in the early stage of field delineation and development. More
specifically, they lead to optimal placement of future wells.

Stage 1 – Exploration. (Illustration modified from Schlumberger sales brochure)

Acceleration Options
 Drill fewer wells
– Understand reservoir better
 Better earth model
 3D data
 Early, conceptual simulation
– Horizontal/multi-lateral wells

 Faster drill times

– Learn from previous wells – drilling practices
– Communicate with RC team
– Team empowerment/ accountability

© PetroSkills, 2013. All rights reserved. 1-11

Reservoir Characterization
Chapter 1

STAGE 2: DELINEATION.
The delineation stage of a reservoir can range from two to five years. Initial wells are
completed in the producing zone with the aim of delineating the size and extent of the
reservoir. Direct and indirect information gathered from these wells and from seismic
surveys significantly improves our understanding of the reservoir structure, productivity
and reserves. During this stage, field development plans are created and a variety of
wells-vertical, horizontal and multilaterals-may be designed and drilled to build up
production and deplete the reservoir efficiently.

Stage 2 - Delineation. (Illustration modified from Schlumberger sales brochure)

Acceleration Options
 Earlier production
– Contemporaneous drilling/
production

 Maximize rates based on preliminary

simulation

 Secondary production techniques from

beginning

 Alliance(s) with product purchasers –

non-conventional funding

1-12 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

STAGE 3: DEVELOPMENT
Reservoir development can take place over a period of five to fifteen years depending
on the size of the reservoir. This is the phase of maximum activity in the field, which
further improves reservoir understanding.

Development drilling is conducted. Surface separation facilities and gathering systems

are put into place. Production builds to peak levels. In many cases, to replenish the
depleted reservoir energy, water or gas injection is implemented. To assist well flow,
operators (or producers) initiate artificial lift in the form of well pumping or gas lift.

Stage 3 - Development. (Illustration modified from Schlumberger sales brochure)

Acceleration Options
 Build alternative plans (workovers/
recompletions) at time of initial
completion
 Empower team to make operational
business choices at fairly high levels
 Alliances with service companies to
update technical level and their local
knowledge
 Involve field and plant personnel in
planning/expenditures
 Visit adjacent facilities for best practices

© PetroSkills, 2013. All rights reserved. 1-13

Reservoir Characterization
Chapter 1

STAGE 4: MATURITY
During the mature stage, which ranges from 15 years to abandonment, the reservoir is
well understood. Oil depletion by production wells causes the gas cap and aquifer to
expand and make the oil column thinner.

Wells are stimulated by fracturing, acidizing or other means to improve productivity and
prolong their economic life. Water or gas injection and artificial lift continue. At the end
of this stage, the wells are plugged and abandoned as their production falls below
economic levels.

Stage 4 - Maturity. (Illustration modified from Schlumberger sales brochure)

1-14 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

RESERVOIR CHARACTERIZATION PROCESS

THEMES FOR SUCCESSFUL RC

 Data to information to knowledge
 Identify and understand critical elements
 Multi-disciplinary input into data planning, collection and interpretation
 Data in a usable, accessible form
 Measured versus interpreted data
 Multiple working hypotheses
 Check perceptions against reality
 Integrated planning to achieve goals of RC team
 Understanding economic impact of activity

RC IS A PROCESS TO BUILD AN RC PLAN

 Understanding process will:
– Accelerate progress toward product
– Build consistency in characterizing reservoirs throughout organization
– Develop common expectations of a quality product

 Planning makes RC more efficient

 Allow RC team to plan its own work

 Put decision-making and knowledge together

© PetroSkills, 2013. All rights reserved. 1-15

Reservoir Characterization
Chapter 1

GENERALIZED RC PROCESS OUTLINE

1. Define the opportunity

2. Evaluate data available

3. Finalize plan and reach agreement on a “performance contract” with organization

----------------------------------------------------------------------------------------------------------------

4. Build the static model

A. Build external geometry of reservoir
B. Build internal geometry of reservoir
C. Add reservoir property distribution

5. Build the dynamic model using flow data

A. Produce forecasts
B. Determine (remaining) reserves

6. Finalize plan and recommendations

Construction of appropriate reservoir characterization (RC) models is an interactive

procedure strongly controlled by reservoir management (RM) context and the cost-
effective availability of information.

A detailed RC process may look as follows:

Match Predict Reservoir Performance

Begin Acceptable and Economics using Model(s)

Specify Initial Incorporate

Resolution/ appropriate Test Model(s) Too much
Uncertainty information Vs. Reservoir Uncertainty
for this into History in Results
RM Plan Step RC Model(s)

No Match Results
Negative
Collect New YES Additional
Data Information Information Acceptable
Available? Results

NO
Proceed to
Identify New Not Next RM
Cost-Effective
Data to Cost-Effective Plan Step
to Collect
Improve Models to Collect

Revise RM
Plan Step

RC Flow Chart, Fowler et al.

1-16 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

DEFINE THE PROBLEM

CHARTER THE PROJECT

 Define Project's Goal(s) / Team Purpose
 Define Outcome (Deliverables)
 Identify Project Resources/Boundaries

 Select Team Members

 Identify Required Skill Sets
 Name Team Leader/ Team Members
 Define Roles and Responsibilities / Team Approach

 Develop Scoping Plan

 Develop Task List & Milestones
 Assign Time to Tasks/ Resources

 Review Charter with Sponsor

 PROJECT CHARTERED AND APPROVED

EVALUATE DATA AVAILABLE

DEVELOP DETAILED CHARACTERIZATION PLAN

 Evaluate Project Data (Scoping - not interpretation - Gather and evaluate)
 Petrophysical data
 Core data
 Structural tops data
 Stratigraphic data
 Seismic data
 Reservoir data
 Production data/injection
 Fluid data

 Are data adequate?

 Additional processing? Collect more?

 Assign Individuals / Duration to Tasks

SET THE BASELINE PLAN

 Develop and Document Predictions

 Document Key Assumptions
 Review the Project Plan
 Develop Required Contingency Plans

© PetroSkills, 2013. All rights reserved. 1-17

Reservoir Characterization
Chapter 1

 Verify Tracking Mechanism

 Team Review

 Review the Project Plan with Sponsor

 Review Plan with:

 Customers, Suppliers, Partners, Interested Parties
 Contract with Required Consultants / Suppliers

 DETAILED PROJECT PLAN COMPLETE

BUILD THE STATIC MODEL

EXTERNAL GEOMETRY OF RESERVOIR

Data Analysis and Interpretation
 Develop integrated database

 Accessible across disciplines

 Petrophysical data
 Normalize/calibrate/edit against cores
 Identify cutoffs for mapping

 Core data
 Fully describe/interpret cores
 Correlate with logs
 Calibrate with test/production data

 Seismic data
 Develop structural interpretation
 Seismic stratigraphic correlation
 Establish correlations with reservoir parameters

INTERNAL GEOMETRY OF RESERVOIR

Data Analysis and Interpretation
 Map distribution of parameters

 Structural tops data

 "Establish tops - all horizons, faults, etc"
 Integrate with seismic data
 Create integrated maps
 "Create tabular display of well data (TVD, TVT, TST, etc)"

1-18 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

 Define facies

 Establish depositional model

 Define reservoir architecture

 Distribute reservoir parameters throughout reservoir

RESERVOIR PROPERTY DISTRIBUTION

Data Analysis and Interpretation
 Reservoir data
 Input petrographic data
 Integrate geology, production, test data"
 Determine morphology of pore system
 Interpret reservoir architecture

 Well History
 Completed interval, recompletions
 Stimulation history
 Production data/injection

 Fluid data

 Develop flow unit correlations

 Redefine stratigraphic intervals

 Modify interpretations as needed

BUILD THE DYNAMIC MODEL WITH DYNAMIC DATA

DATA ANALYSIS AND INTERPRETATION

 Test model within pilot area

 Annotate assumptions for model - range of parameters
 Generate and rank multiple realizations

 Choose best simulation - pursue

 Identify critical elements and likelihood of ranges

 Modify model as needed

 Full field simulation

© PetroSkills, 2013. All rights reserved. 1-19

Reservoir Characterization
Chapter 1

DEVELOP PROJECTIONS AND RECOMMENDATIONS

 Modify plan to fit dynamic model

 Test again as above and confirm fit
 Implement

CRITICAL COMPONENTS TO RC PROCESSES

 Geologic structure
 Stratigraphic model
 Modeling techniques - geostatistics, attribute assessment and modeling
 Fluid distribution - vertical and horizontal
 Production history
 Facilities
 RC team members

1-20 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

AUGER FIELD - SHELL

 Delayed development 1 year for RC study

 220 Million barrels. $1.3 Billion development cost

 2 3D surveys
 1987 - 12,000’ cable, 1990 - 19,000’ cable

 15 appraisal wells, 3 reservoirs

 Integrated 3D with porosity, permeability, saturation, thickness changes,

net/gross, etc

 Cut wells from 27 to 13

 Saved $180 Million

INTEGRATION

To achieve the best results in RC, integration is of vital importance. Usually, RC teams
are set up with participation of the traditional subsurface disciplines. However,
depending on the scope of a RC project it is also important to have less conventional
disciplines on the team, such as legal, environmental and other.

© PetroSkills, 2013. All rights reserved. 1-21

Reservoir Characterization
Chapter 1

Disciplines Data
Geoscience Geologic
Reservoir engineering Geophysical
Production engineering Petrophysical
Drilling Core
Negotiating/legal Pressure
Financial Production
Fiscalists Well test
Field personnel Financial
Facilities engineering Simulation
Management

Technology Tools
PSDM Geostatistics
Horizontal wells Seismic
Multi-lateral wells Logs
Production system simulators Reservoir simulators
Tomography Work stations
Etc. Analogues

Spectrum of disciplines required for effective reservoir management.

1-22 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Introduction

Multidisciplinary Interactions for reservoir delineation, characterization and operation.

© PetroSkills, 2013. All rights reserved. 1-23

Reservoir Characterization
Chapter 1

RC Hint: Never start an RC project before full agreement is

reached between the RC team and their sponsor

SUMMARY OF CHAPTER 1
The purpose of Reservoir Characterisation (RC) is to provide evidence for business
decisions. Therefore, critical elements are being studied following a six staged RC
process. During the third step of this process, agreement is reached between the RC
team and their sponsors regarding the terms of reference for the RC project.

In a RC project it is important to address three aspects, namely the technical issues,

people issues and issues regarding the organization paying attention to processes and
procedures.

In the technical arena the static model needs to be studied in terms of its architecture
and internal structure as well as the dynamic model producing production forecasts and
estimates of reserves. The static and dynamic models are strongly interlinked
representing in fact one single model. Such a model is only complete when in addition
all the uncertainties have been specified.

In RC different stages are recognized, namely an early stage in which the feasibility of
an opportunity is assessed. This stage is characterized by divergent thinking. Stages
two and three are characterized by convergent thinking, when the opportunity is
specified in detail and the way forward is defined including an investment plan.

During projects, there may be opportunities to accelerate them. Value drivers such as
‘advance first production’ or ‘increase plateau rate’ will improve the economics of
projects.

For optimum success, integration of data, people, technology and tools is of vital
importance. With respect to integration of people, disciplines - such as for instance
legal, HSE and public affairs - different from the usual subsurface need to be included
when necessary.

1-24 © PetroSkills, 2013. All rights reserved.

Chapter 2 D ATA FOR C HARACTERIZING
F IELDS
To understand is more important than to know
Alexandru Ioan Cruza (1820 - 1873)
First ruler of Romania (1859 - 1866)
Reservoir characterization is built upon data. A critical challenge to develop an integrated
interpretation of a field or potential field lies in bringing together data from disparate
disciplines and combining it into a cohesive data set that defines the accumulation and
allows for informed decisions about how to manage the property. The traditional approach
was for the interpretations of data to be developed within each discipline with little need to
share with adjacent disciplines. Integrated reservoir characterization requires sharing of
data across disciplines during the interpretative process.

The first step in data analysis involves creating an inventory of the data available and
the quality of those data. This preliminary review is focused upon a scoping of the data
- not interpretation. Key questions are whether the data are viable and adequate for the
purpose of the study.

OBJECTIVES

 Develop common understanding of data

 Understand needs of other team members

 Use common terminology - with same definitions

– Understanding scale of data portrayed

 Understand capability and limitations of each data source

 Quantify the uncertainties in data

 Search for critical elements

 Ensure cost/time efficiency in interpreting and using data

© PetroSkills, 2013. All rights reserved. 2-1

Prospect and Play Assessment
Chapter 

DEVELOP DETAILED CHARACTERIZATION PLAN

 Evaluate Project Data (Scoping - not interpretation - Gather and evaluate)

 Petrophysical data
 Core data
 Structural tops data
 Stratigraphic data
 Seismic data
 Reservoir data
 Production data/injection
 Fluid data

 Are data adequate?

 Additional processing? Collect more?

 Assign Individuals / Duration to Tasks

DATA CONSIDERATIONS
 Sources
 Scale of the data
 Cross-disciplinary applications/integration
 Quality/error minimization
 Data management
 Do data represent larger area?
 Extrapolation to other areas

2-2 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

PROBABILITY DISTRIBUTIONS

An Important Concept - Even though we describe potential values with a range, the
truth is that a single value represents the prospect or play value under consideration.
For example, if we predict that a prospect has most likely gross reservoir thickness of
300 ft (90 m), a minimum gross reservoir thickness of 50 ft (15 m) and a maximum
gross reservoir thickness of 550 ft (165 m), we are describing our assessment of the
geologic conditions based on our best knowledge. The actual gross reservoir thickness
is a single value that – hopefully - falls within the predicted range - ideally near the most
likely expected value.

 Exceedance/Cumulative*
 Normal (gaussian or bell-shaped)
 Lognormal
 Histogram
 Equal
 Rectangular
 Triangular*
 Log-triangular*

Symmetrical triangular distributions are safest distributions to use, because the mean
value or "weight" of the variable in the final assessment is closer to the Most Likely
value (the assessor's Best Interpretation) than it is if the triangle is skewed. The mean
values of both normal and log triangles can be calculated as shown below.

Symmetrical distributions.

© PetroSkills, 2013. All rights reserved. 2-3

Prospect and Play Assessment
Chapter 

CUMULATIVE
In cumulative curve displays, small probabilities values indicate small numbers
(example P10 is a thin reservoir).

120 Economic Threshold

- 40'
100
Cumulative frequency
80 distribution
A chart that shows the
number or proportion (or
60 percentage) of values
less than or equal to a
given amount.
40

20

0
9 18 27 36 45 54 63 72 81 90

Cumulative curve.

EXCEEDANCE
In exceedance curve displays, small probabilities values indicate large numbers
(example P10 is a thick reservoir). Companies use both displays and care needs to be
taken to understand which is displayed. Usually the context will indicate the distribution.
Exceedance curves will be used as the default display of population in this course.
Some cumulative presentations from other authors may be used.

120
Economic Threshold
- 40'
100
Exceedance
80 distribution
A chart that shows the
60 number or proportion (or
percentage) of values
greater than or equal to
40 a given amount.

20

0
9 18 27 36 45 54 63 72 81 90

Exceedance curve.

2-4 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

NORMAL DISTRIBUTION
 Describes many natural phenomena (IQ's, people's heights, the inflation rate, or
errors in measurements)
 Continuous probability distribution
 Parameters are:
− Mean
− Standard deviation
 Some value is the most likely (the mean of the distribution)
 The unknown variable could as likely be above or below the mean (symmetrical
about the mean)
 The unknown variable is more likely to be close to the mean than far away
− Approximately 68% are within 1 standard deviation of the mean
 Some equations:

 And a graph of the normal or Gaussian distribution in normalized form:

© PetroSkills, 2013. All rights reserved. 2-5

Prospect and Play Assessment
Chapter 2

The Gaussian Distribution


  

68% of
Data

95% of
Data

99.7%of
Data

–4 –3 –2 –1 0 1 2 3 4

Normal distributions.

LOGNORMAL DISTRIBUTION
 Widely used in situations where values are positively skewed (where most of the
values occur near the minimum value)
− Financial analysis for security valuation
− Real estate for property valuation
− Distribution of reserves in a play
 Continuous probability distribution
 Financial analysts have observed that the stock prices are usually positively
skewed.
− Stock prices exhibit this trend because the stock price cannot fall below the
lower limit of zero but may increase to any price without limit.
 The parameters for the lognormal distribution
− Mean
− Standard deviation
 Three conditions underlying a lognormal distribution are:
1. The unknown variable can increase without bound, but is confined to a finite
value at the lower limit.
2. The unknown variable exhibits a positively skewed distribution.
3. The natural logarithm of the unknown variable will yield a normal curve.

2-6 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

HISTOGRAMS
Histograms demonstrate the results of many trial of a given set of geologic conditions.
The resulting plots show the frequency of answers within given ranges. This is a
desirable approach to test the models; unfortunately, it is not often possible to generate
many solutions. What do we do in this case?
A frequent answer is to define the potential outcomes of the models by estimating
minimum and maximum values and a central value. There is a great deal of
controversy around how to do this. The most important judgment is to apply sound
geoscience reasoning to the outcome and to assure that whatever technique is used,
that the results are realistic. Some of these approaches are described below and
demonstrated here.

RECTANGULAR DISTRIBUTION
 All values between the minimum and maximum are equally likely to occur.
 Continuous probability distribution.
 The parameters for the uniform distribution are minimum and maximum.
 Three conditions:
1. The minimum value is fixed.
2. The maximum value is fixed.
3. All values between the minimum and maximum are equally likely to occur.

TRIANGULAR DISTRIBUTION
The normal triangular distribution, as well as other distributions, can be transformed into
a cumulative distribution. The triangle is ideal for assessment input because it
emphasizes a Most Likely (Best Interpretation) value, the Minimum and Maximum
limits are fixed (there are no surprises), and its mathematics are relatively simple. The
third leg of any symmetrical triangle can be calculated if the other two are known.

 Shows number of successes when you know the minimum, maximum, and most
likely values.
 Continuous probability distribution.
 The parameters for the triangular distribution are minimum, maximum, and
likeliest.
− For example, you could describe the number of cars sold per week when past
sales show the minimum, maximum, and most likely number of cars sold.
 Three conditions:
1. The minimum number is fixed.
2. The maximum number is fixed.

© PetroSkills, 2013. All rights reserved. 2-7

Prospect and Play Assessment
Chapter 2

3. The most likely number falls between the minimum and maximum values,
forming a triangular shaped distribution, which shows that values near the
minimum and maximum are less likely to occur than those near the most
likely value.

NORMAL TRIANGLE
(e.g., 2 - 4 - 6)

MOST LIKELY = (MIN+MAX) / 2 = (2 + 6) / 2 = 4

MINIMUM = 2 ML - MAX = 2x4 - 6 = 2

MAXIMUM = 2 ML - MIN = 2 x 4 - 2 = 6

MEAN = (MIN + ML + MAX) / 3

= ML (IF SYMMETRICAL)

Normal triangle.

LOG TRIANGLE
(e.g., 2 - 4 - 8)

MOST LIKELY = MIN x MAX = 16 =4

2
MINIMUM = ML / MAX = 16 / 8 = 2
2
MAXIMUM = ML / MIN = 16 / 2 = 8

SYMMETRICAL LOG TRIANGLE

MEAN = ML + 0.06 (MAX - ML) *
*DERIVED BY W. R. JAMES

Log triangle.

LOG-TRIANGULAR DISTRIBUTION
The log triangular distribution is also useful, especially for describing those variables
that seem to be lognormally distributed in nature. The normal and log triangles can
closely approximate the bell-shaped normal and lognormal distributions. These figures
also show the significant difference between normal and log triangular distributions that
have the same Minimum and Maximum endpoints.

2-8 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

Figure 2-1. An asymmetrical data set. The three measures Figure 2-2. A symmetrical data set. The three measure of
of central tendency are different. central tendency are identical.

© PetroSkills, 2013. All rights reserved. 2-9

Prospect and Play Assessment
Chapter 2

MULTIPLE REALIZATIONS OF PERMEABILITY

WELL THREE

MEAN 10.32
0.16
COEF. OF VAR 0.39
MAXIMUM 24.12
MEDIAN 9.08
FREQUENCY

0.12
MINIMUM 4.72

0.08

0.06

0.00
0 5.0 10.0 15.0 20.0
WELL THREE

2-10 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

DATA LIMITATIONS
 Impact of each parameter can’t be predicted
(e.g. the impact of the permeability distribution on the production forecast)

 Important data may not be collected

(e.g. gas relative permeability curves)

 Usually large inter-well regions without any data apart from seismic data

Planning for acquisition is often ignored to the detriment of the overall plan. A multi-
disciplinary approach to planning for data acquisition and usage will maximize the
benefit of the data over time and across disciplines and will minimize expense
associated with re-acquiring data at a later time of acquiring unnecessary data.

It is paramount in this situation to be able to share data in a compatible format and to

understand the critical controls that act upon the data. For example, a reservoir
engineer needs to be able to share production data with a geophysicist and to
understand the scale of investigation for a 3D seismic survey when assisting in the
planning for a 4D survey (indeed, the reservoir engineer needs to understand the
capabilities of the 4D tool prior to sharing data). Raza offers a flow diagram for efficient
data flow as follows:

DATA PLANNING DATA COLLECTION

Where to collect?
Why needed? Users?
How often?
Type, Quantity, At what cost?
Correct procedure
When required, used?
Audit for accuracy
Responsible party?
Database maintenance

DATA ANALYSIS AND REDUCTION DATA SYNTHESIS

Selecting "trustworthy" data
Assessing impact on solution Integration of data
Data confidence range Synthesis of data
Multiple source data

© PetroSkills, 2013. All rights reserved. 2-11

Prospect and Play Assessment
Chapter 2

Information Data Type Supplemental Data

Provided

Structure Seismic Well logs, production model

Reservoir
geometry

Reservoir Well data Logs, cores, correlations,

Framework models
Rock properties

Net pay Wireline Logs Core data

Fluid contacts Wireline Logs Include seismic

Porosity Wireline Logs Calibrate with core

SW Wieline Logs Cores, cap. Press. prod.

Permeability Cores Production,

depositional Model,
simulation

2-12 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

Scale of investigation
(Sech et al, 2009)

© PetroSkills, 2013. All rights reserved. 2-13

Prospect and Play Assessment
Chapter 2

Exercise – Scales of Data

It is important that all participants on the reservoir characterization teams are able to
discuss common terms with a shared understanding of what each term means.

1. For the list of terms below, indicate the scale of investigation (in feet or meters), the
purpose of the tool or technique and whether the data type is measured or
interpreted.

2. It may be necessary to list a range in identifying the scale.

3. The scale of investigation can be listed in units (metric or imperial) or as

microscopic, macroscopic, megascopic, or gigascopic.

4. You may also wish to consider the different scales of investigation laterally and
vertically.

Tool Scale of investigation Purpose of the tool / technique – Measured / Interpreted?

Thin section

Core

Outcrop

Open hole logs

Seismic - 2D

DST

Well test

Production

2-14 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

Seismic methods trade coverage of the reservoir for resolution because it is impractical
today to achieve both high resolution and high coverage. Crosswell methods fill a
resolution “gap” between sonic log measurements and vertical seismic profiles.

© PetroSkills, 2013. All rights reserved. 2-15

Prospect and Play Assessment
Chapter 2

DEPTH OF INVESTIGATION AND RESOLUTION OF LOGGING TOOLS

 The depth of investigation and resolution of the log is in general related

– to the size of the sensors and
– the spacing between sensors.

 In a tool either resolution or depth of investigation has to sacrificed to gain the other.
– Small sensor means better resolution and poor depth of investigation and vice
versa.
– Large sensor spacing means poorer resolution but good depth of investigation.

 The physics of measurement also dictate the resolution and depth of investigation
for some tools (specially radioactivity based tools).

2-16 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

SCALES OF MEASUREMENT

For the following plot, show the scale of processes to be investigated in a reservoir
characterization through the use of a bar scanning the appropriate ranges. Include in
your plots the following processes:

 Diagenesis
 Bedding units
 Facies
 Fracturing
 Flow units
 Oil/gas
 Production

RC Hint: Identify and understand critical elements

SOURCE OF DATA
Most professionals are familiar with the type of data required for reservoir
characterization. There are many sources for this data, but at times, the required
information is not available from measured sources. If it is necessary to estimate a
value or range of values, the professional is well advised to review the model
(geological, geophysical, engineering, etc. - each discussed in a later section). With the
model in mind, better interpretations can be made in selecting the values. There are a
great number of published data sources that may also be used. Satter and Thakur
(1994) provided a table that categorizes data (see below):

© PetroSkills, 2013. All rights reserved. 2-17

Prospect and Play Assessment
Chapter 2

Classification Data Acquisition Timing Responsibility

Seismic Structure, Stratigraphy, Faults, Exploration Seismologists,
Bed Thickness, Fluids, Geophysicists
Interwell Heterogeneity

Geological Depositional Environment Exploration, Discovery Exploration and

Diagenesis Lithology, and Development Development
Structure, Faults and Fractures Geologists

Logging Depth, Lithology, Thickness, Drilling Geologists,

Porosity, Fluid Saturation, Petrophysicists and
Gas/Oil, Water/Oil and Engineers
Gas/Water Contacts and Well-
to-Well Correlations

Coring Drilling Geologists, Drilling

and Reservoir
Engineers and
Laboratory Analysts

Basic Depth, Lithology, Thickness,

Porosity, Permeability and
Residual Fluid Saturation

Special Relative Permeability, Capillary

Pressure, Pore
Compressibility, Grain Size and
Pore Size Distribution

Fluid Formation Volume Factors, Discovery, Reservoir

Compressibilities, Viscosities, Delineation, Engineers an
Gas Solubilities, Chemical Development and Laboratory Analysts
Compositions, Phase Behavior Production
and Specific Gravities

Well Test Reservoir Pressure, Effective Discovery, Reservoir and

Permeability-Thickness, Delineation, Production
Stratification, Reservoir Development, Engineers
Continuity, Presence of Production and
Fractures or Faults, Injection
Productivity and Injectivity
Indices and Residual Oil
Saturation

Production and Oil, Water and Gas Production Production and Production and
Injection Rates and Cumulative Injection Reservoir
Productions, Gas and Water Engineers
Injection Rates and Cumulative
Injections and Injection and
Production profiles

2-18 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

Data Considerations - 2
 Distinguish between measured and interpreted data
 Understand importance of distribution of range of values for reservoir parameters
 Evaluate the error range in each data set
 Evaluate the uncertainty in each data element
 Communicate with data utility and limitations clearly indicated

SCALE OF DATA
Professionals from across the disciplines frequently use common terms to describe the
data sets that are generated within their discipline. Unfortunately, there is often a
communication problem in that the professionals are not describing the same data. For
example, geologists and geophysicists may each describe a formation boundary. The
geologist may well be thinking about a depositionally-defined unit with a scale on the
order of feet while the geophysicist is thinking of a unit defined by reflection
characteristics with a scale of tens of feet. Worthington demonstrates the scales of
investigation and scales of heterogeneity as shown in the two illustrations below:

SCALES OF MEASUREMENT

© PetroSkills, 2013. All rights reserved. 2-19

Prospect and Play Assessment
Chapter 2

GAMMA RAY/OUTCROP/SEISMIC

2-20 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

SCALES OF HETEROGENEITY

Data Considerations -3
 Field scale variations controlled by viscous forces - capillary forces negligent

 Small scale - viscous and capillary forces interact with sedimentary structures
 Viscous forces drive oil and along permeable laminae
 Capillary forces force the wetting phase from hi to low K zones
 Orientation of laminae very important

SCALES OF INVESTIGATION - COUNTESS POOL, ALBERTA, PEIJS-VAN HILTEN, ET AL, AAPG

 Detailed data on sedimentary structures present

 Measurements on thickness of structures and angles of inclination

 Ranges and averages of values – including characterization of distribution –

tabular data

 Upscale data to represent conditions

 Distinguishing facies and flow units

An additional aspect of dealing with scales is that of upscaling. This critical element in
reservoir characterization refers to the process of taking data at one scale and making it
applicable at another. This subject is treated in the section on geostatistics, but needs to

© PetroSkills, 2013. All rights reserved. 2-21

Prospect and Play Assessment
Chapter 2

be considered while evaluating the data for reservoir characterization. An example of

upscaling was presented by Jones et al..

2-22 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

QUALITY/ERROR MINIMIZATION

To be most useful, the data used for reservoir characterization should fully represent
that for which it was measured. Without an understanding of the restrictions on the data
due to the many factors that influence it, that data set is subject to considerable risk -
particularly when those data are to be combined and massaged through several steps
in the reservoir characterization process. Field data is subject to many errors
(sampling, systematic, processing, random, interference etc.). Error and quality can be
understood better when questioned and subjected to testing and validation. Cross-
validation against other disciplines and (where possible) is a useful technique to validate
the data. Sprunt offers the following observations:

In school, most of us are taught to assign error estimates to laboratory measurements.

However, in practice in the oil industry most laboratory measurements are reported
without any confidence limits. Commercial laboratories are loath to report error bars
because their competitors will assert that they can do better. In addition, few published
papers contain error analysis of the reported experimental test results. Thus, many in
the oil industry do not have a sense of the magnitude of errors in laboratory data.

RC Hint: Understand precision and accuracy

Inaccuracy can be removed by calibration. If data are accurate, but imprecise the
solution is to increase the number of data.

© PetroSkills, 2013. All rights reserved. 2-23

Prospect and Play Assessment
Chapter 2

Care should be taken to distinguish between measured data and interpreted data.
Measured data are those that are directly observable and require no interpretations to
derive (ex. depth of a core sample); interpreted data require some sets of assumptions
before deriving the value (ex.. SW saturation based on log interpretation). Those data
that are interpreted are subject to potential error if the assumptions behind their
derivation are flawed or mis-applied.

RC Hint: Distinguish between measured and interpreted

data

A final important point is the manner in which values are chosen to represent reservoir
parameters. Typically, a single value is chosen to represent the entire reservoir or a
significant portion of it. The manner in which values are chosen is critical to the
reservoir characterization process. This topic will be addressed later in the section on
lithology, but is worth emphasizing here. Data forms provided as a part of the course
will assist in making these choices.

DATA MANAGEMENT

Access to data and availability of data sets across the disciplines are critical
characteristics of successful reservoir characterization projects. Part of the problem lies
in the data itself, as addressed above; in addition the framework on which
interpretations and manipulation of data are conducted have an impact. The
geosciences have typically worked on a UNIX platform and the engineers on a PC
platform. Bringing these two, platforms together or networking them--does not eliminate
gaps. To manage petroleum reservoirs more profitably, more efficient methods of
integration between PC and UNIX applications will become essential. Valusek offers
this observation:

The emphasis now is on economics. It is no longer- sufficient to generate a good

geoscience interpretation, and throw the results over- the wall for the engineers to
calculate economics and implement field operations. Any time an asset team
encounters a common decision point associated with two different types of data, they
need to share those data and link the applications that use them. Most technical
decisions made during the life cycle of a reservoir have economic implications. Thus,
most technical decisions are actually business decisions. That's why E&P organizations
need tools to link the subsurface evaluations done on UNIX workstations with the
economics and operations planning done on PCs. Only then will technology integrate
the entire asset team.

DATA BASE DEVELOPMENT

 May require most time of any RC activity
 Needs to be integrated and accessible

2-24 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

 Databases to integrate:
 Base map information
 Geological, geophysical
 Petrophysical
 Engineering, production

Hierarchy of database development

• COMPILATION:
Identifying the sources of information
• EXTRACTION:
Gathering, organizing, and storing data in a manner that allows easy access.
• INTERPRETATION:
Processing, describing, translating, comparing, contrasting, classifying,
correlating, and calculating data
• ANALYSIS:
Interpreting and synthesizing data into general and specific characterizations that
attempt to explain the facts and observations

Within this hierarchical framework there are many tasks, sub-tasks, and iterations.
Fundamentally however, the process involves nothing more than a series of information
flows (i.e., the results of one task are used in the next). This process is highly non-linear
and iterative, and information frequently flows in loops, often more than once. The
discussion that follows is meant to be a general guide. The specific steps that one
should follow are highly dependent on the nature of the study.

DATA BASE STEPS

 Compilation
 Loading
 Validation
 Usage

LOADING PROBLEMS
 Well Name
 Surface Name
 Logs Readings
 Log Orders
 Log Correction
 Deviation Surveys

© PetroSkills, 2013. All rights reserved. 2-25

Prospect and Play Assessment
Chapter 2

DATA FOR RESERVOIR SIMULATION WAGENHOFER

 Geology/Geophysics
– Top Structure*
– Gross Thickness*
– Net/gross*
– Fault locations*
– Fluid contact*
– Depositional environment/features
– Depositional continuity
 In-house Interpretations of:
– Well logs*
– Core data
– Seismic sections
– Reports documents
 Petrophysics
– Permeability*
– Porosity*
– Connate water saturation*
– Relative permeabilities*
– Capillary pressure curves
– Variability in porosity and perm.
– Reports, documents
 Fluid properties
– PVT analysis
– Black oil table*
– Fluid composition*
– Equation of state
– Reports, documents
 Well performance data
– Production/injection rates*
– Completion data/history*
– Well trajectories, well and tubing details*
– Well events*
– Pressures*
– Production/injection logs
– Tubing head, Shut-in casing, flowing pressures
– Damage/skin factors
– Vertical lift curves
– Pressure transient test data

2-26 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

DATA TYPES
 Core
 Geophysical
 Logs - all types (digital and paper)
 Fluid
 Perforations
 Completions and workovers
 Production and well test
 Production

STATIC MODEL - DATA ANALYSIS AND INTERPRETATION

 Develop integrated database
 Accessible across disciplines

 Petrophysical data
 Normalize/calibrate/edit against cores
 Identify cutoffs for mapping

 Core data
 Fully describe/interpret cores
 Correlate with logs
 Calibrate with test/production data

 Seismic data
 Develop structural interpretation
 Seismic stratigraphic correlation
 Establish correlations with reservoir parameters

LOG DATA
1. Must have good correlations

2. Edit data
 Depth edit
 Baseline shifts

3. Normalize data
 Identify by type, contractor, vintage
 Compare to core data
 Calibrate to common response in same lithology
 Normalize to field average and standard deviation

4. Construct petrophysical data base

© PetroSkills, 2013. All rights reserved. 2-27

Prospect and Play Assessment
Chapter 2

CORE DATA
1. Fully describe/interpret cores
 Depth calibrate
 Note location of special analyses
2. Correlate with logs
3. Develop understanding of ranges in parameter values
4. Calibrate with test/production data

SELECTING THE RIGHT SCALE

Figure 2-3.

Permeability profiles on logarithmic (a) and linear (b) scales from plug and probe data
over the interval of DST-3 (2874.5–2882.5 m) in well 30/6-19. Note that the patterns of
permeability seen in the plug and probe data are comparable. Running average (13-

2-28 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

point) probe data are also compared with the core plug data (c) and show the probe
data to be slightly higher.

Permeability profiles on logarithmic (a) and linear (b) scales from plug and probe data
over the interval of DST-3 (2800.0–2809.0 m) in well 30/6-23. Note that the patterns of
permeability seen in the plug and probe data are comparable. Running average (13-
point) probe data are also compared with the core plug data (c) and show the probe and
plug data are similar.

EXTRAPOLATION TO OTHER AREAS

Many reservoir characterization factors are common among several different fields; if
this were not the case, we would have to develop a unique investigation for every
property without the benefit of analogues. The question then follows: What is and is not
transferable from one field setting to another? The transferable items are the principles
and resulting processes that control each of the factors. Specific products (for example
- absolute permeability of an eolian sand) of the processes may or may not transfer, but
should be examined with great care and not applied directly without analysis.

© PetroSkills, 2013. All rights reserved. 2-29

Prospect and Play Assessment
Chapter 2

Depositional environments are the basis of many reservoir characterization

interpretations, but there are many cases in which identical depositional environments
produce radically different accumulations. Variables that control the influence of a
depositional environment on a reservoir include the scale of the accumulation, tectonic
setting, differences in source material (particularly with clastic reservoirs), burial history
and diagenesis. In the Table below, a number of factors are categorized and related to
their influence on reservoir parameters. (X indicates that a property is transferable from
one field to another, O indicates those that are not):

Pressure Temp
Volumetrics

Fluid Types
Saturations
Continuity

Abs Perm

Rel Perm
Porosity
DEPOSITIONAL CHARACTERISTICS
External Geometry O O
Internal Geometry O O O O O
Pattern of Variation
O O O O O
(Quantitative Res. Properties)
Absolute Values
X X X X X X X X
(Quantitative Res. Properties)
DIAGENETIC CHARACTERISTICS
Compaction (including Geopressure) O O O
Cementation O O O O O
Dissolution O O O O O
Recrystallization O O O O O
TECTONIC CHARACTERISTICS
Folding O O O
Faulting O O
Fracturing O O O
FLUID CHARACTERISTICS
Hydrocarbons O
Formation Water O O
Injected Fluids O O

2-30 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Sequence Stratigraphy

HOW CAN WE INTEGRATE DATA

DATA - SUMMARY

 Understand scales, common terminology/understanding - communication

 What is represented (mean, min, max, range, etc)?

 Planning - very important (“Fit for purpose”)

 Measured vs. interpreted data

 Calibrate, normalize, edit

 Error - precision, accuracy

 Accessibility across disciplines

 Upscaling

 Represent multiple working hypotheses

 Represent at “true” scale

© PetroSkills, 2013. All rights reserved. 2-31

Prospect and Play Assessment
Chapter 2

SUMMARY OF CHAPTER 2
Data form the basis for the evidence that is produced by the RC process. The evidence
supports the decisions to be taken. In order to take the best decisions the critical data
elements should be acquired and in addition, the uncertainties in such data must be
specified.

Data must be fit for purpose and the costs of acquisition may not be greater than the
value of that data. Hence, certain data may not be acquired resulting in greater
uncertainties. The uncertainties in the data lead to the risk that is present when taking
decisions.

IN RC, the main question to be asked is what is the evidence that can be produced
based on the available data. Often there is no time or money to acquire additional data
to reduce uncertainties.

2-32 © PetroSkills, 2013. All rights reserved.

Chapter 3 R ESERVOIR M ODELS
We must find clues to the truth from circumstancial evidence
Alfred Wegener (1880 - 1930)
Proposer of plate techtonics

The primary aim of reservoir characterization is to define, in a comprehensive manner,

the details of distribution of reservoir rocks and fluids with the ultimate goal of a
reservoir management scheme to optimally develop a field. The key to modern
reservoir characterization is the integration of data across many disciplines. With the
advent in the last few years of numerous technical applications, integration becomes a
requirement - not just an ideal standard to be approached. These stronger links
between disciplines are driven by the technology, financial pressures, improved
scientific concepts and the changes in business styles of today’s oil and gas
organization.

Some observations:

 Each reservoir is unique

Understanding of the reservoir can be enhance by studying analogues (other
reservoirs, outcrops)
 We may never determine the real “truth”
 The data give clues to the “truth”

Solution:

 Maintain multiple working hypotheses resulting in multiple realisations (models)

in the solution space. If reality (R) is also within the solution space there is
confidence that pursuing the different models will lead to good development
plans.

Notes:

© PetroSkills, 2013. All rights reserved. 3-1

Reservoir Characterization
Chapter 3

 Size of solution space such that the least likely models on the low side (10%) and
the least likely models on the high side (10%) are excluded.

 Hence, 80% of the possible realisations are included.

 Reduce the size of the solution space (if necessary) by appraisal (data
acquisition)

RC Hint: Data  Information  Knowledge

The key challenges for organizations that want to characterize reservoirs accurately
involve several aspects:

 Tying indications from early reservoir performance to the whole reservoir

 Extrapolating data from a few discrete points to the entire field
 Converting data from one specialty to another
 Prediction of local variations within the reservoir
 Calculating the economic impact of technical work
 Conversion of the technical studies into an integrated holistic development plan.

Implied in the discussion above is not only a method to maximize the technical
assessment of a reservoir, but a requirement to communicate across disciplines to
create multi-disciplinary approaches that solve complex problems. Rather than work
sequentially within a specific discipline and then pass the final interpretation to the next
discipline, this new approach succeeds when an asset management team works
through the problem jointly. For example, geologists will talk to reservoir engineers in
order to develop a better understanding of reservoir dynamics from the existing test and
production data and then use that data to improve the depositional model; the
geophysicist may work with the petrophysicist to derive reservoir parameter indicators
from the geophysical data. To summarize, the work of developing models is integrated
and supported by feedback rather than sequentially passing the interpretations to the
next discipline.

This progression is shown below:

3-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Models

Traditional Property Development Sequence

Exploration concept

Research to confirm (geologic concepts, geophysical data gathering)

Drill test well(s)

Petrophysical analysis

Reservoir engineering analysis

Create development plan

Drill development wells

Monitor production

Each of the disciplines is involved throughout the project. For example, in planning the
3D seismic program the reservoir engineer may be consulted about the potential for
imaging reservoir compartments, the petrophysicist may offer suggestions that will
enhance attribute acquisition to be converted to porosity and permeability
measurements and the geologist could offer suggestions for design that would
maximize resolution of individual genetic units. This multidisciplinary approach to
solving the entire problem yields better answers that are more specifically focused on
the issues at hand. Strong potential exists for significant cost savings by focusing on
the specific needs of the other disciplines rather than gathering all of the data possible.

RC Hint: Tie the work of the RC team to a business purpose

The following sections will discuss data interpretation techniques that have been found
effective in oil and gas development around the world. Even though each discipline is
discussed separately, the inference still exists that multidisciplinary approaches are
used throughout where applicable. This does not mean that all decisions or
interpretations need to be made in a team or group setting. The test for whether to use
a synergistic approach is simple - Does the involvement of other disciplines give a
better (cheaper or more comprehensive) answer?

Critical to integrating interpretations across disciplines is the existence of a compatible

data set. An earlier chapter discussed data scaling and compatibility in more detail, but it
is important to keep in mind the needs of other disciplines as the models are considered.
Keep in mind that data gathered are expensive in terms of time, manpower and financial
requirements. In addition, much of the detail derived from the individual data sets is lost
when the models are created; this is a difficult balancing act that is situation dependent.
Advanced project management skills and experienced multidisciplinary team members
can yield a time, cost and manpower efficient reservoir characterization.

© PetroSkills, 2013. All rights reserved. 3-3

Reservoir Characterization
Chapter 3

MODEL TYPES

 Static
 Architecture of the field
 Giga, Mega, Macro and Micro scale

 Dynamic
 All fluid flow within the reservoir
 Production, Pressures, Saturations
 Fluid contacts, Pore system model

STATIC MODEL - DATA ANALYSIS AND INTERPRETATION - 2

 Map distribution of parameters

 Structural tops data

 "Establish tops - all horizons, faults, etc"
 Integrate with seismic data
 Create integrated maps
 "Create tabular display of well data (TVD, TVT, TST, etc)"

 Define facies
 Establish depositional model
 Define reservoir architecture

 Distribute reservoir parameters throughout reservoir

STATIC MODEL - DATA ANALYSIS AND INTERPRETATION - 3

 Reservoir data
 Input petrographic data
 Integrate geol., prod. test data
 Determine morphology of pore system
 Interpret reservoir architecture

 Well History

 Production data/injection

 Fluid data

 Develop flow unit correlations

 Redefine stratigraphic intervals

 Modify interpretations as needed

3-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Models

ANOTHER VIEW OF RESERVOIR CHARACTERIZATION WORKFLOW

One of the PetroSkills Alliance partners with the assistance of PetroSkills consulting
services has mapped the requisite skills for Reservoir characterization. These skills are
applied to each RC project and are a good checklist for workflows in other projects. The
skill groups and specific skills are shown below:

 Reservoir Rock Properties

– Facies/Isopach Mapping
– Porosity
– Permeability
– Special Core Analysis
– Data Acquisition
 Rock Fluid Properties
– Wettability
– Relative Permeability
– Capillary Pressure
– Water Saturation
– Data Acquisition
 Pressure-Dependent Rock Properties
– Compressibility Effects
– Data Acquisition
– Pressure Depletion Effects
 Assessment of Reservoir Heterogeneity
– Displacement Profile
– Structural History Effects
– Fracture Effects
– Develop Conceptual Geologic Model
– Reservoir Compartmentalization
– Parametric Averaging Techniques
– Seismic Reservoir Characterization
 Volumetric Analysis
– Original Hydrocarbons-In-Place
– Probabilistic Techniques

 Reservoir Modeling
– Geocellular Model Development
– Data Analysis
– Geocellular Model Framework
– Facies Modeling
– Attribute Modeling
– Fracture Modeling
– Data Integration Methods
– Upscaling the Geocellular Model
– Model Validation
– Reservoir Simulation Support
– Integrate Dynamic Data

© PetroSkills, 2013. All rights reserved. 3-5

Reservoir Characterization
Chapter 3

STATIC DATA MODEL

Data Form

Reservoir: Ave. Depth: Ave. Thickness:

Type Well: Interval:

Stratigraphy Observed (lithology, source, seal, reservoir?):

Overlying zone:
Underlying zone:

General Characteristics:
Lithology Porosity amt. Porosity Permeability
type
Interp. Method:
Net/Gross Net/Gross Water sol.
Interp. Method:
Texture Fabric
Interp. Method:

Interpreted Depositional Environment:

Primary reasons for interpretation
Analogues:
Modern
Stratigraphy Observed (lithology, source, seal, reservoir?):
Overlying zone:
Underlying zone:
Ancient
Stratigraphy Observed (lithology, source, seal, reservoir?):
Overlying zone:
Underlying zone:

Reservoir Dimensions:
Modern Min ML Max Ancient Min ML Max
Length Length
Width Width
Thickness Thickness

3-6 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Models

Heterogeneities:
Stratigraphic Modern Min ML Max Ancient Min ML Max
Length Length
Width Width
Thickness Thickness

Expected Min ML Max

Length
Width
Thickness

Frequency expected Distribution

Structural Mechanism Distribution

Control

Factors Controlling Reservoir Distribution:

Modern Ancient Reservoir

Production Profile:
Analogue Expected

Area Drained (include method of determining):

Analogue Expected

Critical Elements:
Analogue Expected

Key Mapping Tools:

Assumptions

© PetroSkills, 2013. All rights reserved. 3-7

Reservoir Characterization
Chapter 3

RESERVOIR PROPERTIES
(example only - repeat for each factor modeled)

Modeling objective (ave. for res., range in res., other)

Preferred method to populate grid

Observed Values - Porosity

Data Source #1
Measured / interpreted (circle one)

Lowest value
ML Second mode
Highest value

P10 -1 Std. Dev.

P50 Mean
P90 +1 Std. Dev.

Observed Values - Porosity

Data Source #2
Measured / interpreted (circle one)

Lowest value
ML Second mode
Highest value

P10 -1 Std. Dev.

P50 Mean
P90 +1 Std. Dev.

3-8 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Models

THE INTEGRATED MODEL

SUMMARY OF CHAPTER 3
Each reservoir is unique, but analogues (other reservoirs, outcrops) are used to
understand a reservoir under consideration. The truth about a reservoir may never be
known. Therefore, it is necessary to model multiple realisations of the reservoir to obtain
a complete understanding of the possibilities that may be present. The realisations are
within the solution space, which covers 80% of all the models possible. The solution
space is defined such that reality is also encompassed by it. Development plans for the
modeled realisations will then be meaningful for development of the reservoir under
consideration.

© PetroSkills, 2013. All rights reserved. 3-9

Reservoir Characterization
Chapter 3

3-10 © PetroSkills, 2013. All rights reserved.

Chapter 4 G EOSTATISTICAL M ODELING
Statistics is a means to model reality.
It does not replace thinking!

In his excellent primer on “Geostatistics and Reservoir Geology”, Journel offers the
following description of geostatistics:

In geostatistics, the prefix “geo” clearly links geostatistics to the earth

sciences. The geostatistical glossary (Olea) defines geostatistics as “The
application of statistical methods ... in the earth sciences, particularly in
geology”. Geostatistics provides toolbox for the geologist to use in
analyzing data and transferring such analysis and interpretation to the task
of reservoir forecasting. A tool can never replace data, but it can help build
an interpretation and the corresponding numerical model. Geostatistics is
no substitute for the geologist’s experience in formulating the model
properties, but it may help in creating the model.

Modeling allows the geoscientist to predict reservoir occurrence in much more powerful
ways than previously possible. The capability to offer realizations of the reservoir with
many options built into a three dimensional presentation, bring capabilities that did not
exist previously. Geostatistics lends itself to these multiple realizations allowing effective
economic analysis. Geostatistics also serves as an excellent mechanism for the
integration of geological and geophysical data to generate a three dimensional model of
a reservoir that incorporates all available data; using the modeling process brings the
disciplines together to consider the possibilities that exist within the data sets and ways
to integrate the divergent data into a cohesive model.

Stochastic modeling refers to the generation of synthetic geologic properties in one,

two, or three dimensions. A number of plausible solutions can be created and
simulated, and those results can be compared to see their effect on history matching.

The following figure shows one slice of a geostatistical model of a large Middle East
field. A huge amount of computation is needed to build such models. Efforts are now
underway to speed up the geostatistical modeling process to enable numerous model
scenarios to be tested.

© PetroSkills, 2013. All rights reserved. 4-1

Reservoir Characterization
Chapter 4

INTEGRATED MODELING

4-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

MODELING CONCEPTS

MODELING STEPS
 Definition of Layering Scheme

 Modeling Geological Units

 Modeling Rock-Type Variations within Units

 Distributing Petrophysical Properties variations per Rock-Type or Lithofacies

FROM SEISMIC TO FLOW STIMULATION

1. Load Seismic data, wells, logs, surfaces, tops

2. Manipulate Data & Create Specific data sets

3. Build Relationships between these data

4. Build Facies Models

5. Build Porosity and Permeability Models

6. Generate Saturation Model Using J Functions

and Capillary Pressure Data

7. Calculate volumetrics and Create several Types of Maps

8. Plan for new wells and Build Geobodies

9. Upscale Models Using Flux-Based Simulation

10. Export Model’s Results to Simulator

© PetroSkills, 2013. All rights reserved. 4-3

Reservoir Characterization
Chapter 4

GEOSTATISTICAL APPLICATIONS

 Data distributions
– Representing uncertainties

 Geometry of parameters

 Populating inter-well space

 Multiple working models

 Testing hypotheses

WHY USE GEOSTATISTICAL METHODS?

 Industry tends to under-predict variability, over-predict continuity

 Large volumes of data

 Correlation of data sets

 Predict spatial variations in parameters

 Expand our understanding of data/models

OBSERVATIONS ABOUT RESERVOIRS

 Reservoir dimensions much larger in horizontal direction

 Sampling much more detailed in vertical dimension

 Horizontal permeability distribution poorly sampled - uncertainty about

anisotropic nature of reservoir

 Uncertainty transferred to modeling/simulation

 Data come from:

− Projections of vertical data
− Geologic model projections
− Analog data

RC Hint: Each Field is Unique

4-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

GEOSTATISTICAL OVERVIEW

 Stochastic not deterministic answer

 Method to integrate data sets

 Two categories of geostatistical approaches

1. Estimation methods - best linear estimate

− Kriging, cokriging
− Lacks influence of extremely high or low values

2. Multiple simulation methods

− Reproduces variability
− Many equi-probable answers
− How to rank multiple answers

Wolf et al. describes one application of geostatistical processes:

The geostatistical method is a four-step procedure that calls on several

statistical tools. The first step is to quantify the spatial continuity of the well
data using Variogram analysis. The second step is to find and quantify a
relationship between the well and seismic data. The third step is to use
what has been learned to grid the well data using the seismic as a guide
via kriging with external drift. The last step is to assess the accuracy of the
map just made. Traditionally, a geoscientist creates a map that is
assumed correct until additional information becomes available. Only
rarely is an estimate made of the map's accuracy. A geostatistician
creates an expected value or average map and has a quantitative
estimate of its accuracy. Conditional simulation is a geostatistical tool that
yields a quantitative measure of the error in a map.

The traditional approach to geologic modeling was define the reservoir through the use
of horizon maps, facies distribution maps and top and bottom structural maps to define
reservoir thickness and geometry; sand/shale ratios, porosity, permeability and other
properties are typically interpolated from well to well and displayed in probability
distributions (in the more sophisticated approaches). This approach works reasonably
well with fairly isotropic reservoir, but is lacking in sufficient detail and variability for
reservoirs that are more complex. Geostatistics provides a way for reservoir scientists to
more closely approximate the heterogeneous nature of the earth. It will model both
structural information and reservoir property data into a three-dimensional model which
may he used as a basis for a reservoir simulation study.

© PetroSkills, 2013. All rights reserved. 4-5

Reservoir Characterization
Chapter 4

METHODS TO ASSIGN VALUES

 Cell based (indicator) methods

 Presence/absence at given location
 Subsurface architecture built in

 Object based
 Spatial distribution of “geo-objects”
 Conditioned to well observations and sequence stratigraphy

There are a number of techniques applicable to geostatistical modeling. A good

reference to gain a basic understanding of these applications is the AAPG text,
Stochastic Models And Geostatistics: Principles, Methods, And Case Studies, edited by
Yarus and Chambers. In this text, there are good review papers and case applications.
In one of the review papers, Srivastava explains the uses of stochastic models and an
outline of common stochastic methods; the methods described are shown below
(detailed discussion of each of these is beyond the scope of this course):

 Sequential simulation (useful for selecting values for individual parameters such
as porosity or reservoir thickness)

 Boolean or object based simulation (used to indicate distribution of objects that

have a genetic association - facies for example)

 Turning bands (which adds noise to a “smooth” model)

 Simulation in which the reservoir model is constructed by iterative trial and error

 Probability field simulation that uses original well data to construct controlled
distributions of data

As indicated in the Bashore et al. study (see sequence stratigraphy section), selection
of the model can have significant implications on the modeling approach that is used.
When using a geostatistical approach to developing a reservoir model, geoscientists
distinguish between hard and soft data. Hard data are those data that are reproduced
exactly by all methods, while soft data are reproduced with some degree of tolerance
(Journel). Examples of hard data are well data including thickness of a reservoir,
porosity, and measured permeability; soft data are those that may replicate hard data
such as a seismic attribute that has a relationship to porosity - not a direct reading from
a specific location. Hard data are very desirable, but are much less abundant than soft
data. It is important, then, to establish relationships between data sets so that the more
pervasive soft data can be used to fill in for the less frequent - but more reliable - hard
data. Therefore, selection of the most appropriate model is extremely important to the
success of reservoir characterization.

4-6 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

Srivastava (1994) offers the following warning about selection of a geologic model for
the reservoir:

Though it has become conventional to assume that a lithofacies model is

an appropriate model of reservoir architecture. It is worth considering
whether this assumption is necessarily a good idea. Though the original
depositional facies are easily recognized and described by geologists,
these may not be the most important controls on flow behavior. If the
permeability variations are due primarily to Diagenetic and structural
events that postdate the original deposition, then it is not at all clear why a
model of the original depositional units is an appropriate model of the flow
units. Other criteria, like the responses of a suite of Petrophysical logs,
might better indicate the arrangement of the flow units than would the
more visibly obvious geologic core description.

Several advantages to using detailed geological modeling are mentioned in Tyler et al.
Among them are the following:

 Stochastic simulation of reservoir heterogeneities requires much less time and

fewer resources to build and update the models than a conventional simulation,
which requires much time to iterate history matching.

 Several realizations are available for studies of future predictions, contrary to the
traditional approach of a single simulation with the one probable model for
sensitivity studies.

 Results of the high-resolution sequence stratigraphy, together with further

analysis of unwept oil in these realizations, can provide better knowledge of
reservoir sweep efficiency, infill drilling potential and oil recovery potentials.

© PetroSkills, 2013. All rights reserved. 4-7

Reservoir Characterization
Chapter 4

ARE NEARBY SAMPLES RELEVANT?

• Depends upon purpose
• Must belong to same population
• May need to segment into sub-populations
• Apply same strategy to edges as to center

FACTOR THAT INFLUENCE UNCERTAINTY

• Number of nearby samples
• Proximity of available samples
• Spatial arrangement
• Nature of phenomenon
– Smooth, well-behaved variables – proximity most important
– Very erratic data – number of samples most importantUnivariate Analysis

 Approach
 Organize, present, and summarize a dataset of a single variable

 Purpose
 Quality control
 Better understanding of the data

DESCRIPTIONS
 Graphical
 Histograms
 Cumulative Histograms

 Statistical
 Measures of location
 Measures of spread
 Measures of shape

BIVARIATE ANALYSIS

 Approach
 Comparison of distributions for two different variables

 Purpose
 Investigate the existence and degree of statistical relationship between the
two variables

4-8 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

DESCRIPTIONS
 Graphical
 Crossplots
 q-q Plots

 Statistical
 Correlation coefficient
 Rank correlation coefficient
 Linear regression
 Conditional expectation

Graphical Analysis
 Crossplots - Qualitative feel to how two variables are related; draws attention to
extreme (possibly aberrant) data pairs.

Scatter Plot Scatter Plot

200
200

100
Y Data Value

Y Data Ranks

0 100

-100

-200

0
0 10 20 0 100 200
X Data Value X Data Ranks

 q-q Plot

 Crossplot of the quantiles

 Straighter the line, the more similar the distributions; tails are of particular
interest

 If on the vertical axis the quantiles of an experimental distribution are plotted

and on the horizontal axis the quantiles of the normal distribution, a straight
line relationship would indicate that the experimental distribution is also
normal.

© PetroSkills, 2013. All rights reserved. 4-9

Reservoir Characterization
Chapter 4

q-q Plot
200

100
Y Data Quantiles

-100

-200

0 1000 2000

X Data Quantiles

GEOSTATISTICAL METHODS

 Sequential simulation

 Variograms
 Kriging
 Cokriging
 KED

 Boolean or object based simulation

 Simulation - iterative trial and error

 Probability field simulation

RC Hint: Maintain Multiple Working Hypotheses until

Proven Otherwise

4-10 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

GEOSTATISTICAL TOOLS

KRIGING AND ITS UNCERTAINTY

Well 1 Well 2 Well 3

Uncertainty

VARIOGRAMS
Variograms - also called semi-variograms - are useful tools to assign values from points
of control to points between wells. The method can be used for many attributes and is
often used to develop permeability distribution throughout the field area. Stalkup offers
an excellent explanation of how to calculate a Variogram (emphasis is added):

In this example, permeability values are known at the points shown as

black dots on a regular grid…. To calculate the semi-variograms in the x
direction, the square of the difference between permeability samples is
first evaluated for all samples…. The sum of all the squares is then
divided by twice the total number of data pairs at spacing h, and the
results plotted as the ordinate on a graph where the abscissa is the
distance between sample pairs…. A similar plot is made for samples in
the y direction. …. Values increase with increasing distance between
sample pairs until a distance… is reached at which the Variogram levels
off. This type of behavior shows that at small spacings the samples are
similar, and their differences are small. Put another way, for this type of

© PetroSkills, 2013. All rights reserved. 4-11

Reservoir Characterization
Chapter 4

semi-variograms, as the distance between sample pairs decreases,

samples become more alike. With increasing spacing samples become
less alike until at spacing…. they become completely independent of each
other. The distance… is called the range of influence. The value of the
semi-variograms where it levels off is called the "sill."

Variogram Principle
As the distance between two points decreases, the correlation between data at those
points becomes larger.

Variogram Definition
The variogram is a function that measures one half the mean square error between data
at two sample locations a distance h apart.

 At a distance of zero, the correlation is perfect.

 At any distance larger than the correlation length, the co-variance becomes equal
to the overall variance of the data

Sometimes, however, the semi-variograms appears to extrapolate to a positive value as

sample spacing becomes smaller. In other words, the difference between two sample
values doesn't appear to go to zero as the spacing between sample pairs decreases to
small values…. It indicates a degree of randomness to the sample values at small
spacings and is called the "nugget effect" in geostatistics.

4-12 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

Variogram

MAPS AND MAP MAKING (CHAMBERS ET AL.)

 Map – numerical model of an attribute’s spatial distribution

 Map used to make prediction about reservoir
 Map is poor model if it does not depict characteristics of real spatial distribution
 Geostatistics attempts to improve model through different quantitative methods

BASIC ELEMENTS (CHAMBERS ET AL)

1. Data mining (hard and soft)
2. Spatial continuity analysis
3. Search ellipse design
4. Model cross validation
5. Kriging
6. Conditional simulation
7. Model uncertainty

BUILDING A VARIOGRAM

Consider 4 different wells with coordinates (X,Y) and with a value of a particular quantity
(Z). Z may be net footage in the well, the porosity of a zone or any other quantity
relevant in the well. The distance between two wells can easily be calculated by
sqrt((X2-X1)2 + (Y2-Y1)2) and the semi-variance between two wells by ½ (Z2-Z1)2. In a
variogram, the semi-variances are plotted on the vertical axis against the distances on
the horizontal axis:

© PetroSkills, 2013. All rights reserved. 4-13

Reservoir Characterization
Chapter 4

To remove the scatter (graph on the left), the data are binned (graph in the middle).
Then a variogram model (graph on the right) can be drawn.

We are using semi-variance because for normalized data, ‘semi-variance = variance –

correlation’, or ‘variance = semi-variance + correlation’.

4-14 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

The variogram model should extrapolate back to the origin. Sometimes, already over
short distances the variogram jumps to a finite value near the origin, which is called the
nugget effect. In addition, the distance from the origin over which the variogram reaches
its maximum is the range of influence or correlation length. For distances beyond this
correlation length, correlation between the wells is not meaningful anymore.

Another example with a nugget effect:

© PetroSkills, 2013. All rights reserved. 4-15

Reservoir Characterization
Chapter 4

It may be necessary to derive different variograms in the horizontal and vertical

directions:

4-16 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

Furthermore, the variance in data may differ in different horizontal directions, which is
often the case for permeability (permeability anisotropy):

© PetroSkills, 2013. All rights reserved. 4-17

Reservoir Characterization
Chapter 4

In this example, it is seen that the correlation length in the N-S direction is somewhat
greater than in the E-W direction.

KRIGING
Kriging is a mapping technique that estimates values for a parameter and is consistent
with the Variogram model. According to Wolf et al., kriging answers the question: “What
value should the grid point be assigned that best fits the Variogram model?” Kriging is
developed by mathematically calculating grid values as a weighted average of the
surrounding control points. Hence, the Kriging estimate is a linear combination of the
values at the surrounding control points. The weights become less when the distance
between the point to be Kriged and a control point is larger.

4-18 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

7000 Variogram Model


V
7100 a
3km  r
I
2km a
• n
1km • Point to c
Be Kriged e

7400 1 2 3 4km

KRIGING PROCEDURE

CROSS-VARIOGRAMS
Cross-variograms are used to integrate “soft” data sets with “hard” data sets. For
example the data projections into the unknown areas can be improved over a simple
Variogram approach by using a seismic data set that correlates with the well properties.
Care must be exercised that the data sets have a correlative relationship. Cross-
variograms quantify the spatial relationships between the two data sets.

© PetroSkills, 2013. All rights reserved. 4-19

Reservoir Characterization
Chapter 4

DEFINITION OF SEMI-CO-VARIANCE

COKRIGING
An improvement on the kriging process involves the integration of a second, denser
(“soft”) data set with the data to be gridded. Typically, this is a seismic data set that
correlates with the original data set. The second data set (the guide data) offers
weighting for the calculated relationships and honors the variograms, and cross-
variograms. Wolf et al. present the excellent graphical depiction of this process shown
below:

4-20 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

KRIGING WITH EXTERNAL DRIFT (KED)

Cokriging is computer intensive and therefore expensive and time consuming to apply.
Geostatisticians have developed KED to overcome these limitations.

In this approach, a relationship is sought between the guide data (the seismic data) and
the well data. For instance, if the well data are net footages (isochores), seismic
amplitude at the well locations may be used to derive a relationship between peak
amplitude and sand thickness. The model used in the KED consists then of the
variogram of the well data and the relationship between amplitude and footage.

Practical applications have shown little difference between cokriging and KED. This
process is illustrated by Wolf et al. in the diagram on the next page:

© PetroSkills, 2013. All rights reserved. 4-21

Reservoir Characterization
Chapter 4

CONDITIONAL SIMULATION
Rather than developing a single "best fit" answer as in the kriging or cokriging process,
conditional simulation offers a number of plausible solutions - each of which fits the
conditions described; each of these equi-probable solutions is a potential answer to the
conditions. Typically, a great number of solutions are derived, and statistical analyses
are performed. From these analyses, the probability of any occurrence can be
determined. For example it might be determined that 32% of the solutions to the sand
thickness at a specific location might be greater than 40 feet.

4-22 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

Wolfe et al. demonstrates a geostatistical approach to assessment. In their method, Realizations are generated for
thickness of sand at a prospect location, and, then ranked from low to high.

MAPPING TECHNIQUES

Hand contour Kriging KED Cond. Simul.

Output One model One model One model Equiprobable
solutions
Data used Well data - Well data Well, seismic Well, seismic
single value/well
Appearance Smooth Smooth Noisy Noisy, highly
variable
Advantages Familiar More Combines Honors working
rigorous data sets models
Disadvantages Too simplistic? Mapping Relationship Ranking
algorithm btw. datasets realizations
Usage Quick model, Mapping, Mapping, Flow
simple setting volumetric volumetrics simulation,

© PetroSkills, 2013. All rights reserved. 4-23

Reservoir Characterization
Chapter 4

Base map with hand contoured thickness estimates.

These data can also be computer contoured.

4-24 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

The seismic horizon equivalent to the sandstone – note amplitude changes along
the interval.

Relationship between Seismic amplitude and sand thickness:

Distribution of seismic amplitudes:

© PetroSkills, 2013. All rights reserved. 4-25

Reservoir Characterization
Chapter 4

One of the realizations from the geostatistical analysis.

SAND ISOPACH MAP

KRIGING WITH
EXTERNAL DRIFT
USING PEAK AMPLITUDE
AS A GUIDE

TWO KILOMETERS

80 FEET

40 FEET

-
0 FEET

4-26 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

Conditional Simulation Kriging

Output Equi-probable realizations One deterministic solution

Properties Honors wells, variogram, Honors wells, minimizes

range of properties errors, average solution

Image Noisy Smooth

Usage Flow simulation, Mapping, volumetrics

uncertainties

Six of 100 realizations – each one is an equi-probable solution:

The thicknesses were arranged in ascending order and plotted on a graph (see
following illustration). This is a compilation for sand thickness estimates at the location.
In the prospect risk assessment, there were many trials run to predict the thickness of
reservoir at the prospect drill site. These predictions are shown here in a histogram.
Thirty-five percent of the outcomes were below the economic minimum of 40 feet of
sandstone (this economic minimum was derived separately in a parallel evaluation
process). The risk for reservoir thickness in this prospect is therefore 0.35; this means

© PetroSkills, 2013. All rights reserved. 4-27

Reservoir Characterization
Chapter 4

that if you drill a prospect like this, there is a 35% probability of failure (less than 40 feet
of sand). The adequacy for reservoir at this prospect location is 0.65 (1 – risk or 1.0 -
0.35). Keep in mind that when this prospect is drilled, only one thickness of sandstone
will be found!

S a n d D is trib u tio n

30
Economic
25 Threshold 40'

20

15

10
35% 65%
5

0
8

0
9

The following map simply displays the data generated in the last few slides at all points
on the map and computes the possibility or the risk of having less than 40 feet of sand.
One could also display the probability of having more than the economic minimum (a
more positive view). This probability is called the adequacy.

4-28 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

BOOLEAN OR OBJECT BASED SIMULATION (SRIVASTAVA)

OBJECT BASED SIMULATION

This geostatistical technique populates an area with shapes to attempt to replicate
reservoir conditions. Population of the interwell area follows a set of rules defined by
the team.

This technique is typically used to develop a distribution of reservoir and non-reservoir

facies rocks prior to a simulation process. As illustrated below, a set of rules describing
the anticipated reservoir geometry is formulated. Next, a grid is populated randomly
according to those rules. The statistics of the distribution is then compared with the
desired. Lacking and adequate set of statistics, cell locations are exchanged and
statistics re-run. This is repeated until the desired result is obtained.

EXAMPLE OF RULES FOR MODELING

1. Minimum channel belt thickness: 2m

2. Channel belt width = A x Thickness

3. Channel belts may erode all other facies types

4. The channel belt length is infinite

5. Channel belts thinner than 3m have no associated crevasse splays

6. Minimum crevasse splay width: 200m

7. Minimum crevasse splay thickness: 0.5m

8. Crevasse splay width = B x Thickness + 160

9. Crevasse splay length = 1.5 x Width

Random numbers are used to generate the bodies in the reservoir model. The shape
and the size of the bodies are determined by the rules, whereas the selected bodies
must be such that the well data (i.e. the control data) are being obeyed.

Below a theoretical display of the technique is shown followed by a case study.

© PetroSkills, 2013. All rights reserved. 4-29

Reservoir Characterization
Chapter 4

4-30 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

Application of this technique is demonstrated in the following pages for the Statford field
in the North Sea. In this field five different ‘valey fills’ can be distinguished, which were
modeled using geostatistical conditional simulation:

© PetroSkills, 2013. All rights reserved. 4-31

Reservoir Characterization
Chapter 4

4-32 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

© PetroSkills, 2013. All rights reserved. 4-33

Reservoir Characterization
Chapter 4

4-34 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

© PetroSkills, 2013. All rights reserved. 4-35

Reservoir Characterization
Chapter 4

SIMULATED ANEALING (SRIVASTAVA)

4-36 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

© PetroSkills, 2013. All rights reserved. 4-37

Reservoir Characterization
Chapter 4

464

4-38 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

SOME EXAMPLES

© PetroSkills, 2013. All rights reserved. 4-39

Reservoir Characterization
Chapter 4

GEOSTATISTICS CASE STUDIES

 N. Sea
 one well with water contact, 3D survey
 Stochastic depth maps predicted net/gross, SW’s, oil volumes
 Oklahoma Morrow
 Risk maps to predict presence of economic sand thickness
 N. Cowden Ranch
 7, 14, 55 well cases
 Each accurately anticipated porosity trends

4-40 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

COWDEN RANCH POROSITY – 7 WELLS

9.1

9.9

7.8 9.4

7.5

7.1
8.3

© PetroSkills, 2013. All rights reserved. 4-41

Reservoir Characterization
Chapter 4

POROSITY / ACOUSTIC IMPEDANCE CROSS PLOT

4-42 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

To demonstrate the differences in interpolation between mature and immature fields, we

simulated the early development of the North Cowden Unit using the actual order and
location of the wells drilled. The first seven wells drilled were used as the initial data,
and a simple map of porosity has been constructed using a standard computer mapping
algorithm. As expected, increasing detail is seen as additional wells are drilled.
Subsequent figures depict the map of porosity after 14 wells were drilled and finally, 55.

© PetroSkills, 2013. All rights reserved. 4-43

Reservoir Characterization
Chapter 4

THE TRUTH

VARIOGRAM – 17 WELLS

400
Binned data points
300
Variance

200

Variogram model
100

0
0 3 9 12 18 27 36 45

Distance between two data points

4-44 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

HAND-CONTOURED DATA – 17 WELLS

30

15
Elevation

-15

-30
0.0 Distance 100.0

KRIGING SOLUTION

“The Truth”
30
Kriged Solution

15
Elevation

Oil/Water
0
Contact
-15

-30
0.0 Distance 100.0

AVERAGE OF 100 SIMULATIONS

Kriged Solution
30 Average of 100 conditional simulations
15
Elevation

-15

-30
0.0 Distance 100.0

© PetroSkills, 2013. All rights reserved. 4-45

Reservoir Characterization
Chapter 4

SIX SIMULATIONS

4-46 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

AVERAGE WELL SPACING (M.) REQUIRED FOR DETERMINISTIC CORRELATION

Pattern Layer Cake Jigsaw Labyrinth

Rectangular 1000 600 200

Triangular 1200 800 300

© PetroSkills, 2013. All rights reserved. 4-47

Reservoir Characterization
Chapter 4

4-48 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

ANDRE JOURNEL
 “Different equiprobable numeric reservoir models can be built all honoring the
same hard and soft data”

 Puts geoscience data in language of engineers.

 Significant application in Chevron, Amoco and Shell, and in many more

companies. Geostatistics is well established and growing all the time.

 Areas to improve
- model size,
- how to select simulated models from a set of equiprobable models for flow
simulation and development planning.

SUMMARY OF CHAPTER 4
Geostatistics is a technology which produces results that are in line with all the data
used in the analysis (all well data are obeyed, seismic data may be used to guide the
evaluation). As such results have become independent of the interpreter.

Geostatistics starts with analyzing the data in order to define variograms. Variograms
show the correlation length. Correlation beyond this correlation length is not meaningful.

There are two main categories to model reservoirs geostatistically, namely

- Linear estimation techniques

- Simulation

Linear estimation techniques (such as Kriging, Co-kriging and KED) produce one single
solution, whereas simulation techniques (such as conditional simulation, Boolean
simulation and simulated annealing) produce many equiprobable solutions.

Geostatistics is quite involved and therefore requires specialists. However, it offers an

opportunity to bring scientists from different disciplines together.

The application of geostatistics has grown over recent years and is nowadays widely
used.

© PetroSkills, 2013. All rights reserved. 4-49

Reservoir Characterization
Chapter 4

GLOSSARY OF VARIOGRAM TERMS, KUPFERSBERG AND DEUTSCH

Anisotropy Ratio: The ratio of the range in the direction of maximum continuity and in
the direction of minimum continuity.
Correlogram: A chart of correlation coefficients vs. lag separation distance (see “lag”
below). In general, the correlation between petrophysical properties separated by
some distance decreases as the separation distance increases. Given some
assumptions, the correlogram is 1.0 minus the standardized or relative variogram
(see below).
Correlation Scale: The distance scale at which the correlogram correlation is zero.
From a geostatistical modeling perspective, well data have no influence on modeling
beyond the correlation scale. The correlation scale is commonly called the range of
correlation.
Covariance Function: The covariance is a measure of correlation. Recall that the
correlation coefficient is the standardized coviariance, that is, the covariance divided
by the square root of the product of the variance of each variable. So the covariance
function is like the correlogram (see “correlogram”), except that it has not been
standardized or normalized to a correlation coefficient between -1 and 1.
Distance Classes: The variogram is calculated as the average squared difference in
the petrophysical property for a given distance. In practice, well data are not on a
regular grid and the calculation of an experimental variogram must consider some
tolerance (say, 100 m ±20 in). Such a range of distance is called a distance class.
Experimental Variogram: The ultimate goal of a variogram study is to arrive at a licit
three-dimensional (3-D) variogram model that must obey specific mathematical
properties to be useful in further studies. Experimental variogram points for specific
distance classes (see “distance classes”) are calculated from the available well data
before fitting such 3-D variogram models. They are called experimental in the sense
that they precede the determination of an ultimate variogram model.
Facies: In this paper, facies simply may be considered as different rock types or
groupings of data that share certain properties. For example, limestone and dolomite
would be considered two different facies. A variogram study would have to be
conducted within each significant facies.
γ (h) or gamma (h): See the text for a formal definition of the variogram. We should
note that γ (h) is technically the semi-variogram, whereas 2 γ (h) is the variogram.
Gaussian Variogram: Depending on the context there can be two meanings: (1) a
continuous petrophysical property such as porosity or permeability commonly is
transformed to a Gaussian or normal distribution prior to variogram calculation; the
experimental variogram in this case is sometimes referred to as a Gaussian
variogram or (2) more commonly, there is an analytical function that is used to fit
experimental variograms known as a Gaussian variogram model. This analytical
function is called Gaussian because it has an ‘exponential-squared” term, which is
the same as the Gaussian probability distribution.

4-50 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geostatistical Modeling

Lag Increments: When the experimental variogram is calculated, the data pairs are
grouped into classes of similar distances. Any of these classes is also called a lag,
and the difference between the average distance of the data pairs in one class and
the average distance of the data pairs in the subsequent class is referred to as the
lag increment.
Lag Separation Vector: The vector that describes the change of direction and average
distance between one lag and the subsequent lag is called a lag separation vector.
Lag Tolerance: To rigorously define a distance class, the two parameters lag increment
and lag tolerance are used. If the distance of a data pair falls into the distance range
described by the lag increment plus or minus the lag tolerance, the data pair is
sorted into this distance class. The most common choice for the lag tolerance is one-
half the lag increment.
Nested Structure: The experimental variogram that is defined only at discrete
distances is modeled by an analytical function called the theoretical variogram. The
theoretical variogram can be composed of a sequence of individual analytical
functions each of which describes only the dissimilarity between data pairs within a
particular distance interval. In this case, the theoretical variogram is said to have a
nested structure. The term “structure” refers to each analytical subcomponents.
Normal Score Transform: The transformation of the given data so that their cumulative
distribution function corresponds to a standard Gaussian cumulative distribution
function with zero mean and unit variance.
Positive Definiteness: Each analytical function that can be used to model the
experimental variogram has to be positive definite. Positive definiteness is a
mathematical condition that ensures (1) existence of the solution of the kriging
matrix, (2) uniqueness of the solution, and (3) that the variance of any linear
combination of the data values will be positive.
Relative Variogram: The relative variogram measure is computed as one-half of the
squared difference between two data standardized by the squared mean of the data
used for the lag. This standardization distinguishes the relative from the traditional
variogram (see equation 1 in the text).
Sensitivity Study: If a model consists of several parameters that need to be
determined, it is useful to figure out those parameters that influence the model result
the most. Typically, the parameter values are varied, and the model response is
monitored. The model is most sensitive to parameters where a little change causes
a significant different model result. This procedure is called a sensitivity study.
Stratification: In this paper, stratification refers to the sequence of different fades in the
vertical direction. Inter-facies variation, in this context, is considered as the variation
of facies properties from one facies to the next fades.
Vertical Trend: A variable is said to have a vertical trend if the variation of data within a
neighborhood (vertical direction) can be described as a smoothly varying function of
the coordinates.

© PetroSkills, 2013. All rights reserved. 4-51

Prospect and Play Assessment
Chapter 4

4-52 © PetroSkills, 2013. All rights reserved.

Chapter 5 S EQUENCE S TRATIGRAPHY
Uniformitarianism
“The present is the key to the past”
James Hutton (1726 - 1797)
The first geologist ever

The recent geologic literature is full of examples of sequence stratigraphic models that
incorporate geologic models with the high quality of recent seismic data. Even though
emphasis has focused on sequence stratigraphic approaches, the basics were
developed in the last century through the work of Wagner and many others that
followed. It is not necessary to develop new and burdensome terminology to explain
the facies distribution in a sedimentary sequence.

Critical to the reservoir characterization process is being able to communicate across

disciplines in a manner that maximizes data applicable to the property to be developed.
The geoscientist should keep in mind the audience for the interpretation and translate it
into usable format and terminology. The key considerations in developing a sequence
stratigraphy model are to:

 Define the distribution of reservoir and non-reservoir facies

 Determine the continuity of the facies

 Establish the lithostratigraphic correlations

 Develop an interpretation of the chronostratigraphic correlations

 Formulate a depositional model that adequately explains that accumulation

RC Hint: Keep a multiple working hypothesis in mind

Haldorsen and Damsleth (1990) explained reservoir-description techniques:

“A reservoir is intrinsically deterministic. It exists; it has potentially

measurable, deterministic properties and features at all scales; and it is
the end product of many complex processes ... that occurred over millions
of years. Reservoir description is a combination of observations (the
deterministic component), educated aiming (geology, sedimentology, and
the depositional environment) and formalized "guessing" (the stochastic
component).”

© PetroSkills, 2013. All rights reserved. 5-1

Reservoir Characterization
Chapter 5

SCENARIO TREES
Data

Fully
Interpretation
Correlatable Meandering Distributary
Channels Channels

Realization (facies)

Realization (K,  )
(Facies related distributions)

Flow Simulator

In reservoir modeling it is important to consider multiple scenarios. The geoscientist

must make an inventory of possible models; he should think as widely as possible. All
these possible models form the “solution space”. If the solution space is wide enough,
one can be sure that reality, the sub-surface situation as it is, is part of the solution
space.
Development plans should be made for all modeled possibilities in the solution space.
At least the plans should be such that it is known what should be done if a particular
scenario in the solution space turns out to be reality. Clearly, this reasoning implies that
if reality falls outside the solution space, then modeling of the scenarios in the solution
space becomes irrelevant.

The solution space was earlier discussed in chapter 3 ‘Reservoir Models’:

5-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

RESERVOIR CHARACTERISATION WORK FLOWS

After Bergomano et al.

© PetroSkills, 2013. All rights reserved. 5-3

Reservoir Characterization
Chapter 5

After Borgomano et al.

STATIC RESERVOIR MODELS

1. Establish stratigraphic concepts and framework

2. Define facies

3. Perform petrophysical analysis

4. Define flow units

5. Apply throughout field (geostatistical methods)

6. Test concepts and modify as needed

5-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

FACIES

 Lithologic and biologic characteristics of a sedimentary deposit

 Result of processes
 Change from place to place
 Deposited only in areas occupied by its specific depositional environment

 Facies spread by the shift of depositional environments with time

 Relationships of this spread determined by:

 Sediment supply
 Relative sea levels
 Subsidence
 These relationships build the basics of STRATIGRAPHY

STRATIGRAPHY

OBJECTIVES OF STRATIGRAPHY
 Define distribution of reservoir and non-reservoir facies

 Determine continuity of facies

 Establish correlations

 Develop interpretation of time-stratigraphic correlations

 Formulate depositional model to explain accumulation

TIME AND ROCK UNITS

 Time units (not directly observed)

 Rock units (no time implication)

 Time-stratigraphic units (must be interpreted)

BASIC CONCEPTS OF SEQUENCE STRATIGRAPHY

 Depositional Sequence is the basic unit for stratigraphic analysis

 An individual sequence is the product of cyclic changes in relative sea level

 Sea level changes occur on global scale

 We can recognize sequences based on Outcrops, Seismic, Cores and Well data

© PetroSkills, 2013. All rights reserved. 5-5

Reservoir Characterization
Chapter 5

SEQUENCE STRATIGRAPHY INTERPRETATION TOOLS

 High resolution seismic data to interpret reflector geometries and sequences.

 Well logs, cuttings, and cores to indicate specific lithofacies.

 Biostratigraphy to identify and date sequence boundaries

 Geochemistry and microfossils analysis to indicate depositional environments

and ages.

 Paleo-water depth to determinate depositional environments.

BASIC IDEAS & DEFINITIONS

 Viewing rock units in terms of Time rather than Depth.

 Depositional environments vary laterally, so strata deposited during the same

time interval will contain different rock types.

 In sequence stratigraphy we assume seismic reflectors represent surfaces of

constant times that cross lithologic boundaries.

5-6 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

EMPHASIS
Facies Analysis - emphasizes internal character of strata.

Sequence Stratigraphy - emphasizes the discontinuities that interrupt facies

successions.

Unconformity
Correlative Conformity

Unconformity
Correlative Conformity

Allows recognition of discrete stratigraphic packages whose depositional origin,

geometry, and sedimentary makeup change through time.

© PetroSkills, 2013. All rights reserved. 5-7

Reservoir Characterization
Chapter 5

WALTHER’S LAW

POWER OF SEQUENCE STRATIGRAPHY

 Provides a time framework for correlating and mapping sedimentary strata.

 Gives a greater degree of reservoir predictability than we previously had.

 Fosters integration of stratigraphy, sedimentology, paleontology, geochemistry,

and petroleum geology.

 Teaches us that there is a systematic and orderly arrangement of facies, whose

locations can be better predicted because we pay greater attention to geologic
time, stratal geometry, facies trajectories, and the relationship of facies to
bounding surfaces.

5-8 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

HISTORY

Stratigraphy Pioneers Depositional Processes,Systems

Udden, Grabau, Barrell, Weller, and Facies Pioneers
Wanless, Lowman, Sloss,
Wheeler, Van Siclen, McKee Walther, Fisk, LeBlanc, Allen, Coleman, Wright, Walker,
Mutti, Ginsburg, Shinn, Harms, Fisher, McGowen, Brown,
Galloway

Sequence Stratigraphy Early Work (1970s)

Sloss, Vail, Mitchum, Todd, Sangree, Todd

The 80’s and 90’s

Van Wagoner, Haq, Posamentier, Jervey,
Galloway, Cross, Schlager, Sarg, Goldhammer,
Handford, Hunt, Helland-Hansen, Shanley,
McCabe, Walker, Nummedal, Aitken, Flint,
Armentrout, Plint, and many others

CONTROLS ON SEQUENCE DEVELOPMENT

 Eustasy

 Subsidence/Uplift

 Relative Sea Level

 Accommodation

 Frequency of Sea Level Changes

 Sediment Supply

 Base Level

 Transit Cycles

 Volumetric Partitioning

 Facies Differentiation

© PetroSkills, 2013. All rights reserved. 5-9

Reservoir Characterization
Chapter 5

VALIDITY OF INTERPRETATION DEPENDENT ON TIE TO WELLS (CORES)

SEISMIC STRATAL GEOMETRY

5-10 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

SEQUENCE STRATIGRAPHIC CYCLES

Our Tools
1. Seismic interpretation of stratigraphy and depositional systems – 2d and 3d
2. Core and well-log interpretation of stratigraphy and environments – 1d stacking
patterns, 2d-3d correlation and mapping
3. Modern analogs – depositional systems, sediment patterns, bathymetry, physical
processes, object models, 1d-3d
4. Ancient analogs (outcrops) - stacking patterns, stratal patterns, object modeling,
1d-3d
What More Do We Need?
1. Better understanding of depositional systems (processes and products) –
modern/ancient
2. Improved understanding of geobody shapes, dimensions, composition, etc. –
catalog
3. Improved understanding of stacking patterns
4. Improved recognition of lithofacies from indirect detection – logs (image), seismic
5. Improved confidence in correlation strategies (refine our understanding of
sequence stratigraphy)
6. Higher frequency seismic data to improve resolution

© PetroSkills, 2013. All rights reserved. 5-11

Reservoir Characterization
Chapter 5

E+T=S+W
Eustasy minus Tectonic Subsidence equals Sediments plus Water depth.

DEFINITIONS
 Hiatus:
 Fall in relative sea level or base level or
 Change in base level rise or
 Climate or tectonic processes in source area influencing erosion and clastic
influx
 Parasequence: Basic building block of sequences (Van Wagoner et al., 1988)
 Characterized by aggradation, progradation, and retrogradation
 Affected by sediment supply and minor base level fluctuations
 Accommodation space:
 Space available for potential sediment accumulation (Jervey, 1988); space
available below base level (Shanley and McCabe, 1994)
 Provided by tectonic subsidence and eustasy

5-12 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

WEST AFRICA EXAMPLE

TVD (m) Thickness Duration

(m) (years)

Marine 959 Top salt 33

Marine 992 Top 21 500,000

Vembo shale

Coastal 1013 Top 15 300,000

Gamba sst

Fluvial 1028 Top 129 2.5

Dentale sst million

Fluvial 1157 Base

Dentale sst

© PetroSkills, 2013. All rights reserved. 5-13

Reservoir Characterization
Chapter 5

5-14 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

PARASEQUENCE-STACKING PATTERNS

© PetroSkills, 2013. All rights reserved. 5-15

Reservoir Characterization
Chapter 5

FACIES ARCHITECTURE (SECH ET AL., 2009)

5-16 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

FACIES AND MODERN ANALOGUE (SECH ET AL., 2009)

© PetroSkills, 2013. All rights reserved. 5-17

Reservoir Characterization
Chapter 5

MAJOR SEQUENCE STRATIGRAPHIC AND CHRONOSTRATIGRAPHIC SURFACES

CAPTION BELONGING TO DIAGRAM ON NEXT PAGE

Summary diagram showing for each third-order sequence the characteristics of the
sedimentary system, the sequence stratigraphic model, and the typical reservoir, source
rock, and seal facies. The differences between these sequences are largely due to the
dominant controlling factor: eustatism for sequence I, environmental change through
clay influx for sequence II, and tectonism for sequence III. HST = high-stand systems
tract; TST = transgressive systems tract; LST = low-stand systems tract.

5-18 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

Sequence Stratigraphy Exercise

The following diagram indicates two well logs with an SP curve drawn to represent the
lithology.

1. Construct a tentative correlation for this sequence that is prograding to the right.

2. The sand represented by the SP deflection is a delta-front sand and overlays

pro-delta silt mudstones.

Hint: Draw time lines that might result from the depositional processes prior to
interpreting lithologic distribution.

© PetroSkills, 2013. All rights reserved. 5-19

Reservoir Characterization
Chapter 5

Typically, support for the interpretation of the sequence stratigraphic model comes from
the geological depositional models, study of analog environments (including outcrops
and modern environments), seismic studies and reservoir engineering data (principally
well test data). The distribution of the reservoir and non-reservoir facies provide the
basis for many of the interpretations, but it is essential to include an interpretation of the
time distribution of the facies in order to develop a complete reservoir characterization
model. Numerous interpretations of the physical data are possible, but an integrated
approach including data from other disciplines and a chronostratigraphic approach will
give a truer representation of the reservoir.

Case Study. Bashore, et al. have studied the affects of using a

lithostratigraphic interpretation versus a chronostratigraphic correlation (von
Wagoner et al.) in an Upper Cretaceous depositional environment including
continental shelf and upper slope deposits on a passive margin. The deposits
were primarily siliclastics that ranged from 50-150 m in thickness. The
depositional sequence was buried at a depth of about 1300 m. The study area
2
was 200 km with 12 wells and 8 seismic lines; Log suites and paleontologic
data were adequate for evaluation. Bashore et al interpreted the data using
strictly a Lithologic model and also using and integrated chronostratigraphic
model.

copied from Bashore et al.

Fluid flow interpretations results indicate that the geologic model had little affect
on the breakthrough time or oil recovery, but fluid displacement patterns may
be substantially different. Correlation strategies significantly affect how the
seismic (soft) data are applied within the model and therefore have significant
affects on developing continuity relationships when interpreting porosity and
permeability distributions. They indicate that an integrated model may prove
crucial for solving swept and unwept areas, developing infill drilling programs,
conformance control and designing tertiary recovery processes.

5-20 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

BIOSTRATIGRAPHY

“FORTIES” CORRELATION (NO BIOSTRATIGRAPHIC CONTROL)

“FORTIES” CORRELATION (WITH BIOSTRATIGRAPHIC CONTROL)

© PetroSkills, 2013. All rights reserved. 5-21

Reservoir Characterization
Chapter 5

LITHO- AND CHRONSTRATIGRAPHY

LITHOSTRATIGR. VS. CHRONOSTRATIGRAPHIC - MOUTH-BAR CORRELATION, AISNSWORTH, ET AL

GR
GR GR
GR
Flooding shale - datum

Flooding shale - datum

GR GR
Flooding shale - datum

Water Injection Only sand with

pressure support

Good reservoir quality

Hetherolithics
Shale

Schematic two-dimensional Illustration of lithostratigraphic vs. chronostratigraphic

mouth-bar correlation techniques between two adjacent wells. Note that in (A) and (B)
the same well data are used; however, they are correlated differently. Also note that in
(B) the thin-bedded heterolithics are correlated with the thicker, good-quality upper
mouth-bar sands and hence could contribute to reservoir performance. In this example,
irrespective of the structural dip (to the right or to the left), there is no possibility of
bypassing hydrocarbons with the lithostratigraphic correlation technique (A); however,
an infill well may be justified for either a structural dip to the right or the left If the wells
are correlated as in (B). Another impact of correlating as in (B) is that the predicted
aquifer support would be restricted relative to a lithostratigraphic correlation (A).

5-22 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

From Olson, et al.

The following figure is a schematic longitudinal section along axis of the Gironde incised
valley illustrating facies and stratigraphic patterns of the transgressive fill.

From Allen Posamentier

 Initial flooding surface

 Tidal ravinement surface
 Wave ravinement surface
 Bayline: onlap during rapid sea level rise, e.g., estuarine over alluvial

© PetroSkills, 2013. All rights reserved. 5-23

Reservoir Characterization
Chapter 5

RC Hint: Take great care in selecting the length of

correlation for stratigraphic and lithologic components

5-24 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Sequence Stratigraphy

© PetroSkills, 2013. All rights reserved. 5-25

Reservoir Characterization
Chapter 5

FLOW CHART FOR MODELING

Techniques suggested for developing a facies model for use in reservoir

characterization are shown below (MacDonald and Aasen):

SUMMARY OF CHAPTER 5
Sequence stratigraphy provides a time framework for correlating and mapping
sedimentary strata. In this work, correlation lengths needs to be considered. Since in
any stratigraphic evaluation, there are uncertainties, it is essential to maintain multiple
working hypotheses and build multiple different models.

An understanding of the difference between time units and rock units assists in the
definition of flow units.

Inclusion of biostratigraphic data improves the sequence stratigraphic analysis.

Ultimately, sequence stratigraphy intends to define the depositional environment.

5-26 © PetroSkills, 2013. All rights reserved.

Chapter 6 G EOLOGY
Field-scale characterization of reservoirs is the pivotal link between the exploration/ discovery
process and the development/reservoir management process.
Cross

Developing an understanding of the petrophysical characteristics of the reservoir(s) is a

critical aspect of reservoir characterization. This aspect typically is an effort of the
geologist, petrophysicist and reservoir engineer. The key objective is to describe the
reservoir properties that relate to producibility. Critical factors include:

 Lithology

 Net/gross ratio

 Porosity

 Permeability (horizontal and vertical)

 Hydrocarbon saturation

 Hydrocarbon types

 Diagenetic alteration

 Facies distribution

When coupled with the analyses of the facies derived from sequence stratigraphic
studies, analyses of the presence of barriers to flow, interpretation of the physical
characteristics of the reservoir facies and consideration of structural implications, flow
units can be defined. These flow units are the basic building blocks of the model which
will predict reservoir behavior. Cross describes the process well:

… three essential elements must be evaluated in field scale reservoir

characterization. The first is describing or predicting the spatial
arrangement of sedimentary facies within the entire reservoir unit. The
second element is evaluating the contributions that lithologic
heterogeneities of varying scales and characteristics make in dividing the
reservoir into compartments. The third is devising a way to convert
lithostratigraphic units into lithohydraulic units such that fluid-flow
pathways in the reservoir are described.

Cross in the same paper discusses the usage of analog data in the application of facies
models form other areas. He makes several significant points:

© PetroSkills, 2013. All rights reserved. 6-1

Reservoir Characterization
Chapter 6

 “… empirical observations and numerical models indicate that the

arrangements of facies in identical depositional systems vary with
changes in first-order controls of stratigraphic architecture: e.g.,
changes in rates of tectonic movement, base level, sea level and
sediment supply”

 “A catalog of measured sedimentary facies attributes, collected only

in the context of similar depositional systems, will amalgamate
originally discrete and distinct populations into a heterogeneous,
mixed assemblage of attributes that are similar only in name.”

 “… if these measured attributes are considered representative…

and if they are measured independently of a high-resolution time
frame, without assessment of the fundamental controls on the
observed stratigraphic architecture, then their indiscriminate
application in a reservoir simulator can introduce a false confidence
about the accuracy of the geologic framework used in the
simulation.”

A process showing the important steps is shown below:

Like all processes in reservoir characterization, this process shows a feed-back loop. A
reservoir characterization is never finished. It is always important to test concepts and
modify them where necessary, e.g. when new data become available.

The steps in the process are vital steps required to define reserves. An elegant way to
arrive at reserves resulting from static evaluations is applying Monte Carlo simulation.
With Monte Carlo simulation volumes in place and reserves can be calculated in a
probabilistic way taking into account the uncertainties in all contributing variables
captured by probability density functions.

The Monte Carlo method is highlighted in the figure on the next page:

6-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

MONTE CARLO SIMULATION

Monte Carlo simulation (probabilistic modeling) showing inputs with their uncertainties, the various stages
in the 3D modeling process and the corresponding outputs. Final result is the exceedance probability
curve of reserves.

© PetroSkills, 2013. All rights reserved. 6-3

Reservoir Characterization
Chapter 6

TRAPS

The following figure illustrates several types of traps that might be encountered.

LITHOLOGIC COMPONENTS

Building static models involves investigations and understanding of:

 External geometry of reservoir

 Internal geometry of reservoir

 Distribution of reservoir parameters

LITHOLOGIC MODELS - ELEMENTS

 Lithology

 Net/gross ratio

 Porosity

 Permeability (horizontal and vertical)

 Hydrocarbon saturation

 Hydrocarbon types

 Diagenetic alteration

 Facies distribution

6-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

STATIC MODEL – DATA ANALYSIS AND INTERPRETATION - 2

• Map distribution of reservoir
 Structural tops data
 "Establish tops - all horizons, faults, etc"
 Integrate with seismic data
 Create integrated maps
 "Create tabular display of well data (TVD, TVT, TST, etc)"
 Define facies
 Establish depositional model
 Define reservoir architecture
• Distribute reservoir parameters throughout reservoir
STATIC MODEL - DATA ANALYSIS AND INTERPRETATION - 3
• Reservoir data
– Input petrographic data
– Integrate geol., prod., test data"
– Determine morphology of pore system
– Interpret reservoir architecture
• Production data/injection
• Fluid data
• Develop flow unit correlations
• Redefine stratigraphic intervals
• Modify interpretations as needed

© PetroSkills, 2013. All rights reserved. 6-5

Reservoir Characterization
Chapter 6

ROLE OF THE GEOSCIENTIST IN RESERVOIR STUDIES

DEPOSITIONAL ENVIRONMENTS

Depositional environments can be subdivided into three categories, namely nonmarine

environments, coastal environments and marine environments:

Knowledge of the depositional environment adds to the understanding of the reservoir

and assists in the prediction of reservoir characteristics. Moreover, data is usually
limited and therefore, uncertainties are always present. Knowledge of the depositional
environment assists in coping with these uncertainties.

The next diagram summarizes these ideas:

6-6 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

OUTCROPS

Outcrops are essential for obtaining statistical data on sand bodies. In the following
picture the channel sand body is clearly visible:

Reynolds (AAPG) produced an excellent paper providing many statistical data for sand
bodies.

Petroleum reservoirs in paralic successions commonly comprise a wide range of sand-

body types. To estimate the volume of petroleum that is percent and the percentage

© PetroSkills, 2013. All rights reserved. 6-7

Reservoir Characterization
Chapter 6

that is recoverable, and to optimize development and production schemes, it is vital to

know the dimensions and orientation of each sandstone body Although the sedimentary
facies and the orientation of paralic sand bodies have been extensively studied, there is
little quantitative data on sand-body dimensions. Reynolds’ paper addresses that data
gap and reports an extensive database of widths, lengths, and thicknesses of paralic
sandstones.

Following an approach that has been successful in fluvial successions, an attempt is

made to develop predictive relationships between sand-body thickness, which is
commonly known from well data, and the key unknowns of width and sand-body length.
In pursuing this goal, in addition to classifying the data by sand-body type, grain size,
basin type. etc., key aim of this study has been to test the hypothesis that sand-body
dimensions are controlled by their sequence-stratigraphic setting. Crossplots of sand-
body width against thickness show that discrimination of the data by sand-body type
produces the tightest set of clusters; therefore, this should be the first step in choosing a
realistic range of dimensions for a given sandstone thickness. Further analysis reveals
that sequence stratigraphic setting is a useful means of refining the choice of analog
and reducing the range of dimensions.

The following graphs present the data:

Dimensions of Sand Bodies

Reynolds, 1999 AAPG

A log-log plot of width vs. thickness for all available data with sandstone
bodies differentiated by sand-body type. Sand-body types show clear
clustering and only limited overlap of dimensions. Two lines recording
thickness to width ratios of 1:100 and 1:1000 are shown for reference.

6-8 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

Dimensions of Sand Bodies

Reynolds, 1999 AAPG

Cumulative frequency plots of sand-body widths differentiated by sand-body type.

Specific conclusions are:

(l) Valleys are much wider than channels. Crude averages show that channels
(distributary channels, crevasse channels, etc.) are narrow, less than 1 km in width, with
aspect ratios of 1:100. Whereas valleys average around In km in width and have aspect
ratios on the order of 1:1000.

(2) Shoreline-shelf sands are huge sheets tens to hundreds of kilometers in length, with
mean widths that range from 7 to 25 km and vary according to systems tract. Shoreline-
shelf sands deposited in highstand systems tracts are, on average, twice as wide as
those deposited in transgressive systems tracts. Thicknesses of shore-line-shelf sands
relate strongly to their position within a sequence set.

(3) Valleys and shoreline-shelf sands have areal extents comparable to giant oil fields.
Flood tidal deltas and mouth bars are comparable in area to small fields.

(4) Distributary channels crevasse channels, tidal flats, and crevasse splays are small
areally.

(5) Some systematic trends occur (e.g., in flood tidal deltas), allowing the prediction of
width and length from thickness. Other data sets show a high degree of scatter, but
maximum, mean, and minimum values can be determined for width, length, and
thickness. There is a clear partitioning of certain sand-body types into certain systems
tracts.

Reynolds also presented a schematic with dimensions of sand bodies:

© PetroSkills, 2013. All rights reserved. 6-9

Reservoir Characterization
Chapter 6

Typical shapes of sand bodies are shown in the next two diagrams:

ALLUVIAL FANS

In many environments of deposition, water plays a major role. Depending on the

velocity of water certain sizes of particles can be carried by the water:

6-10 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

The relationship between particle size and velocity of the water can be used to
understand the deposition of particles in the various environments.

An alluvial fan deposit occurs on the edge between the mountains and the continental
plain. At the transition, a sudden change in velocity happens:

Using the relationship, a range of particles that can not be carried any more by the
lower velocity water will be deposited. Hence, the sorting in this deposit is relatively poor
implying a reduced permeability.

RIVERS

Two types of rivers occur, namely braided rivers and meandering rivers. The braided
river is usually somewhat higher in the mountains, whereas the meandering river
meanders across the continental plain.

© PetroSkills, 2013. All rights reserved. 6-11

Reservoir Characterization
Chapter 6

The river banks of the braided river are relatively strong and well defined causing the
river to flow along a relatively well-defined path. The meandering river can meander
over an area of several kilometers wide. Consequently, the deposition in the braided
river is localized whereas deposition in the meandering river can be found spread-out
over the pain as visualized in the following diagram:

The velocity of the water in the braided river is fairly constant, whereas for the
meandering river the velocity in the inner bends of the river where the deposition of
material takes place, is reducing with time. As a result, the sorting in a braided river
deposit is very good, whereas in the meandering river deposit a fining-upward sequence
is found.

For a braided river, the distance of transport of the particles has been relatively short so
that the size of the particles is still relatively large. This leads to high permeabilities,
which may as high as 2 D. In the meandering river deposit the quality of the deposits is
reducing when going up implying deteriorating permeability when moving up (fining-
upward sequence).

DELTAIC DEPOSITIONAL ENVIRONMENTS

There are three types of deltas:

– River-dominated

– Tide-dominated

– Wave-dominated

6-12 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

When a river flows into the sea or ocean, the velocity of the water is gradually reduced.
As a result, the finest particles are transported further away from the river mouth so that
larger particles are stacked on top of smaller particles. In this way, a coarsening-upward
sequence is deposited. Also when particles are picked up by the waves and transported
e.g. along shore (see Galveston picture), first the finest particles are picker up and
transported over the largest distance leading again to coarsening-upward sequences.
Consequently in coastal deposits, the best developed reservoirs are found near the top
of deposits.

In the next diagram, more examples of deltas are given:

In summary:

© PetroSkills, 2013. All rights reserved. 6-13

Reservoir Characterization
Chapter 6

EOLIAN DEPOSITIONAL ENVIRONMENT

Eolian deposits are found in deserts. The material is picked up from beaches by the
wind and transported over sometimes very large distances to the place of deposition.
Cap rocks of Eolian deposits are often transgressive shale in combination with salt
deposits.

TURBIDITES

Some material is carried across the continental shelf and dropped into deep water
forming turbidites.

The material has been transported over a long distance. Hence, turbidites consist of fine
often shaly material and have a thinly-laminated character. Nevertheless, very good

6-14 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

production can be obtained from turbidites. Bonga offshore Nigeria, the first turbidite
development offshore Africa, produces 240,000 bbls per day.

Turbidites are often found in deep water and therefore, often developed using FPSOs.

GEOMETRIC COMPLEXITY OF RESERVOIRS

An excellent paper was presented by Weber and van Geuns (JPT), who proposed a
categorization of reservoirs in three different types so as to distinguish their complexity.
These types are:

 Layer cake reservoirs

– example: stacked barrier bars

 Jigsaw puzzle reservoirs

– example: deltaic complex

 Labyrinth reservoirs

– example: stacked fluvial sequence

The next diagram shows their different geometries:

© PetroSkills, 2013. All rights reserved. 6-15

Reservoir Characterization
Chapter 6

LAYER
–CAKE RESERVOIR TYPE

6-16 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

JIGSAW PUZZLE RESERVOIR

Mugu lagoon tidal channel, California

© PetroSkills, 2013. All rights reserved. 6-17

Reservoir Characterization
Chapter 6

LABYRINTH RESERVOIR TYPE

Meandering river, Congo

6-18 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

Reservoir DATA FOR DEFINITION

Type Key Heterogeneities Seismic Wells Production Analogue
Layercake High continuity, Sequence Cores, logs Vertical pulse Shale length
wedge outs, stratigraphy (facies, K), test, long tests, stats,
permeability RFT/MDT PLT depositional
models
Jigsaw Thief zones, flow 3D seismic, Cores, borehole Long tests, Sand body
unit orientation, seq. Strat. imaging, logs PLT, horizontal char acts.,
K contrasts, local (facies, k), tests depositional
baffles RFT/MDT models, shale
length stats
Labyrinth Orientation of flow 3D seismic, Cores Horizontal Sand body
units, interconnectivity, VSP, well to (oriented), Pulse tests, size,
dead end gravity traps well borehole prod. Tests, Depositional
tomography imaging, RFT, PLT model
logs for
probabilistic
modeling,
horizontal wells

In summary:

Heterogeneity Wells per km2

Layercake High continuity 1

Jigsaw K-contrast 2 to 4

Labyrinth Inter- 10 to 30
connectivity

Tools that can be used for definition of the different types are:

Seismic
- sequence stratigraphy
- 3D seismic, VSP
- well-to-well seismic (possibly)

© PetroSkills, 2013. All rights reserved. 6-19

Reservoir Characterization
Chapter 6

Wells
- logs and cores
- borehole imaging tools
- RFT / MDT

Production
- production tests
- well interference tests, pulse tests
- tracer tests

Outcrops
- shale length
- san body characteristics and size
- depositional models

Recovery from clastic reservoirs may vary between 5 and 80% depending on many
variables. In the next diagram, some statistics are shown depending on depositional
environment and drive mechanism.

In the diagram also a comparison is made with recovery efficiencies in carbonates.

Carbonates will be discussed later on in this chapter.

6-20 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

MODELING FLUVIAL SANDSTONES (BRIDGE AND TYE)

The following figure shows three channel belts present in part of the Colville River flood
plain, Alaska. Note the juxtaposition of the narrow channel belt of the Kogosukruk River
(left) formed by a single, sinuous channel and the multiple, sinuous-to-braided channels
in the much wider Colville River channel belt (center). Active and abandoned channels
and bars are easily discernible. To the right is another abandoned channel belt. Each of
these channel belts could be identified in the subsurface using core and wireline-logs,
and accurate predictions of their dimensions could be made. Photograph from July 1979
in the National Petroleum Reserve Alaska, approximately 40 km (25 mi) northeast of
Umiat.

And the next figure shows simplified plan forms, cross profiles in various positions and
orientations, and large-scale stratal geometries of channel belts for braided and
unbraided sinuous rivers. Braid bars and point bars migrated downstream and
expanded laterally. Large-scale inclined strata are shown schematically on cross
profiles, and upper bar and lower bar deposits are distinguished. The total thickness of
bar deposits varies laterally and is generally less than the maximum bankfull depth.

© PetroSkills, 2013. All rights reserved. 6-21

Reservoir Characterization
Chapter 6

(A) Two-dimensional stratigraphic simulation of fluvial channel belts (stippled) in the Travis
Peak Formation, North Appleby field, East Texas basin (zone I of Tye, 1991! Davies et al).
(B) Wireline logs and channel-belt sandstone bodies of zone I as originally correlated by Tye
(1991). (C) Revised well-to-well correlations using same data as in (B), but constrained by
recalculated channel-belt widths. Note that limiting channel-belt widths decreased sandstone
body continuity in two dimensions.

6-22 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

Net-to-gross ratios and accommodation:

Width to thickness ratios:

Note that width/thickness ratios of sand bodies of some of the meandering systems can
in places be relatively small, because of the thick packages of cohesive sediments in
which these systems develop. Apparently lateral accretion in meandering systems can
be restricted under such conditions, just as in anastomosing systems. Note also that
multistory sand bodies have a significantly different w/t trend than single-story sand
bodies.

© PetroSkills, 2013. All rights reserved. 6-23

Reservoir Characterization
Chapter 6

Lithofacies in fluvial systems:

After Allen

SEDIMENTARY
FACIES
ENVIRONMENT
A. Fine-grained sandstones Fluvial channels
A2: Course - V. coarse sandstones Braided Stream
B. Rooted siltstones Paleosol
C. Very course - to fine - grained sandstones with clay drapes Upper Estuarine channels
D. Parallel laminated black shales with climbing ripples Central Estuarine bay
E. Flaser and wavy bedded sandstones and siltstones Restricted tidal flag
F. Inclined deformed sandstones and siltstones Tidal channel
G. Laminated calcareous mudstones Lower estuary
H. Poorly to moderately fossiliferous planar cross-bedded Estuarine mouths
sandstones and pebble conglomerates
I. Highly fossiliferous planar cross-bedded very coarse- to medium- Upper shoreface
grained sandstones and pebble conglomerates
J. Rarely to moderately burrowed planar cross-bedded medium- to Proximal middle shoreface
fine- grained sandstones
K. Moderately to thoroughly burrowed rippled fine-grained Distal middle shoreface
sandstones
L. Thoroughly burrowed fine- to very fine-grained silty sandstones Lower shoreface
and fading ripples
M. Thoroughly burrowed very fine-grained silty sandstones and Offshore transition
sandstones with interbedded normally graded sandstones
N. Thoroughly burrowed siltstones Offshore
O. Parallel laminated black shale Shelf

6-24 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

BRACKISH CHANNEL LITHOFACIES

LAGOONAL LITHOFACIES

© PetroSkills, 2013. All rights reserved. 6-25

Reservoir Characterization
Chapter 6

LOWER ESTUARY LITHOFACIES

TIDAL INLET LITHOFACIES

6-26 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

UPPER SHOREFACE LITHOFACIES

Incised Estuary – dominated valley-fill Regressive marine-dominated succession of

transgressive succession of facies (Gaskill, facies ( Kendrick, Gentzler Field)
Arroyo Field)

© PetroSkills, 2013. All rights reserved. 6-27

Reservoir Characterization
Chapter 6

TYPICAL GAMMA RAY RESPONSES

GR and grain size in river deposits:

Typical vertical sequences of lithofacies from different parts of sandy channel bars and channel fills.
Idealized gamma-ray logs also given.

6-28 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

Legend for sedimentological logs:

kH,max

© PetroSkills, 2013. All rights reserved. 6-29

Reservoir Characterization
Chapter 6

Measurements of paralic sandstone bodies. (A) Sandstone bodies were measured at the parasequence scale.
An individual parasequence may yield a number of measurements in this example three channel
measurements (1 to 3) and a shoreline-shelf sandstone body measurement (4). (B) In plan view in each case,
width is the minimum sand-body dimension and Length is the maximum sand-body dimension. (C) Sand-body
thickness data have been collected by measuring maximum thicknesses. A randomly placed well penetration is
likely to encounter a reduced thickness, closer to the mean thickness. By contract, in sheet sands a randomly
placed well is more likely to penetrate a thickness close to the maximum. (D) Multistory and multilateral channel
sandstone bodies, respectively, record numerous stacked and adjacent channel positions. At a parasequence
scale, multistory and multilateral channels appear to be rare in paralic successions, and as a result, it was
possible to collect data from individual channel sandstones.

6-30 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

From: Eschard et al, 1998 AAPG

© PetroSkills, 2013. All rights reserved. 6-31

Reservoir Characterization
Chapter 6

Models of major depositional environments. The curve on the left shows the SP or gamma ray response and
the curve on the right shows the relative grain size profile. The size of the dots next to the vertical profile
indicates the relative magnitude of permeability expected in such a sequence. (Parts c and d are from Walker,
1984, and parts f, h, and I are from Galloway and Hobday.)

GR responses in meandering channels and coastal deposits:

6-32 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

After C. Kendall, 2003 (modified from Emery)

GR response in stacked channels:

© PetroSkills, 2013. All rights reserved. 6-33

Reservoir Characterization
Chapter 6

PREDICTING DYNAMIC RESERVOIR RESPONSE

The following case study is a study using the Paria river outcrop. The Paria River is a
contributory to the Colorado River in the US. The study simulated a water flood under
various permeability distributions.

The study objective was to understand the effect of permeability heterogeneity on the flood
front.

(a) The exposure of the Paria River a sandstone incised valley fill used in the reservoir simulations. (b) A
lithological interpretation is made initially, before this interpretation is corrected for any perspective distortion
developed when acquiring the outcrop photomosaic. (c) The corrected panel is then gridded into 661 × 81
rectangular simulation cells of 0.74 × 0.61 cm. Corrected interpretation is used as input for Eclipse reservoir
simulations. The Paria River photomosaic is a section through the A sandstone incised valley fill unit,
perpendicular to paleoflow, that is, paleocurrent was into the picture.

6-34 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

Water saturation plots at 0.25 pore volumes injected for four realizations of the full model with an injector-
producer pair injecting to obtain an interstitial velocity of 12 cm/day where sand is (a) homogeneous or has
a stochastic distribution with a volumetric average permeability of (b) 238 md, (c) 280 md, or (d) 182 md.
Arrows indicate the location of wells and direction of flow within.

© PetroSkills, 2013. All rights reserved. 6-35

Reservoir Characterization
Chapter 6

Water saturation plots for the same permeability distribution plotted at 0.25 pore volumes injected where
(a) heterolithics have been replaced by shale and where sand flow properties have been assigned to (b)
heterolithics, (c) shales, or (d) both heterolithics and shales. Arrows indicate the location of wells and
direction of flow within.

The results of this study lead to a very important conclusion that needs to be realized
when upscaling models for reservoir simulation purposes

RC Hint: When model parameters (e.g. permeabilities) are

averaged, the performance of the reservoir model is
improved

6-36 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

A second case study involves data from the Lower Jurassic Kayenta outcrop. In this
study, the objective was to evaluate the contribution of very low permeability (~ 1 mD)
conglomerates on the recovery efficiency.

© PetroSkills, 2013. All rights reserved. 6-37

Reservoir Characterization
Chapter 6

Fluvial Deposits – Complexity

After North and Taylor
Width:
Facies Thickness Width
Thickness

Non-channelized sheet 0.5-2 30-100 15-200

Poorly channelized sheet 3-4 50-100 13-33

dominated by interbedded facies

Poorly channelized sheet 4-6 100-350 17-88

dominated by facies Cpc/Cpf

Poorly channelized sheet 4-6 100-350 17-88

dominated by facies Cxd/Cxp

Channelized sheet 3 10-50 3-17

Compound ribbon 25 200-370 8-15

Ephemeral-fluvial sandstones form the main reservoirs in many oil and gas fields.
Production histories and development difficulties suggest these deposits are more
complex than is commonly supposed in the petroleum industry. There is inadequate
published information to account for this observation. Relevant sedimentological
literature on such deposits is qualitative and unsuitable for detailed reservoir
characterization studies. Therefore, integrated sedimentological, petrophysical and fluid-
flow simulation studies have been conducted on outcrop analogs in the southwestern
United States. Our key objectives were to define flow units and to find areas of
sensitivity inherent both in the sediments and in the analytical and statistical procedures
used to describe and model such systems.

We report results of integrated studies on the Lower Jurassic Kayenta Formation in

southeastern Utah. The Kayenta Formation is a relatively proximal, sand-rich
succession dominated by channel-fill sandstones. Sedimentological analysis identified
eight main genetic units. In contrast, reservoir simulations, based on probe-
permeameter data collected at outcrop show that six of the main genetic units have
uniform flow behavior. Intra-formational mudclast conglomerates are the dominant
permeability barriers, both because of the clasts themselves and because they act as
sites of preferential cementation. Stochastic permeability models conditioned to the
outcrop data produce results little different from models based on averaged values for
each facies. The most sensitive feature of the models concerns the assumptions made
about the mud-clast conglomerates. The model results show that only by integrating
analytical approaches can we understand the full character of an outcrop analog in such
a way that we can use it predictively.

6-38 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

WATER BREAKTHROUGH SIMULATION

 Model 1
 Each cell assigned mean permeability value
 Model 2 - Assess influence of mudclasts on flow
 All conglomerates impermeable (Cgu, Cgc)
 Model 3A - assess permeability range in facies
 Stochastic permeability model - resembled #1
 Model 3C - A.A., but set conglomerates to 1 md
 Model 4 - Utmost realism
 Added spatial variability to sandstones
 Note under-running in basal sequence

© PetroSkills, 2013. All rights reserved. 6-39

Reservoir Characterization
Chapter 6

Conclusions of the study are:

1. Similar to the previous study, if permeability heterogeneities are introduced, the

performance of the reservoir as seen from the stability of the flood front is
deteriorating.

2. Low permeability (~ 1 mD) in the conglomerates is sufficient to recover the

hydrocarbons from the conglomerates.

6-40 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

DIAGENESIS

Major diagenetic processes are:

Two major components of the sedimentary model:

© PetroSkills, 2013. All rights reserved. 6-41

Reservoir Characterization
Chapter 6

GEOLOGIC FACTORS AFFECTING SUBSURFACE RESERVOIR

GEOLOGIC FACTOR EFFECT ON POROSITY
1. Ancient destructive diagenesis Reduces porosity
(sediment infill, cementation,
recrystallization, mechanical
and chemical compaction)
2. Ancient constructive diagenesis Enhances porosity
(dissolution, fracturing)
3. Other ancient diagenesis May reduce or enhance porosity
(mineralogic replacement,
authigenic clay growth,
brecciation, tectonic deformation)
4. Framework composition/provenance May control post-depositional diagenesis
Environment of deposition Controls pre-diagenesis porosity
5. Paleoclimate Affects EOD, weathering, karstification
6. Depth of burial Indirectly related to porosity loss
7. Pressuring and overpressuring Early overpressure may enhance porosity
8. Thermal maturity Indirectly related to porosity loss
9. Erosional events/unconformities May reduce or enhance porosity
10. Pore fluid migration (water) Enhances cementation/dissolution
11. Pore fluid migration (oil) Inhibits cementation/dissolution
12. Associated rock strata—seal Affects pore fluid entrapment
13. Associated rock strata— source rock Controls type of migrating pore fluids
Modified from: Tobin

VARIABLES AFFECTING RESERVOIR QUALITY

Comparison of variables affecting reservoir quality in sandstones and carbonates.
After Choquette and Pray.
Variable Sandstones Carbonates

Sediment High variability (depending on Low variability [variations of

composition provenance and depositional CaC03 and MgCa(C03)2]
environment)

Cement Quartz, calcite, dolomite, clay Aragonite, high- and low-Mg

mineralogy and anhydrite, etc. calcite, dolomite

Original pore Intergranular Intergranular predominates,

geometry but intragranular is important

Ultimate pore Intergranular = intercrystalline > Intergranular = intercrystalline

6-42 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

geometry moldic = moldic > microporosity

Uniformity of Fairly uniform within a facies Ranges from fairly uniform to

pore size, extremely heterogeneous,
shape, and even within a facies
distribution

Influence of Minor to major Usually major

diagenesis

Schematic diagram illustrating common diagenetic environments – Moore.

CARBONATE DIAGENETIC ENVIRONMENTS

The following illustration makes it easier to understand the role of different water
chemistries in changing the mineralogy of carbonates.

It is easy to see that diagenesis in carbonates can take place in the depositional
environment; it is not necessary to bury it or expose it to start this process.

© PetroSkills, 2013. All rights reserved. 6-43

Reservoir Characterization
Chapter 6

CARBONATE MINERALS AND THEIR THERMODYNAMIC STABILITY

The key to understanding reservoir properties in carbonates relates to the variety of
component mineral stability. Particles made from aragonite are most unstable and
prone to alteration even prior to burial. It is this diagenesis that makes carbonate
reservoirs a challenge to model.

6-44 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

FACTORS CONTROLLING SANDSTONE DIAGENESIS

Depositional Environment Tectonics

(determines)

Sand/Shale Sand Body Sedimentary Hydraulic Sorting Provenance

Ratio Geometry Textures

Fluid Flux Fluid Chemistry Detrital Mineralogy

Time,
Diagenesis Temperature,
Pressure
(which ultimately determines porosity and permeability)

Major diagenetic processes - impact on porosity

Major Diagenetic Processes
Zone Temp.
Preserves or Enhances
Destroys Porosity
Porosity
Shallow <80°C or  Grain coatings (inhibit  Clay infiltration
176° F later overgrowths)  Carbonate or silica cement
(<5,000 to  Non-pervasive (in some cases
10,000 ft) carbonate cements that irreversible)
can be dissolved later  Authigenic kaolinite
 Compaction of ductile
grains
Intermediate 80-140°C  Carbonate cement  Kaolinite, chlorite, and illite
Or 176- dissolved precipitate as a result of
284°F  Feldspar grains feldspar dissolution
dissolved  Ferroan carbonate and
quartz cement
Deep >140°C or  Feldspar, carbonate,  Quartz cement (most
284°F and sulfate minerals destructive)
dissolved  Kaolinite precipitation
 Illite, chlorite form as
products of feldspar
dissolution
 Pyrite precipitation

© PetroSkills, 2013. All rights reserved. 6-45

Reservoir Characterization
Chapter 6

Cements vs. facies/environments

Environment Facies Cement

Dune Quartz overgrowths dominate; also clay coats

Eolian Grain-coating clays; in areas that were

Interdune alternately moist and dry anhydrite, dolomite, or
calcite common

Flood plain Chlorite, illite, smectite, and mixed-layer clay

Fluvial Calcite at base, grading u p into calcite plus

Channel quartz, quartz plus clay minerals, and clays plus
minor carbonate

Carbonate minerals such as calcite or siderite;

Near shore marine All illite in sands deposited where fresh and marine
water mix

Calcite mainly; also dolomite, illite, chlorite rims,

Marine shelf All
and quartz

Greater variety than other environments;

Deep marine All cements include quartz, chlorite, calcite, illite,
and occasional siderite or dolomite

6-46 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

POROSITY THROUGH TIME

© PetroSkills, 2013. All rights reserved. 6-47

Reservoir Characterization
Chapter 6

Characteristics of Dry, Freshly Broken Surfaces

at 20x Magnification
Pore
Visible Pinpoint Consolidation Permeability
Type
Type I Subclasses
I Abundant to Very Needle prove
IA:>100 md
common; abundant to easily dislodges
IB:>10-100 md
interconnection common many grains
IC:>1-10 md
visible on many from rock
ID:>±0.5-1 md
pores surface
± 0.5-1.0 md
II Scattered Abundant to Needle probe
common can only (depending on
particle size,
occasionally
sorting and clay
dislodge a grain
from rock mineral content)
surface
Too tight to
III None to very isolated None to a few Usually very
produce gas at
pores well
commercial rates
consolidated
and/or pores even when
filled with clays fractured or
interbedded with
or other pore-
type I rocks
filling material

SNEIDER’S PORE CLASSIFICATION FOR CLASTICS

Sneider and King developed a simple method of classifying pore types from cuttings.
The classification of clastic rock pore types from cuttings is made by comparing pore
types with production tests and log analysis. The pore types are as follows:

Type/Description
I. Rocks with pores capable of producing gas without natural or artificial fracturing.
II. Rocks with pores capable of producing gas with natural or artificial fracturing
and/or interbedded with type I rocks.
III. Rocks too tight to produce at commercial rates even with natural or artificial
fracturing.

Procedure: Predicting Sandstone Permeability

The procedure below, with steps and recommended action, is for predicting the
permeability of sandstones from cuttings using 20x magnification (from Sneider and
King.

6-48 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

Step/Action
1. Estimate grain size and sorting using standard size-sorting comparators, thin
section and SEM photomicrographs, and rock photographs.

2. Estimate volume percentages using Terry-Chillingar charts made for volume

estimates.

3. Estimate consolidation using the scheme described in the preceding table.

4. Describe the visible and pinpoint porosity and interconnectedness.

5. Estimate permeability from rocks on comparators and/or using rock

characteristics described in the preceding table. (Comparators can be made or
purchased.)

6. Predict permeability for the formation in prospective areas where petrophysical

characteristics are believed to be similar.

Lateral Heterogeneity
Low Moderate High
Wave dominated Delta-front Meander belt,
delta, Barrier mouth bar, Fluvially
core, Barrier Proximal delta dominated
Low
shore face. front, Tidal delta, Back
Sand-rich strand deposits, mud- barrier
plain rich strand
Eolian, Wave- Shelf bars. Braided
modified delta Alluvial fan, Fan stream, Tide –
Vertical
Moderate (distal) delta. dominated
Heterogeneity
Lacustrine delta
delta, Distal
Basin-flooring Coarse-grained Back barrier,
turbidites meander belt, Fluvially-
Braided delta dominated
High
delta, Fine-
grained
meander belt
After Galloway et al.

© PetroSkills, 2013. All rights reserved. 6-49

Reservoir Characterization
Chapter 6

CARBONATE RESERVOIRS

 Typically thick intervals

 Tend to be made of one lithology
 Complex relationship between porosity and permeability
 Extremely heterogeneous
 Require detailed, complex development plans

Carbonate Sediments Siliclastic Sediments

The majority of sediments occur in Climate is no constraint, sediments occur
shallow, tropical environments worldwide and at all depths
The majority of sediments are marine Sediments are both terrestrial and marine
The grain size of sediments generally The grain size of sediments reflects the
reflects the size of organism skeletons hydraulic energy in the environment
and calcified hard parts
The presence of lime mud often indicates The presence of mud indicates settling
the prolific growth of organisms whose out from suspension
calcified portions are mud size crystallites
Shallow water lime sand bodies result Shallow water sand bodies result from the
primarily from localized, physiochemical interaction of currents and waves
or biological fixation of carbonate
Sediments are commonly cemented on Sediments remain unconsolidated in the
the seafloor environment of deposition and on the
seafloor
Periodic exposure of sediments during Periodic exposure of sediments during
deposition results in intensive diagenesis, deposition leaves deposits relatively
especially cementation and unaffected
recrystallization
Signature of different sedimentary facies The signature of sedimentary facies
is obliterated during low grade survives low-grade metamorphism
metamorphism

CARBONATES SUMMARY
Clastics Carbonates
Deposition after transport Deposition in situ
Climate no constraint Sediments occur in tropical areas
Worldwide at all depths Shallow water
Both marine and terrestrial Marine
Grain size related to distance of transport Grain size reflect size of skeletons

6-50 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

Production from carbonates

 50% platform edge grainstones
 20% reefs - platform margin and isolated
 10% platform slope deposits
 5-10% shoaling upward cycles
 2% basinal limestones
 5-10% diagenetic traps

DEPOSITIONAL MODEL
Under low-stand conditions and in a humid environment, a rimed platform may undergo
karstification due to widespread subaerial exposure. This may lead to subterranean
drainage and the development of caves near the water table. At the shelf margin, low-
stand reefs and/or grain shoals may form. The shallow-marine carbonate factory is
greatly reduced in size. Although slope failure and the generation of sediment gravity
flows can occur during any sea-level position, some triggering mechanisms may favor
low-stand conditions. Debris-flow deposits may form lobes or aprons at the toe-of-
slope, whereas turbidites may extend further out onto the basin floor as apron or fan
lobes. Handford and Loucks.

© PetroSkills, 2013. All rights reserved. 6-51

Reservoir Characterization
Chapter 6

Generalized depositional models for Clear Fork and lowermost Glorieta formations, in North Robertson unit,
Gaines County, Texas.

CARBONATE ROCKS CONSTITUENTS

 Grains
 Matrix
 Cement
 Fossils
 Pellets
 Ooids
 Interclasts

6-52 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

Fossils

Pellets
 Sand-size, spherical or ellipsoidal particles
 Formed by deposit-feeding organisms that eat mud.
 These organisms digest organic matter and excrete the non-digested mud as
pellets.

Ooids
 Composed of Aragonite
 Nuclei in the center
 Concentric or radial
 Well Sorted

© PetroSkills, 2013. All rights reserved. 6-53

Reservoir Characterization
Chapter 6

Intraclasts
 Large rounded grains
 Angular sometimes
 Internal structure
 Reworked carbonates

6-54 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

CLASSIFICATION OF CARBONATE ROCKS

 Limestone

 Dolomite

LIMESTONE
While most carbonate grains are easy to recognize in hand specimen when seen in thin
section they can be difficult to identify. The problem is, as Majewski (1969) remarked,
that a "variety of shapes are produced by random cuts through a single geometric
pattern, and shapes can be duplicated in such cuts by different designs; also,
characteristic features may be obscured and others may become apparent" when the
grains are exposed in a specific plane. Never the less the carbonate grains of this
collection are separated from one another on the basis of their shape, size and internal
structure.

Carbonate grains can be separated from one another on the basis of their shape, size
and internal structure. Because the grains commonly collect near their site of origin,
they can be used, in conjunction with other rock characteristics including vertical and
lateral facies relationship and sedimentary structures, to determine the depositional of
the rocks they occur in. Information about grain types and the manner in which they
occur in rocks can be communicated by means of limestone classifications.

The interpretation of the depositional setting of carbonates is based on grain types,

grain packing or fabric, sedimentary structures, and early diagenetic changes. The
identification of grain types is commonly used in subsurface studies of depositional
setting because, unlike the particles in siliciclastic rocks, the grains making up
carbonates generally formed within the basin of deposition. This oversimplification or
rule of thumb doesn’t always apply. For example, ooid sands, which characteristically
form on highly-agitated shoals, may also accumulate on beaches, islands, sand flats,
deltas and even turbidite fans. Thus, while most carbonate geologists will use grain
types to make an interpretation of depositional setting they should further test their
hypothesis using the other criteria listed above.

Carbonate sedimentary particles may be subdivided into micrite (lime mud) and sand-
sized grains. The grains can be separated on the basis of their shape and internal
structure and can be subdivided into two major groups: skeletal and non-skeletal.

© PetroSkills, 2013. All rights reserved. 6-55

Reservoir Characterization
Chapter 6

Carbonate rock textures: Dunham classification

The following figure is Dunham’s classic subdivision of depositional textures mudstone
– grainstone/boundstone. This figure is critical to explain different terms used to explain
range of energies in carbonate depositional settings.

It is still used to describe the fabrics in core today. Note some clear limitations in
reservoir description for petrophysical responses; limestone versus dolomitized fabrics;
and challenges in characterisation of interpartical porosity present in packstone (Lucia
1999). In this context, the figure indicates where REEF fits in addition to more familiar
Clastic split of SAND and MUD - grain size separator for carbonate sands and muds -
so muds BOTH very shallow in lagoon and very deep in the basin/slope

6-56 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

© PetroSkills, 2013. All rights reserved. 6-57

Reservoir Characterization
Chapter 6

The figure below shows the interaction of geological processes - deposition and
diagenesis - in determining the microgeometry of the present-day samples studied for
the Khuff and L-III samples. The column on the left shows the initial sediments, their
fabrics are differentiated by the four small cartoons: ranging from grainstone (top)
through packstone, wackestone and mudstone (bottom).

After deposition the sediments are altered by the physical and chemical processes of
diagenesis. For these cases diagenesis proceeded along one of four paths. The fabric
which results (take a close look at the cartoons) may be very different from the fabric at
the time of deposition and, for a given rock, depends on the depositional class and on
the diagenetic path.

There are many more paths than this. There are even subpaths and variations within
the paths within the case studies, but the simplification represented here is sufficient to
make the point that diagenesis plays a major role in determining microgeometry and
fabric.

In path 1 the original sedimentary fabric was obliterated by recrystallization associated

with the process of dolomitization. The result is an aggregate of crystals fabric
regardless of the depositional facies. Compaction and cementation significantly reduced
the porosity.

In path 2 the dolomitization process simply replaced the original material retaining the
depositional fabric, including pore space both within and between grains. Compaction
and cementation significantly reduced the porosity.

6-58 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

In path 3 the primary porosity was reduced by cementation. There was a later
dissolution phase, which preferentially dissolved some of the original grains and
resulted in the formation of moldic porosity. Compaction and cementation significantly
reduced the porosity.

In path 4 the limestone was leached by a process that did not follow the depositional
texture. The leached channels which result cross grains and matrix.

Geological Processes.

© PetroSkills, 2013. All rights reserved. 6-59

Reservoir Characterization
Chapter 6

The microgeometry classes which we have recognized in these case studies are shown
here in the context of their depositional and diagenetic history.

Most geologic schemes for carbonate classification emphasize the solid phase of the
rock – types of fossils and mineralogy – seeking to predict the reservoir scale
distribution of rock textures through an understanding of the depositional and diagenetic
processes. In many cases depositional class impacts diagenesis, but this is by no
means consistent.

Unfortunately existing geologic schemes are not optimal for the purposes of
petrophysics. Knowing only the depositional fabric does not specify the present day
microgeometry. Knowing in addition the diagenesis still does not (in general) specify the
microgeometry well enough.

Present Day Rock Fabrics.

The following figure shows cartoons of idealized end member fabrics recognized in case
study samples. Each of these fabrics is associated with a particular microgeometry
class in our case studies. However there may be other fabrics, which we did not
encounter, corresponding to the same microgeometry classes in other formations.
Petrographically we found these fabrics to be dominant, however other fabrics do occur
on the thin section scale. The cartoons are drawn to help visualize the NMR and flow
properties of the samples associated with the microgeometry classes in our case
studies and to help geologists relate to microgeometry classes.

6-60 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

© PetroSkills, 2013. All rights reserved. 6-61

Reservoir Characterization
Chapter 6

6-62 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

© PetroSkills, 2013. All rights reserved. 6-63

Reservoir Characterization
Chapter 6

Timing of diagenetic events – a. isopachous cement and compaction b. cement

encases compacted grains & spalled isopachous cement is relatively late c. Cross
cutting relationships such as fractures and mineral replacements. Calcite filling is last
event replacement anhydrite came after poikilotopic cement

DOLOMITE
Dolomite = Dolostone
Magnesium replace Calcite

6-64 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

Dolomite Texture
The distribution of the size of crystals may be:

 Unimodal (equigranular fabric)

 Bimodal (inequigranular)
 Polymodal

The fabric of inequigranular dolomite may be Porphyrotopic or Poikilotopic.

Recrystallization in Carbonates
Anhedral
 Crystal is not bounded by crystal faces
 Fabric is named Xenotopic

Subhedral
 Crystal is partly bounded by crystal faces
 Fabric is named Hipidiotopic

Euhedral
 Crystal is wholly bounded by crystal faces
 Fabric is named Idiotopic

Idiotopic dolostone is one of the most important carbonates reservoirs for oil and gas. It
is named as Sucrosic Dolostone.

© PetroSkills, 2013. All rights reserved. 6-65

Reservoir Characterization
Chapter 6

USING DYNAMIC DATA TO CONFIRM STATIC MODELS

From: Dehghani, et
al, 1999, AAPG

A comparison between historical field production data (hist.) and the results from a scaled-up permeability model
not considering vuggy zones (Model). Water-cut (bbl of water per day/total of oil and water) and oil rate (stock tank
bbl/day) are plotted for 60 yr of production history. The model results do not match the water-cut and oil rate
histories.

A comparison between historical field production data (hist.) and the results from a scaled-up permeability model
with superimposed high-permeability vuggy zones; in this case 4 d of permeability was added to the secondary
porosity cube using a longer correlation length of 400 ft (122 m) (4SP2). Water cut (bbl of water per day/total of oil
and water) and oil rate (stock tank bbl/day) are plotted for 60 yr of production history. The model closely matches
the oil rate, as well as the water cut.

6-66 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geology

SUMMARY OF CHAPTER 6
Since data are often sparse, modeling of reservoirs is significantly improved if
knowledge about the environment of deposition is known. It is important in this modeling
to test concepts and to modify models where necessary. Particularly, when new data
becomes available re-modeling becomes necessary.

Monte Carlo simulation is a powerful technique to quantify uncertainties.

Reservoir complexity can be described using concepts such as layer cake, jigsaw
puzzle and labyrinth. In this order, reservoir development requires more wells and a
denser well spacing.

Permeability heterogeneity has a significant impact on the fluid flow in reservoirs.

After deposition, diagenesis starts to play a role causing either porosity increases or
porosity decreases.

The main difference between carbonate and siliciclastic deposition is transport. In

carbonates the deposition is in situ in the same place where the carbonate was formed.
In siliciclastic deposition, rock material has been transported over sometimes long
distances.

In carbonates, different types of porosity are recognized, namely inter-granular, fine

inter-granular and intra-granular porosity. In this order, the permeability is reducing.
Usually, permeabilities in carbonates are less than in siliciclastics.

© PetroSkills, 2013. All rights reserved. 6-67

Reservoir Characterization
Chapter 6

6-68 © PetroSkills, 2013. All rights reserved.

Chapter 7 P ETROPHYSICAL D ATA
No single discipline can provide complete reservoir knowledge, but by integrating
technology from several disciplines, operators have the best chance for success.
Soliman, et al.

Aims of the petrophysical analysis are to develop models that can be used to project
data across the study area. The data that are available for critical analysis is
concentrated in a very small area when considering the breadth of the reservoir. It is
incumbent, then, to be able to project data from the well bores to the large areas
between the wells.

A necessary requirement of these projections is to integrate the information that is

available from the various disciplines. Several methods are available to make these
projections; traditional deterministic approaches are probably the most frequently used
methods. More recently, geostatistical approaches have been gaining favor as
computing power grows, software applications are more available and the techniques
are more widely understood. Discussions of these methods are included in sections on
geostatistical modeling and geophysical modeling (example of permeability derived from
multiple realizations).

WHAT IS PETROPHYSICS

Petrophysics is the science of measuring rock properties and the study of their
relationships to each other.

 Locate the reservoir

 Type of reservoir (Lithology. e.g. sandstone, carbonates)
 Ability of the reservoir to have hydrocarbon
 Type of hydrocarbon (Oil or Gas)

STATIC MODEL - DATA ANALYSIS AND INTERPRETATION

• Develop integrated database

 Accessible across disciplines

 Petrophysical data
 Normalize/calibrate/edit against cores
 Identify cutoffs for mapping

 Core data

© PetroSkills, 2013. All rights reserved. 7-1

Reservoir Characterization
Chapter 7

 Fully describe/interpret cores

 Correlate with logs
 Calibrate with test/production data

 Seismic data
 Develop structural interpretation
 Seismic stratigraphic correlation
 Establish correlations with reservoir parameters

RC PETROPHYSICAL PROGRAM
 Open Hole Log Evaluation

 Special Open Hole Log Evaluation

 Evaluation of Shear Data from Sonic Logs

 Cased Hole Log Evaluation

 NMR Logging

 Reservoir Surveillance

 Quantitative Seismic Studies

 Generate Well Synthetics
 Model Seismic Amplitudes

 Interpretation of Borehole Images and Dip Meter Data

 Petrophysical Core Description

OPEN HOLE LOG EVALUATION

 Discriminate Reservoir/Non-Reservoir
 Identify Mineralogy/Lithology
 Porosity Determination
 Saturation Determination
 Estimate Permeability
 Evaluate Wireline Testing
 Analyze and Integrate Routine Core Analysis
 Analyze and Integrate Special Core Analysis
 Integration Of Core and Log Data
 Analyze and Integrate Petrography and Mineralogy Data

7-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

STATIC MODEL - DATA ANALYSIS AND INTERPRETATION - 2

• Map distribution of parameters

 Structural tops data

 "Establish tops - all horizons, faults, etc"
 Integrate with seismic data
 Create integrated maps
 "Create tabular display of well data (TVD, TVT, TST, etc)"

 Define facies

 Establish depositional model

 Define reservoir architecture

 Distribute reservoir parameters throughout reservoir

QUICK LOOK EVALUATION

 Do I have a reservoir?
 Do I have hydrocarbons?
 What is the type of hydrocarbons?
 Can we produce it (permeability)?
 How much is there (porosity and saturation)?

© PetroSkills, 2013. All rights reserved. 7-3

Reservoir Characterization
Chapter 7

AMOCO APPROACH TO PETROPHYSICAL MODEL (GUNTER, SPWLA )

 Define rock types

 Relate to geologic framework, lithofacies

 Relate petrology to porosity, permeability, capillarity
 Products: rock, pore, fluid models

 Integrate with formation evaluation

 Define compartments, flow units

 Product: petrophysical model

 Calibrate seismic and/or geostatistical model

 3D reservoir description
 Calibrate with production tests, pressure transient tests, decline curves
 Product: Input into simulation model

It is clear in this approach that integration is vital in petrophysics. ‘Define the rock types’
must be done in close cooperation with geologists and sedimentologists, ‘Integrate with
formation evaluation’ is an area where petrophysicists, geologist and reservoir
engineers meet, and for “Calibrate with seismic and/or geostatistics’ petrophysicist,
seismologist, seismic intepreters and geologists need to come together.

7-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

THE LOGGING ENVIRONMENT

Petrophysics is involved in data acquisition. A major part of this data acquisition is done
in borehole by logging (while drilling or after drilling). The issue is that measurements
are done in an environment (the borehole) that is different from the virgin formation, for
which data are required. In addition, borehole fluid invasion complicates the situation:

The properties are measured in the borehole, whereas the conditions in the uninvaded
zone need to be known. The challenge is:

 Discover unique relationships between reservoir properties and log responses.

 Steps:

1. Identify reservoir property of interest (primary)

2. Identify properties measured by logs (secondary)
3. Establish relationships from primary to secondary

 Wireline Logs

- large amounts of data (but indirect) of in-situ properties

- require involved interpretation
- Methods – x-plots, Pickett plots, pattern recognition approaches

 Unique solutions required

– not “worldwide average” equations

 Confirm with data from multiple sources (e.g. core data)

© PetroSkills, 2013. All rights reserved. 7-5

Reservoir Characterization
Chapter 7

PETROPHYSICAL PERSPECTIVES

WHAT WE MEASURE: SUMMARY

Tool Source Measures Application

Gamma Natural Radioactivity Reservoir detection

Ray

Density Gamma Ray Electron density Porosity

Reservoir Detection
Lithology
Identification
Hydrocarbon Type

Neutron Neutrons Hydrogen index Porosity

Reservoir Detection
(Number of Hydrogen atoms
Lithology
per unit volume)
Identification
Hydrocarbon Type

Sonic Sound Transit time Porosity

Seismic Calibration
(from transmitter to receiver)

PETROPHYSICAL DATA AND SOURCES

 Mudlog (cutting and drilling parameters)

 Core (drilled core and sidewall core)

 Open hole log (After or while drilling)

 Cased hole log (during or after production)

 Borehole Seismic

7-6 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

ANOTHER CHALLENGE: RESOLUTION

The scale of petrophysical data varies from the micrometer scale up to as much as the
100 of metres scale:

HOW TO INTEGRATE PETROPHYSICAL DATA

Integration of data is a main responsibility of petrophysics. This integration takes place

across the entire hydrocarbon life cycle. In the exploration phase borehole acoustic data
(check shots, velocities, synthetic seismograms) are integrated with seismic data. In the
reservoir assessment phase many data from different sources are brought together
(mudlog, core and log data). During drilling, the most important measurement is made,
namely cement bond evaluation (acoustic measurement). While producing a reservoir,
the reservoir needs to be monitored in order to understand the reservoir performance.
Cased hole logging plays an important role in this phase for understanding of fluid flows
in the borehole (flow meter), monitoring of fluid levels (pulsed neutron capture) and
corrosion inspection of casings (acoustic devices).

At all stages the measurements assist to define the static and dynamic models Whereas
often the static and dynamic models are referred to as separate models, it must be
emphasized that they form one single reservoir model. Changes in the static model
impact on the dynamic model and vice versa. The following diagram provides a
summary:

© PetroSkills, 2013. All rights reserved. 7-7

Reservoir Characterization
Chapter 7

FLUID AND FACIES DETERMINATION

The combination of density and neutron logs can be used to distinguish between the
major facies. In addition, the combination assists in distinguishing between oil and gas
zones. When the neutron and density logs are overlain while the scale on the neutron
log is reversed, a large separation is shown between the logs in a gas zone, whereas
this separation is small in a oil zone in fact being marginally larger than in a water zone
(see the diagram on the next page).

It should be noted that in gas condensate reservoirs, the density of the fluid in situ is
such that hardly any separation is visible. There is a risk that in such a reservoir the
hydrocarbon is interpreted as oil. A production test will of course resolve this issue.
However, if under circumstances no production test can be done (say the rainy season
started early), there is a log evaluation technique which will distinguish between the
different options of oil or gas. This technique will be demonstrated below.

The density and neutron logs are a combination of two logs. More logs can be used in
cluster analysis, which is in fact a ‘crossplot in more than two dimensions’.

Fluid densities can also be evaluated using pressure versus depth plots.

7-8 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

© PetroSkills, 2013. All rights reserved. 7-9

Reservoir Characterization
Chapter 7

CLUSTER ANALYSIS

In the following case study (Tucker et al, AAPG) six different log responses are used to
distinguish between different lithologies, called clusters. The logs used are:

 Gamma Ray
 Density
 Neutron
 Sonic
 Resistivity
 Photo-electric effect

As six logs are being used also six cluster can be distinguished. The cluster need to be
defined using core data. The geologist is needed to describe the core and the
petrophysicist is needed to interpret the logs and to add the fluid data, again a very
integrated job.

Once the clusters have been defined and the model is calibrated using the core data,
the model can be used to simulate a core in other wells in a reservoir. In the
development phase when many wells (several hundreds of wells is easily possible) this
technology provides a means to evaluate lithologies and facies in detail.

7-10 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

Highest gamma ray Non-reservoir cluster 3

lowest velocity, low density, (8 dolomitic sandstones).
intermediate neutron, lowest resistivity,
lowest photoelectric.

Highest velocity, highest density, Non-reservoir cluster 1

lowest neutron, (7 tight peloid dolopackstones)
highest resistivity,
highest photoelectric
lowest gamma ray.

Highest neutron, lowest density, Reservoir cluster G

low velocity, (41 porous peloid dolopackstones;
low resistivity, low photoelectric, 31 fusulinid-peloid dolopackstones).
intermediate gamma ray. dolowackestones; 30 bivalve-peloid

Intermediate velocity, density, neutron, Reservoir cluster 4

Intermediate gamma ray, (25 fusulinid-peloid dolowackestones;
low to intermediate resistivity. 16 peloid dolopackstones; bivalve-peloid
dolopackstones).

Intermediate velocity, low density, Reservoir cluster 5

intermediate to high neutron, (14 bivalve-peloid dolopackstones;
Intermediate gamma ray, 14 peloid dolopackstones;
low resistivity. 9 fusulinid-peloid dolowackestones).

High velocity, high density, Non-reservoir cluster 2

low to intermediate neutron, (21 tight peloid dolopackstones).
intermediate to high resistivity,
high photoelectric.

© PetroSkills, 2013. All rights reserved. 7-11

Reservoir Characterization
Chapter 7

7-12 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

PRESSURE AND FLUID SAMPLING

Objective of the WFT

 Determination of reservoir pressures
 Confirmation of the fluid type, evaluated from other logs
 Calculation of the oil, gas and water densities at reservoir conditions
 Determination of the
– Free Water Level (FWL)
– Gas Oil Contact (GOC)
– Gas Water Contact (GWC)
 Indication of the reservoir permeability

© PetroSkills, 2013. All rights reserved. 7-13

Reservoir Characterization
Chapter 7

THE QUEST FOR PERMEABILITIES FROM LOGS

 Conventional: K - φ plots using density logs calibrated with cores.

Fancy from cores: Mini-permeameter tool providing a continuous K-log.

Note that permeability can be related to grain size distributions and sorting.
Sneider’s procedure for this purpose is given below.

 Qualitative:

Is there a mud cake? Mud cake is being built up against permeable zones.

From micro-resistivity logs (they measure porosity in the invaded zone, and
hence permeability; low resistivity implies high permeability).

 From fancy logs:

Nuclear Magnetic Resonance (NMR) logging

The NMR log is supposed to measure the effective porosity = total porosity minus
porosity occupied by the connate water; the effective porosity is the pore space
through which the flow of fluids is taking place.

Array sonic log

The array sonic tool allows to record the full wave train arriving at the acoustic
sensors. In this wave train the arrivals of the compressional, sheer and Stoneley
waves can be recognized. The Stoneley wave is a wave that is polarized in a
circular fashion around the direction of propagation. As such, the idea is that
Stoneley wave energy is dissipated into the formation. This dissipation is more
severe when the rock is porous and hence has permeability.

Remember:

Logs measure static properties and permeability is a dynamic property!

THE PORO-PERM SHOW

We will look at SEM (Scanning Electron Microscopy) pictures of rock samples that have
approximately the same porosity, but significantly different permeabilities. The pictures
are followed by Sneider’s pore classification system and a cross-plot of the K - φ data.

7-14 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

SEM photographs of pore types IA, IB, IC, ID in sandstones.

SEM photographs of pore types II and III in sandstones.

© PetroSkills, 2013. All rights reserved. 7-15

Reservoir Characterization
Chapter 7

Sneider’s Pore Classification for Clastics

Characteristics of Dry, Freshly Broken Surfaces
at 20x Magnification
Pore
Visible Pinpoint Consolidation Permeability
Type
Type I
I Abundant to Very Needle prove
Subclasses
common; abundant to easily dislodges
IA:>100 md
interconnection common many grains
IB:>10-100 md
visible on many from rock
IC:>1-10 md
pores surface
ID:>±0.5-1 md
± 0.5-1.0 md
II Scattered Abundant to Needle probe
common can only (depending on
particle size,
occasionally
dislodge a grain sorting and clay
from rock mineral content)
surface
Too tight to
III None to very None to a Usually very
produce gas at
isolated few pores well
commercial rates
consolidated
even when
and/or pores
filled with clays fractured or
or other pore- interbedded with
type I rocks
filling material

The K - φ demonstrate that

over a small range of porosity
the permeability changes over
several orders of magnitude.

In such a situation straight line

relationships between log(K)
and φ must be used with
caution.

7-16 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

Sneider’s procedure to estimate permeability from grain size distributions and sorting
Sneider and King developed a simple method of classifying pore types from cuttings.
The classification of clastic rock pore types from cuttings is made by comparing pore
types with production tests and log analysis. The pore types are as follows:

Type/Description
I. Rocks with pores capable of producing gas without natural or artificial fracturing.
II. Rocks with pores capable of producing gas with natural or artificial fracturing
and/or interbedded with type I rocks.
III. Rocks too tight to produce at commercial rates even with natural or artificial
fracturing.

Procedure: Predicting Sandstone Permeability

The procedure below, with steps and recommended action, is for predicting the
permeability of sandstones from cuttings using 20x magnification (from Sneider and
King).

Step/Action
1. Estimate grain size and sorting using standard size-sorting comparators, thin
section and SEM photomicrographs, and rock photographs.

2. Estimate volume percentages using Terry-Chillingar charts made for volume

estimates.

3. Estimate consolidation using the scheme described in the preceding table.

4. Describe the visible and pinpoint porosity and interconnectedness.

5. Estimate permeability from rocks on comparators and/or using rock

characteristics described in the preceding table. (Comparators can be made or
purchased.)

Predict permeability for the formation in prospective areas where petrophysical

characteristics are believed to be similar.

THE GHAWAR FIELD IN SAUDI ARABIA

This also a permeability story.

© PetroSkills, 2013. All rights reserved. 7-17

Reservoir Characterization
Chapter 7

7-18 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

REGIONAL EAST-WEST CROSS-SECTION

EAST-WEST SEISMIC DEPTH SECTION – SOUTH GHAWAR

© PetroSkills, 2013. All rights reserved. 7-19

Reservoir Characterization
Chapter 7

Ghawar

Discovered in 1949, Oil column 1300 ft, Np = 80 billion bbls (2011), Potential = 5-6
million bbls/d, pressure maintenance by water injection, simulated with a 1.032 billion
grid block black oil simulator.

In the picture, the colors vary from red (oil) to blue (water)

7-20 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

COMMON ARAB D FABRICS

© PetroSkills, 2013. All rights reserved. 7-21

Reservoir Characterization
Chapter 7

Uthamaniyah field (Saner and Sahin, AAPG)

7-22 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

Ghawar as a whole (428 samples)

© PetroSkills, 2013. All rights reserved. 7-23

Reservoir Characterization
Chapter 7

ACTIVATION LOGGING

In activation logging high energy neutrons are

transmitted into the formation. These neutrons
interact with the atoms in the formation. During
the interaction the neutrons are deflected and
slowed down while a gamma-ray is produced
from the energy lost by the neutron (inelastic
scattering spectrum). This process continues
until the neutron has slowed down to thermal
energy when it is completely absorbed. During
this absorption also a gamma-ray is generated
(thermal capture spectrum).

The generated gamma-rays, which have an energy typical for the elements they
interacted with, are counted by the tool at least those that reach the detector. Tools can
be designed to measure specific gamma-ray energies or a complete gamma-ray energy
spectrum. The investigation depth is a couple of inches. The tool can be run both in
open and in cased hole.

The responsibility for the data acquisition using this technology lies in Petrophysics, but
the technology has applications in many areas of the business. Hence, close
cooperation between disciplines is again required to make optimum use of this
technology.

7-24 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

Typical applications are (refer Schlumberger information on the web):

 Production logging (or gravel pack logging) to evaluate the quality of gravel
packs.
 Production logging to measure water flow in and around the borehole.
 Aluminum activation log as a measure of clay volume.
 Elemental capture spectroscopy to determine the mineral content of the
formation.
 Pulsed neutron spectroscopy to measure typical ratios representative for oil,
salinity, lithology, porosity and clay, respectively.
 Conventional neutron logging to measure porosity.

Production logging
Activation of silicon and aluminum. The detector is placed below the neutron source and
the log is run slowly so that oxygen and other activated elements have decayed before
the detector passes the activated interval. A high count rate indicates a high quantity of
silicon in a sand pack or aluminum in a bauxite pack.
A somewhat fancier application is oxygen activation to detect and quantify the flow of
water in and around the borehole. Oxygen can be activated to become a nitrogen
isotope, which upon decay (half-life only 7.1 sec) produce gamma-rays of 6.13 MeV.
There are detectors above and below the source to monitor flow upwards and
downwards.

Aluminum activation log

Activation of aluminum with relatively low energy neutrons. The gamma-ray spectrum of
the formation will be measured before activation and after activation. The difference
provides the aluminum content of the formation. This type of log is a direct indicator of
the volume of clay, since clay minerals are alumino-silicates.

Elemental capture spectroscopy

This log uses the capture spectrum only. Output are the relative yields of silicon,
calcium, iron, sulfur, titanium and gadolinium. To get absolute elemental concentrations
calibration is necessary using cores an oxide-closure model. In the oxide-closure model
the assumption is made that sedimentary minerals are oxides so that the sum of the dry
weight percentages of all oxides must be 100%.
Wi = F * Yi / Si , where W is the dry weight percentage, Y is the relative elemental
gamma ray yield, S the tool sensitivity to a specific element as measured in the
laboratory, and F the unknown normalization factor. The dry weight percentage of
oxygen is Oi = F * Xi * Yi / Si , where X is the oxide association factor given by the
chemical formula. Since Σ Oi = 1 , F can be calculated and all W’s can be determined.
This tool is often run in combination with the natural gamma-ray spectroscopy log,
which provides information on the Thorium, Uranium and Potassium content of the

© PetroSkills, 2013. All rights reserved. 7-25

Reservoir Characterization
Chapter 7

formation. The combination of induced and natural spectroscopy logs is often referred to
as the geochemical log used in formation evaluation.
Gamma-ray spectroscopy is applied in gas- and oil-shales to determine the TOC (total
organic content) of the formation.

Pulsed neutron spectroscopy log

Since in neutron spectroscopy produces relative gamma-ray yields from atoms in the
formation, in many application ratios are considered. These ratios are C/O for oil
detection, Cl/H for salinity determination, Si/(Si+Ca) for lithology evaluation, H/(Si+Ca)
for porosity determination and Fe/(Si+Ca) for clay determination.
The C/O ratio is measured by what is called the carbon-oxygen tool. It is used for
monitoring the movement of the OWC during production.

Conventional neutron logging

Conventional neutron logging is a type of activation logging in which two detectors are
used at different spacings from the source. The signal recorded is the ratio between the
responses of the far and near detectors. This ratio is calibrated in the workshop in a
known formation with a known borehole environment (typically the tool placed against
the side of an 8 inch borehole in a limestone block filled with fresh water at surface
temperature and pressure).
The main principle behind this log is that the neutrons loose their energy most efficiently
by collisions (of an elastic nature) with nuclei of similar mass, which are mainly the
hydrogen atoms. As such, the tool is a measure of the hydrogen content of the
formation referred to as the hydrogen index.

Example Barnett Shale

After Zhang of Baker Hughes, 2010 SPWLA Topical Conference.

Shale gas characteristics

 Organic shales with high kerogen (TOC) content
 Both source and reservoir rock of the gas
 Both adsorbed and free gas
 There are natural fractures present, but fracture stimulation is required to
produce at economical rates.

7-26 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

Shale gas evaluation: an integrated approach

Evaluation method
• Method 1
- total porosity from density log
- mineralogy from spectroscopy log
- carbon content from mineralogy
- excess carbon content = carbon content minus carbon associated with minerals
- TOC is proportional to excess carbon content

• Method 2
- effective porosity from NMR log
- total grain density (ρgr) from density logh and NMR porosity
- matrix density (ρm) from spectroscopy log
- TOC volume fraction of matrix = (ρm-ρgr)/(ρm-ρTOC)
- Toc weight fraction = (ρTOC/ρgr) x TOC volume fration

© PetroSkills, 2013. All rights reserved. 7-27

Reservoir Characterization
Chapter 7

Barnett Shale

Well 1 Well 2

7-28 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

FORMATION MICRO-SCANNING

Formation micro-scanning (FMS) provides an image of the borehole wall. It is an

electrical measurement with electrode spacing of only 2 cm. As such, it has a resolution
of some 2 cm, which is much finer than of any of the other logging tools. The readings
need to be calibrated to reflect reservoir and non-reservoir rock. For this purpose, color
coding is being used. Calibration is best be done in wells that also have cores.

On the images sinusoidal features are recognized, which highlight planes that intersect
the borehole. The higher the amplitude of sinuses the greater the relative dip between
the plane and the borehole.

Applications are:

 Evaluation of formation dips.

 Evaluation of thinly laminated sequences.
 Fracture detection.

In the following two displays FMS images are compared with layering seen in outcrops
showing that indeed the FMS provides similar information as what can be seen in
outcrops.

Comparison of Shale-Clast Conglomerates in an outcrop with formation micro-scanner images

© PetroSkills, 2013. All rights reserved. 7-29

Reservoir Characterization
Chapter 7

Comparison of stratification of a sand dune in an outcrop with formation micro-scanner images

FRACTURES

One of the most difficult aspects of characterizing reservoirs is how to deal with fractures.
Some reservoirs are totally controlled by fractures, others are unaffected and most are
influenced to some degree. Recognition and prediction of these fractures is difficult at best –
although modern imaging logging techniques are major improvements. Antonellini and
Molemna have published an important paper that analyzes fracture systems in carbonates
at each level of occurrence – regional, fault-based, local fractures, brecciation, micro-
fractures. RC teams might find a similar approach useful in their reservoirs.

Burial diagenesis: fracturing

“Burial” diagenesis can significantly enhance
the productivity of tight reservoirs, particularly
carbonates. When these fractures are open
they become very important as flow conduits.
Note that the permeability in the matrix is
usually negligible compared to the fracture
permeability.

7-30 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

Fahud (Oman) field example

FMI image, 3.3 m (10 ft) in length, shows fractures and bedding planes. Image log with interpretations is shown as
folded image shown as borehole cylinder in 3-D space with fracture planes (a) and unfolded image with sinusoid
fracture curves (b). Basic interpretations include fracture dip, dip azimuth, and depth.

The above example is characterized by the different fracture spacings in the different
limestone layers (which one would expect for different layer thicknesses). This fracture
spacing is a key parameter to model along with fracture orientation.

The sequence is a typical shelf carbonate interbedded with shale. Compression to

anticline fractured the carbonates.

Borehole image logs provide important sub-seismic images. Current data analysis
methods, however, do not fully use the information contained in these images. Instead
of using borehole images to extract a three-dimensional (3-D) visualization of fracture
networks, most interpretations are limited to rose diagrams, stereonets, and histograms,

© PetroSkills, 2013. All rights reserved. 7-31

Reservoir Characterization
Chapter 7

thus ignoring the rich 3-D information contained in the image logs. The ignored
information can be used for representation of

(1) the surrounding fracture network in a borehole section,

(2) fracture orientations in 3-D, and
(3) intersections of different fracture sets in 3-D.

There are other borehole image logs in addition to the FMS. They have been widely
used for detecting fractures and have become the most important tool for imaging sub-
seismic features. These logs include electric, acoustic, radar, and ultrasonic methods.
Oriented cores also provide firsthand, reliable fracture data, and efforts have been
made to integrate data from core and different imaging tools to validate image
interpretation. Core, however, is commonly of limited extent compared to borehole
images, and is not oriented in most cases. Therefore, fracture orientation is not always
measurable in cores. A new logging while drilling (LWD) density log ROSI provides a
complete 360 degrees image, but its resolution (16 pixels/circumference and 2-10
pixels/ft) is still not yet adequate for most fractures.

Fracture and fault characterization

 Characterize nature, distribution
and orientations of potential fluid-
pathways or baffles

 Improve understanding of sub-

seismic and seismic scale reservoir
compartmentalization

 Generate input parameters for

reservoir fracture and flow models

Logged fractures from borehole images

characterize nature, distribution and
orientations of potential fluid-pathways or
baffles. They improve understanding of
sub-seismic and seismic scale reservoir
compartmentalization. In addition, they
generate input parameters for reservoir
fracture and flow models.

Further details on the geology of fractured

reservoirs are presented in a paper
entitled “Geology of Fractured Reservoirs”
that can be found on the flash drive, which
is part of the RC material handed out.

7-32 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

CASED HOLE LOGGING

In the wellbore
Production logging

At the casing / tubing

Perforation

Corrosion inspection

Between the casing and the formation

Cement bond evaluation

In the formation
Reservoir monitoring

CORES

So far, we saw how cores were used to obtain mini-permeameter logs, which provide a
continuous profile of permeability versus depth. Such profiles need to be calibrated with
routine permeability measurements on one inch core samples.

In addition, cores were used to calibrate the model in cluster analysis. After such
calibration the cluster model can be applied in other wells to simulate cores in uncored
wells.

In general, core data are used to calibrate logs. Porosities from density logs are
calibrated using core porosities. Resistivities can only be translated into hydrocarbon
saturations after the saturation exponent (n) and lithological exponent (m) have been
measured on core samples. Such measurements represent in fact the calibration.

Obviously, for good calibration, cores need to be correlated with logs. This is done by
correlating gamma ray logs run during the logging operations and those run when the
cores were taken.

Systematic observations from cores starts with completing a list such as:

1. Core Number, Box, Depth 6. Pore Association/Connection

2. Lithology 7. Cements
3. Pore Type 8. Fractures
4. Pore Size 9. Breccias
5. Pore Shape

© PetroSkills, 2013. All rights reserved. 7-33

Reservoir Characterization
Chapter 7

Cores are also used (in combination with logs run in boreholes) in studies to determine
depositional environments. In the geology session, we saw already how the gamma ray
log is used to identify fining-upward and coarsening-upward sequences representative
for meandering fluvial deposits and coastal deposits. In braided river environments we
were expecting fairly constant gamma ray logs showing little radio-activity.

A couple of guidelines for optimum usage of core data:

1. Fully describe/interpret cores

– Depth calibrate
– Note location of special analyses
2. Correlate with logs
3. Develop understanding of ranges in parameter values
4. Calibrate with test/production data
5. Sample representative intervals - not just the unique

The following chart is an example flow chart for a coring programme:

7-34 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

In this chart supplementary or special core analysis is mentioned. Here, three of them
are discussed in more detail, namely relative permeability, wettability and capillary
pressures.

There are many core applications. To mention a few, they are:

• Calibration of logs (e.g. density log).

• Provision of model parameters to translate log measurement into quantity of
interest (e.g. m and n in resistivity log evaluation).
• Provision of data that cannot be measured in situ (e.g. relative permeability
curves).
• Calibration of special logs (e.g. FMS, gamma ray spectroscopy log)
• Calibration of models (e.g. cluster analysis models)
• Determination of depositional environments
• Assistance in biostratigraphy by showing fossil records

Realtive permeability

Wettability
Cores brought to the surface are made dirty by the drilling fluid. In the laboratory, core
samples cut from cores are cleaned by cooking them in a solvent, e.g. toluene. After
this treatment, the core samples are fully water wet. Then, they are saturated with
formation brine, and subsequently they are aged by flooding the hydrocarbon from the
reservoir through the cores until residual oil saturation is reached.

© PetroSkills, 2013. All rights reserved. 7-35

Reservoir Characterization
Chapter 7

The question is whether by this treatment the in situ wettability will be restored? And in
addition, is it possible that in situ the rock was oil wet and that the original wettability can
be restored.

In the literature there is an ongoing debate. Some authors claim that they encountered
reservoirs that are oil wet, whereas others believe that all anomalous reservoir
behaviour can be explained without having to introduce oil wetness.

Oil wet reservoirs can be recognised on logs if high resistivities leading to oil saturations
of over 98% are observed, or during the production phase, if unexpected early
breakthrough of water and anomalously high watercuts are observed.

Capillary pressure
Capillary pressure data are used to determine hydrocarbon saturations as a function of
height above the free water level (FWL). This application is particularly useful in thinly
layered reservoir for which the height of the thin layers is less than the resolution of the
logging tools. Clearly, core is required for this application.

7-36 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

Knowledge of the pore size distribution of rocks as measured by capillary pressure

curves are valuable in reservoir characterization to help calculate or estimate:

 Reservoir rock quality

 Net pay - non-pay
 Seal capacity of rocks, faults, and expected hydrocarbon columns
 Barriers to fluid flow within reservoirs
 Thickness and location of the hydrocarbon-water transition zone
 Recovery of hydrocarbons in a reservoir composed of rocks with variable pore
sizes, porosity, and permeability
 Residual oil after primary or supplemental recovery
It is helpful to know the capillary pressure relationships for the hydrocarbon-brine
system in the reservoir of interest. In practice, capillary pressure tests on samples can
be run using a variety of wetting (e.g. brine or air) and non-wetting (e.g. oil or mercury)
fluids. Hydrocarbon-brine capillary pressure measurements are difficult, slow and costly.
A number of hydrocarbon-brine capillary pressure curves have been made and the
results compared with the more rapid simpler, less costly air-brine or air-mercury
measurements for the same rock type. Based on these comparisons, one can estimate
the hydrocarbon-brine capillary pressure relations from air-brine or air-mercury
measurements.

Pore-size distribution of rocks aids in understanding and predicting entrapment and

production of hydrocarbons. The following are considered:

 The key capillary pressure relationships and equations.

 What are laboratory air-mercury and hydrocarbon-brine capillary pressure

curves?

 How to convert laboratory air-mercury (or water-hydrocarbon) curves to reservoir

conditions?

 How to use capillary pressure curves to evaluate

 seal and trap capacity
 reservoir rock - non-reservoir rock
 pay - non-pay
 productivity and permeability

© PetroSkills, 2013. All rights reserved. 7-37

Reservoir Characterization
Chapter 7

SUMMARY OF CHAPTER 7
Petrophysics plays a central role in data acquisition across the entire hydrocarbon life
cycle. As such, petrophysics is a key player in integration.

When data are acquired, the logging environment must be taken into account.
Corrections have to be made or the impact of the logging environment on the data
needs to be specified.

Fluid determination can be done using logs and pressure versus depth measurements.

Permeability is a major parameter to be known, but has to be measured on cores.

Permeabilities from logs must be calibrated with core and are then still not very widely
applicable.

Activation logging is a powerful technology with applications in many parts of the

business such as production technology, geology and reservoir engineering, and in
formation evaluation in general.

Formation micro-scanning may replace cores after the FMS images have been
calibrated with core. Application are in thinly bedded formation and in fracture detection.

Petrophysics also plays an important role in both routine and special core analysis.
Special core analysis include saturation en lithological exponents required for resistivity
interpretation, relative permeability and wettability, and capillary pressure curves.

7-38 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

RAPID EVALUATION OF KEY WELLS BY A MULTIDICIPLINARY TEAM

© PetroSkills, 2013. All rights reserved. 7-39

Reservoir Characterization
Chapter 7

7-40 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

© PetroSkills, 2013. All rights reserved. 7-41

Reservoir Characterization
Chapter 7

7-42 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

DATA REQUIREMENTS AND SOURCES FOR ANALYSIS

Variable Primary Source Secondary Source

Reservoir pressure RFT, offset well DST, PBU, assumed gradient
Water salinity water analysis Log analysis
Gas Gravity Gas chromatograph Assume 0.65
Gas impurities Gas chromatograph Assume none
Filtrate salinity Mud sample Log header
Well radius Log header, bit record Caliper log
Cementation factor (a) Assume a=1 Log analysis
Formation temperature Log analysis
Cementation exponent Log analysis Core data
Saturation exponent Log analysis Core data
Shale resistivity Log analysis
Porosity Log analysis Core data
Shale volume Log analysis
Water saturation Log analysis
Mud solids content Mud sample Mud system description
Mudcake compressibility (v) Dynamic/static mud test
Erosion factor Dynamic/static mud test
Mudcake reference perm. Dynamic/static mud test
Equilibrium invasion rate Filtration test, drilling parameters
Wellbore pressure at zone Mud weight and hydraulics
Critical gas saturation Assume Sgc=0%
Pore compressibility Correlation Assume 3.5x10-6 _
Capillary entrance pressure Correlation
Dispersivity Match Parameter
Irreducible water saturation Log analysis
Pore geometry exponent (X) Log analysis Assume 2
Permeability Match parameter

© PetroSkills, 2013. All rights reserved. 7-43

Reservoir Characterization
Chapter 7

ANSCHUTZ RANCH WORKSHOP

1. THE PROBLEM
– Declining liquid rates
– High operating costs
– Increasing injection gas costs

2. CURRENT MODELS CANNOT MATCH HISTORICAL PRODUCTION RATES

3. Assignments
1. Designing an approach to solve problem
2. Developing common data base with consistent measurements
3. Adequately representing inter-well heterogeneities

DESIGNING AN APPROACH TO SOLVE PROBLEM

 Six operators at beginning of development

 Amoco is operator following unitization

 Each company has information not yet shared with all

 Suggestions:
 Consider the support material
 Clarify the problem to use as a basis for planning the study
 Choose the style for the team

7-44 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

Anschutz Ranch East Field - USA

Assignment
Your responsibility is to design a reservoir characterization plan for the Anschutz Ranch
field. Your plan should include:

 Goals of the reservoir characterization

 Key considerations - factors that will influence your plan including limitations,
constraints, objectives, participants, timing demands, financial constraints, etc.
 Makeup and structure of the reservoir characterization team
 Data sources and needs
 Expected technical problems to be encountered
 Studies to be conducted
 Timeline
 Method of implementation including means of approval

Discovery
The Anschutz Ranch East field was discovered in 1979 by Amoco Production Co. with
the completion of the No. 1 Bountiful Livestock well. The well spudded 29 March 1979
on a farmout lease from BWAB, Inc. and Tom Brown, Inc. and was completed 31
December 1979. The discovery well penetrated 940 ft (286.5 m) of gross pay in the
Jurassic Nugget Sandstone in what is now the West Lobe of the field. The East Lobe
was discovered by The Anschutz Corporation, Anschutz Ranch East No. 12-26W (26-
13~121W, Uinta County, Wyoming, completed 23 December 1980. The two lobes
produce gas and gas condensate from the Nugget Sandstone and have separate gas-
water contacts. The West Lobe discovery well is perforated from 12,816 ft (3909 m) to
13,515 ft (4122 m) and had an initial flowing rate of 1054 BCPD, 4053 MCFD, and 28
BWPD. The East Lobe discovery well is perforated from 14,750 ft (4499 m) to 14,910 ft
(4548 m) and flowed 141 BCPD, 750 MCFD, and 96 BWPD (Lelek, 1983).

The No. 1 Bountiful Livestock well was drilled to test for the possibility of an anticlinal
fold in the Nugget Sandstone as interpreted from seismic data and inferred from
shallower folding in the Champlin Petroleum Co. No. 404 "A" No. 1(35-13N-121W, Uinta
County, Wyoming).

Discovery Method
Approximately 25 mi. (40 km) of conventional 2D seismic and shallow correlations to a
nearby Cretaceous Bear River Sandstone test (Champlin

Petroleum Co. No. 404, Amoco "A" No. 1) were interpreted to define the Anschutz
Ranch East prospect and locate the discovery well. The Champlin No. 404 well was
used to tie shallow seismic reflections, and it revealed an overturned fold in the Jurassic
Preuss Formation. The presence of folding in the Preuss, stratigraphically above the
Nugget-Twin Creek interval, and the seismic interpretation suggested the possibility of
an anticlinal fold at the deeper Nugget level (Lelek, 1983). A 3D seismic survey

© PetroSkills, 2013. All rights reserved. 7-45

Reservoir Characterization
Chapter 7

including 40 mi. (102 km) of 12-fold data was acquired in 1981 to help detail the
structure for further field development.

The refinement in seismic acquisition and processing characterized thrust belt

exploration throughout the 1970s and early 1980s. This enabled interpreters to resolve
structures below thrust detachment planes as well as below disharmonically folded salt
in the Preuss Formation. However, structural resolution remains difficult for some parts
of the fold even with the most advanced techniques. The discovery well at Anschutz
Ranch East penetrated the Nugget interval significantly off the structural crest and close
to both the eastern overturned flank and the northeastward plunging terminus of the
field (Figure 2). Extensive well control and 3D seismic have led to detailed
interpretations and a better definition of the structure.

Post-Discovery
Shortly after the field's discovery it was determined that the West Lobe of Anschutz
Ranch East field contained a rich retrograde gas-condensate with a dew point only 230
Psia (1586 kPa) below the original reservoir pressure. Pressure depletion below the
dew point would have resulted in a condensate liquid dropout of up to 40% of the

7-46 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

hydrocarbon pore volume (HPV). Such a dropout could have led to a 75% loss of the
reserves in place. Because of this, a specialized plan-of-depletion was designed to
optimize production while maintaining the reservoir pressure. This plan called for
nitrogen generation and injection to replace the hydrocarbons produced from the
reservoir.

The original injection plan in the West Lobe utilized an inverted nine-spot pattern and
proposed a 10% HPV "slug" of injected buffer mix containing lean residual field gas and
no greater than 35% nitrogen. Subsequent pressure maintenance has resulted in the
total injection of approximately 50% of the hydrocarbon pore volume with similar
nitrogen and field gas mixtures. In 1986, the injector layout was modified to an
"imperfect line-drive pattern" to take advantage of permeability trends inherent in the
eolian reservoir sandstones. The continuing challenge is to maintain a constant sweep
of the reservoir to minimize hydrocarbon loss and to maintain field-wide pressure above
the dew point. Extensive coring (approximately 12,000 ft [3658 m. of core), core
analyses, and reservoir modeling programs have been undertaken to better predict the
field's response to various depletion schemes.

 Injection and production profiles for up to 10 years of history

 Unitization maps
 Gross stratigraphy
 K, porosity, SW calculations
 Pressure transient tests
 Unsuccessful history match attempts
 SEISMIC – 2-D, 3-D, inadequate resolution below major stratigraphic unit level

Currently, the West Lobe of Anschutz Ranch East field contains 30 producing wells and
18 nitrogen-gas injection wells. Production from the West Lobe is expected to continue
into the early part of the twenty-first century.

 48 wells (29 producers, 18 injectors, 1 shut-in)

 LOGS - 6 operators, different vintages, contractors, parameters, quality
 Conventional log suites, dipmeters
 11 cored wells
Whole core descriptions (different operators, nomenclature), horizontal & vertical
K, 1500 core plugs with Kh max data

In the East Lobe, sufficient differences exist between the original and dew point
pressures to tolerate partial pressure maintenance (cycling) and pressure depletion
alternatives. Furthermore, the amount of reserves contained in the East Lobe cannot
economically justify full pressure maintenance. The East Lobe contains seven
producing wells and is entering into the pressure depletion phase of production.

Significant drilling problems were encountered during the development of the field. Salt
in the Jurassic Preuss Formation often caused hole problems and casing failures. Early
in the field's development, a method was devised to drill and simultaneously under-ream

© PetroSkills, 2013. All rights reserved. 7-47

Reservoir Characterization
Chapter 7

the salt to minimize salt encroachment around the drill pipe (Holt and Johnson, 1985).
More recently, oil-based drilling fluids have proved effective for controlling the thicker
sections of salt. As a result of these improvements and of lower day-rate rig and
subcontractor costs, drilling expenses are now half of those in the early1980s.

7-48 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

Rock Properties

The Nugget reservoir is anisotropic and inhomogeneous with respect to fluid flow. Many
of the reservoir properties are tied to the original depositional architecture and only
slightly modified by subsurface physical and chemical diagenesis (Lindquist, 1983).
The Nugget reservoir in the West Lobe can be effectively divided into three
Petrophysical zones. The upper two-thirds of the formation contains large-scale eolian
dune sandstones and is the best reservoir rock. The lower third of the interval is
characterized by smaller-scale eolian sandstones, some with water-influenced
depositional textures. This lower zone comprises the poorest Nugget reservoir rock in
the field.

The average matrix porosity of the Nugget reservoir rock is 12% with the highest
observed porosities approaching 22%. The average reservoir matrix permeability is
10.9 MD with some permeability’s ranging as high as 400 md. Correlation quality of the
Kmax plot (reservoir porosity plotted against maximum permeability parallel to the
bedding plane) and the Kh90 plot (reservoir porosity vs. permeability perpendicular to the
bedding plane) for this data set are 31% and 51%, respectively. This high variability
makes reservoir modeling predictions difficult.

Lindquist (1983) gives a detailed description of the Nugget eolian sandstones and their
relationship to reservoir petrophysics. Zone 1 is characterized by laminated sandstones
that are interpreted as stacked climbing wind-ripples and grainflow deposits in relatively
small cross-bed sets. Zone 2 is the best reservoir interval with generally higher
porosities, thicker cross-bed sets, and more grainflow deposits. Zone 3 is the poorest

© PetroSkills, 2013. All rights reserved. 7-49

Reservoir Characterization
Chapter 7

reservoir owing to its finely laminated wind-rippled and water-influenced depositional

textures.

Fractures probably influence the directional permeability within the field. Core
descriptions commonly identify gouge-filled fractures in Zone 1, only minimal gouge-
filled fractures in Zone 2, and numerous open fractures in Zone 3. The distribution of
fracture morphology is similar to that discussed by Nelson (1985), in which higher
porosity sandstones commonly shear and create a low-permeability gouge. Fracturing
in the lower-porosity sandstones more often creates open fractures that can enhance
reservoir permeability. The fracture sets in the Nugget Sandstone at Anschutz Ranch
East field are interpreted to be of dominantly tectonic, fold-related origin with generally
predictable senses of motion or offset, as set forth by Nelson (1985). There is also a set
of open fractures oriented at low angles to the structural axis that are thought to be
related to the period of post-thrust extension.

7-50 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

© PetroSkills, 2013. All rights reserved. 7-51

Reservoir Characterization
Chapter 7

Partners
The field is owned by six companies with differing interests throughout the field.

Regulation
The field straddles the Utah-Wyoming line. Each has jurisdiction for a portion of the
field. The field is located in a fairly sensitive environmental area.

7-52 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Petrophysical Data

© PetroSkills, 2013. All rights reserved. 7-53

Reservoir Characterization
Chapter 7

7-54 © PetroSkills, 2013. All rights reserved.

 Chapter 8 G EOPHYSICAL M ODELS
Seismic data densely sample the entire interwell region, while well log data are available only at the
well locations and the measurements in the well only probe centimeters into the near wellbore
region.
Araktingi et al.

Oil and gas companies are using an increased number of three-dimensional (3D) seismic
surveys in reservoir management because the detail needed to develop reservoirs is
increasing greatly. The amount of data requisite to find an accumulation is insufficient to
develop that same accumulation. Robertson (1991) describes the reservoir
characterization process and the application of 3D seismic to the resolution:

A good working definition of reservoir management is maximizing the

economic value of a reservoir by optimizing recovery of hydrocarbons while
minimizing capital investments and operating expenses.

The first thing to note is that this definition is not geophysical or geological;
in fact, it's not even an engineering definition. Reservoir management really
is the economic process of raising the worth of a property to the highest
possible level….

He continues:
A 3D seismic survey is one of the tools in the evaluation tool kit. An initial
interpretation of the survey impacts the original development plan. As
subsequent events like the drilling development wells occur, the added
information is used to revise and refine the original interpretation. Often, as
time passes and the database builds, elements of the 3D data that were
initially ambiguous begin to make sense, and the interpretation becomes
more detailed and sophisticated.

KNOWLEDGE ABOUT SEISMIC THEORY

 Wave principles
 Seismic acquisition
 Seismic processing
 Seismic interpretation
 Seismic attributes
 4D Seismic
 Well seismic

Petrel model Norg field, The Netherlands

Courtesy Shell (2012)

© PetroSkills, 2013. All rights reserved. 8-1

Reservoir Characterization
Chapter 8

WAVE PRINCIPLES

For the lithological rock column, an acoustic impedance (ρ x v) trace is made. At layer
boundaries, where ρ x v changes, acoustic waves are reflected and transmitted depending
on the contrast in ρ x v. Reflection coefficients are spikes at the layer boundaries. This
reflection trace is convolved with the seismic wavelet yielding the seismic trace. Because the
wavelet is not a spike but rather an acoustic disturbance of finite duration, averaging takes
place, which becomes visible in the seismic trace.

In seismic processing, as much as possible of this averaging is removed by a process

called deconvolution.

ACQUISITION

In the old days, seismic lines were acquired onshore using dynamite as the energy
source. Today, most onshore seismic data is acquired using vibroseis, where the energy
source is a giant vibrator truck in which a pad in the middle of the truck rests directly on
the ground to raise the truck so that the pad supports the full weight of the vehicle. The
truck then vibrates to put energy into the ground.

This technology was developed to acquire seismic e.g. in housing areas, where dynamite
was of course prohibited. However, because the energy transmission into the subsurface
is very controlled with vibroseis, the technique is nowadays also being used in remote
areas such as deserts.

8-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

Vibroseis truck

Offshore acquisition configuration

© PetroSkills, 2013. All rights reserved. 8-3

Reservoir Characterization
Chapter 8

Offshore acquisition configuration

Schematic figure illustrating the principles of reflection seismology. From an artificial

source, waves propagate in the earth and when reaching a surface corresponding to a
change in acoustic impedance (rocks with contrasting velocities and/or densities) they
will be reflected, refracted and diffracted.

8-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

Vessels towing a number of long streamers of listening devices record low-frequency

sound waves bounced off layers of rock far below the ocean floor. In this way a so-called
3D seismic survey is being acquired. Geophysicists process and interpret this seismic
data with proprietary computer software to create high-resolution images that give us
much more information than the images used by most of the industry.

The main advantage of 3D surveys is that “out-of-plane” reflections can be recorded and
treated properly. In 2D surveys it is assumed that all the sound from the subsurface
comes from a vertical plane underneath the line of receivers. This assumption is OK
when dips are relatively small. In faulted reservoirs when reflections against the faults are
also recorded, the assumption breaks down completely. 3D seismic solves this problem.

Usually, in 3D seismic much denser sampling is realized which is another advantage

leading to better resolution.

Applications of 3D seismic are in resolving complex subsurface structures and

stratigraphy with many faults and compartimentalization. Later in this chapter, 4D seismic
will be discussed, which allows flood fronts to be monitored and bypassed hydrocarbons
to be identified.

In the next two pictures a 2D and 3D seismic survey are being compared.

In the 2D survey a section is obtained which may be disturbed by out-of-plane

reflections.

In the 3D survey, a seismic cube is obtained providing a 3D image of the subsurface.

© PetroSkills, 2013. All rights reserved. 8-5

Reservoir Characterization
Chapter 8

Modern seismic acquisition technology includes three component seismic including shear
waves, wide azimuth seismic allowing for instance viewing underneath salt packages,
OBC (Ocean bottom cables) avoiding the negative impact of the sea water layer,
wireless acquisition offshore using ROVs (Remotely operated vehicles) placing acoustic
receivers on the sea floor, and wireless acquisition on land e.g. in rain forests minimizing
the operational footprint in pristine environments.

8-6 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

PROCESSING

PSTM = Pre-stack Time Migration

PoSDM = Post-stack Depth Migration
PSDM = Pre-stack Depth Migration

© PetroSkills, 2013. All rights reserved. 8-7

Reservoir Characterization
Chapter 8

8-8 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

In the two examples above, it is clearly seen that the Pre-stack Depth Migration or
Imaging is by far to be preferred.

Stacking is an averaging process removing statistical noise. Migration is required to

correct for the assumption that the sound recorded arrives from a direction vertically
underneath a receiver. This assumption distorts the image and therefore migration of the
image is required to remove this problem.

In PSTM the migration is done in time, the domain in which seismic is recorded.
Basically, this means that a constant velocity is being assumed for the entire subsurface,
which is not the case. In PoSDM, the migration is done after stacking. Indeed stacking
removes noise, but also part of the signal. In PSDM, every trace recorded is considered
separately, which removes the negative effects of the stacking. The term ‘imaging’ is
used when a forward modeling approach is applied while iterating until the constructed
seismic image resembles the recorded image sufficiently.

SEISMIC PROCESSING – SUMMARY

Deconvolution Improve vertical resolution

Stacking Remove statistics

Migration Improve the image

The latest development in seismic

processing is Full Waveform Inversion
(FWI). In a forward modeling approach
not only the first arrival times are being
considered, but the complete wave
trains are being inverted.

Clearly, this requires significant

computer power, but that is nowadays
available and is still also being
improved.

FWI also requires very special

processing algorithms which are
usually of a proprietary nature.

© PetroSkills, 2013. All rights reserved. 8-9

Reservoir Characterization
Chapter 8

GEOPHYSICAL APPLICATIONS

Geophysical data is applicable to reservoir characterization in four general applications:

analysis of fluid and rock properties, understanding the structural and stratigraphic
relationships of the accumulation, prediction of spatial variation of reservoir parameters,
and monitoring fluid flow and pressure within the reservoir(s). In the following diagram
these applications are indicated also demonstrating which disciplines are mainly
contributing in the various areas:

The seismic signal allows geological interpretation as follows:

8-10 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

INTERPRETATION

An interpreter needs to realize that as a result

of the frequency used, the vertical resolution of
the seismic signal is reduced.

The two pictures show a seismic signal compared

to a building and to a channel seen in an outcrop.
To resolve a feature in the seismic signal it needs
a size of at least a quarter of a wave length.
Clearly, the channel can not be resolved by the
seismic in this way. Later it is shown that by
studying seismic attributes, an improvement can
be achieved in this regard.

Increasing the frequency would improve the

resolution, but the disadvantage is that this
would cause a faster attenuation of the acoustic
signals reducing the depth of investigation. The
frequency needs to be selected to balance
resolution and depth of investigation. In this
way, frequencies around 50 herz have been
choosen.

An excellent study was carried out by Gartner and Schlager (AAPG), who constructed
seismic sections for a synthetic example using different frequencies:

© PetroSkills, 2013. All rights reserved. 8-11

Reservoir Characterization
Chapter 8

8-12 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

Nowadays seismic interpretation is done on workstations that allow full integration with
the business of other disciplines. The example shown is a display from Landmark’s
‘Decision Space 3D’.

This integration on workstations allows inclusion of all data in a consistent and uniform
way including comparison of seismic with geological models, assessment of drained
areas, well planning, etc.

STRUCTURAL INTERPRETATION

SEISMIC CUBE
After processing of 3D seismic, the ‘seismic cube’ is obtained. The seismic cube is an
areal array of seismic traces with time in the third dimension. These traces may be
compared with synthetic seismograms obtained from well data, if at a particular areal
location a well was drilled.

The cube can be sliced in any vertical dimension to create vertical sections or sliced in
horizontal dimensions to create time slices, which represent constant time. In addition,
sections through the seismic cube may be created following a particular loop in the
seismic, providing a horizon slice possibly representative for say the top of a reservoir.

An example of a seismic cube demonstrating the structural interpretation in the upper

part of the cube, is given below:

© PetroSkills, 2013. All rights reserved. 8-13

Reservoir Characterization
Chapter 8

STRUCTURAL AND STRATIGRAPHIC RELATIONSHIPS

The traditional role of seismic data as it relates to reservoir characterization is to define
the details of the structural aspects of each reservoir. Fault placement, reservoir
boundaries, dip relationships and other subtleties of the field have been - and are still
being - mapped in most E/P companies throughout the industry. The detailed
examination of sequence stratigraphy has added a great deal of value to reservoir
characterization. Because of the explosion of 3D technology in the last decade, these
applications have become routine. In particular, the ability of high-resolution to define
reservoir limits has had significant positive impacts on improving the quality of the
reservoir model.

8-14 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

Robertson offers the following warning: “minor character changes in 3D seismic data
don’t need to correlate with real geologic changes”. It is, therefore, critical that a multi-
disciplinary approach to reservoir characterization includes interpretation of the seismic
data. The greater the understanding of the reservoir model characteristics such as the
expected Lithology, stratigraphic sequences, continuity and structure the better the
resulting interpretation. This requires detailed communication across the disciplines.

The next example compare a 2D and a 3D interpretation. The 2D is smooth, whereas the 3D
shows a lot more detail. This is not necessarily noise, but a result of the much denser
acquisition of the 3D survey. In the 2D interpretation much more averaging is required.

Clearly, the saddle in the 3D interpretation becomes much more visible.

© PetroSkills, 2013. All rights reserved. 8-15

Reservoir Characterization
Chapter 8

The next example shows the interpretation of a reef. At various times, time slices are cut
from the 3D cube displaying the variation of the reef with time, or after a time-to-depth
conversion with depth.

8-16 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

The last example shows a subsea structure map of the top of a reservoir sandstone. The
map was created by multiplying the time structure map, as interpreted from the 3D
seismic data set, by the velocity gradient map:

The color transition from white to green represents the pool’s original gas/water contact.
The size of the grid is 400 x 400 ft.

SEISMIC COHERENCE

Seismic coherence is a seismic attribute, which is discussed under ‘structural

interpretation’, because it is a powerful technique to extract structural information from
the seismic data. It is a measure of the trace-to-trace similarity (correlation) of the
seismic waveform within a small time window. If the correlation is high, the continuity of
the stratigraphy between the traces is high; if the correlation is low for instance a fault or
fracture may be present in between the traces. From the seismic cube a coherence cube
is constructed that highlights structural features in the seismic cube.

For each point in the 3D seismic volume, the wafeform of adjacent traces (e.g. red trace
compared to the blue traces) over a short vertical time window:

© PetroSkills, 2013. All rights reserved. 8-17

Reservoir Characterization
Chapter 8

8-18 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

WHY USE SEISMIC COHERENCE?

 Excellent tool for delineating geological boundaries (faults, lateral stratigraphic
contacts, etc.)
 Allows accelerated evaluation of large data sets
 Provides quantitative estimate of fault / fracture presence
 Often enhances stratigraphic information that is otherwise difficult to extract

SEISMIC ATTRIBUTES

In addition to seismic coherence, there are many other seismic attributes, such as
amplitudes, phase changes, frequency variations, dip and azimuth data, interval travel
times between events, velocity ratios (vp/vs). Typical information to be derived from these
attributes include the presence of gas (bright spots) from amplitudes, lateral continuity
and bedding configuration from phase information, bed thicknesses from frequency data
using spectral decomposition, and as we discussed, faults and fractures and other
stratigraphic information from coherence.

Attribute analysis is particularly powerful if well control is available. Correlations of

attribute with well data such as porosity, fluid type, lithology, net pay thickness and other
reservoir parameters can be utilized. Robertson describes the process well:

“The correlations usually require borehole control (well logs, cuttings,

cores, etc.) both to suggest initial hypotheses and to refine, revise, and
test proposed relationships. An interpreter develops a hypothesis by
comparing a seismic parameter in the 3D volume at the location of a
well to the well's information, often through the intermediary of a
synthetic seismogram match. The hypothesis is then used to predict
rock/fluid properties away from the borehole control, and subsequent
drilling validates (or invalidates) the concept. For example, 3D seismic
surveys are commonly used in the productive Pleistocene trends of the
offshore Gulf of Mexico to directly map gas saturation. One correlates
seismic amplitude anomalies with gas-saturated sandstones, and then
maps the areal extent (and sometimes net feet of pay) of these bright
spots laterally and vertically through the data volume.”

Arestad et al. have demonstrated an innovative multi-disciplinary approach to the

derivation of rock properties; their usage of 3D, 3-C (multi-component) seismic surveys
result from input from the reservoir characterization team at the Colorado School of
Mines. Their focus is the discrimination of porosity and lithology in the upper Devonian
Nisku dolomite reservoir. Compression seismic data was unable to characterize the
Nisku, but shear wave applications imaged the productive reef. The p to s wave plots
(Vp/Vs) also correlate with porosity variations within the reservoir.

© PetroSkills, 2013. All rights reserved. 8-19

Reservoir Characterization
Chapter 8

Integration of the seismically derived parameters with other data is often a difficult
procedure - especially if the desire is to be able to incorporate hard data rather than soft
data. The geostatistical section above offers some suggestions on methods to integrate
data sets. Chambers et al. provide the following observations:

“The general motivation behind data integration is to combine data in

an effort to obtain more accurate results having reduced uncertainty
than would be achieved by using only a portion of the data. In theory,
this is great; however, in practice we often find it difficult to combine
data measured by different tools and at different scales. Several
geostatistical methods are available that provide the fundamental
framework for quantitative data integration.”

The integration of secondary data sets, such as acoustic impedance data from the
inversion of 3D seismic data, can significantly reduce interwell estimation uncertainty.
The soft data also add small-scale variability not sampled from well data alone. Extra
care is required in the data processing and inversions steps if seismic data are to be
quantitatively integrated into a reservoir description, especially if the final product is a
fluid flow model. To realize the full potential of the secondary data set, data types must
be properly correlated and scaled.

Loren et al. offer methodology that is applicable to areas of minimal log control. The lack
of control may be due to a lack of enough wells drilled in the area or due to old logs with
inadequate information. The method they suggest is:

“Synthetic acoustic and density logs can be created from commonly

available resistivity logs and a uniquely derived sand fraction curve.
Examination of drill cuttings, sidewall, and whole cores enhances the
accuracy of the sand fraction curve, defines reservoir quality and
delineates exploration opportunities. Rock, log, and calculated data are
transformed into the time domain and integrated into the seismic data
by means of the synthetic acoustic transit time. The process algorithms
are project-calibrated with a learning set of diversely representative
high quality log data. After verifying algorithm calibration, the synthetic
log generation process can then be applied regionally with confidence,
even in older wells with "ancient" resistivity logs. The synthetic acoustic
and density logs are the basis for models of variable Lithology and fluid
content, from which synthetic seismograms and AVO response are
generated.”

Araktingi et al. offer a comprehensive review of methods of integrating seismic and well
logs. This presentation offers a primer for those who want to apply seismic data sets in
the reservoir characterization process. Included in their paper is a compilation of
limitations, problems and pitfalls. Significant points include:

1. Expected minimum resolvable thickness from seismic is ¼ of the

wavelength. For example, in clastic rocks with an average velocity of

8-20 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

2500 meters/sec (8200 feet/sec.) with a 50-Hertz high frequency end,

resolution is about 12.5 meters ( 41 feet).
2. Seismic is recorded in time rather than depth. Conversion requires
time-depth information for usage. Scarcity of velocity information and
its approximate nature causes application problems. (see discussion
below)
3. Low range frequencies (below 8 to 10 Hertz) are frequently not included
in the data sets. These frequencies are important to porosity derivation
and must be supplied from elsewhere.
4. Most processing programs assume a stationary waveform. Processing
algorithms may violate the stationary waveform making trace-based
convolution at fault and producing poor to inaccurate inversion results.
5. Integrated seismic logs often mistie seismic data by 5-10% due to poor
hole conditions, mud invasion and seismic dispersion.

As a result of these limitations, extreme care and in-depth understanding of the geologic
models are very important to deriving appropriate data from the seismic data set.

Oz Yilmaz in “Thinking in Depth” has presented thoughts on the time-depth relationships

in developing an earth model. Portions of his abstract are reproduced below for
consideration:

“Processing and inversion are intended to estimate an earth model from

seismic data. Nevertheless, they differ in one fundamental respect -- the
output domain. Processing yields an earth model in time, whereas
inversion yields an earth model in depth.”

“A way to judge accuracy of an earth model derived from processing and

inversion is by way of time and depth migration, respectively. In fact, we
apply these two migration methods to post- and prestack data to derive an
earth model, itself. Representation of an earth model in time usually is in
the form of a velocity field for time migration, which has to be smoothly
varying both in time and in space. Representation of an earth model in
depth usually is in the form of a detailed velocity-depth model for depth
migration, which comprises two sets of parameters – layer velocities and
reflector geometries.”

“A fundamental problem with inversion is velocity-depth ambiguity, which

requires independent estimate of layer velocities and reflector geometries.
Due to velocity depth ambiguity, an output from inversion is an estimated
velocity-depth model with a measure of uncertainty in layer velocities and
reflector geometries. It now is widely accepted in the industry that results of
inversion are geologically plausible only when there is a sound
interpretation effort put into the data analysis. In this context, inversion can
be viewed as interpretive processing in contrast with conventional

© PetroSkills, 2013. All rights reserved. 8-21

Reservoir Characterization
Chapter 8

processing.”
In the following a number of examples are presented:

FLUID EFFECTS ON AMPLITUDES

The graph shows the fluid dependence of the P-wave velocity for gas and oil bearing
sandstone. Clearly, for gas the gas effect is significant between say 85% and 100%
water saturation causing a significant effect in acoustic impedance.

As a result, when entering a gas zone there is a strong reflection yielding high amplitudes
compared to a much smaller reflection down structure in the water zone.

In the seismic section, this effect may be observed. Such a reduction in amplitude when
moving off-structure is often accompanied bij a flat reflector indicative for a gas-water
contact. Schematically, this is shown in the next picture:

8-22 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

The next areal map is a field case from the Imo River M1.0 reservoir in Nigeria (courtesy
Shell). The red color represents the high amplitudes. Noticeable in this map is that the
amplitude anomalies conform very well with the fault pattern in the reservoir.

The next example compares a synthetic seismogram derived from well data with the
seismic section. Using the well data a stratigraphic interpretation of the section can be
made:

© PetroSkills, 2013. All rights reserved. 8-23

Reservoir Characterization
Chapter 8

The next picture shows the Spectral Decomposition of a 50 ms window centered on the
"D" sand of an example reservoir. The 30 Hz slice appears to be the best at imaging the
"D" sand.

8-24 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

© PetroSkills, 2013. All rights reserved. 8-25

Reservoir Characterization
Chapter 8

To obtain the best results from attribute analysis, a combination of attributes should be
used. The following field case from the C5 Project (Crave Norte area, Colombia,
September 2011) is an excellent example using the horizon slice itself (Picture A),
seismic coherence (Picture B) and spectral decomposition (Picture C).

Picture D presents the channel interpretation based on the attribute analysis.

C5 Project (Columbia, September 2011) – seismic attributes

8-26 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

The next example compares a stratigraphic section with a geologic impedance model. In
the impedance picture some of the internal structure of the reef van be observed.

MONITORING FLUID FLOW – 4D SEISMIC

Nowadays 4D seismic data acquisition is common practice. The method is outlined

below:

4D SEISMIC MONITORING
 Time-lapse reservoir monitoring
 Fluids alter impedance in reservoir
 Change in fluid content changes impedance
 Differences may result from fluid changes
 Technical challenges:
 Matching datasets
 Compatible processing

© PetroSkills, 2013. All rights reserved. 8-27

Reservoir Characterization
Chapter 8

 Positioning
 Quantification
 Good reservoir model
 Good analysis of the Value of Information (VOI)

Staging 3D seismic surveys over time to monitor a producing reservoir can determine the
position of fluids being moved by production and can predict future fluid movement more
accurately. Monitoring the reservoir in this way helps:

 Identify bypassed oil and hydraulic barriers

 Predict premature water or gas breakthrough
 Prevent lost wells and possibly reduce number of infill wells
 Enhance recovery.

Baseline survey across a field prior to production is the first step in the 4D process.
Operators in mature areas may use seismic reservoir monitoring to prevent premature
abandonment. The technical concept behind 4D is to relate changes in seismic
attributes with time to changes in the reservoir due to production. 4D will not work for all
fields: it depends on the rocks, the pore fluids, reservoir thickness, structure and depth,
and data quality and consistency. 4D seismic is often combined with other forms of time-
dependent information. Other 4D datasets can include pressure histories, temperature
monitoring from producing wells, cased-hole worker logs such (pulsed neutron) and
reservoir simulations of the expected vs. actual behavior of the field. Companies now
expect to remotely sense the drainage pattern in subsurface reservoirs. Time-lapse
images of producing reservoirs are acquired through the repeated acquisition of 4D
seismic surveys to track hydrocarbon movement with time. The promise of identifying
bypassed and poorly drained reserves are very encouraging. 4D seismic truly
represents an integrated approach that brings together disciplines that were separated
from each other almost entirely in approaches that are more traditional.

Xu et al present a case history of a 4D application in Eugene Island 330 Field in offshore

Louisiana. Their description follows:

“We can demonstrate the utility of 4D seismic monitoring in the Eugene

Island 330 Field of offshore Louisiana. Four generations of 3D seismic
surveys have been acquired over the Eugene Island 330 field, first in 1985,
then in 1988, 1992 and again in 1994. These datasets contain seismic
snapshots of the field as it has been drained of more than 100 MMBOE.
The 4D study was begun in 1992. At that time, a traditional, gravitational
water drive was assumed to be sweeping the LF reservoir in the four
corners area of Blocks 330/331/337/338. A 1992 snapshot shows
considerable seismic amplitudes remaining downdip of the supposed
oil/water contact. Up until 1992, primary wells draining the fault block were
the A-11 and A-6 of Block 331, the B-2, B-5 and B-7 of Block 330, and the
A-13 and 1-15 of Block 338.

After the 3D seismic survey of 1992 was acquired, there were only three

8-28 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

wells active in the fault block: new wells B-5ST, B-6ST in Block 330 and
the A-12 in Block 338. A new 3D seismic survey was acquired in 1994,
and it clearly shows drainage, or dimmed amplitudes, caused by
production in the intervening years. About 2.4 MMBOE were produced
from these wells, and about 600 acres were dimmed. Assuming a 100 foot
pay sand with 30% porosity and a drop in Sw, of 30%, that would require
that each acre-foot produce about 400 bbl, quite reasonable for clean
Pleistocene sands in the GOM.

The bypassed oil remaining after 1994 will require the placement of yet
another well into the fault block, and therein lies the power of 4D seismic
monitoring. Each new well recovers an additional percentage of oil-in-
place, increasing the ultimate recovery efficiency. In addition, the 4D
seismic changes can be modeled within a reservoir simulator to risk the
new well. Amplitudes drained by the new well are predicted to account for
an additional 2 MMbbl for a well placed horizontally along the 330/338
property line to the four corners. In another two years, that well will water
out, and additional amplitudes are predicted to remain after that. Then a
well projected more north-south across the property line will be required to
recover an additional 2 MMbbl, and so on into the future.”

4D SUCCESS FACTORS
 Best applied in soft unconsolidated reservoirs
 Works best when gas phase is appearing or disappearing
 Oil / water transitions more difficult to image

All related to:

- compressibility of the system

- moduli of the system (bulk and shear) and density

TECHNICAL SUCCESS FACTORS - SHELL (AFTER DE WAAL AND CALVERT)

 A good pre-production base survey is vital.
 Preferably, the same acquisition geometry for base and repeat surveys.
At least, a sub-set where acquisition geometry is the same.
 Joint reprocessing of base and repeat surveys.
 Use of proprietary Shell in-house software.
 Training of specialist 4D seismic geophysicists.

As an example of 4D seismic modeling, a recent study by Statoil is shown. This study is

considered a best practice example of how geophysicists, geologists and reservoir
engineers must work together. The study also demonstrates that excellent benefits can
be derived from this integrated 4D reservoir modeling.

© PetroSkills, 2013. All rights reserved. 8-29

Reservoir Characterization
Chapter 8

CROSS-DISCIPLINARY GEOPHYSICAL RESERVOIR MODELING (GRM) ON VISUND

NORD – MODEL UPDATE AS A BASIS FOR NEW WELLS

This study was carried out by S. Njerve and I. Mæland of Statoil, Norway. The study was
presented at the 74th EAGE Conference in Copenhagen, Denmark, 4th to 7th June 2012.

SUMMARY
The decision of sub-sea redevelopment of the Visund North field that was closed in since
2006, was made. GRM was to be applied to update existing models. By thorough cross-
disciplinary use of production data and geological and geophysical interpretation of 4D
seismic, the Visund Nord dynamic reservoir model was updated. Based on the new
model, two new oil producers east in the area and an additional exploration pilot were
planned. Drilling started in Q3 2012. It is likely that another two wells will be drilled from
the sub-sea template in the future.

- Northern part of the Visund field in

block 34/8, Brent reservoir with 40m
rim and gas cap.

- Produced oil from sub-sea wells

between 2002 and 2006.

- Closed-in in 2006.

Before the GRM was done, there was a reservoir

simulation of the northern Visund area. A history match
of production and pressures could be made, but

uncertainties remained in terms of the fluid movement in

the reservoir also related to uncertainties in the areal

8-30 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

communication within the reservoir. This simulation model is summarized as follows:

As the reservoir simulation led to conservative recommendations (new wells too close to
existing wells), the project remained marginal and it was decided to do a full GRM study
in order to determine whether integration with the seismic data would make the project
feasible.

The seismic database consisted of four different surveys in 1990/1991, 2004, 2007 and
2010. An example of the quality of the seismic can be seen from the following dip line
(blue line in map view):

In the difference picture, the anomaly caused by the water flooding is clearly observed.

© PetroSkills, 2013. All rights reserved. 8-31

Reservoir Characterization
Chapter 8

Since the oil production started in 1999, the flooding indicated by the seismic has
developed from north to south. Between 1991 and 2004, the strongest flooding has been
mapped in the north, while between 2004 and 2007, the anomaly has moved further
south. This flodding pattern was confirmed by production data from the two previous
wells, of which the most northern one observed high watercut quite early.

Example RMS amplitude maps are as follows:

Since the height of the flooded part of the oil rim is far below the seismic resolution, 1D-
modeling of different flooded cases was used to understand the tuned 4D signal and to
quantify the movement of the OWC. The principle of this 4D tuning works as follows:

The two stratigraphic columns are convolved with a wavelet to produce two synthetic
seismograms, of which the difference is taken (note the polarity difference in the two
synthetics).

With this principle synthetics for different levels of the OWC were constructed:

8-32 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

Using these synthetics, relationships were derived between the OWC movement (m) and
time difference (ms), and secondly, between the OWC movement and the amplitude ratio
(amplitude at time t divided by amplitude of base) difference:

Detailed analysis of the various time-lapse seismic surveys led to maps showing the
flooding development as derived from the seismic. These maps were the main input from
the geophysicist to the reservoir engineer:

© PetroSkills, 2013. All rights reserved. 8-33

Reservoir Characterization
Chapter 8

The objective of the history matching was to match the produced volumes, the observed
reservoir pressures and the flooding maps obtained from seismic. The match was
achieved by varying the size of the gas cap and of the aquifer, by adjusting the
transmissibility across the faults in the model and by varying the vertical communication.
The results of this reservoir simulation are shown in the next picture:

In order to increase the confidence of the quantitative 4D interpretation, the history-

matched simulation model was used to derive synthetic seismograms, which were
compared with the recorded seismic:

The base case for redevelopment of Visund Nord is to drill new oil producers in the north
and the south away from the identified 4D seismic anomalies in the area. Although the
4D affected areas are not considered to be completely flooded, it is expected that
production east of the previous wells will drain the remaining reserves without too high a
watercut.

8-34 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

In addition, an exploration well will be drilled on the eastern side, where hardly any 4D
anomalies have been observed.

There are two more slots on the sub-sea template, which may be used in future for two
additional development wells.

The success of the project can be summarized as follows:

The process of Geophysical Reservoir Modeling (GRM) can be summarized as follows:

In this process, iteration through the various steps may be required, which makes a good
integration between geophysics, geology and reservoir engineering more than vital.

© PetroSkills, 2013. All rights reserved. 8-35

Reservoir Characterization
Chapter 8

PetroSkills is most grateful that Statoil made the above study available for the benefit of a
broader audience. The main authors, S. Njerve and I. Mæland, are commended for an
excellent piece of work. Their integrated approach to 4D reservoir modeling
demonstrates best practice, which PetroSkills wishes to recommend to all in the
integrated reservoir characterization arena.

ADDITIONAL EXAMPLES OF 4D SEISMIC

AUGER FIELD - GOM

4D - Equalizing the Data
 1997 onward, the surveys were shot using metric units - but Auger's earlier
surveys were shot in English units
 2002 survey utilized latest acquisition technology - a substantial improvement over
the 1990 baseline survey
 State-of-the-art processing to create a separate data set for additional
stratigraphic and structural details from the seismic

 New acquisition techniques allowed denser lateral sampling and better vertical
resolution
– Better resolution from the newer data images layers 40 feet instead of 80 feet
– Lateral spacing of 50 feet instead of 100 feet
 Structural or stratigraphic details that explain an amplitude remaining isolated from
producing wells

Auger Field 4D results

 Potential for stranded attic opportunities and unswept bypassed reserves
 Geochemistry provided evidence that barriers broke down over the life of a
reservoir
 Helped validate Shell's overall development plan
 Helped the team discover additional deeper targets

8-36 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

 Identified redevelopment opportunities

– undrained attic hydrocarbons
– unswept down-dip and laterally-isolated hydrocarbons
– newly delineated deeper reservoir potential
 Confirmed course of action was on track

CARBONATE RESERVOIR – MALAYSIA

During production of gas from this carbonate reservoir, a tilted gas-water contact
developed, causing some wells to produce water. A theoretical feasibility study showed it
was just possible that 4D monitoring would give usable results. Seismic profiles recorded
in 2001 were compared with their 1990 equivalents. The results exceeded expectations,
the difference between the two surveys revealing the original (OGWC) and present gas-
water contacts (PGWC). The test showed that time-lapse techniques can be used,
provided the data are recorded with techniques that match each other closely.

NORTHERN NORTH SEA - UK

Three 3D seismic surveys have been acquired on a field in the Northern North Sea, in
1980, 1995 and 2000. Differences seen mainly between the 1995 and 2000 surveys
allowed detailed fluid movements to be mapped. Infill wells were targeted in areas where
oil had not been produced. The 4D results even allowed engineers to re-perforate
existing wells, producing oil from reservoirs where oil had been left behind.

In the picture on the next page, the original oil water contact (OOWC) is shown. The
brighter colors represent areas with hydrocarbons. It is seen that the flooding of the
reservoir can be well observed. In the lower left corner of the picture, remaining
hydrocarbons are observed:

© PetroSkills, 2013. All rights reserved. 8-37

Reservoir Characterization
Chapter 8

OMAN - MIDDLE EAST

A novel application of 4D reservoir imaging involves placing sequences of seismic

detectors down boreholes. 2D or 3D vertical seismic profiles (VSPs) can be recorded to
monitor changes in the rocks, or detect fluid movement during oil production.
Alternatively, the detectors can just listen in `passive' mode to detect very small acoustic
events (micro-seismic), which are induced during hydrocarbon extraction due to the
changes in reservoir stress. These surveys are relatively inexpensive and can be
repeated frequently. Five projects are ongoing in Oman to monitor steam-flooding, water
injection and natural depletion. Oman is ideal to test these applications with downhole
geophone arrays as conventional surface seismic data in this region typically lacks the
resolution associated with marine seismic data. Placing the geophones subsurface
dramatically increases the data quality and hence facilitates reservoir monitoring.

8-38 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

WELL SEISMIC

CROSS-WELL SEISMIC

Crosswell data are collected by placing a seismic source in one well and a receiver string
in a nearby well. Energy which propagates directly between wells without being scattered
serves as the basis for constructing velocity images (tomograms). Energy which is
reflected is used to construct reflection images.
Cross-well seismology has the capability to define reservoirs with much more detail.
Paulsson demonstrates this capability using computer tomography modeling. The
process can add detail not available previously and has implications in identifying non-
continuous reservoirs, location of pitchouts and details of the interwell structure. The
following illustration is a model of such an application:

© PetroSkills, 2013. All rights reserved. 8-39

Reservoir Characterization
Chapter 8

Note that in this latter example the well distance is only 56 m.

The cross-well reflection and velocity images (center) fill the resolution gap between
modern surface seismic data (left), and the sonic log data (right center) and core
measurements (far right). These data are from a West Texas carbonate reservoir.

8-40 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

Texaco EPTD evaluated a number of candidate projects, seeking a gas reservoir in an

area where crosswell had previously worked well and where a variety of well separations
were available to better characterize the effect of increasing interwell distance on the
resulting crosswell images. Texaco selected a West Texas sandstone reservoir 40 to 60
ft thick, within a carbonate section.

A crosswell acquisition system developed by Texaco EPTD using an air gun source
produces greater low-frequency output than the GRI/TomoSeis RCPTM piezoelectric
source used in high-speed, commercial operations. The air gun and RCP (resonant
cavity piezoceramic) sources are similar in many ways: both are fluid coupled and can be
operated "on-the-fly," firing while the source is moving. As a result, GRI, Texaco EPTD
and TomoSeis worked together to identify a gas reservoir application in which both
sources could be operated at well spacings greater than 40 acres

Example - McElroy Field, W. Texas –Tucker, et al, 1998 AAPG. S-Wave images between
wells 1068 and 1202.

© PetroSkills, 2013. All rights reserved. 8-41

Reservoir Characterization
Chapter 8

COMPARISON OF SURFACE SEISMIC, 3D VSP DATA AND SONIC LOGS

8-42 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

WELL PLANNING WITH SEISMIC

This example is from the Clark Wilson sandstone in the Mist field. This reservoir was
selected for a gas storage project.

2D SURVEY BEFORE PRIMARY PRODUCTION

The following figure is a vertical seismic section from 3-D survey parallel to 2-D seismic
section of figure 8-44. The top of the Clark and Wilson Sandstone (C&W) has a strong
trough (red) amplitude anomaly. There is a “relict” amplitude anomaly near the downdip
edge of the original gas-water contact (GW). There is no amplitude anomaly associated
with the “new” gas-water contact (NG/W).

© PetroSkills, 2013. All rights reserved. 8-43

Reservoir Characterization
Chapter 8

3D SURVEY AFTER PRODUCTION

CROSS-SECTION BASED ON 3D SURVEY

Cross section through a Mist gas pool after primary production. Note the position of the
“new” gas-water contact. Discovery well on left; injection/withdrawal well on right. Vertical
exaggeration 1.7-to-1.

The next figure is a vertical seismic section parallel to the path of a "horizontal” well.
Looking at many sections like this, and at time slices and flattened time slices, is
invaluable when planning a well path. The amplitude anomaly continuity gives a "feel” for
a variability of internal reservoir stratigraphy or structure.

8-44 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

SEISMIC WHILE DRILLING

Seismic-while-drilling (SWD) is one of the latest applications of seismic acquisition

technology. Using the energy from the drill bit as it penetrates the earth, seismic
reflections are recorded and processed on-site for immediate interpretation. The quality
of the output is equivalent to that obtained from more expensive VSP surveys. Also,
because one is acquiring seismic ahead of the bit, anomalous readings caused by over-
pressure or faulting can be used to alert the drilling crew to potential problems ahead.

© PetroSkills, 2013. All rights reserved. 8-45

Reservoir Characterization
Chapter 8

8-46 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

GAS POOL EXERCISE - DEBRIEF

© PetroSkills, 2013. All rights reserved. 8-47

Reservoir Characterization
Chapter 8

PROJECT ECONOMICS

year 2011 2012 2013 total

production per 35.2
7665 7665 1533 = Gp, well rate 7 MMscf/d
year (MMscf) Bscf
million BOE 1.3 1.3 0.3 1 BOE = 5800 scf
income
33.0 33.0 6.6 1 BOE gas = US$ 25
(million US$)
OPEX 5 5 2.5 5 million per annum
Depreciation Plant cost US$ 40 million,
10 10 10 10 40
Gas Plant depreciation over 4 yeras
Depreciation Well cost US$ 8 million,
4 6 8 8 6 4 4 40
Wells depreciation over 4 yeras
Income before
22.0 24.0 0.1
tax (million US$)
Tax 16.5 18.0 0.1 Tax rate 75%
Income after tax
5.5 6.0 0.0
(million US$)
Discounted
income 5.5 5.2 0.0 Discount rate = 15%
(million US$)
Discounted
income 5.5 5.2 0.0 10.8 = NPV (million US$)
(million US$)

The NPV is US$ 10.8 million. OK, a small project but profitable.

What will be the strategy for the years to come?

8-48 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

INTEGRATION CASE HISTORY

The following is a flowmap of modeling procedures showing the feedback loops with 3-D
seismic processing (via 3-D velocity model) and 3-D seismic interpretation (volume
analysis, dine-depth conversion). As a final step, 3-D synthetic seismic can be used for
quality control of the reservoir model This process also allows a quantitative assessment
of the likely probabilities to be made for the different modeling scenarios.

Input Modelling
1-D Well Data

Multiple Scenarios Modelling

3-D Depositional Model
2-D Analogue Data
3-D Diagenetic Model

3-D Seismic Interpretation

•Horizon mapping 3-D Rock Type Based Model
•Volume analysis Time

3-D Seismic Processing 3-D Velocity Model

3-D Rock Type Based Model

Geological Modelling
Depth
to Seismic Feedback
Loops
3-D Synthetic Seismic

Grotsch and Mercadier, 1999

© PetroSkills, 2013. All rights reserved. 8-49

Reservoir Characterization
Chapter 8

Grotsch and Mercadier, 1999

Schematic overview of the modeling steps Illustrating the Integration of different data
sources into the rock type based 3-D reservoir model As an example, the two-
dimensional cross section in the northern part of Malampaya with wells MA-1 and MA-2
Is used. (A) Cross section of the 3-D seismic reflectivity data. (B, C) Body rendering
using 3-D volume analysis allows identification of large-scale reservoir bodies of similar
properties, such as the pervasively cemented oceanward western margin. Integration of
seismic and well data allows establishment of a 3-D diagenetic model (D) Depositional
areas as outlined for the seismic horizon N1/2 (‘most likely” scenario). The location of the
cross section used is indicated as a dashed line. (E) Growth stages as outlined by
seismic horizons are mapped from the 3-D reflectivity data set (A) (F) Depositional model
derived from well and 3-D seismic data showing the large-scale depositional areas that
are laterally constrained by 3-D seismic volume analysis and well data. (G) Diagenetic
model indicating the pervasive early marine cementation along the western margin and
the cemented horizons at the buildup tops. (H) The reservoir rock type (RRT) subdivision
derived from petrophysical data and neural network analysis provides the high-resolution
vertical variability of ERTs in the 3-D reservoir model (see Figure 9). (1) Cross section
through the RRT-based reservoir model, which is used for saturation modeling and as
input for dynamic simulation.

8-50 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

COMPOSITE TECTONIC AND SEDIMENTOLOGIC HISTORY

The above figure is a schematic drawing showing a composite section that illustrates the
tectonic and sedimentary history of the Malampaya/Camagopla~formations. The
Oligocene part of the Nido Limestone overgrows a preexisting rug geomorphology for
nearby top-ended clastics deposits prior to the Nido Limestone. Note that the Eocene nor
mal faults along the western margin of the Malampaya~ buildup were reactivated during
the middle Miocene transgressional movements, causing tectonic inversion Some
reverse fault movements also are observed along the eastern margin of the platform. The
figure is not to scale; distances between the wells are given for reference.

© PetroSkills, 2013. All rights reserved. 8-51

Reservoir Characterization
Chapter 8

DIAGENESIS

8-52 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

SUBSIDENCE AND RELATIVE SEA LEVEL HISTORY

This figure shows the subsidence and relative sea level history of the
Malampaya/Camago platform as derived from Sr isotope stratigraphy (absolute ages)
and carbon isotope measurements (suggesting repeated exposure). High resolution of Sr
isotope dating in the Oligocene—early Miocene allows recognition of time gaps of 0.5—
2.0 m.y. during the growth history of the platform. Constrained data points are indicated
as circles on the curve. Note the change in tectonic subsidence during the late Rupehan
with rates changing from about 10 m/m.y. to 2O m/m.y. (not compensated for absolute
eustatic sea level change).

© PetroSkills, 2013. All rights reserved. 8-53

Reservoir Characterization
Chapter 8

RESERVOIR ROCK TYPES

The above figure shows thin-section photomicrographs illustrating the reservoir rock types
(RRTs) for the Nido Limestone in the Malampaya/Camago area. Five RRTs have been
defined based on thin section analysis, prosily, permeability, and saturation-height
functions. Note that RRTs do not necessarily follow the classical texture-based carbonate
classification schemes, but rather combine the effects of depositional texture and
diagenetic overprinting into classes of common petrophysical properties. This approach
was necessary because the diagenetic overprinting is not controlled by fabric or lithofacies.
Average properties for porosity and permeability and main pore types are listed for each
RRT Well name and sample depth are given below each photomicrograph.

8-54 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

TIME-DEPTH CONVERSION
The following figure shows time-to-depth conversion in carbonates is predominantly a
function of the 3-D porosity distribution within the carbonate platform. Reservoir modeling
was first performed in the time domain (A). This approach allowed construction of a 3-D
Nido limestone velocity model that honors the actual 3-D reservoir architecture. Based on
this velocity model, the porosity dependent 3-D depth conversion of horizons was
performed, matching in all scenarios the well data (B). For this approach, porosity-
velocity functions for gas- and oil-bearing zones and the gas-oil contact (GOC) were
provided to the model (steps 1 and 2 in the box). Using this information, depth (z) was
calculated on a cell-by-cell basis (step 3 in the box).

© PetroSkills, 2013. All rights reserved. 8-55

Reservoir Characterization
Chapter 8

RESERVOIR ROCK TYPE MODELS IN DEPTH

INTEGRATION CASE STUDY RESULTS

 Multi-scenario models were built in depth

 Time-depth conversion done with velocity-porosity model

 Iteration loops in such a study can not be avoided

8-56 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Geophysical Models

SUMMARY OF CHAPTER 8
The seismic sequence consists of acquisition, processing and interpretation. The three
major processing techniques are deconvolution to improve seismic resolution, stacking to
remove noise and migration to improve the seismic image.

PSDM (pre-stack depth migration) processing enhances the seismic record significantly.

The latest development in seismic processing is Full Waveform Inversion (FWI), which
has become possible as a result of ever growing computer power. Clearly, with FWI,
more data are considered and a better image of the subsurface may be expected.

Interpretation of seismic using seismic attributes such as amplitude, coherence and

frequency assists in highlighting many details sometimes hidden in the seismic record.

Seismic visualization has made major steps forward thanks to the rapid improvement in
seismic work stations. Workstations allow fully integrated reservoir characterization,
because all data is being dealt with in a consistent way.

4D seismic has become a powerful technology to monitor fluid flow in reservoirs. The
Geophysical Reservoir Modeling (GRM) process consists of three steps, namely 1)
interpretation of seismic differences in terms of flooding patterns; 2) history matching of
the flooding patterns derived from seismic together with history matching of produced
volumes, pressures, fault sealing patterns, etc; and 3) construction of synthetic seismic
using the history matched reservoir model and comparing this synthetic seismic with the
recorded seismic.

Well-to-well seismic is powerful in cases where well distances are relatively small.

Seismic while drilling may be a means to view ahead of the bit, e.g. for overpressure
prediction.

© PetroSkills, 2013. All rights reserved. 8-57

Reservoir Characterization
Chapter 8

8-58 © PetroSkills, 2013. All rights reserved.

Chapter 9 R ESERVOIR E NGINEERING
Classical volumetric, material balance and decline curve analysis methods, and high-technology
black oil, compositional and enhanced oil-recovery numerical simulators are used for analyzing
reservoir performance and estimating reserves. The accuracy of the results is dictated by the quality
of the reservoir model used to make reservoir performance analysis.
Satter and Thakur

Assembly of the pieces of the reservoir characterization puzzle often occurs within the
reservoir engineering process. The traditional approach was to assemble the pieces
from the other disciplines and put together a model that explains as much as possible.
Typically, little feedback was given to the other disciplines for some very good reasons:

 The organizational structure tended to minimize interdependence and

communication

 Geoscientists and engineers focus on different aspects of the problem (the

exploration process was typically ending for geoscience experts and just
beginning for the reservoir engineering group - the development phase)

 Poor understanding of the other discipline’s objectives and tools produces

significant communication gaps

 The software used to process the models was based on different platforms (pc
environment for engineers, Unix for geoscientists)

For these reasons and others, a multi-disciplinary approach to developing a reservoir

model was a long time in coming. Needs for better integration and the access to better
modeling programs that are linked more easily have begun to modify the traditional
model. The reservoir engineer can now bring together more data, in both quantity and
detail, from the geoscientist and production engineer.

History matching of the reservoir leads to a feedback of geological information to the

geologist to, in turn, improve the geological model. The reservoir engineer can now
routinely and readily update the model with new data or interpretations and quickly
provide consistent maps and sections for assessing results of activities like infill drilling.
Three-dimensional geological modeling programs can provide maps and cross-sections
in large numbers permitting the engineer to become thoroughly familiar with the geology
prior to designing the simulation model.

Specific responsibilities of the reservoir engineering function related to reservoir

characterization include:

© PetroSkills, 2013. All rights reserved. 9-1

Reservoir Characterization
Chapter 9

 Estimation of the original hydrocarbons in place in the reservoir

 Analysis of reservoir performance

 Prediction of the future performance under the prevailing reservoir conditions (i.e.
estimation of the reserves and recovery factors)

 Estimation of the recoverable reserves and recovery factors under various other
producing methods that are presently known or projected.

 Periodic updating of the model as the quality and quantity of the data improve
during the life of the reservoir

However, such responsibilities can only be carried out successfully in close cooperation
with other disciplines. Ultimately, there is one single reservoir model with its
uncertainties (leading to multiple realisations) for which input of all disciplines is
required. Every time that new data is obtained, the reservoir model needs to be updated
using the input from the different disciplines.

Weber showed that input from all disciplines is required to enable integrated reservoir
studies as displayed in the next diagram:

Nevertheless, although inter-disciplinary sharing is strongly promoted, separate technical

work by individual disciplines is required. Therefore, a process of de-coupling and re-
coupling is also important. The following diagram demonstrates this in some detail:

9-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

FORCES CONTROLLING FLOW

 Gravity (buoyancy)
- segregates fluids

 Capillary forces
- Increase wetting phase saturation
(effect more in tighter rocks)

 Viscous forces
- impedance to flow

© PetroSkills, 2013. All rights reserved. 9-3

Reservoir Characterization
Chapter 9

DRIVE MECHANISMS

Critical data appropriate for these responsibilities include the analysis of the drive
mechanism active in the reservoir. Variations in reservoir pressure, production rates,
gas-oil ratios, water influx and gas cap expansion are determinants in describing the
drive mechanism present. Important contributions in analyzing the drive mechanism are
enhanced by integrating lithologic, geophysical and sequence stratigraphic data. Drive
mechanisms recognized include:

 Water drive

 Gas cap

 Dissolved gas

 Compaction drive

 Gravity drainage

 Gas depletion

 Combinations

The capacity to produce oil from a reservoir is dependent upon the reservoir fluid
pressure available and the use of this pressure in moving fluids through the reservoir
pore system into the well bore. Production of reservoir fluids is dependent upon a
pressure draw down; therefore, the pressure drop that is produced between the static
reservoir pressure and the well bore flowing pressure. The relation between reservoir
draw-down pressure, flow rate, and permeability will be discussed in the next chapter
when individual oil well performance is considered. The use of pressure to move fluids
into the well bore, however, does not explain reservoir performance, and, hence, the
natural energy system used to produce oil during the primary production period.

The natural energy or reservoir drive that is used during primary production operations
can be visualized by considering that each unit volume of oil produced must be replaced
by something in the reservoir since a vacuum cannot exist. In describing the various
drive mechanisms that do exist, therefore, each mechanism will be discussed
concerning the material that is replacing the oil being produced. As a consequence,
three major primary drive mechanisms will be discussed, which are: (1) the solution
gas-drive reservoir; (2) the gas-cap drive reservoir; and (3) the water-drive reservoir.
Additionally, a combination-drive reservoir will be discussed that is, in effect, a
combination of all three primary drive mechanisms. Finally, an additional energy source
called gravity segregation (potential energy) will be discussed although it is not
considered a primary drive mechanism, but a source of additive energy that can be
available under certain circumstances. Each of these drive mechanisms will be

9-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

discussed individually, and it should be remembered that for all practical purposes, the
reservoir is a constant temperature tank.

WATER DRIVE
A water-drive type reservoir is one in which the oil column is associated with a very large
aquifer. The oil column can be either an undersaturated oil or a saturated oil having a gas
cap. For the system to be specifically a water-drive reservoir, however, the gas cap does
not play a part in the energy drive mechanism. For a saturated oil reservoir to be
performing as a true water drive and not a combination drive, therefore, the pressure
would have to be maintained such that the gas cap does not expand. For this to be the
case, it would be necessary for an equal volume of water to encroach and replace the
voidage created by oil production. As can be observed, a constant pressure head exists
in the outcropping aquifer. With maximum access area available for water encroachment
into the oil column (resulting from the bottom-water drive) as oil is produced, water could
readily replace the voidage created. This would result in an optimal reservoir performance
during primary recovery operations and should lead to good primary recovery depending
upon the degree of reservoir homogeneity and oil mobility. It is easy, of course, to
observe the energy in an artesian-type water-drive reservoir since a pressure head exists.
However, for the non-artesian type of associated aquifer the question arises as to why
water would influx into the oil column as oil is produced.

© PetroSkills, 2013. All rights reserved. 9-5

Reservoir Characterization
Chapter 9

The principal reasons for water encroachment during oil production would be due to: (1)
the cumulative result of aquifer water expansion resulting from the initial pressure draw
down and (2) aquifer pore volume compression in response to any pressure reduction.
Of course, both water expansion and pore volume compression in response to each psi
of pressure draw down would be small unless magnified many times. The volumetric
change in water and the volumetric change for rocks would require an associated
aquifer that is very large, hence, humongous in relation to the oil reservoir itself. A third
factor that could play a part in forcing water to encroach into the oil system would be the
evolution of gas from the aquifer water once the pressure in the aquifer is lowered
below the saturation pressure of the water. This condition may not occur, however,
unless the other two mechanisms are not sufficient to maintain pressure in the system.

If the pressure of a strong water-drive reservoir is maintained it indicates that water is

replacing the voidage created by production at an equal rate. For an undersaturated oil
reservoir, the gas-oil ratio would also remain constant and approximately equal to the
initial solution gas-oil ratio. A reasonably constant production rate can generally be
maintained for a period of time, declining as more water is produced and as wells tend
to water out and are shut-in such that fewer wells are producing. All wells will be
produced to the maximum economic limit as controlled by water production.

Ultimate recovery by this type of primary production drive can most commonly be
expected to range between 40 and 60%, although higher recoveries have been
observed. A good rule of thumb for primary recovery efficiency by a water-drive type
reservoir would be 50%. The many factors that could influence the recovery efficiency
will be discussed in later chapters. Since recovery is usually quite high as a result of this
drive mechanism it would be a poor risk when considering enhanced recovery
operations. Since the oil has already been displaced by water, a more exotic and
expensive operation, as compared to waterflooding will have to be initiated.

The quality of performance of a water-drive type reservoir can be significantly influenced

by the rate of water production. It would be possible to produce at such a high rate that
pressure in the reservoir is drawn down considerably or continues to decline because
the water cannot encroach at the same rate as oil is produced. This could be due to
limited access for water to enter the oil column, such as with an edgewater drive or to a
lower capacity of the aquifer to transmit water expansion and pore volume compression
responses within the aquifer pore system.

Water Drive Reservoirs

Characteristics Trend
1. Reservoir pressure Remains high
2. Surface gas-oil ratio Remains low
3. Water production Starts early and increases to appreciable amounts
4. Well behavior Flow until water production gets excessive
5. Expected oil recovery 35 to 75 percent

9-6 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

Idealized water drive performance. (After Torrey, courtesy of Prentice-Hall, Inc.)

Pressure production performance of some water-drive pools. (After Elliott, courtesy of AIME.)

© PetroSkills, 2013. All rights reserved. 9-7

Reservoir Characterization
Chapter 9

GAS CAP DRIVE

Characteristics Trend
1. Reservoir pressure Falls slowly and continuously
2. Surface gas-oil ratio Rises continuously in up-structure wells
3. Water production Absent or negligible
4. Well behavior Long flowing life depending upon size of gas cap
5. Expected oil recovery 20 to 40 percent

The difference in this reservoir as compared to the solution-gas drive-type reservoir is

that in this case the oil column is a saturated oil in equilibrium with a free gas phase
existing as a gas-cap located structurally high within the reservoir. It is the presence of

9-8 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

this free gas volume that exists initially in the reservoir at initial reservoir pressure and
temperature conditions that substantially alters the performance behavior of this system
during the primary producing life of this reservoir. Wells are drilled and completed in the
oil column, and, as oil is produced and conceptual voidage created, the contained
reservoir fluids must expand to fill that voidage. This will be, as one might expect,
associated with a reservoir pressure decline. The dominant fluid expansion, however, is
associated with the free gas present in the gas cap. In fact, since the reservoir oil is a
saturated oil, the remaining oil volume will shrink with production, and gas will evolve
within the reservoir from this remaining oil volume. The voidage conceptually created by
production is, therefore, replaced by gas-cap expansion and solution-gas evolution
within the oil column. If the gas cap is of any consequential size, the dominant influence
in voidage replacement would be due to the gas-cap expansion in effect overshadowing
the influence of solution gas in the replacement of the voidage created by oil production.
It should also be pointed out that if any associated water exists, it is a very limited
amount and not a factor in this producing mechanism nor would be the influence of pore
volume compression.

The expected performance behavior for a gas-cap drive reservoir is considerably

different from that encountered with a solution-gas drive reservoir. As production
occurs, reservoir pressure will decline such that the remaining fluids within the total
system can expand to fill the conceptual voidage created by production. With a sizeable
volume of free gas already existing within the reservoir, the decline in reservoir pressure
is much more moderate than the decline observed initially with regards to the
solution-gas drive-type reservoir. In fact, the major loss of pressure associated with this
primary producing system will be in the later stages of the primary production operation
when depletion of the gas-cap gas occurs. Compared with this cushioning of pressure
decline due to the presence of the free gas-cap gas volume is the more stable early
production rate of oil that may remain reasonably constant for a period of time during
the early production period. At some point in time, however, this daily oil production rate
will start to diminish gradually as reservoir pressure continues to drop, the mobility of
the oil in the oil column continues to decline (associated with diminishing effective
permeability of the oil as the solution-gas saturation slowly increases), and viscosity of
the oil slowly increases. Additionally, of course, as the gas cap expands into the oil
column, depending upon the reservoir configuration and well locations, producing wells
will be shut-in or converted into gas injectors for pressure maintenance. Fewer wells will
be producing oil toward the later stages of the primary producing period, thereby also
reducing the capacity to produce oil. Therefore, the daily oil production rate can be
expected to decline. Finally, and quite differently, the gas-oil ratio behavior, when
compared to the gas-oil ratio behavior of the solution-gas drive system, can be seen to
increase continuously throughout the life of the primary producing period. This is due to
the domination of the produced gas-to-oil ratio by the ever increasing production of free
gas-cap gas throughout the life of this primary producing period. Production of gas-cap
gas will overshadow the influence of solution gas being evolved in the oil column. In the
later stages of the primary producing life of the reservoir, the gas-oil ratio will start to
rise quite rapidly as the gas cap is being blown down. It is during this blow down of the
gas cap that the reservoir pressure will start to diminish quite rapidly.

© PetroSkills, 2013. All rights reserved. 9-9

Reservoir Characterization
Chapter 9

In comparison to the performance behavior of the solution-gas drive reservoir, the

gas-cap drive reservoir system behaves differently, and a sizeable difference in
performance behavior relative to ultimate recovery is also observed. The recovery
efficiency of a gas-cap drive reservoir can be expected to fall between 20 to 40% of the
initial oil in place, a sizeable improvement when compared to the solution-gas drive
system. A good rule of thumb for the ultimate recovery to be expected is 30% of the
initial oil in place and could be higher depending upon numerous factors to be
discussed later. Of course, ultimate recovery by primary operations of from 20 to 40% of
the initial oil in place also means that this drive system is inherently inefficient with 60 to
80% of the initial oil in place remaining within the reservoir once the system is pressure
depleted. As a result, this reservoir is also a good candidate for post-primary injection
operations and will probably be waterflooded.

Idealized performance of a gas-cap-drive reservoir. (After Torrey, courtesy of Prentice-Hall, Inc.)

9-10 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

Typical gas-cap-drive performance, Goldsmith San Andres Dolomite Pool in West Texas. (After Craze, courtesy of
AIME.)

DISSOLVED GAS DRIVE

© PetroSkills, 2013. All rights reserved. 9-11

Reservoir Characterization
Chapter 9

Characteristics Trend
1. Reservoir pressure Declines rapidly and continuously
2. Surface gas-oil ratio First low, then rises to maximum and then drops
3. Water production None
4. Well behavior Requires pumping at early stage
5. Expected oil recovery 5 to 30 percent of original oil in place

The solution-gas drive mechanism can be visualized as an enclosed volumetric tank.

This tank contains an undersaturated oil and possibly a small volume of associated
water, which is not a factor in this drive mechanism. Wells are drilled and completed in
the oil column. One can visualize that for each barrel of oil removed something must
replace the void space created. Since this system is a closed tank, the only possible
source for oil replacement is that all the remaining oil in the tank expands an amount
equal to the volume of oil removed (if the associated aquifer is not a factor due to its
limited size and the pore volume remains constant). For this conceptual void space
resulting from production to be filled, therefore, all the remaining oil in the reservoir must
expand an equivalent amount, and, for this to occur, reservoir pressure must decline
proportionately. As long as reservoir pressure remains above the bubble point, the
energy mechanism is a liquid-oil expansion drive. Of course, since liquid oil does not
expand much per unit of pressure decline, it can be visualized that reservoir pressure
must decline rapidly. This pressure behavior shows the idealized performance behavior
for the solution-gas drive reservoir. Besides this pressure behavior, the typical
performance for the daily oil-production rate and the producing gas-oil ratio is also
shown. As can be seen, the initial daily oil-production rate is high, and this production
period where the pressure remains above the bubble point is referred to as flush
production. Additionally, the producing gas-oil ratio behavior reflects a producing gas-oil
ratio essentially equivalent to the initial solution gas-oil ratio. This gas-oil ratio behavior
would be expected since all the oil being moved into the well bore contains all its initial
solution gas, which is subsequently evolved in the production process and measured at
the surface as the producing gas-to-oil ratio.

Once the bubble-point pressure is reached in the reservoir, however, the performance
characteristics change drastically. The voidage created by production must still be filled
by expansion of the remaining reservoir fluid, but now the associated pressure drop is
below the bubble point and the oil will shrink. Of course, associated with this shrinkage
will be evolution of solution gas within the reservoir, which readily expands with
pressure reduction and readily fills the voidage created by both production and
shrinkage of the oil. The evolution of solution gas in the reservoir strongly influences the
performance characteristics. It can be noted that rapid initial decline in pressure to the
bubble point is somewhat arrested due to the expansion capacity of the gas, which

9-12 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

tends to support the reservoir pressure and cushion its decline. One would also expect
that the producing gas-oil ratio would reflect this gas evolution in the reservoir with a
reflected decline in the producing gas-oil ratio since not all the initial gas in solution is
retained in the oil entering the well bore. This lower gas-oil ratio will exist as long as the
gas being evolved in the reservoir is below the critical saturation that is the saturation of
gas required for gas to flow. Once this reservoir gas saturation builds up to the point
where gas starts to flow as a free gas phase into the well bore, a rapid rise in the
producing gas-oil ratio will be observed. This rising gas-oil ratio will continue over time
as the gas saturation in the reservoir continues to increase. It will not continue to rise
throughout the primary production period, however, as a function of three combined
influences. These influences are: (1) the reservoir pressure is declining, and, hence, the
capacity of the gas to expand will decline; (2) the oil production rate will become more
stabilized later in the primary producing life of the reservoir; and (3) the oil that is being
produced later in time contains less gas in solution. This gas-oil ratio behavior becomes
a fingerprint for a solution-gas drive reservoir. Finally, the oil production behavior is also
quite sizably altered once the bubble- point pressure in the reservoir is reached. The oil
production rate will undergo a rapid decline caused by the loss of mobility resulting from
the combined loss of effective oil permeability and the increase in oil viscosity. This
rapid loss of the capacity to produce oil should be expected as a result of the loss in
mobility and pressure decline. In time, the rate of loss will diminish as the production
history goes through a transitional period and ultimately reaches a somewhat stabilized
but low oil producing rate. If this stabilized oil production rate is still economic, it may be
projected to continue for many years.

This drive mechanism, however, is not efficient with expected primary performance to
result in recoveries ranging from as low as 5% to 20%. A good rule of thumb to apply as
a generalized recovery factor would be about 15% of the initial oil in place recovered
during primary production operations. Higher recovery efficiencies might be
encountered if certain conditions exist, as will be discussed in later chapters. One mode
of improved primary recovery performance would be that of pressure maintenance,
hence the injection of water and/or gas into the reservoir to maintain pressure. Pressure
maintenance would keep gas in solution, thereby optimizing oil mobility as well as
maintaining pressure and ultimately improving primary recovery. The concept of
pressure maintenance will be elaborated upon in a later chapter. It should also be
obvious that with such low expected primary recovery efficiency that this type of
reservoir is a prime candidate for post-primary injection operations and in particular, a
waterflood, since it is the most economic of the post-primary injection operations. These
operations will be discussed in a later chapter.

© PetroSkills, 2013. All rights reserved. 9-13

Reservoir Characterization
Chapter 9

Generalized performance of a solution gas-drive pool. (After Torrey, of Prentice-Hall, Inc.)

Performance of the Slaughter San Andres Limestone Pool, West Texas, under predominantly solution gas drive
(After Sessions, courtesy of AIMT).

9-14 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

COMBINATION DRIVE

© PetroSkills, 2013. All rights reserved. 9-15

Reservoir Characterization
Chapter 9

RESERVOIR PRESSURE TRENDS FOR RESERVOIR UNDER VARIOUS DRIVES

RESERVOIR GAS-OIL RATIO TRENDS FOR RESERVOIR UNDER VARIOUS DRIVES

9-16 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

COMPACTION DRIVE
When the reservoir pressure is declining during production, the effective stress (being
the difference between the overburden stress and the pore pressure) is increasing. This
effective stress is acting on the rock frame as such decreasing the pore space. This
phenomenon is called compaction. Effectively, the hydrocarbons are squeezed out of
the pore space, which is called compaction drive.

GRAVITY DRAINAGE

Gravity segregation, or as it is sometimes called, gravity drainage, is not considered as

a primary drive mechanism but is a form of energy (potential energy) that can sizably
improve the performance behavior of all pressure declining reservoirs under certain
circumstances. Gravity segregation occurs when gas evolved from solution in the oil
migrates up structure with subsequent drainage of oil down structure. This redistribution
of fluids in the reservoir can play a major role in improving performance but will only be
effective if there is considerable structure within the reservoir system and if the oil is
mobile. The movement of gas up structure can effectively add to the performance of an
existing gas cap or may create a secondary gas cap if one already does not exist. To
optimize the influence of gravity segregation, it would be desirable to produce the wells
under low draw down pressures to minimize gas production and optimize gravity
segregation. The importance of gravity segregation will be discussed in a later chapter.

© PetroSkills, 2013. All rights reserved. 9-17

Reservoir Characterization
Chapter 9

GAS RESERVOIRS

Gas reservoir fluid systems were described in Chapter 4 and consist of the gas
reservoir (whether defined as a wet or a dry-gas system) and the retrograde gas
condensate reservoir. These reservoirs are the non-associated type, do not exist in
equilibrium with an oil column and, therefore, are not gas-cap systems. The
non-associated gas reservoirs, however, may coexist with an aquifer that can provide
some degree of water drive.

When producing a gas reservoir, the reservoir engineer's goal is to obtain maximum
recovery just as it is when producing an oil reservoir. This optimal recovery includes
maximum recovery of both gas and condensable liquid hydrocarbons. For the reservoir
engineer, this presents several problems: (1) the loss of hydrocarbon liquid in the
reservoir when producing a retrograde gas condensate reservoir by pressure depletion,
and (2) low recovery efficiencies (particularly sweep and conformance efficiencies)
when producing, in particular, a retrograde gas condensate reservoir by water drive.

When producing a retrograde gas condensate reservoir by pressure depletion, as

reservoir pressure declines to the upper dew point pressure, liquid will start to condense
in the reservoir at reservoir temperature conditions. A retrograde gas system is usually
rich in intermediate and heavier components and most commonly encountered in
reservoirs deeper than 5,000 ft where pressures exceed 2,000 psi. Pressure depleting a
volumetric, retrograde gas condensate reservoir (no water drive) can result in the loss of
sizable liquid condensate volumes within the reservoir. This liquid dropout besides
creating production problems due to gas mobility loss would also, for the most part, be
unrecovered and remain as a residual liquid condensate in the reservoir once the
system is depleted and abandoned.

There are several methods available for obviating this liquid loss in a
pressure-depleting, volumetric-type reservoir, which are: (1) gas cycling, and (2) water
injection pressure maintenance. The most desirable of these would be gas cycling since
it will usually result in higher recovery efficiency when compared to water injection
pressure maintenance. This process shows how the retrograde gas condensate fluid is
produced from the reservoir and replaced with a dry gas. The original reservoir
retrograde phase envelope is one in which, when pressure depleting Reservoir A at
constant reservoir temperature, the dew point would be crossed and liquid condensate
can be expected to occur within the reservoir. By producing this system while it is at
high reservoir pressure conditions, stripping the condensable liquids in the surface
processing system, and then re-injecting the dry gas the problem would be eliminated.
The re-injected dry gas miscibly displaces the retrograde gas. This results in a reservoir
gas system that can be pressure depleted, with no liquid condensation occurring in the
reservoir, once cycling is completed. This is because the phase envelope for the dry
gas system has a cricondentherm less than the reservoir temperature.

The injection of water to maintain reservoir pressure above the dew point pressure for
this volumetric system having no water drive would lead to lower recoveries due to the
inevitable bypassing and entrapment of pockets of gas as the system is depleted. Initial

9-18 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

cycling of a water-drive type, retrograde gas reservoir followed by water-drive

production of the dry-gas system is preferable to straight water-drive production of the
retrograde-gas system. Cycling followed by dry-gas production is preferred since
volumes of dry gas instead of rich gas would be entrapped by the encroaching water.

Phase diagram – dry gas

© PetroSkills, 2013. All rights reserved. 9-19

Reservoir Characterization
Chapter 9

Phase diagram – wet gas

Phase diagram – retrograde gas condensate

9-20 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

DYNAMIC PERFORMANCE OF GAS WELLS

Some equations:

PV  nzRT

z factor introduced to compensate for deviations from ideal gas law. z is a function of
pressure and temperature.

G p  GIIP  GIP

gas produced equals gas initially in place minus gas remaining in place.

These two equations lead to:

P Pi  Gp 
 1  
z zi  GIIP 

This equation is represented in the following graph by the red line.

P/Z WITH TIME

© PetroSkills, 2013. All rights reserved. 9-21

Reservoir Characterization
Chapter 9

FIELD EXAMPLE

RESERVOIR FLUID TYPES

9-22 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

SECONDARY RECOVERY MECHANISMS

If the primary reservoir energy is insufficient, then energy may be added to the reservoir
by injection of water or gas into the reservoir.

When injecting gas usually hydrocarbon gas is being utilised. However, particularly in
retrograde gas condensate reservoirs, also nitrogen injection is useful, When injecting
dry gas or nitrogen into the reservoir, the composition of the hydrocarbon is shifting to
lighter components. As a result, when at the end of the producing life the reservoir is
blown down, liquid saturation in the reservoir will be negligible or at least limited to a
minimum.

RC TEAM INPUT INTO A WATERFLOOD

 Observation:
Many potential candidate reservoirs for waterflooding are poorly defined even in
mature stages of field development.
 For efficient waterflooding it is essential to have a good reservoir model, of
course with its uncertainties.
 When defining waterflood projects it is essential to look for critical elements.
Some of these critical elements are compartimentalisation, fault sealing, flow
units and also the economics of a project.

WATERFLOOD PATTERNS

© PetroSkills, 2013. All rights reserved. 9-23

Reservoir Characterization
Chapter 9

WATERFLOODING: GEOLOGICAL PERSPECTIVES OF RESERVOIR

ENGINEERING
BOTTOM LINE
Since the future of a waterflood is based on the extrapolated price of oil, the economic
analysis must accurately predict price trends if the project’s economic goals are to be met.

PROBLEM ADDRESSED
Prior to the onset of a waterflood, there must be an accurate evaluation and prediction of reservoir
condition. This includes an analysis of the architecture, structure, presence and orientation of
fractures, as well as reservoir limits, rock, and fluid property and pressure distributions.

TECHNOLOGY OVERVIEW
Since waterfloods are typically implemented in mature developed properties, it might be assumed
that the reservoir is well defined, with few potential surprises. Experience indicates otherwise.
However, operators who are aware of common pitfalls can avoid mistakes that will reduce recovery
and profitability when implementing waterfloods.

The growth of a waterflood prospect, from conception to abandonment, is divided into two stages.
The first stage, which is the topic of this workshop, deals with the initial analysis up to the
development of the reservoir model. It is especially critical that engineers and geologists work
together during this initial conceptual stage. The second stage, which is more engineering- oriented,
deals with development of the reservoir model, implementation, and operation until project
abandonment.

Typically, an operator’s first step in evaluating a waterflood candidate should be to assemble and
evaluate preliminary reservoir data. Information about the gas:oil ratio (GOR) and water production
provide key information about whether a project is an attractive candidate. Operators should examine
GOR data, both on the basis of the well’s initial potential and cumulative production basis. Oil
reservoirs exhibiting cumulative GORs that are less than 3000/1 are generally attractive candidates.

Wells with a GOR less than 3000/1 are generally attractive candidates. Gas reservoirs do not make
favorable waterflood candidates because of the lack of sufficient oil to bank. The operator should,
however, be aware that a well’s high GOR can be caused by comingling of fluids from different
stratigraphic layers. Detailed stratigraphic correlation should be able to distinguish the types of fluids
coming from various zones.

Operators should take special care looking for indications of a gas-oil contact or gas cap. If present, a
gas cap will distort GOR data and its presence must be considered when designing a waterflood. It
also is critical to note water-oil contacts, calculate fluid saturations, and identify sources of water
production. In analyzing gas and water data, operators should be aware that they are not always
accurately reported, so considerable judgment may be required.

A two-step analysis of reservoir data is used to evaluate a waterflood candidate. First, the operator
reviews initial production potential, gas-to-oil ratios (GORs), amount and type of water production,
pressure levels, and core information. If this information meets the requirements, a second, more
thorough review of data of the reservoir’s fluids pressure, volume, and temperature (PVT) properties
should be initiated in cooperation with the company’s engineering department. Ii also is important to
understand the source of formation water within the reservoir and calculate its saturation. Formation
water production can be caused by many processes, such as water saturation above irreducible water
saturation, water entering the wellbore accidentally from another reservoir (e.g. hole in casing), water

9-24 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

introduced mechanically (e.g. an injection well), or natural encroachment due to migration of the oil-
water contact.

Production history should be analyzed, too. An analysis of primary recovery will determine not only
if sufficient oil saturation remains for successful water-flooding but also if there are potential
problems.

Determining net pay for a waterflood is important because the calculations affect original -oil-in-
place (OOIP) values, reserve estimates, economics, and performance characteristics. Properties that
can have an impact on net pay include porosity-permeability correlations, average porosity and
thickness, impermeable parts of the reservoir, and fluid contacts. Mapping net pay may be the
primary contribution of geologists to evaluating waterflood candidates.

Although it’s often the least-evaluated aspect of a waterflood project, the reservoir’s structure can
have major influences, such as those caused by faulting (and fracture); degree of dip; and
determination of overall homoclinal, anticlinal, or synclinal nature of the reservoir.

The proper techniques for core analysis begin even before drilling and continue through laboratory
examination. Cores can be used to significantly increase data available about lithology, depositional
environment, diagenesis, and rock type. Core samples are required to define pore geometry changes
with depth, areal distribution, permeability, and porosity.

But fractures can be a positive benefit for wells under primary production because of increased
permeability accessibility. This holds true whether the fractures are
natural or induced, and whether they are produced from stress caused by local, regional or basinal
factors.

Fractures also can be a detriment to hydrocarbon production because they may connect to downdip
water. Because fracture permeability is often several orders of magnitude higher than the matrix
permeability, injected water will generally travel through fractures. This can be disastrous for a
waterflood if water is injected into a fracture that is in communication with a producing well.

Economic Implications:
Economic analysis, with all its incorporated data, predicts the success of the waterflood. As a result,
neither the geologist nor the engineer alone is responsible for the prediction. A realistic price
structure for the economic evaluation is often the deciding factor for economic success of
waterfloods.

Long-range price trends are important, since the future value of a waterflood project is based on an
extrapolated price of oil. Because the future price of oil is unknown, project economic analysis
should incorporate a conservative price structure. Further, prices must be adjusted, taking inflation
into account, as part of the economic analysis.

Based on a workshop sponsored by P7TC’s South Midcontinent Region and the Oklahoma Geological Survey on
June 10-11, 1998, in Oklahoma City, OK

ENHANCED OIL RECOVERY

In order to improve the performance of a reservoir and to maximize its ultimate

recovery, fluids that are normally not present in the reservoir may be injected into the

© PetroSkills, 2013. All rights reserved. 9-25

Reservoir Characterization
Chapter 9

reservoir. Such Enhanced Oil Recovery schemes are aiming at increasing the mobility,
i.e. the ratio of permeability over viscosity, of the oil in the reservoir.

The main EOR schemes are:

 Steam injection
By injecting steam the temperature in the reservoir is being increased.
That means that the viscosity of the oil is being decreased and
consequently that the oil mobility is being increased.

Two approaches are distinguished, namely steam flooding when steam is

injected in a injector well and oil is produced in another well, and steam
soak when in an alternating way steam is injected in a well and
subsequently oil is produced from that same well.

 Polymer flooding
It often is the case in a water drive that the water mobility is greater than
the oil mobility (unfavourable mobility ratio). In this situation water will
finger through the oil and eventually bypass the oil leaving remaining oil
behind in the reservoir.

To improve the recovery in this situation, a slug of polymer is injected

before water injection is started. The mobility of the polymer is less that
the mobility of the oil. Hence, the polymer will drive the oil through the
reservoir as a piston.

In this situation will the water mobilty clearly be greater than the polymer
mobility. Hence, the water will finger through the polymer and eventually
bypass the polymer. The polymer slug has, however, been designed such
that breakthrough of the water through the polymer will only occur when
the head of the polymer slug has arrived at the production well.

 Surfactant flooding
Introduction of surfactants in the reservoir will reduce the interfacial
tension between oil and water. In this way the relative permeability of the
oil with respect to water will increase resulting in a higher oil mobility.

 CO2 flooding
CO2 is injected into a reservoir at residual oil saturation. Two effects occur,
namely the saturation of the oil containing the CO2 increases above
residual saturation and becomes mobile and secondly, the viscosity of the
mixture is reduced. Both effect contribute to improved oil recovery.

 In-situ combustion
When oxygen is being injected into the reservoir, combustion of the oil
around the injector well may occur. Such in-situ combustion increases the
temperature (hence reducing oil viscosity) and the pressure (hence adding

9-26 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

drive energy in the reservoir).

 Microbial EOR
Microbes are injected in the formation together with nutrients such as
sugars, phosphates or nitrates. The microbes ferment the hydrocarbons
and also produce surfactants and CO2 which lead to processes as
described above.

UNCONVENTIONAL RESOURCES

 Gas
- shale gas
- tight sandstones
- coal bed methane
- coal gassification
- deep HP gas
- methane hydrates

 Oil
- tar sands
- oil shales
- bio-fuels
- GtL, Gas to Liquids

Different sources amongst which JPT (July 2011)

© PetroSkills, 2013. All rights reserved. 9-27

Reservoir Characterization
Chapter 9

VOLUMES OF HYDROCARBONS IN PLACE

Volumes of hydrocarbons in place commonly are derived through techniques that are
based on:

 Volumetric calculations

 Decline curves

 Material balance

 Mathematical simulation

Each of these is well known, and details of their implementation are outside the scope
of the course. It is worth pointing out, however, that the geostatistical techniques
described in a preceding section can be used to provide multiple realizations
(simulations) based on the expected ranges of components; these realizations can
provide the basis for a Monte Carlo type estimation of reservoir volumes (see section on
economics and risking).

One of the most important functions of the reservoir engineer is to determine the reservoir
oil and gas in place and predict the recovery anticipated under the prevailing reservoir
mechanism. This determination of the initial hydrocarbon-in-place (oil and/or gas) and the
projected recoverable reserves are essential for economic planning. These evaluations, of
course, are a continuing process throughout the life of the reservoir. Hydrocarbon volumes
in place and reserve estimates are made initially to optimally develop the reservoir system.
Continued monitoring of initial hydrocarbons-in-place and reserve predictions are necessary
to confirm earlier estimates and for long-term planning pertaining to maximizing oil and/or
gas recoveries.

The initial reservoir volume of hydrocarbon-in-place is always determined using the

available geophysical and geological data. This volume is generated from the best quality
maps that can be produced relative to the reservoir bulk, or hydrocarbon saturated, pore
volume. The limitations of this volumetric estimate can, of course, be visualized concerning
the difficulty in defining reservoir limits and the estimate of net pay variations within the
apparent reservoir volume.

To accomplish these reservoir evaluations, the engineer requires an array of data falling in
four broad categories. These categories include: (1) reservoir rock properties, (2) reservoir
fluid properties, (3) a reservoir pressure history, and (4) the reservoir production history.
Previous chapters have discussed the acquisition and to some degree the application of this
data. This chapter will further emphasize the need for accurate and representative data as
required in reservoir evaluation.

The material balance is a tool the reservoir engineer can use to predict initial hydrocarbons-
in-place using production history or reserves using initial hydrocarbon-in-place. It can be
expressed as a balance between the voidages created by production being equal to the

9-28 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

change in volume of all the reservoir fluids, plus the volume of all the extraneous fluids
which enter the reservoir. As such, it can be visualized as the reservoir engineer's
accounting system.

The material balance is a tool the reservoir engineer has used to simulate reservoir
performance. The problem with the material balance approach is, however, two fold. First
allowances cannot be made for the variation of rock and fluid properties with location in the
reservoir and secondly, the dynamics of fluid movement within the system are overlooked.9
Material balance applications are, of course, still used since they are technically correct and
economically justifiable.

In order to overcome these limitations analog models were used. This approach, however,
also has many drawbacks some of which were that they were expensive, difficult to make
and operate and lacked adaptability.

Introduction of computer evaluation of numerical models replaced analog models and is

now extensively used to analyze reservoir behavior. The complex partial differential
equations governing fluid flow behavior in porous media are solved numerically by
computer. The process involves dividing the reservoir into discrete blocks and
performing mass and energy balances on all the blocks simultaneously. The gridding
process enables a more realistic representation of rock and fluid properties which can
vary in any manner. The stages of reservoir simulation are: (1) development of the
geologic model, (2) reservoir properties upscaled to produce the static reservoir model,
(3) optimizing the grid and calibrating with dynamic data to obtain the simulation model.

The recent application of 4-D seismic for reservoir monitoring has greatly enhanced
reservoir management and will enable even greater recovery to be achieved. 4-D
seismic monitoring is the use of repeated 3-D seismic surveys obtained over time.
Differences between 3-D seismic surveys can be evaluated to reflect changes in fluid
distribution over time. To do this competently, however, all factors affecting the seismic
data must be considered. When monitoring fluid movement in a reservoir, therefore, it is
necessary to integrate the engineering process and geologic model with the seismic
survey. 4-D seismic provides an insight into the interwell area that cannot be obtained
from any other source. This insight enables the engineer to "see" the reservoir as it
responds to production operations. Reservoir drainage patterns can be observed,
bypassed oil and gas defined, fluid influx more optimally controlled, and additional wells
(or laterals) drilled to optimize recovery.

RESERVOIR MODELING
Additionally, characterization of the reservoir as individual flow unit or - most likely - a
series of interrelated flow units (zones within the reservoir with similar fluid flow) is a
key step in understanding the reservoir and developing simulations. Flow units are best
described with a thorough understanding of the geologic model developed through
collaboration by the reservoir characterization team.

© PetroSkills, 2013. All rights reserved. 9-29

Reservoir Characterization
Chapter 9

THE INTEGRATED MODEL

Clearly, modeling reservoirs in this detail is an integrated effort. The contributing pieces
(the geological model, the petrophysical model, the dynamic model and also the
economic model) are so interwoven that an continuous discussion between
geoscientists from different disciplines is essential. Such discussion also guarantees
that the integrated model is being developed in an iterative way every time taking into
account new data and new insights.

RESERVES

Future quantities of petroleum expected to be recovered from naturally occurring

underground accumulations

 Attempts to standardize reserves terminology began in the mid 1930's

 Working separately, the Society of Petroleum Engineers (SPE) and the World
Petroleum Congresses (WPC) produced definitions for known accumulations –
adopted in 1996

 Preferred standards across the industry.

9-30 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

CONCEPTS
 “Proved reserves should be based on current economic conditions, including all
factors affecting the viability of the projects …. the term is general and not
restricted to costs and price only”

 “Probable and Possible reserves could be based on anticipated developments

and/or the extrapolation of current economic conditions”

Reserves are those quantities of petroleum which are anticipated to be commercially

recovered from known accumulations from a given date forward.

All reserve estimates involve some degree of uncertainty

 Uncertainty depends chiefly on the amount of reliable data available at the time
of the estimate and interpretation of these data

 Relative degree of uncertainty may be conveyed by placing reserves into either

proved or unproved

 Unproved reserves are less certain to be recovered than proved reserves

ESTIMATION OF RESERVES
 Method of estimation is called:
 Deterministic if a single best estimate of reserves is made based on known
geological, engineering, and economic data

 Probabilistic when the known geological, engineering, and economic data are
used to generate a range of estimates and their associated probabilities

VOLUMETRIC DETERMINATION
NR = GBV N/GF(1-Sw)

Where:
NR = hydrocarbons in place at reservoir conditions
GBV = Gross Bulk Volume of reservoir
N/G = Net to Gross ratio
F = Porosity, fraction
Sw = Water saturation, fraction

Conversion to surface volume - oil

Shrinkage factor (1/FVF – formation volume factor)

© PetroSkills, 2013. All rights reserved. 9-31

Reservoir Characterization
Chapter 9

MORE ON RESERVES
 Estimates will generally be revised as additional data becomes available or
economic conditions change

 Reserves are attributed to either natural energy or improved recovery methods

– Improved recovery methods -all methods - to increase ultimate recovery
– Examples: pressure maintenance, cycling, waterflooding, thermal methods,
chemical flooding, and the use of miscible and immiscible displacement fluids

PROVED RESERVES
 “Those quantities of petroleum which, by analysis of geological and engineering
data, can be estimated with reasonable certainty to be commercially recoverable,
from a given date forward, from known reservoirs and under current economic
conditions, operating methods, and government regulations”

 Proved reserves can be categorized as developed or undeveloped

 If deterministic methods high degree of confidence that the quantities will be

recovered.

 If probabilistic methods - at least a 90% probability that the quantities actually

recovered will equal or exceed the estimate

 In general, reserves considered proved if supported by actual production or

formation tests

 In certain cases, proved reserves may be assigned on the basis of well logs
and/or core analysis analogous to other proved reservoirs

 Area considered as proved includes:

– area delineated by drilling and defined by fluid contacts
– undrilled portions of the reservoir that can reasonably be judged as
commercially productive on the basis of available geological and engineering

 In the absence of data on fluid contacts, the lowest known occurrence of

hydrocarbons controls the proved limit unless otherwise indicated by definitive
geological, engineering or performance data

UNPROVED RESERVES
 Similar to proved reserves, but
– Technical, contractual, economic, or regulatory uncertainties preclude such
reserves being classified as proved

9-32 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

 May be further classified as probable reserves and possible reserves.

 May be estimated assuming future economic conditions

– Expressed by allocating appropriate quantities of reserves to the probable
and possible classifications.

PROBABLE RESERVES
“Those unproved reserves which analysis of geological and engineering data suggests
are more likely than not to be recoverable”

 Probabilistic methods – at least 50% probability that the quantities actually

recovered will equal or exceed the sum of estimated proved plus probable
reserves

Types
 Formations appear to be productive based on well log characteristics - lack
definitive data/tests

 Reserves anticipated to be proved by normal step-out drilling

 Established improved recovery methods

 Area of the formation separated from the proved area and structurally higher than
the proved area

POSSIBLE RESERVES
“Those unproved reserves which analysis of geological and engineering data suggests
are less likely to be recoverable than probable reserves”

 When probabilistic methods are used, there should be at least 10% probability
that the quantities actually recovered will equal or exceed the sum of estimated
proved plus probable plus possible reserves

 Reserves which, based on geological interpretations, could possibly exist beyond

areas classified as probable

 Reserves in formations that appear to be petroleum bearing based on log and

core analysis but may not be productive at commercial rates

 Incremental reserves attributed to infill drilling that are subject to technical

uncertainty

© PetroSkills, 2013. All rights reserved. 9-33

Reservoir Characterization
Chapter 9

 Reserves in an area of the formation that appears to be separated from the

proved area by faulting and geological interpretation indicates the subject area is
structurally lower than the proved area

RESERVES SUMMARY
 Proven reserves, 0 to P90
(developed or undeveloped

 Probable reserves, P50 to P90

 Possible reserves, greater than P50

RESOURCE CLASSIFICATION SYSTEM

9-34 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

FLOW UNITS

Log-based stratigraphic cross-section of the F37 reservoir.

FLOW UNITS - DEFINED

 A flow unit is a continuous volume of the total reservoir within which
geological and petrophysical properties that affect the fluid flow, are more or
less similar and different from those in other flow units.

 Particular properties of interest are permeability and capillarity

© PetroSkills, 2013. All rights reserved. 9-35

Reservoir Characterization
Chapter 9

Major types of geological and petrophysical data applied to flow unit zonation of a well.

As seen in this Figure, the flow units are delineated on the basis of permeability
contrasts due to lithofacies changes and to the presence of a laterally continuous barrier
to vertical permeability (flow unit 4). According to the definition of flow unit, it is
permissible to define a flow unit that exhibits only weak flow or no flow through it. This
property of flow units makes it possible to use a single numbering system for identifying
both obvious flow units and probable permeability barriers that can be mapped at the
same scale as reservoir quality flow units.

9-36 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

Some examples of lithofacies and flow unit subdivisions of clastic and carbonate reservoirs, (a) Lithofacies and (b)
flow unit subdivision of the Shannon Sandstone body in the Hartzog Draw field, Powder River basin, Wyoming.
Modified from Hearn et al., 1984.) (c) Lithofacies and (d) reservoir facies (flow unit) subdivision of the Rainbow Lake
reef reservoir (“A” Pool), Alberta, Canada. (Modified from Langston and Chin.)

FLOW UNITS - RIGOROUS

 Hydraulic quality of rock controlled by:
– Pore geometry
– Mineralogy (abundance, type, morphology, location relative to pore throats)
– Texture (grain size, shape, sorting, packing)
 Mean hydraulic radius concept
 Calculate RQI (reservoir quality Index)
 Reference:SPE # 26436, Amaefule, et al.

RQI (micro m) = 0,0314 sqrt(K (mD) / por (%))

Successful reservoir simulation models incorporate geologic features that have a
significant effect on reservoir performance. A model should be developed for each
distinct geologic flow unit-defined area. Future reservoir performance will indicate
whether the model is the appropriate or needs modification.

© PetroSkills, 2013. All rights reserved. 9-37

Reservoir Characterization
Chapter 9

HETEROGENEITY

Reservoir continuity is an important consideration in developing an effective model for

simulation. The influences of reservoir heterogeneity include:

 Baffles to flow
 Permeability heterogeneity
 Contrasts in vertical and lateral permeability

Weber presents an interesting classification of reservoir heterogeneity types as shown

below:

9-38 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

IMPEDIMENTS TO FLOW
 Barrier
 Baffle
 Hi K contrast
 Low K contrast

HETEROGENEITY CAUSES
 Causes:
– Depositional and stratigraphic controls
– Diagenesis
– Pore system connectivity – diagenetic or depositional
– Intersecting structural and stratigraphic patterns
– Hydrodynamics and overpressure

 Techniques to recognize (integrated studies):

– Sedimentologic – cores, diagenetic analysis, pore system study
– Wireline logs, DST, production tests
– Geochemical analyses

HETEROGENEITY – FIELD RESULTS

 Geochemical – reservoir fluid continuity
– Whole gas and oil chromatographic data in GOM

 Detailed study of faults and clay smearing in Denver Basin (K) – Shale Gouge
Ratio (SGR) used to calculate connectivity
– Supporting data – BHP bombs, pressure transient tests, GOR, production
history

 Diagenetic and oil saturation differences in Red River in Williston basin identified
two trends
– (NE-SW pinchout of porosity and NW plunging nose)
– best wells at intersection of trends

 Detailed structure and stratigraphy in Green River Basin (K), Wyoming

– Abnormally pressured gas condensate in deep basin
– Detailed 3D and well-bore data identified complex structural trap within thin
alluvial fan SS’s separated by silty mudstones

 Compartmentalization noted in cores, gas injection discrepancies and anomalous

oil production in K in Wind River, Wyoming
– Detailed stratigraphy successive valley-fill sequences
– Up to 12 flow-separated SS’s
– Not recognizable in well logs
– Shales separate reservoirs that form baffles

© PetroSkills, 2013. All rights reserved. 9-39

Reservoir Characterization
Chapter 9

Reservoir DATA FOR DEFINITION

Type Key Heterogeneities Seismic Wells Production Analogue
Layercake High continuity, Sequence Cores, logs Vertical pulse Shale length
wedge outs, stratigraphy (facies, K), test, long tests, stats,
permeability RFT/MDT PLT depositional
models
Jigsaw Thief zones, flow 3D seismic, Cores, borehole Long tests, Sand body
unit orientation, seq. Strat. imaging, logs PLT, horizontal char acts.,
K contrasts, local (facies, k), tests depositional
baffles RFT/MDT models, shale
length stats
Labyrinth Orientation of flow 3D seismic, Cores Horizontal Sand body
units, interconnectivity, VSP, well to (oriented), Pulse tests, size,
dead end gravity traps well borehole prod. Tests, Depositional
tomography imaging, RFT, PLT model
logs for
probabilistic
modeling,
horizontal wells

Haldorsen and Chang present an interesting discussion of the effect of discontinuous

shales on productivity. Their paper presents and overview of continuous and
discontinuous shale affects, statistical methods to generate reservoir continuity models
and, also, offers statistical analytical procedures to model the affects.

Heterogeneities can occur at all scales within the reservoir. To appropriately simulate
the reservoir, an adequate understanding of these heterogeneities and their affects are
needed. Lassater et al. discusses reservoir heterogeneity in detail; in their paper, they
classify the heterogeneities into small (centimeters in scale), medium scale
(depositional units) and large scale (many depositional units). They summarize their
investigation as follows:

At the small scale, heterogeneities can result in oil entrapment in small

cul-de-sacs where there is a significant contrast in capillary pressures
between the porous media inside and surrounding the cul-de-sacs. Other
than this effect, small-scale heterogeneities do not appear to have a
significant effect on recovery efficiency. The large number of pore volumes
that sweep a small region during a typical flood results in the rapid
recovery of a significant fraction of the mobile oil.

At the medium scale, gravitational forces are more important and capillary
forces less important. Relative to the small scale, fewer pore volumes
sweep a medium-scale region. Heterogeneities tend therefore to have a
greater effect. In comparing the channel sand and bar sand examples, we

9-40 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

showed that while the channel appears to have a less favorable saturation
profile, behind the front there is probably little difference in sweep
efficiency. Additional cases need to be examined in order to reach a more
general conclusion.

In our example, the medium-scale heterogeneities result in much earlier

water breakthrough. Medium-scale and even small-scale heterogeneities
may be extremely important in highly adverse mobility situations such as
gas and liquid.

The most important effect of heterogeneities is at the large scales where

barriers, faults, and major contrasts in permeability result in the complete
or partial bypassing of large regions of the reservoir. The determination of
reservoir-scale estimates of vertical and horizontal permeability remains
an important challenge for the industry.

Systematic permeability variations resulting in high permeability channels

or layers such as at that for the medium-scale model, are undoubtedly
more significant in reducing recovery than a more random distribution in
permeability such as the small-scale model.

A potentially important mechanism in reducing recovery that has not been

included in our models is the mechanical trapping of oil by the
juxtaposition of impermeable barriers. While such mechanism has not
been seen at the small scale, outcrop maps of shale barriers show that it
might be operative at larger scales.

CATEGORIES OF COMPARTMENTS
By this point in the course, the student should understand several methods to identify
each of these categories. The producing and depleted reservoirs seem obvious; the
question here remains the size and distribution of the reservoir (or flow unit).
Stratigraphic modeling techniques allow a prediction of the reservoir dimension. These
dimensions can be verified with reservoir simulation techniques. This is an important
practice to develop confidence in the predictive capability of the reservoir
characterization techniques being used.

The incompletely drained compartments may be indicated through production analysis

techniques combined with a good static model. New in-field reservoirs and deeper pool
reservoirs can be predicted through the stratigraphic modeling or geostatistical
methods. Finally, good integrated multi-disciplinary analysis of the well log data can
lead to a better understanding of productive and non-productive reservoirs leading to
the by-passed zones.

© PetroSkills, 2013. All rights reserved. 9-41

Reservoir Characterization
Chapter 9

SCALES OF HETEROGENEITY

9-42 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

HETEROGENEITY AND FACIES (Reynolds, et al, AAPG)

FLUVIAL FACIES

© PetroSkills, 2013. All rights reserved. 9-43

Reservoir Characterization
Chapter 9

DELTA-PLAIN FACIES

The next figure shows a reservoir model based on the proximal delta-front fades
association. The association is characterized by alternation between laterally extensive
delta-front siltstones, and mouth-bar and channel sandstones. The result is a stacked
succession of reservoirs vertically sealed by the delta-front siltstones. Internally, the
ands display low heterogeneity. Coarsening-upward and fining-upward profiles occur
and are likely to be reflected in systematic permeability trends. In addition, individual
reservoirs are likely to be compartmentalized by sealing faults.

9-44 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

PROXIMAL DELTA-FRONT FACIES

DISTAL DELTA-FRONT FACIES

© PetroSkills, 2013. All rights reserved. 9-45

Reservoir Characterization
Chapter 9

The previous figure shows a reservoir model based on the distal delta-front facies
association. The association is characterized by tubular sandstone beds interpreted to
be the product of hyperpycnal underflows separated by fine-grained siltstones and
mudstones. This type of reservoir Is likely to have zero effective vertical permeability
and to have horizontal permeability strongly controlled by vertical faults.

RESERVOIR MANAGEMENT

 Maximize profitability from known asset

 Optimum take-off points
 Fewest wells
 Most production

 Maximize recovery
 Treatment (fracturing, acidizing = inflow improvement)
 Secondary recovery techniques, rates, patterns
 EOR

 Tools
 Reservoir Characterisation (static model, flow units)
 Simulation
 Reservoir monitoring

When optimizing the business, there is a choice of three options as shown in the
following diagram:

VIR = Net present value of the cash flow divided by the net present value of the investments

9-46 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

FIELD SURVEILLANCE

 Pressure
Tools: MDT, permanent downhole pressure gauges

 Rates
Tool: PLT

 O/W contact
Tools: c/o logging, RST

 Front monitoring, bypassed hydrocarbon detection

Tool: 4D seismic

RC Hint: Make sure that the FDP contains a

comprehensive reservoir surveillance plan

© PetroSkills, 2013. All rights reserved. 9-47

Reservoir Characterization
Chapter 9

HORIZONTAL WELLS

PERFORMANCE IMPROVEMENT FACTOR (PIF)

PI hor Lhor kv
PIF  
PI vert Lvert kh

where
PI productivity index
Lhor net length of the reservoir drained in horizontal well
Lvert net perforated interval in the vertical well
kv vertical permeability
kh horizontal permeability

PRO’S AND CON’S OF HORIZONTAL WELLS

- improved PI, reduced drawdown
- changed drainage geometry with much larger drainage area

9-48 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

- more reservoir rock exposed

- laterally-distributed reservoir compartments connected
- reduced foortprint

- limited (or no) recompletion options

- lack of control over water and gas inflow
- more inflow at the heel of the well than at the toe (possible solution DCVs)

After: Hurst and Bong Poh Yuk, SPE IPTC 15199 (Nov 2011)

© PetroSkills, 2013. All rights reserved. 9-49

Reservoir Characterization
Chapter 9

Courtesy Dana Petroleum Company  March 2012

RC Hint: Always consider to consult well engineers

9-50 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

UPSCALING

Reservoir simulation models consist of the assembly of individual blocks or cells that
represent discrete portions of the reservoir. Reservoir properties and fluid movement
dynamics then are calculated for each block over time. These complex calculations are
time consuming and computer intensive, therefore, in the interest of minimizing time and
cost, the total number of blocks often is reduced. Upscaling is the term that applies to
averaging the detail from a data source - usually for reasons of efficiency. The need to
diminish the model size is balanced against the need to adequately represent the detail
available and that required to accurately characterize the reservoir. Achieving balance
requires tough decisions and significant effort within the reservoir characterization team.

Mansoori describes the upscaling process well:

In this process, two considerations must be given the highest priority.

First, the upscaling method used must be consistent with the degree of
geologic complexity of the formation of interest. For simple
heterogeneities, calculating effective permeability may not require full
tensorial treatment. This, however; may become extremely important for
complex cases of high anisotropy and heterogeneity over multiple scales.
Second, the degree of upscaling should be carefully chosen such that an
optimum balance is achieved between flow simulation time on the coarse
grid and preservation of important geologic features in the fine grid
reservoir model.

Most of the techniques discussed in the literature consider only absolute

permeability (or hydraulic conductivity) as the parameter to be upscaled
for flow calculations. Note that for cases where changes in rock properties
in a heterogeneous medium have a significant effect on relative
permeability, capillary pressure, and rock dispersivity, the upscaling of
absolute permeability under a single-phase flow assumption may not be
adequate to ensure satisfactory agreement between the fine- and the
coarse-grid multiphase flow simulation results. The results of flow
simulations in an up-scaled system should be checked for excessive
numerical dispersion, proper rate allocation for injection and production
wells, and performance predictions inconsistent with major geologic
features.

Ideally, all of the detail available to the reservoir characterization should be used in an
actual simulation model, but modern reservoir simulation software running on computer
workstations has a practical limit of about 100,000 blocks (models may consist of a
million or more blocks.). This limit is set by current software products and the enormous
memory requirements needed by most simulation software. Therefore, without extreme
care being exercised, much of the detail generated by the reservoir characterization
team could be eliminated from the simulation model. The conclusions to be drawn are:

© PetroSkills, 2013. All rights reserved. 9-51

Reservoir Characterization
Chapter 9

 Significant care needs to be exercised in choosing data included in the reservoir

characterization simulation

 Cross-discipline understanding of the controlling processes in the reservoir can

promote efficient usage of time, effort and data

 Planning prior to data gathering - with the needs of the simulator in mind - is time
well spent

This upscaling process has the most potential for creating inaccuracies in large
heterogeneous reservoirs. The challenge is to develop a balance between providing
details required to depict the reservoir versus minimizing computational requirements. It
also is important to be able to develop multiple realizations of the simulation in order to
look at a wide range of possibilities and to reach a good history match. Lassater et al.
demonstrate upscaling graphically:

9-52 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

UPSCALING – CELLS

THE STRUCTURES
reflect the physical processes of sediment transport and deposition and include planar
lamination, cross lamination and cross bedding ( A), which can be described as 2D or
3D, depending upon the shape of the original sedimentary bedform ( B).

The transport mechanisms responsible for the different structures can lead to variation
in sediment texture both within and between laminae, and between adjacent structures,
with porosity and permeability controlled by grain size, packing and sorting. Permeability
can vary by an order of magnitude or more between adjacent foreset laminae. In recent
years, the impact of such small-scale permeability variation on hydrocarbon recovery
has been demonstrated, and upscaling procedures such as the Geopseudo method
have been developed to take advantage of the hierarchy of geological length scales.

The Geopseudo Atlas library of lamina, lamina set and bed upscaling models is
intended to be a comprehensive representation of the types of sedimentary structures
likely to be encountered in reservoir rocks. Dependent upon the 3D geometry of the
sedimentary structures, 3D ( C) and 2D ( D) models of sedimentary structures can be
constructed, by use of geologically relevant input.

© PetroSkills, 2013. All rights reserved. 9-53

Reservoir Characterization
Chapter 9

9-54 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

UPSCALING CHOICES

© PetroSkills, 2013. All rights reserved. 9-55

Reservoir Characterization
Chapter 9

RESERVOIR SIMULATIONS

One of the goals of reservoir characterization is to develop data needed for input into a
computer-based simulation or characterization of the reservoir. Reservoir simulators
are generally classified as black oil, compositional, thermal and chemical, depending
upon fluid flow, mass and heat transport behavior as discussed below (from Satter and
Thakur1994):

 Black-oil models are most frequently used to simulate isothermal, simultaneous

flow of oil, gas and water due to viscous, gravitational and capillary forces.
"Black oil" is a term used to signify that the hydrocarbon phase is considered as
a single liquid and gas and not by chemical composition. The phase composition
is constant even though the gas solubility in oil and water is taken into account.

 Compositional simulators account for variation of phase composition with

pressure in addition to flow of the phases. They are used for performance studies
of volatile-oil and gas-condensate reservoirs.

 Thermal simulators account for both fluid now and heat transport and chemical
reactions. They are used for simulating steam-flood and in-situ combustion
processes.

 Chemical simulators account for fluid flow and mass transport due to
dispersion, absorption, partitioning, and complex phase behavior. They are used
for surfactant, polymer, and alkaline flooding.

9-56 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

HISTORY MATCHING
 Starting point: The static and the dynamic model are one integrated model.

 Assumptions necessary to match historical data must be reasonable taking into

account the physics of the system.

 Changes in the simulation model must be checked against what is possible in the
petrophysical geological model.
 It is still an integrated effort

SIMULATORS - OBSERVATIONS
 Running a simulator requires special expertise.
 Simulators require significant computing power.
 The up-scaling may have simplified the model.

HYDRAULIC DIFFUSIVITY EQUATION (IN RADIAL FORM)

SIMULATORS - METHODS
 Differential equations (hydraulic diffusivity equation) relate pressure, volume and
time at every location in the reservoir. The hydraulic diffusivity equation conbines
three basic principles, namely the conservation of mass, the equation of state
and Darcy’s law.

© PetroSkills, 2013. All rights reserved. 9-57

Reservoir Characterization
Chapter 9

 Differential equations transformed into difference equations.

 Solutions indicate reservoir performance

– Fluid flow
– Pressure and saturation changes as a function of time and spatial
location.
– The main outcome is a production forecast that can be used in economic
analyses.

SECTOR VERSUS FULL FIELD MODELLING

 It is not always necessary to model the full field.

 No-flow boundaries may be recognized that subdivide the reservoir into sectors.

 Sector models are useful to study concepts.

 Before starting a simulation, the purpose of the simulation should be formulated.

What is the question that needs to be solved.

 Ideally, reservoir simulation should be preceeded by material balance

calculations.

MECHANICS OF SIMULATION
 Reservoir divided into grid cells using a X, Y, & Z coordinate system

 Rock and fluid properties are assigned to each cell

 Computations are carried out for all phases in each cell at discrete time steps

 Simulation output consists of production volumes and rates, pressure and

saturation distributions & material balance errors

 Other process specific information provided at selected time steps

SIMULATORS – INPUT
 History matching
 Geologic modeling
 Porosity, thickness, saturation, permeability
 Upscaling
 Technology integration
 4 D seismic
 Logs, tests, performance data

 Reservoir data

9-58 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

APPLICATION OF BLACK OIL SIMULATORS

 Aquifer behavior
 Up-dip gas injection
 Flank water injection
 Vertical water influx
 Single well operations
 Simulation of large multi-well structures
 Effects of reservoir heterogeneity
 Simulation of large reservoirs of several non-communicating producing horizons
 Analysis of lease unitization

Parallel simulation use new, parallel architecture computers that break the simulation
model into smaller pieces and distribute them among multiple processors. In this way,
each processor effectively solves a much smaller model. As a result, the operator can:

 Run larger models;

 Run existing models more quickly;
 Reduce cycle time for a simulation study;
 Significantly reduce, and sometimes eliminate, the need for upscaling;
 Improve the simulation of near-wellbore phenomena; and
 Quantify uncertainty by running multiple realizations of simulation models.

Parallel simulation allows the operator to perform full physics simulation on a geologic
scale and receive the results in a reasonable timeframe. An initial model for the geologic
scale (200,000 grid blocks), broken into 16 segments and simulated with a multinode
IBM 590 processor is performed in less than 12 hours.

Because of improved computer technology, including Risc-based Unix processors and

parallel multiprocessors, larger simulation models can be run, and larger, more complex
and more realistic geological models are being created. For example, engineers can
create geologic models with more than 10 million cells, even though a simulation model
with 500,000 gridblocks is considered large. An upscaled model simulation can reduce
simulation time dramatically. As an example, a small geologic scale model with 200,000
grid blocks run on a midrange workstation needs a simulation time of 3 days. The
upscaled version of this model containing only 15,000 grid-blocks runs with a simulation
time of 20 minutes. The results from the upscaled simulation are comparable to those of
the 200,000-cell model.

CONDITIONING DATA
 Fit data to model
 Integrate all data sets
 Fit to local conditions
 Normalize throughout field
 Carefully correlate data sets
 Determine sensitivities to known variables

© PetroSkills, 2013. All rights reserved. 9-59

Reservoir Characterization
Chapter 9

DATA FOR RESERVOIR SIMULATION (WAGENHOFER)

 Geology/Geophysics
 Top Structure*
 Gross Thickness*
 Net/gross*
 Fault locations*
 Fluid contact*
 Depositional environment/features
 Depositional continuity

 In-house Interpretations of:

 Well logs*
 Core data
 Seismic sections
 Reports documents

 Petrophysics
 Permeability*
 Porosity*
 Connate water saturation*
 Relative perm.s*
 Capillary pressure curves
 Variability in porosity and perm.
 Reports, documents

 Fluid properties
 PVT analysis
 Black oil table*
 Fluid composition*
 Equation of state
 Reports, documents

– Well performance data

 Production/injection rates*
 Completion data/history*
 Well trajectories, well and tubing details*
 Well events*
 Pressures*
 Production/injection logs
 Tubing head, Shut-in casing, flowing pressures
 Damage/skin factors
 Vertical lift curves
 Pressure transient test data

9-60 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

STREAMLINE SIMULATION

Streamline simulation is a powerful technology complementary to conventional reservoir

simulation, which applies finite difference techniques. The purpose of streamline
simulation is to visualize the flow of hydrocarbons (in particular liquid flow) through the
reservoir. This is being done by tracing streamlines through the reservoir. Streamlines
are curves that are tangent to velocity direction in the velocity field, which is a gradient
of the pressure field.

Streamline simulation is most effective when there are clear velocity direction, i.e. in
convective processes. If the velocity field is not so clear, as in diffusive processes such
as initial depletion generating a secondary gas cap or primary gas cap expansion.

Streamlines are being traced in terms of Time-of-Flight. In fact a translation of

coordinates takes place from space coordinates in 3D space to coordinates being the
Time-of-Flight and the streamline number. The next figure presents this translation in
two dimensions:

There a number of applications for streamline simulation, namely:

 Swept and drainage volume calculations

 History matching

© PetroSkills, 2013. All rights reserved. 9-61

Reservoir Characterization
Chapter 9

 Using streamline simulation to define sector models

 Infill well targetting
 Flow visualization
 Flood front management
 Rate allocations (injector-producer relationship

In the following picture, a permeability distribution is shown (red is relatively high

permeability and blue relatively low) as well as streamline graphs with streamlines
traced from the injector suggesting a picture of volume swept by the injector and a
streamline graph traced from the producer suggesting a picture of the drainage volume
of the producer. In the last graph the swept and drained volumes are combined clearly
showing from where the recovery in this example is coming from. The latter graph
clearly shows where infill drilling should be targeted.

Another example of the application of streamline simulation is Assisted History

Matching. The picture below shows again a permeability distribution (again red is
relatively high permeability and blue relatively low) in a five-spot pattern, which is
translated into a streamline pattern.It is clearly seen in the streamline graph that the
relative density of streamlines is high where the permeability is also relatively high.

The last two graphs in the picture concentrate on the situation between the injector and
well 4. The streamlines are coloured blue signifying the shortest Time-of-Flight between
the injector and the producer and pink the longest Time-of-Flight. The watercut graph

9-62 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

is also coloured with blue showing the early watercut and pink the late watercut. With
these two graphs the watercut performance in years is related to a particular set of
streamlines. Let say that the watercut performance in the yellow period does not fit the
history, then changes in those parts of the reservoir model need to be made that are
penetrated by the yellow streamlines.

RC Hint: If you want to connect geology and reservoir

engineering, apply streamline simulation

Advantages of streamline simulation are:

 Powerful for convective processes.

 Transport equations are being solved along 1D streamlines, which is numerically

very efficient so that much more detail can be incorporated in the reservoir
model.

 Excellent tool for multi-disciplinary reservoir characterization.

Disadvantages are:

 Not really suitable for diffusive processes.

 The assumption that streamlines are fully independent is not fully correct.

© PetroSkills, 2013. All rights reserved. 9-63

Reservoir Characterization
Chapter 9

 It can not easily capture physics that is transverse to the flow direction (e.g.
gravity, capillarity, diffusion, compressibility, transverse thermal effect), although
this disadvantage has to some extent be solved by ‘operator splitting’.

For further detail on streamline simulation, reference is made to

Datta-Gupta, A. and King, M.J., Streamline Simulation, Chapter 1 of Textbook

Series (SPE, 2007)

Thiele, M.R., Batycky, R.P. and Fenwick, D.H., Streamline Simulation for modern
reservoir engineering workflows, JPT, January 2010

RESERVOIR MANAGEMENT

The amount of oil and gas likely to be recovered from a new field depends partly on
where the wells are drilled, the type of wells used (vertical or horizontal), the type of
hydrocarbons being produced, and whether gas or water is injected to maintain
pressure. ExxonMobil's reservoir engineers are known for their ability to get the most
value from oil and gas resources. That is an important advantage in the high-cost
business of deepwater development.

LOOKING DEEP INTO THE FUTURE

To predict how reservoirs will perform 10, 20 or even 30 years from now, we conduct
detailed studies using some of the same proprietary modeling tools that helped us find
the oil and gas to begin with. These tools are an integral part of our reservoir
management technology. They allow us to accurately describe the resource, select the
best well locations and design the optimal depletion rate for the field.

3D RESERVOIR IMAGING - EXAMPLES

 Bass Strait, Australia -increased oil recovery to

more than 70 percent in several ExxonMobil-
operated fields by applying advanced reservoir
modeling techniques, reworking existing wells,
drilling new wells, and upgrading surface
processing equipment.

 Jay Field, Florida - recovering more than 60

percent of the oil from a geologically complex carbonate reservoir with the aid of
advanced reservoir modeling techniques and by injecting nitrogen gas.

 Groningen Field, Netherlands - embarked on a program to keep facilities up-to-

date with respect to safety, environmental requirements, technology and
operating efficiency. We expect to recover 97 percent of the gas in this very large
and mature field.

9-64 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

 Ubit Field, Nigeria - increased the amount of recoverable oil by 500 million
barrels through the application of 3-D seismic, advanced reservoir modeling
techniques, horizontal wells, and by upgrading surface processing equipment

 Cold Lake, Canada - increased the recovery of heavy (tar-like) oil from 10
percent to more than 25 percent by injecting steam into the formation to heat the
oil and make it flow toward the producing wells.

 Arun Field, Indonesia - recovering about 94 percent of the gas and 67 percent
of the condensate from a mature, lean gas condensate reservoir using
compositional reservoir modeling to balance gas injection and maximize
condensate recovery.

 Prudhoe Bay, Alaska - increased the amount of recoverable oil by 4 billion

barrels through the development of advanced geologic models, injecting gas to
improve oil recovery, upgrading surface processing equipment, and by using
horizontal wells.

 Means Field, Texas - increased reserves by 66 million barrels in a mature field

by drilling new wells and by injecting water and carbon dioxide to improve oil
recovery.

 Brent Field, North Sea - increased the amount of recoverable oil and gas by
400 million oil-equivalent barrels through the early reduction of reservoir pressure
to liberate gas from residual and bypassed oil. The use of advanced reservoir
modeling, reworking existing wells, drilling new wells and upgrading surface
equipment have contributed to the increase in reserves.

 Topacio Field, Equatorial Guinea - installing electrically driven multiphase

subsea pumps to increase production from four wells by 6,000 barrels a day
(2000).

To meet the world's growing demand, energy companies must be able to find oil and
gas from increasingly difficult locations and bring them to market at a competitive price.
Thanks to 3-D seismic imaging and advanced reservoir modeling, we're doing that now
and will continue to do so well into the future.

© PetroSkills, 2013. All rights reserved. 9-65

Reservoir Characterization
Chapter 9

COMPLEX INTERPLAY OF FACIES IN ARAMCO SHUAIBA

Complex interplay of facies in Aramco Shuaiba (Cretaceous) - Rudist barrier (orange), lagoonal (blue), and basinal
(green).

9-66 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

The depositional model is a prograding carbonate ramp with local patch reef
accumulations in the shallower waters. The reefs have lower permeability than the
grainstones surrounding them and are "baffles" to vertical communication within the
reservoir. The unique placement of these reefs caused the lagged oil to form where it
did. The simulation model results, which are shown in the diagram as changes in oil
saturation in 5-year increments, reproduced the lagged oil correctly. This was
considered a positive feedback on the geology of the model, which continues to be
predictive many years later.

SEQUENCE STRATIGRAPHY HELPS PREDICT LAGGED OIL

© PetroSkills, 2013. All rights reserved. 9-67

Reservoir Characterization
Chapter 9

HANIFA RESERVOIR
The Hanifa reservoir in Berri field has been on production under peripheral water flood
since the mid-1970's. During that time, a significant quantity of the reserves has lagged
the water encroachment. In order to evaluate alternative strategies to recover this
lagged oil, using reservoir simulation modeling, a detailed geological study and
geocellular modeling effort was launched in the late 1980's. Previous simulation efforts
had been unable to model the lagged oil in the field. This time, a sequence stratigraphic
framework was developed which changed the layering and facies distribution within the
geological model. This was the first sequence stratigraphic framework model
constructed in Saudi Aramco and the subsequent successful reservoir simulation
history-match and predictions have confirmed its power as a predictive geological tool.

Structural growth has raised more of the low quality facies into the oil column, leaving
more of the high permeability grainstones in the water let. Facies were distributed in the
model using a stochastic algorithm constrained by well control and parasequence
isopachs.

Porosity was distributed in the model using a stochastic algorithm constrained by facies.
As a result, the porosity model distribution by facies matches that of the well data.

9-68 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

© PetroSkills, 2013. All rights reserved. 9-69

Reservoir Characterization
Chapter 9

VERIFYING STATIC MODELS WITH DYNAMIC DATA

9-70 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

© PetroSkills, 2013. All rights reserved. 9-71

Reservoir Characterization
Chapter 9

INTEGRATION CASE STUDY

HORSE CREEK FIELD – AIR INJECTION PROJECT

 Strat. trap in North Dakota, USA

 Integrated multi-disciplinary team

 Detailed geologic studies

 Reservoir simulation studies

 Laboratory experiments

9-72 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

RESERVOIR PROPERTIES
 Porosity – 16%  3800 acres

 Perm – 10-20 md  Depth – 9125’

 Net pay – 20’  Red River Fm.

 SW - .35  OOIP – 45.7 MBO

 Temp. - 220°  Primary – 4.5 MBO – 9.9%

 Initial press. – 4000#  Incremental – 7.6 MBO – 16.62%

 Gravity – 32°  Total – 12.136 MBO – 26.53%

 Bubblepoint – 625#

© PetroSkills, 2013. All rights reserved. 9-73

Reservoir Characterization
Chapter 9

HORSE CREEK STUDIES

 PVT of oil
 Waterflood conceptual
 Black oil S/2
 Full field black oil
 ARC test
 Combustion tube experiments
 Thermal model of combustion tube
 Radial air injection
 Full field thermal simulation

WATERFLOOD STUDY
 Feasibility for waterflood at Horse Creek
 Not a good candidate due to poor water/oil relative permeability
 Rejected as possibility

BLACK OIL MODEL

 Initial model of S/2 was encouraging
 confirmed volumetric assumptions
 estimated incremental recovery

 Full field fully tested static model

 history match
 indicated lack of water influx
 tested response to gas injection
 served as model for later simulations
 incremental reserves 7.6 million barrels

ARC TESTS
 Assessed oil oxidation parameters under reservoir conditions
 Qualitative tests indicating oil reactions with air
 Indicated oil would react at low temperature conditions

COMBUSTION TUBE BURNS

 Quantified combustion characteristics of rock, oil, brine
 Supported previous work and assumptions
 Indicated efficient recovery process

9-74 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

AIR INJECTION RADIAL MODEL IN THERMAL SIMULATOR

 Matched combustion tube results to reality
 No coke formation in formation
 Tested radius of burn
 Predicted:
 oxygen behavior in reservoir
 tested sensitivities to varying permeability
 indicated oxygen radius of less that 700 feet

FULL FIELD THERMAL SIMULATION

 Incremental reserves 7.2 - 7.9 MMBBLS
 Oxygen front less than 3500 feet from injectors
 Oxygen breakthrough did not occur
 Air injection increased recovery by 10% over nitrogen injection
 Subsequent economics confirmed feasibility of project

RESULTS AFTER 9 MONTHS OF OPERATION

 Facilities operated at 99.7% efficiency
 2 BCF air injected
 Current rate 8.5 MMCF/D
 BHP increased by 550 PSI
 GOR decreased
 Ave. prod. rate from 293 BOPD to 400 BOPD
 36% increase
 Encouraged that target of 100-1300 BOPD will be met

© PetroSkills, 2013. All rights reserved. 9-75

Reservoir Characterization
Chapter 9

RESERVOIR ENGINEERING DUTIES

 Review well data for compartments/inadequate drainage

 Estimate incremental reserves
 Estimate reservoir size
 test data
 production data
 Estimate reservoir geometry
 Refine/focus geoscience models

EXERCISE – CONTRIBUTIONS TO RC

 For each RC discipline, list the following:

 Critical elements contributed to the RC process

 Limitations of the discipline

 Challenges for the future

 Primary and secondary customers on the RC team

 Scale(s) of data contributed

9-76 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

1.
Establish Stratigraphic Framework

2.
Establish Facies, Flow Units and
Geologic Prototype Areas in Cored Wells

3.
Develop Digital Log-Based Statistical Models for
Facies, Flow Units and Permeability in Cored Wells

4.
Test Statistical Models Against Core Data

5.
Map Facies, Flow Units and Geologic
Areas Field-Wide with Log Data

© PetroSkills, 2013. All rights reserved. 9-77

Reservoir Characterization
Chapter 9

9-78 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Reservoir Engineering

© PetroSkills, 2013. All rights reserved. 9-79

Reservoir Characterization
Chapter 9

SUMMARY OF CHAPTER 9
Reservoir studies require a great deal of integration between disciplines. However,
there is also a lot of work to be done that is typically carried out by disciplines.
Therefore, study teams have to decouple, do their specific discipline-type of work, and
then recouple to share information.

Understanding of the drive mechanisms that play a role in the reservoir is necessary to
enable accurate estimates of ultimate recoveries and reserves.

Unconventional resources both in oil and in gas have become a major part of
hydrocarbon development.

In the evaluation of hydrocarbon resources, volumetric calculations, decline curve

analysis, material balance and reservoir simulation are all essential. Nowadays, the
value of material balance calculations is often underestimated. Any reservoir simulation
should be preceeded by material balance calculations to get an overall feel for the
reservoir, and also to judge whether the production data (in particular the gas
production data) can be trusted.

In (finite difference) reservoir simulation, it is not always necessary to model the full
field. No-flow boundaries may subdivide the reservoir into sectors that can be modeled
separately. In sector models concepts (such as size of drainage areas per well, water
and gas coning, mobility ratio effects) can be studied providing guidelines for optimum
reservoir development.

Streamline simulation is a powerful technology complementary to finite difference

reservoir simulation. Streamline simulation is particularly useful to model convective
processes (e.g. injection schemes with pressure maintenance), whereas diffusive
processes (e.g. depletion with secondary gas cap development) are less easily
modeled.

Identification of flow units, which requires very much an integrated disciplinary

approach, is of vital importance to define appropriate development plans.

Horizontal wells have become bread and butter in hydrocarbon resource development.

In gas field development, aquifer presence is a dominant factor. If there is no aquifer

influx in a gas reservoir, recoveries as high as 97% of GIIP can be realized.

The objective of a reservoir study is to build a reservoir model including all geology
(geomodel), all petrophysics (petromodel) and all dynamic properties (dynamodel) in an
integrated way while the model is only complete when all the uncertainties are also
specified. In addition to all technical work, the economics (econmodel) of the reservoir
development need to be considered. The close integration of these four models into the
reservoir model, will lead to the best understanding of the reservoir and consequently,
to the best development of the reservoir.

9-80 © PetroSkills, 2013. All rights reserved.

Chapter 10 E CONOMICS AND R ISKING
Commercial oil and gas fields result from the complex interplay of the geologic controls
of hydrocarbon source, reservoir, and trap. At least 11 factors… must each and every
one be adequate at one location, or no significant volumes can accumulate or be
produced. It is no wonder that commercial oil and gas fields are so rare, and that there
are so many dry holes, and that there is so much risk in exploration.
David A. White

The primary objective of any reservoir characterization project is to maximize

profitability from the property. There are many methods applicable to achieving this
goal, but two common threads among them include the ability to assess the risk of an
outcome being accurate (a simulated reservoir production scheme for example) and the
financial impact of the project on the company or partnership. Modern techniques
including geostatistical approaches provide these indications for better decision-making.

RISKING/ECONOMICS CHALLENGE

 Portray the work done in (financial) terms that customers understand

 Cost < Benefit
 Suggestion: Consider decision-tree presentation

0.56 A Pool, B Pool

B = Pool
0.70

A = Pool
0.30
0.80 0.24 A Pool, B Dry
B = Dry

0.20
A = Dry
0.14 A Dry, B Pool
B = Pool
0.70

0.30 Both Dry 0.06

B = Dry 

© PetroSkills, 2013. All rights reserved. 10-1

Prospect and Play Assessment
Chapter 10

OBJECTIVES FOR ECONOMICS

Objectives for economic and risking portions of the reservoir characterization process
are to:

 Provide knowledge and tools for practical, systematic pre-drill assessment of

potentially recoverable oil and gas volumes in developing fields

 Provide optimum Reservoir Characterisation by using the best available methods

to obtain:

 Trap, volumetrics and hydrocarbon content.

 Potential reservoir and size.

 Quantify all geologic, engineering and production risks and uncertainties

 Provide insights for the organization for:

 Assessing assessments
 Avoiding pitfalls
 High-grading opportunities
 Planning selectively

UNCERTAINTY AND RISK

 Uncertainty - imprecision in estimating the size range of factors

 Risk - the probability that no field or related volume factor exists

 Relates to minimum required for an economic accumulation

 Adequacy - one minus risk

 Probability - the perceived likelihood of the occurrence of an event

CHANCE OF ADEQUACY
 (1- RISK)
 Used for individual factors or combined for entire accumulation
 Adequacy of prospect or play equals product of individual control factors:
 Source, Reservoir, Trap, Seal, Recoverability
 Adequacy minimum economic accumulation not presence or absence

RISKING GUIDELINES
1.0 Certain
0.7 Likely
0.5 Equally likely/unlikely
0.3 Unlikely
0 Impossible

10-2 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Economics and Risking

Used for individual controls or combined for a prospect or play likelihood

RISK REDUCTION TECHNIQUES

 Production seismic technologies
 3-D
 4-D
 Interwell
 Production data
 Simulation
 Actual
 Good models
 Technology transfer programs

Exercise - Geological risk factors

The first and most crucial step in assessment is to have a firm understanding of the
geologic factors controlling oil and gas occurrence. These control factors become the
geologic risks. List below all the factors that you can think of that are essential to the
geologic creation of an oil or gas field having significant potentially recoverable
reserves. You may list more than 10 if you want to.

1.

2.

3.

4.

5.

6.

7.

8.

9.

10.

© PetroSkills, 2013. All rights reserved. 10-3

Prospect and Play Assessment
Chapter 10

RC Hint:
Always understand the economic impact
of your work or recommendation
– Act as a business person not just a technical expert

Economics and risking are critical to every oil and gas project. Each oil and gas
company has competing demands for its limited funds, and those that are most
successful are the ones that evaluate their business decision most accurately over time.
Underestimating the value - or overestimating the risk - of a project means that the
project may be funded at the expense of a more suitable investment. The converse is
true as well. An ideal assessment scheme is one in which the sum of the risked means
of all project assessments equals the sum of the reserves found.

During the Second International Reservoir Characterization Conference held

Dallas, a work group offered the following on economics:

The group decided that risk could best be defined as "the difference between
the value of an acceptable outcome and the value of the possible range of
outcomes at any given time. The definition itself gives clues as to the
elements involved in estimating risk: there must be an estimate of the
acceptable outcome followed by an estimate of the range of possible
outcomes. Unfortunately, neither task is especially easy. An estimate of the
acceptable outcome entails a reasonable understanding of the economic
climate under which the project is being undertaken. This factor is highly
process specific inasmuch as the requirements of acceptable rate of return
(short-term benefit) must be balanced against the needs of reserve additions
(long-term benefit). Unfortunately, the acceptable outcome estimate is itself
uncertain owing to variances in economics, politics and costs.

Estimating the range of possible outcomes is simpler, at least conceptually,

and it bears more directly on the themes of this conference. The group thought
it important to begin this process by understanding the reservoir and process
in question. There are two main parts to this: identify the individual parts of the
system, then understand how these parts relate to each other. Since reservoir
systems are usually very complex, this understanding must usually come from
a variety of sources, hence the need for often-quoted interchange of
information and insights among disciplines. Specific technologies involved at
this step are production seismic techniques (both three-dimensional and
interwell), production data (interpreted perhaps by large simulation models),
and a concerted program of technology transfer. Specific factors for estimating
the range of uncertainty in enhanced oil recovery then would be original oil in
place, recovery efficiency, rate of recovery and the reservoir description. The
culmination of this understanding would be a workable model of the reservoir
and process.

10-4 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Economics and Risking

Estimating the range of uncertainty involves using the model developed above
to translate the effects of individual uncertainties into the cumulative
uncertainty of the model output. How to do this efficiently is the subject of
active research, as evidenced by the number of presentations dealing with
mathematical modeling and stochastic assignments at this conference. The
cost of many of these techniques is large; however, this cost, like those
associated with fathering and analyzing data, must be factored into the notion
of a acceptable outcome as discussed above.

As a final point, the group noted that the idea of risk manifests itself in numerous ways in
current practice. The focus of this discussion has been on economic risk, but there is now
a risk, incompletely manifest in economics, that is associated with environmental factors
and a risk associated with spending too much money on data gathering and analyses.

EVALUATION PROBLEMS
Critical to the evaluation process is understanding the models: modeling the right
concepts with the appropriate methodology. The problems come when:

 the wrong concept is modeled

 the right concept is modeled incorrectly
 interdisciplinary sharing of the model components is inadequate
 no modeling is done

Risking of reservoir characterization projects is similar to risking for exploratory

prospects except that more details are available and the risks are lower (the discovery
has already been made). Each of the variables that contribute to trap volume and
producibility need to be evaluated in an appropriate manner. Factors that should be
evaluated include:

 trap volume
 net/gross ratios
 porosity
 permeability
 HC saturation
 oil and gas fractions
 HC fill
 oil shrinkage factors
 free gas volume factors
 recovery factors

Care should be taken when assigning values to each of the assessment factors.
Assistance from other disciplines can provide better answers to such factors as whether
the factors are dependent or independent or what an informed estimate of a value might

© PetroSkills, 2013. All rights reserved. 10-5

Prospect and Play Assessment
Chapter 10

be. Even with the great advances in geostatistical approaches and other new methods it
is important to remember the old computer mantra “garbage in, garbage out”.

Wolfe et al. demonstrate a geostatistical approach to assessment. In their method, 200

realizations are generated and ranked from low to high. If, for example, sand thickness
is being investigated, the thicknesses are arranged in ascending order and plotted on a
graph (see illustration below).

Cumulative probability distribution of sand thickness at the proposed location. Each asterisk represents sand
thickness for one map. Vertical line intersects the distribution at the mean (45 ft) which equals the KED solution.

In this example, the median sand thickness expected is 465 feet and there is and 80%
probability of having 32 feet or more of sand at the modeled location. This type of
analysis can be conducted for any location on the map or for any number of
parameters. Another possibility with the data set is to plot the economic threshold for a
specific location and then directly read the probability of exceeding that value. The next
illustration demonstrates that approach:

10-6 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Economics and Risking

SAND THICKNESS - THRESHOLD

This presentation indicates that there is a 35% probability that the sand thickness at the
proposed location is less than the economic threshold (40 feet). A risk map could
demonstrate the probability of reaching the economic threshold for sand at all of the
locations on a map. This map could be the contoured probability of finding less that the
threshold. In the case Wolfe et al. presented there appeared to be a lot of error in sand
thickness map and offered suggestions of means to increase confidence in the
interpretation:

 using a better seismic data set as the guide

 increasing the control

To quote them:

The narrower the probability density function, the less error or uncertainty.
Primary factors that control the narrowness of the probability density
function or geostatistical estimate are:

 Correlation coefficient between the seismic and well data, where the
higher the correlation coefficient, the narrower the probability density
function (PDF)

© PetroSkills, 2013. All rights reserved. 10-7

Prospect and Play Assessment
Chapter 10

 Variogram model range of the well data, where the longer the range, the
narrower the PDF

 Distance of the proposed location to the nearest well control, where the
lesser the distance, the narrower the PDF (if it is at an existing well, the
PDF is a spike)

RISKED BUSINESS VENTURES

 Usage of risked analyses

 Effective for portfolio management
 Very dangerous for single business venture
 What is risk?
 What is uncertainty?
 Tools
 Decision trees
 Multiple simulation techniques (Monte Carlo, etc.)

10-8 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Economics and Risking

CERTAINTY EQUIVALENT/ RISKED RESERVES

MONTE CARLO SIMULATION

© PetroSkills, 2013. All rights reserved. 10-9

Prospect and Play Assessment
Chapter 10

Forecast: RISKED SUM

1,000 Trials Frequency Chart 992 Displayed
.028 28

.021 21

.014 14

.007 7

.000 0
29.33 38.75 48.16 57.58 67.00
BCF

NET PRESENT VALUE (NPV)

Present value is a function of

- The amount A in real terms

- The discount rate r (fraction/annum)
- The time t (years)

The net present value is the summation of the present values (income and expenses)
over all years of the project. Note that always a reference date needs to be specified.

In this analysis, concepts such as money of the day (simply the money one can produce
from his/her purse), real term money (money that maintains its spending power and
therefore requires a reference date) and inflation (devaluation of money over time) are
relevant.

10-10 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Economics and Risking

CASH SURPLUS AND ACCOUNTING PROFIT

 Cash surplus = Revenues - Royalties - CAPEX - OPEX - Tax

 Accounting profit = Revenues - Royalties - depreciation - OPEX – Tax

EARNING POWER (IRR)

© PetroSkills, 2013. All rights reserved. 10-11

Prospect and Play Assessment
Chapter 10

VALUING RESERVOIR STUDIES/DATA (VOI)

25% Hi Case 80
EMV 1 Dev’mnt 50% Mid Case 50
No info (mid
case) 25% Low -20
Case
EMV 1 = 0.25 x 80 + 0.5 x 50 + 0.25 x -20= 40

25% Hi Case 100

EMV 2
With info
50% Mid Case 48
25%
Low 0
Case
EMV 2 = 0.25 x 100 + 0.5 x 48 + 0.25 x 0= 49

Value of data = EMV 2 – EMV 1 = 9

 Concepts:
 Acquiring knowledge reduces uncertainty (requires investment/delay)
 Increases value

 “Real option analysis” indicates maximum-value benefit

 Function of:
 Projects value (EMV)
 Characterization analysis time
 Cost of delay

10-12 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Economics and Risking

IMPACT OF APPRAISAL

© PetroSkills, 2013. All rights reserved. 10-13

Prospect and Play Assessment
Chapter 10

RC EXAMPLE

 Design program to recover oil:

 Minimum number of wells, ASAP without damaging productive life
 Three scenarios:
1. Appraisal only
2. Appraisal + reservoir studies
3. 3D + appraisal + reservoir studies

FIELD ASSUMPTIONS
 Discount factor 12
 Recoverable reserves (MBO) 1000
 Appraisal wells (cost $10M/well) 4
 Reservoir studies (2 year expenditure - $M) 11.5
 3D seismic (2 year expenditure - $M ) 26
 Development wells (cost $8M/well) (cost $10M/well) 16
 Expected well productivity (BOPD) 20,000
 Development drilling program – years 4
 Facilities (2 year expenditure - $M ) 250
 Oil price margin (after tax $) 3

SCENARIO RESULTS:
1. Appraisal only $1057M
2. Appraisal + reservoir studies $1097M
3. 3D + appraisal + reservoir studies $1235M
4. Perfect knowledge $1605M

 Value of all potential options exceed price

 Sensitivity study of Option 3 indicates threshold cost of 5 times estimated cost
 Timing critical, delay of 9 months destroys value

10-14 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Economics and Risking

SUMMARY OF CHAPTER 10
A Reservoir Characterisation study is only finished when the economics and risks of the
recommendations have been fully assessed. In other words, economic analysis is an
integral part of the Reservoir Characterisation study.

As in any development expenditures (investments) precede revenues, it is essential to

analyse the economics in terms of net present value (NPV). In NPV calculations a
reference date has to be specified. In such calculations, the Earning Power or Internal
Rate of Return (IRR) can be determined.

Cash surplus and accounting profit are different quantities. In cash surplus CAPEX
appears, whereas in accounting profit depreciation plays a role.

The uncertainties in the evidence on which basis the decisions are taken, is usually
significant giving rise to risks in the decisions. Data acquisition or appraisal may reduce
the uncertainties and hence the risks. However, data acquisition requires expenditure
and causes delay of the development, which must be justified by evaluating the Value of
Information (VOI).

© PetroSkills, 2013. All rights reserved. 10-15

Prospect and Play Assessment
Chapter 10

10-16 © PetroSkills, 2013. All rights reserved.

Chapter 11 O RGANIZATIONAL S TRUCTURE
If it were up to me, I would obviously design the company for integrated high tech.
teams to look for previously invisible fields. You have to be good, very good in a
modern sense of advanced technology; highly motivated. innovative, flexible quick.
John Masters - Canadian Hunter

TEAM STYLES

Teams outperform individuals working alone or in traditional organizational groupings

especially when the tasks require the integration of skills, judgment and experience as
in reservoir characterization. Effective reservoir characterization teams are deeply
committed to their purpose, goals, fellow team members and their approach. They are
united around a significant performance challenge - maximizing the productivity of a
producing field.

TEAM BUILDING:

Managed, planned, systematic coordination of effort by group with a common

goal, for the purpose of achieving that goal in an optimally productive way.

The product that it produces defines a team.

MANAGING QUALITY
 Traditional approach
 Quality through inspection

 Team-based approach
 Quality through adherence to high standards

HIGH STANDARDS APPROACH ADVANTAGES

 Flatter organization

 Lower overhead

 High profitability

 Lower product cost

 Employee ownership of the product

 Higher job satisfaction

© PetroSkills, 2013. All rights reserved. 11-1

Reservoir Characterization
Chapter 11

HIGH STANDARDS APPROACH ELEMENTS

 Shared vision

 High levels of trust

 Agreed definition of quality

 Training to acquire skills to measure quality

 Goal setting / measurement system

 Hands-off approach by management

REASONS TEAMS PERFORM WELL

 Synergize complementary skills

 Add social dimension that enhances performance

 Joint goals and objectives support effective joint problem solving

 Team members have more fun!

OBSERVATIONS
 Performance challenges create teams

 Teams integrate performance and learning

 Most organizations prefer individual accountability

 Disciplined approach to team building often ignored

Most teams also fit into one of two functional categories:

 Permanent teams: continue for long periods and take on a continuing series of
assignments.

 Task forces: temporary groups brought together to work on a specific objective;

once it's achieved, the group usually disbands.

Most reservoir characterization teams fit into one of two relationship categories:

 Reporting teams: vertical groups, made up of a manager/team leader and staff.

 Peer teams: horizontal groups, made up of individuals from various functional

disciplines at similar levels on the organization chart frequently with an assigned
leader who takes on the role of team leader.

11-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

RC Hint: Always ask: Does the involvement of other

disciplines give a better
(cheaper or more comprehensive) answer?

ORGANIZATIONAL MANAGEMENT

The organization may take a number of approaches to forming teams. The methods
will vary with the company's culture, the setting and the nature of the field to be
described. Some of the approaches require reducing the controls that the company has
over its own destiny through the inclusion in a partnership or an alliance. The benefits
of any approach should outweigh the costs of doing the work in the new way (the costs
include both the financial and intangible costs associated with the change process).
Approaches that companies have taken include:

 Employee teams - traditional in-house multi-disciplinary work teams

 Example - a team assigned to develop a new discovery

 Partnerships - formal arrangements with another company to develop the field

they own in cooperation
 Example - a joint venture to develop or unitize a field may involve sharing of
data, interpretations and/or resources

 Strategic alliances - arrangements with other E/P companies or service

companies to share capabilities and services while retaining core strengths
 Example - an agreement to share some technology in developing the
property, but retain prospect development as a proprietary pursuit

 Outsourcing - to commission part of a service through a third party

 Example - purchasing services from an outside organization in a portion of
the reservoir characterization process

The bottom line is forcing many oil companies to redefine their core activities. Oil and
gas companies are reviewing costly research and development programs and retaining
only proprietary technology that gives them a competitive advantage. The rest is being
outsourced to service companies or university consortia. Many companies see less
advantage in owning technology, more in being able to access it quickly and to use it to
improve efficiency.

Outsourcing has the added advantage of eliminating duplication and raising the
industry's overall efficiency. The trend is toward closer relationships with the service
industry. This provides dramatic opportunities for improved communication. The oil
company can guide and follow service industry developments from beginning to end
and, conversely. the service industry gets the chance to understand a client need from

© PetroSkills, 2013. All rights reserved. 11-3

Reservoir Characterization
Chapter 11

the inception of each project. Partnerships greatly facilitate success in the new world of
cooperation.

A natural outgrowth of this trend that has been growing for a long time is the formation of
formal agreements to jointly conduct reservoir characterization projects through a strategic
alliance. These alliances typically consist of the participating partners, service companies,
suppliers and possibly regulatory agencies. For these approaches to be successful, a
number of elements need to exist. A strong alliance or partnership must have:

 Alignment of purpose among all parties

 Detailed planning prior to inception
 Management commitment to the project
 Complimentary skills
 Avoidance of weak partners
 Clearly defined roles and responsibilities
 Commitment to the long term
 Measurement mechanisms
 Common corporate cultural values
 Strong interpersonal relationships
 Mutual trust among the team members

It is apparent from this list of criteria that companies cannot simply form a strategic
alliance on an ad hoc basis; the formation of successful alliances requires a significant
amount of planning and commitment and development of skills that have not been
required for development projects in the past.

THEMES TO DEVELOP SUCCESSFUL RESERVOIR CHARACTERIZATION TEAMS

Through a long period of observing reservoir characterization processes, a series of
common characteristics have become apparent. These include:

 Shared vision - organizational vision and project vision

 Alignment - knowledge and decision making authority at the source of the
information
 Compatible organizational structure to support team activity
 Assist teams through team maturity stages
 Focus on team’s product -the reservoir characterization process - to the
exclusion of other concerns
 You get what you reward
 Redefine success for individuals
 Skill acquisition is critical to success of the team. Individuals are responsible for
their own career

11-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

The changes needed for a transition to a team-based environment require modification

in how organizations are structured. Traditional approaches are reliant upon heavy
control of strategies and tasks from the upper layers of management in the organization;
the new approaches imply that the teams are charged with heavy responsibility for the
control of tasks that are congruent with strategies designed by senior management (in
many organizations the lowest levels in the organization also have input into strategy
formulation). This change in relationships is illustrated in the models below:

Hierachical Model Team-Based Model

S
Exec. T Exec.
R
Mgmt.
A Mgmt.
T
E
G Team
Team
Y
TASK TASK

On their way to becoming mature and fully functioning, reservoir characterization teams
go through a predictable five-stage transition process that often lasts from three to five
years.

© PetroSkills, 2013. All rights reserved. 11-5

Reservoir Characterization
Chapter 11

 Stage 1: KICKOFF
 Stage 2: STORMING
 Stage 3: NORMING
 Stage 4: FORMING
 Stage 5: INTEGRATED
 SUPER TEAM

Kick-off Confusion Dependency Team loyalty Integrated

Team No Unsure about Learning to work Intense loyalty to Integrated into work
commitment to roles/responsibilities together, team, conflicts with of org., other teams,
other team direction/ other parts of org. offer ideas for better
members inspiration from ways to work
leader
Individuals No Confused about future, Dependent on Functionally Fully committed to
commitments questioning leader(s), competent, highly vision of org., willing
to team or developing new loyal to team to challenge
project relat./skills processes
Team Leader Understands Identifies conflicts & Focuses on Shares leadership, Meets with other
org. & issues project, team focuses on org. leaders to share
processes mbrs., processes relat.s learning

Management Shares team Overwhelmed by Pulls away from Conflict with team, Coordinates,
model confusion and needs, daily op.s, offers focus on org. vision facilitates cross team
threatened guidelines comm.
Support Skeptical Feel left out of process, Relat. improve as High level of coop. w/ Active in offering
Groups about teams confused auth. is defined team integrated sol.s
Organization Traditional Ident. systems that Modifies: perf. Monitors team Expands team
approaches need to be changed eval., auth., prof. activity and adjusts approach throughout
growth, rewards, as needed organization
training, discipline
systems

Confusion & High High Lessening Minimal within team, Low

conflict may be high with rest
of organization
Productivity As in Falling Rising above Continuing to rise Maximizes
traditional traditional levels
Training Understanding Interpersonal, cross Problem solving, Leadership, Company and
Focus team model training conflict resolution evaluating industry trends
performance

11-6 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

CHALLENGES OF STAGE 1
 Overall focus - Sharing info about:
 model
 expectations
 identifying issues

 The training focus

 explaining model
 interpersonal skills
 initial administrative
 technical overview of all team tasks

CHALLENGES OF STAGE 2
 Overall focus - identifying confusion and building plans for:
 managers letting go
 team empowerment
 identifying systems to modify

 The training focus

 setting team rules
 learning resource locations
 communication techniques

CHALLENGES OF STAGE 3
 Overall focus - becoming a unit
 leadership exertion
 defining authorities with management, support, other teams
 trust building

 The training focus

 building leadership in Team Leader and others
 learning new organizational systems

CHALLENGES OF STAGE 4
 Overall focus - effective team, fully integrated, rotational task leadership
 working within the organization
 sharing results with other teams
 quality management

 The training focus

© PetroSkills, 2013. All rights reserved. 11-7

Reservoir Characterization
Chapter 11

 working across team boundaries

 building a learning organization
 maintaining team passions

CHALLENGES OF STAGE 5
 Overall focus - developing new organizational norms
 sharing ideas up, across and down
 participation in strategic initiatives

 The training focus

 learning more about industry
 dealing with outside parties - partners and contractors

TRADITIONAL ORGANISATION

11-8 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

MATRIX ORGANISATION

MATRIX TEAMS
 Work is conducted through team

 Administrative/professional growth through department structure

 Individuals have two “bosses”

 Participants may serve on several teams

 Team leader coordinator - not supervisor

DEDICATED TEAMS
 Reporting is solely through team structure

 Discipline heads have an advisory role or are members of review teams

 Requires significant organizational change

 Produces a flatter organization

© PetroSkills, 2013. All rights reserved. 11-9

Reservoir Characterization
Chapter 11

DEDICATED TEAM ORGANISATION

11-10 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

WHITNEY CANYON FIELD TEAM - AMOCO - HISTORY

 Overthrust belt - Wyoming, USA

 Two completely separate operations:

 Field operations
 ineffective completion techniques
 no involvement in plant

 Gas plant
 operating at 1/2 capacity
 considering merger with Chevron facility

 Expensive plant turnarounds - not successful

WHITNEY CANYON FIELD TEAM - AMOCO - SOLUTION

 Formed single team using systems dynamic approach
 Interconnection of field and plant operations
 Included 2 production engineers & geologist on plant team

 Reviewed field potential prior to action

 More gas reserves than originally estimated
 Ineffective completions, formation damage
 Attempted revitalization - unsuccessful
 Committed to new well ($6-8 million)

 successful due to coordination with Amoco technology group and

contractors
 two more successful wells

 Visited other facilities for “best practices”

 Involved plant personnel in planning for next turnaround

 Major cultural change in operational style

 Ownership, involvement, systems approach, benchmarking

 Successful project
 Production increases, faster turnarounds, better drilling/completions,
employee buy-in, faster hookups

© PetroSkills, 2013. All rights reserved. 11-11

Reservoir Characterization
Chapter 11

TEAM LEADERSHIP SKILLS

 Kick-off
 Understanding organization and team model, high trust level
 Confusion
 Conflict resolution, commitment to team and project
 Dependency
 Multi-disc. Skills, planning, decision-making
 Team loyalty
 Sharing of leadership, org. Integration, team learning
 Integrated
 “Big picture” orientation

Girgis et al. presented a matrix that defines roles for an AMT (asset management team)
Leader:

TEAM LEADER - Objective - Coordinate the team's

efforts to accomplish goals/objectives.

The leader is: The leader is not:

Coordinator/facilitator Boss
Engineer/geoscientist who spends less than 15% of Full time MTD manager
time on administrative tasks
Focuses majority of effort on technical work Administrative tasks overshadow technical
responsibility
The leader does: The leader does not:
Coordinates goal setting process w/ tea ands Determine team G/O's
management
Calls, chairs, documents all formal team meetings, Allow too much idle discussions, interdisciplinary
arranges informal work stations discussions and wasted time in meetings
Actively seeks input form all appropriate members and Only look to apparent discipline for answer
disciplines
Encourages new ideas and innovation form all Criticize or allow negative criticism to dominate
disciplines discussions or reviews
Strives to create an atmosphere of unity and Allow trivial discussions to continue
cooperation
Arranges meetings and work sessions to maximize Expect attendance form all members at every meeting
efficiency
Searches out ideas from all members Allow a few strong personalities to dominate
Leads effort to formalize team G/O's \ seeking input Wait for management deadline to gather data
form all team members and management

11-12 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

THE LEADER
One member of the team may be formally designated as leader either by a
management or by the team itself. Unlike a manager, a leader may have no power to
discipline, punish or reward; the leader is usually empowered to do only three things:
preside over meetings, see to it that the team completes its assignment on time and
report the team's conclusions to higher management. Nevertheless, even though the
leader's role is procedural, not hierarchical, the leader can gain the strong respect of the
team and emerge as a genuine inspirational leader.

 The informal leader(s): On teams with or without a designated leader, an

informal leader or leaders may emerge. The emergence of several competitive
leaders may be bad for the team; what's best is the emergence of one strong
leader or two who can work in partnership. Informal leaders with solid leadership
skills can sometimes make a team quite productive.

 The process: A team without a designated leader or an informal leader can still
have leadership. The leadership is provided on a situational basis by the use of
process skills, goal setting and problem skills by all members.

Leaders' responsibilities
 Envision opportunity - identify and anticipate change
 Portray the vision and mission both internally and externally and articulate new
strategic directions
 Inspire total belief and commitment to the new directions to make the team
culture responsive
 Strive to develop and lead the team
 Achieve measurable results
 Communicate with the rest of the organization

Leadership contrasted to management

Leadership is responsible for formulating and communicating a compelling vision for
the team. (Leaders do the right thing)

Management is responsible for organizing, scheduling and implementing the vision.

(Managers do things right)

The role of the leadership and the management of the organization is to:

 Develop, know, implement and revise the shared vision.

 Assess the current and future needs of the team and its clients in concert with
the shared vision.
 Identify, organize, develop and mobilize available resources.
 Develop and modify goals based on the needs, resources, vision.
 Design and implement specific action plans.

© PetroSkills, 2013. All rights reserved. 11-13

Reservoir Characterization
Chapter 11

TEAM LEADER ROLES

 Keep teams tied to organization’s shared vision
 Provide resources team needs
 Link team to other teams
 Share experience from background/ perspective, technical audits
 Be the team advocate with management and the organization
 Keep team insulated from outside distractions

MANAGEMENT'S JOB
 Prepare the environment, discipline to maintain direction
 Select and empower talented people
 Create an inventory of challenging problems
 Create the vision for the future

SKILLS FOR MANAGING TEAMS

 Understand impact of decisions on teams
 Share opinions, decisions, positions
 Know real issues - and demonstrate it
 Decisive in team management decisions, influence openly, encourage
collaboration
 Provide constructive, candid feedback

CANDID FEEDBACK
 Clean feedback:
- Specific
- Focus on behaviour
- Descriptive
- Designed to help the receiver of the feedback

 Dirty feedback
- General
- Guessing motives of receiver
- Prescriptive
- Designed to help the giver

RC Hint: Candid feedback is a ‘present’ from the person who

gives the feedback, to the receiver of the feedback

11-14 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

RECEIVING FEEDBACK
 Clean:
- Listen
- Be silent
- Ask questions for clarification

 Dirty
- Argue
- Justify
- Attack / defend
- Rationalise (“rational lies”)
- Withdraw
- Seek revenge

© PetroSkills, 2013. All rights reserved. 11-15

Reservoir Characterization
Chapter 11

ROLES FOR MIDDLE MANAGERS

 Coach and mentor for the teams
 Train less experienced employees
 Participate on a managers team
 Serve on the teams as a full team member
 Become resource team member

SERVING LEADERSHIP
 Vision owned by the leader
Cannot be disputed!

 The serving leader creates an environment in which available talent naturally

develops the business towards the visionary future.

INTEGRATED TEAMS WITH MODERN COMPUTER METHODS ARE

KEY TO RESERVOIR CHARACTERIZATION

Bottom LINE
Embracing computer technologies can increase efficiencies and provide a positive economic
impact through integrated reservoir characterization studies.
PROBLEM ADDRESSED
Many oil and gas fields in mature basins have not benefited from interdisciplinary char-
acterization studies. These reservoirs provide an opportunity for innovative independent
operators to maximize the economic potential through integrated teams using low-cost PC-based
software packages.
TECHNOLOGY OVERVIEW
Reservoir heterogeneity, often due to compartmentalization, has a profound impact on oil and
gas recovery. Heterogeneities occur at the wellbore, interwell, and on field-wide scales. Types of
compartrnentalization can include: stratigraphic sedimentologic, structural, petrophysical/fluid,
diagenetic, pressure, and combinations. The use of integrated, multidisciplinary teams has
proven to be an effective way to increase recovery and profits. Experience confirms that putting
the right people with the right skills (and tools) in the right projects—with firm management
support—is the key to success.

Computer-related data acquisition, manipulation, and interpretation techniques must be used to

develop efficient reservoir management schemes that will maximize economic potential. Routine
map, production, and log data are available from a wide variety of sources, along with mudlogs
and borehole imaging logs. In addition, Internet resources, such as state oil and gas commission
websites, provide a wealth of useful information. In general, digital data allow easier data
manipulation, the choice of display orientations, a variety of features, and interactivity.
Computerized data manipulation is highly efficient. Database management (DBM) software has
also become more user friendly with the implementation of easier interfaces. Today, large
databases can export to applications designed for specific analyses. Raster data can be input

11-16 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

using scanners; vector data is input using a digitizing tablet or stylus. The average cost of
digitizing well data in-house is $42 to $57 per well.

FIELD RESULTS
Field studies have demonstrated several ways to use this digitized data. Examples of integrated
reservoir characterization are described below for two fields:

Sorento Field in Colorado produces primarily from flu-vial sandstones deposited in incised
valley fills cut into marine mudstones of the Morrow Formation (Pennsylvanian) on the eastern
flank of the Denver Basin. To understand compartrnentalization of the field, a reservoir
characterization study was conducted integrating geology, geophysics, and petroleum
engineering information.

Sequence stratigraphy was used to define the development of reservoir architecture. The fluvial
architecture is highly heterogeneous and compounded by several minor fluctuations of sea level
that stacked and partially eroded individual depositional stages. High-frequency 3-D
compressional seismic data were used to distinguish the incised valley outline and lateral barriers
to flow. Low-permeability carbonate-cemented zones and floodplain deposits are associated with
stacked channel sequences within the valley. These deposits created baffles and barriers to fluid
flow that were mapped using high-frequency 3-D compressional data.

When the distribution of diagenetic features was combined with sandstone geometry, the
reservoir compartments became apparent. Production data were used to corroborate geological
and geophysical interpretations. It was found that the reservoir is comprised of four flow units
that display separate fluid contacts but maintain a baffled pressure relationship. This implies that,
in developing similar fields, drilling a location structurally lower than a known oil-water contact
may be justified. A large 3-D seismic survey may pay for itself in a complex reservoir setting if
it can decrease the number of dry holes. Based on a dry hole cost of $115,000 per well and a
seismic cost of $50,000 per square mile, a 3-D survey of five square miles that prevents two dry
holes would cost the same as drilling them. If more than two dry holes were prevented, which is
likely in a large field, the 3-D survey is even more economic. Also, the operator gains
understanding that should result in more efficient development and management of the field.

Slattery Field in ~4~oming contains approximately 65 wells and produces primarily from
aeolian sandstones of the Minnelusa Formation (Permian) on the eastern flank of the Powder
River Basin. The study of South Slattery sought to: maximize oil recovery, demonstrate an inte-
grated approach to reservoir characterization, and create a “cookbook” that independent
operators could use to develop their own methodologies for reservoir characterization.

Geological work for this study was done using a desktop computer loaded with GeoGraphix GES
(well data base manager, mapping and cross section software) and GeoQuest QLA2 and RIS
LESA (log analysis software programs). Digital log data was used, even though it tends to be
expensive to acquire.

The geologist and engineer were physically located in facilities 1,000 miles apart and relied upon
e-mail for communication and file transfer. This arrangement illustrates that operators can use

© PetroSkills, 2013. All rights reserved. 11-17

Reservoir Characterization
Chapter 11

remote expertise without having to hire additional staff to complete reservoir characterization
work.

Most of the time was spent mapping, correlating, and working on cross-sections. While the
geologist had a working knowledge of the software at the beginning of the project, nearly 25% of
the time was spent learning even more about the software. If the same software were used in
future projects, the learning curve should decrease.

If digitized log data had been purchased, about 20% of the geologist’s time could have been
conserved. The amount of time the geologist spent preparing for simulation (grid preparation)
was small, suggesting that simulation would not add a significant time commitment.

Based on this study, a realistic estimate for the time a geologist needs to characterize a reservoir
is approximately one well per day. Using the mean salary for a petroleum geologist ($300 per
day according to the American Geological Institute), the cost of a geologist’s time for reservoir
characterization is $300 per well. The cost of software (amortized over the first year. with one
well per work day and 250 work days/year) is about $82 per well. The cost of data works out to
be $27 per well. Thus, the total cost of developing the geological model for reservoir
characterization is approximately $400 per well.

The highest cost consideration most likely is the geologist’s time, which is three times that of the
data and software combined. Efforts to save money should, therefore, be directed at equipment,
software, and other methods to make the time spent by the geologist more productive. Simple
suggestions for increased geological efficiency include purchasing data and using geotechnical
assistants.
Based on a workshop sponsored by PTTC’s Rocky Mountain Region and the American
Association of Petroleum Geologists, in Denver, CO.

TEAM BUILDING ELEMENTS

 Performance appraisals*
 Reward systems*
 Simplifying systems*
 Effective communications
 Clear roles /responsibilities / assignments
 Fostering innovation
 Asking for appropriate help
 Need-based training

TRAINING CATEGORIES
Interpersonal skills training
Without the conventional channels of authority for giving and receiving direction, work
team members need to talk with, explain to, agree with, disagree with, decide, listen to

11-18 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

and convince more people more frequently than ever before. In other words, they need
to be skilled communicators - both one-on-one and in groups. Interpersonal skills
training starts with the basics, like listening, giving feedback and problem solving. As
the teams evolve, training moves on to more advanced skills, such as influencing
others, handling diversity, resolving conflicts and peer reviews.

Administrative skills training

Work teams need to learn to perform many of the administrative tasks that managers
formerly did for them. Dealing with other parts of the organization is frequently needed.
Departments including accounting, marketing, land, purchasing and budgeting are
frequent contacts. Learning to deal with these groups and fulfill reporting and record
keeping needs is an essential part of establishing empowered teams. As the team
matures, tasks including performance appraisal, hiring/firing and compensation may be
assigned.

Technical skills training

Teams must develop a better appreciation for other disciplines through cross-training in
other reservoir characterization tasks. Work process skills upgrades and information
system training will support the success of the team.

TYPICAL DISCIPLINES REPRESENTED ON RESERVOIR CHARACTERIZATION TEAMS

 Geology
 Geophysics
 Petrophysics
 Reservoir engineering
 Production engineering
 Gas/chemical engineering
 Drilling
 Land
 Production operations
 Environmental
 Design/construction engineering
 Legal
 Economics
 Research/development
 Service contractors
 Management

© PetroSkills, 2013. All rights reserved. 11-19

Reservoir Characterization
Chapter 11

TEAM COMMUNICATION

DECISION MAKING TOOLS

Problem solving
One of the most important activities of teams is the solution of problems. The potential
to develop creative and insightful solutions through group process is one of the major
benefits of team building. A process to consider a resolve problems is shown in the
following steps:

 Define the problem (by identifying the symptoms and the probable cause); list
the criteria of a good solution
 Generate trial solution
 Evaluate each trial solution in the light of pertinent data
 Choose the best of the trial solution in the light of pertinent data
 Develop an implementation plan and arrange follow-up

Brainstorming
A frequently used term in analyzing issues is "brainstorming". Frequently the term is used
informally to indicate a general discussion of an issue. A more rigorous application of the
technique is shown below. This very useful technique can yield a great many useful
solutions. It is recommended that teams use the more formalized approach until they
become more skilled at brainstorming. Frequently individuals are assigned additional
responsibilities (recorder, facilitator, and spokesman). The steps are:

 Define a specific problem or a question to focus on.

 Have each team member individually write down as many ideas as possible that
might solve the problem or answer the question.

 Record and post all ideas without discussing or evaluating any them.

 Encourage people to build on other ideas and continue recording and posting all
ideas.

 Clarify the ideas. Give everyone a chance to look at the ideas and ask questions
about the meaning of any idea. Do not evaluate any of ideas at this time.

 Accept and enjoy moments of silence - often a new burst follows

 Set a time limit for the brainstorming session

Roles of Facilitator
1. Help keep the group on task

11-20 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

2. Watch the time to assure that the group completes its task
3. Assure that everyone is able to participate - no one dominating, no one
excluded
4. Remind people to listen as others are talking
5. Encourage people to respect and use their different perspectives and views

Roles of Recorder
1. Listen for key words; do not edit - use exact words
2. Capture the basic ideas
3. Write rapidly, legibly (1 to 1-1/2 inches high)
4. Number each sheet; reference topic, group name, etc.

Selection grid
A method for selecting one option out of several possibilities. It involves deciding what
criteria are important and using them as a basis for reaching a acceptable decision.

What it can be used for

 Choosing a single problem from a list of problems
 Choosing a single solution from a list of solutions

Steps in creating a selection grid

 Narrow the list of potential choices
 Choose criteria, each with a scoring system
 Make a grid with the criteria across the top and the options on the left side. Fill in
the grid to evaluate how well each option satisfies each criterion.
 Use the information on the grid to help you select a best option

Things to keep in mind

 List your criteria without regard to the options
 Use consensus rather than voting whenever possible
 The selection grid may not give you a clear-cut decision, but it provides
information helpful in making the decision

Impact analysis
A procedure to discover what impact a situation has on the organization. It involves
getting specific impacts to the organization and other available information about the
decision.

What it can be used for

 Confirming whether a project is worth pursuing
 Indicating how extensive a problem is
 Uncovering new aspects of a problem

© PetroSkills, 2013. All rights reserved. 11-21

Reservoir Characterization
Chapter 11

 Letting team members share their viewpoints and come to agreement about what
exactly is happening
 Suggesting how things should be

Steps in creating an impact analysis

 Ask each team member to describe an impact of the current situation or the
proposed change
 Discuss the information that has been listed.
 Based on you discussion, write a problem statement. This includes:
 The current problem situation
 Its impact
 The desired state of affairs
 The expected impact of correcting the problem

Possible dangers of groupthink

Groupthink is the tendency of a team or other group to be so eager for cohesion and
unanimity that it impairs its ability to think effectively, face reality and make sound
judgment. The dangers are:

 An illusion of invulnerability
 A tendency to discount warnings
 Unquestioning belief in the team
 Stereotyped views of other groups
 Pressure on any member who contradicts the team consensus
 Self-censorship by individuals
 An illusion of unanimity

Groupthink: The cure

 Encourage openness: Make it plain the team values candor, and you don’t
consider anyone exempt from criticism.
 If you are the manager or team leader, don't state your views until the rest of the
team state theirs. Remember that your voice carries weight, so much that it may
stifle discussion.
 On major decisions, consider setting up several groups to consider the same
problem.
 As a variation on this idea, break the team into sub-teams at various steps of the
decision-making process.
 If practical, let members of the team discuss the team's thinking with interested
persons who are not on the team, get their reactions and report the reactions to
the team. This is a good way to introduce new viewpoints.
 Occasionally invite qualified outsiders to sit in on a team session and encourage

11-22 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
Organizational Structure

them to challenge the team.

 Make a deliberate attempt to create "devil's advocates." In each phase of the
determination process. Make it plain that the only job during that phase will be to
pick holes in the team's thinking
 Where competition is involved, take the time to develop scenarios of their
intentions. This can best be done when the team evaluates alternative solutions.
The questions each member should consider are: If I worked for the competition,
what would I suggest doing to counteract the solution we're now considering? In
the light of such possible counteraction, how good does the solution look?
 If time permits and the problem is sufficiently weighty, hold a "second chance"
session after the team has chosen the "best" alternative. Ask each member to
express as strongly and cogently as possible any lingering doubts about the
alternative selected. If serious doubts emerge, re-open the entire process.

TOP MEETING PROBLEMS

Meetings are a necessary evil of the team process. Frequently meetings do not work as
well as planned. The reasons for unsuccessful meetings are :

 Getting off the subject

 No goals or agenda
 Too lengthy
 Poor or inadequate preparation
 Inconclusive
 Disorganized
 Ineffective leadership/lack of control
 Irrelevance of information discussed
 Time wasted during meetings
 Starting late
 Not effective for making decisions
 Interruptions from within and without
 Individuals dominate discussion
 Rambling, redundant, or digressive discussion
 No published results or follow-up actions
 No pre-meeting orientation/canceled or postpone meetings

Skills needed to lead a participate meeting

 Designing and planning
 Focusing the meeting
 Encouraging participation
 Recording people's ideas
 Managing the group process
 Organizing, connecting and summarizing data
 Bringing the group to consensus and closure

© PetroSkills, 2013. All rights reserved. 11-23

Reservoir Characterization
Chapter 11

Considerations concerning meetings

 Is the meeting necessary?
 Can the objective be accomplished more efficiently in another way?
 Who should attend the meeting?
 How long do we need?
 When should the meeting be held?
 What environment and atmosphere does the meeting call for?
 Is any pre-meeting work necessary?
 Will the group accomplish more if members prepare for the meeting ahead of
time?
 What should members bring to the meeting?
 What processes will accomplish the objective? Brainstorming? A problem-solving
model? Small-group work? An informative presentation? A short training
session? Prioritizing?
 What adjustments need to be made during the meeting?
 As the meeting progresses, is the group making progress toward its objective, or
does it need to do something different?
 Was the meeting successful?
 Did the group achieve what it set out to achieve? If not, why?
 How might the next meeting be improved?

SUMMARY OF CHAPTER 11
The dedicated team structure is very suitable for the Reservoir Characterisation
process, because it facilitates the necessary inter-disciplinary sharing that is so
important for optimum RC. Both team members and management need to have the
appropriate skills to make dedicated teams successful. Suitable training is required for
this purpose. Three training categories need to be addressed, namely interpersonal,
administrative and technical training.

In an organization with dedicated teams, giving and receiving feedback is essential.

Both must be clean. Clean feedback is owned by the giver of the feedback. Clean
receiving of feedback ensures that improvements will be made.

11-24 © PetroSkills, 2013. All rights reserved.

Chapter 12 P ROJECT M ANAGEMENT
Unfortunately. the skills, organization and toolsets of many companies have not caught
up with demands of value management, and the focus of their legacy systems and
processes still bias their thinking toward volume.
Vaughn

Project management skills are much needed in the reservoir characterization process.
Much of the work previously done by management personnel will be passed along to
the reservoir characterization team. The skills include traditional project management
skills plus the ability to understand the whole of the project within the context of the
company’s overall strategies and tactics. The process consists of a number of elements
including:

 Goals of the project within the organizational context

 Formation of the team (through alliance or partnership or in-house)
 Inventory of resources available (people, financial, data, physical)
 Projecting project timelines
 Oversight of outsource contractors

There is computer software available to facilitate these activities, but the task confronting
a reservoir characterization team charged with a large project can be quite daunting. Bob
Sneider demonstrates the activities that a reservoir characterization team needs to
undertake to accomplish the reservoir characterization goals:

 Reservoir Characterisation Team planning

– What is the team trying to do?

– Who will accomplish those tasks?

– How should people work together?

– How do the goals, roles, procedures and relationships affect people's work?

© PetroSkills, 2013. All rights reserved. 12-1

Prospect and Play Assessment
Chapter 12

GENERALIZED RC PROCESS OUTLINE

1. Define the problem
2. Evaluate data available
3. Finalize plan and reach agreement on a “performance contract” with organization
4. Build the static model
A. External geometry of reservoir
B. Internal geometry of reservoir
C. Reservoir property distribution

5. Testing model with dynamic data

6. Finalize plan and recommendations

With most reservoir characterization projects, one of the most difficult activities will be to
bring together a team that consists of representatives from several partner companies,
service companies, suppliers and the government representatives (Refer to the section
above that outlines the methods for establishing an effective strategic alliance.)
Appropriate milestone along the way to completion of the project and rewards
appropriate to the contributions expected are significant success factors. Paramount
among success factors are the focus on the team product and the adherence to the
shared vision for success of the project.

12-2 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Project Management

PROJECT PLANNING AND IMPLEMENTATION

Tyraskis, et al. have charted the RC process for Saudi Aramco. Their conclusions
provide insights to the project planning process:

Improving reservoir characterization has a direct positive impact on the company’s core
mission — the more efficient production of hydrocarbons. The perceived benefits to the
Saudi Aramco E&P organization include the following:

 Increased management awareness of the breadth of these projects;

 Better planning model for the acquisition and allocation of key resources,
such as computer hardware/software and professional expertise.

A thorough understanding of the interaction among the multiple departments included in

the characterization process. Consequently, identification of expertise and minimization
of potential bottlenecks in the data flow are obtainable.

A detailed knowledge of the time, physical resources, and people necessary for these
projects also provides a better platform from which to make decisions concerning
outsourcing of related work.

The existing model was constructed in the most general way possible to cover the
widest range of contingencies. The methodology employed in defining the IRCPM is
easily adaptable to other organizations and processes. Consequently, it does not apply
in detail to any current project, although it does come closer to some than others. The
next step in the implementation of a more rigorous planning methodology is to apply the
existing framework to a specific project and to predict project milestones, resources and
costs (Tsingas et al.).

In summary, an improved planning process when combined with conventional and

technical tools, directly benefits resource management as well as the reservoir
characterization process itself.

The chart shown below illustrates an outline for the processes of a typical reservoir
characterization project. Larger and more complex projects or those with multiple
partners will be more complex. The basic outline consists of definition of the problem
usually by a management group and more senior discipline leaders followed by
development of a preliminary plan. Resources and constraints help to define the plan
that is the next step. An important and frequently forgotten step is to identify all of the
stakeholders in the reservoir characterization process; stakeholders are any parties
that are affected by the project. Typically include are internal stakeholders
(management, professionals, support staff), partners, service companies, suppliers,
environmental groups and the government agencies. Successful plans take into
account the needs of the company plus the needs of these stakeholder groups.

© PetroSkills, 2013. All rights reserved. 12-3

Prospect and Play Assessment
Chapter 12

The plan for the reservoir characterization project should always be considered a
working document subject to change based on new data or conditions. It is very
important for the team members to have a significant hand in developing the plan which
should include not only activities, but should also define working relationships on the
reservoir characterization team and its relationship to management. The most
successful reservoir characterization teams are those with a clearly-defined and shared
vision for the project; the team members should be empowered to make routine daily
decisions and more substantial decisions in their own discipline - in other words to act
as the experts that they are. The shared vision for the project is the driving force behind
the work that is done; it is the unifying factor that all decisions should revolve around.

After development of the plan, the more technical activity of reservoir characterization
begins. Each discipline member should be charged with developing the best model
tested by the most recent techniques within the constraints of the project (time,
resources, data, and financial constraints). As has been stated many time before in this
manual, cross discipline coordination will enhance the final product immensely and
will lead to significant efficiencies. Each team member will need to develop and
exercise new ways of doing business.

Finally, the team will be given the opportunity to present its findings and
recommendation; these findings should come as little surprise since perceptive
managers will already have sampled the findings on an ongoing basis throughout the
project. The successful completion of the reservoir characterization project is only the
first step in a long multi-disciplinary approach to producing the most reserves at the
least cost.

RC Hint: Team-based project planning and accountability is

a key factor in the success of projects

VALUE CREATION

12-4 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Project Management

CHARTER THE PROJECT

 Define project's goal(s) / team purpose
 Select team members
 Identify required skill sets
 Name team leader
 Name team members
 Define roles and responsibilities

 Define outcome (deliverables)

 Identify project resources/boundaries

 Develop team approach

 Develop scoping plan
 Develop task list & milestones
 Assign time to tasks
 Assign resources

 Review charter with sponsor

 Project chartered and approved

DEVELOP DETAILED CHARACTERIZATION PLAN

 Identify Milestones and Deadlines

 Evaluate Project Data (Scoping - not interpretation)

 Petrophysical data - Gather and evaluate
 Core data - Gather and evaluate
 Structural tops data - Gather and evaluates
 Stratigraphic data - Gather and evaluate
 Seismic data - Gather and evaluate
 Reservoir data
 Production data/injection - Gather and evaluate
 Fluid data

 Assign Work / Duration to Tasks

 Assign Individuals to each Task

© PetroSkills, 2013. All rights reserved. 12-5

Prospect and Play Assessment
Chapter 12

SET THE BASELINE PLAN

 Develop and document predictions
– Document key assumptions
– Review the project plan
– Review plan with customers
– Review plan with suppliers
– Review plan with interested parties
 Develop required contingency plans
 Verify tracking mechanism
 Team review
 Review the project plan with sponsor
 Contract with required consultants / suppliers
 Detailed project plan complete

The planning process described above seems long and cumbersome, but can save a
great deal of time later in the project. Even though there are a great many steps to the
planning process, these are accomplished rather quickly –m particularly if experienced
professionals are accessed in a collaborative manner. The steps that follow are the
heart of the RC process and comprise the bulk of the time in characterizing reservoirs.
The process follows from development of a static model to development of a dynamic
model to test the model. Many of the steps in the process are contemporaneous and
not sequential. Note that there are many iterative loops built into the RC process so
that interpretations can be enhanced based upon other data sets or subsequent
interpretations.

STATIC MODEL - DATA ANALYSIS AND INTERPRETATION

Develop integrated database

 Accessible across disciplines

 Petrophysical data
 Normalize/calibrate/edit against cores
 Identify cutoffs for mapping

 Core data
 Fully describe/interpret cores
 Correlate with logs
 Calibrate with test/production data

 Seismic data
 Develop structural interpretation
 Seismic stratigraphic correlation
 Establish correlations with reservoir parameters

12-6 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Project Management

 Map distribution of parameters

 Structural tops data

 "Establish tops - all horizons, faults, etc"
 Integrate with seismic data
 Create integrated maps
 "Create tabular display of well data (TVD, TVT, TST, etc)"

 Define facies

 Establish depositional model

 Define reservoir architecture

Distribute reservoir parameters throughout reservoir

STATIC DATA MODEL

Data Form
Reservoir: Ave. Depth: Ave. Thickness:
Type Well: Interval:

Stratigraphy Observed (lithology, source, seal, reservoir?):

Overlying zone:
Undelying zone:

General Characteristics:
Lithology Porosity amt. Porosity type Permeability
Interp. Method:
Net/Gross Water sol.
Interp. Method:
Texture Fabric
Interp. Method:

Interpreted Depositional Environment:

Primary reasons for interpretation

Analogues:
Modern
Stratigraphy Observed (lithology, source, seal, reservoir?):
Overlying zone:
Undelying zone:

Ancient
Stratigraphy Observed (lithology, source, seal, reservoir?):
Overlying zone:
Undelying zone:

© PetroSkills, 2013. All rights reserved. 12-7

Prospect and Play Assessment
Chapter 12

Reservoir Dimensions
Modern Min ML Max Ancient Min ML Max
Length Length
Width Width
Thickness Thickness

Heterogeneities
Stratigraphic Modern Min ML Max Ancient Min ML Max
Length Length
Width Width
Thickness Thickness

Expected Min ML Max

Length
Width
Thickness

Frequency expected Distribution

Structural Mechanism Distribution

Control

Factors Controlling Reservoir Distribution

Modern Ancient Expected

Production Profile
Analogue Expected

Area Drained (incude method of determining)

Analogue Expected

Critical Elements
Analogue Expected

Key Mapping Tools

Assumptions

12-8 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Project Management

STATIC MODEL - DATA ANALYSIS AND INTERPRETATION - 3

 Reservoir data
 Input petrographic data
 Integrate geol., prod., test data"
 Determine morphology of pore system
 Interpret reservoir architecture

 Production data/injection

 Fluid data
 Develop flow unit correlations

 Redefine stratigraphic intervals

 Modify interpretations as needed

DYNAMIC MODEL - DATA ANALYSIS AND INTERPRETATION

 Test model within pilot area
 Annotate assumptions for model - range of parameters
 Generate and rank multiple realizations

 Carry out material balance calculations

 Carry out sector model simulations to understand concepts

 Identify critical elements and likelihood of ranges

 Modify model as needed

 Full field simulation

DEVELOP PROJECTIONS AND RECOMMENDATIONS

 Modify plan
 Test again as above
 Implement

© PetroSkills, 2013. All rights reserved. 12-9

Prospect and Play Assessment
Chapter 12

ID Name Start Tools Duratio % DonInRespons Predec

RESERVIOR CHARACTERIZATION PROJEC Jul 17 '9 680.66 0%
1 Charter the Project Jan 03 ' 2.75d 0%
17 Develop Detailed Characterization Plan Jul 17 '9 669.41 0%
45 STATIC MODEL - Data Analysis and Inte Feb 09 '0 5d 0% 17
81 DYNAMIC MODEL - Data Analysis and In Feb 16 '0 1d 0% 78
90 Evaluate Plan Appropriateness Feb 14 '0 4.25d 0% Team Lea77
103 Check if Major Deviations Require Replann Feb 16 ' 8h 0% Team Lea102
104 (ACTIONPL.MPD) - Project Implementation Feb 14 ' 4d 0% 91,92
105 Analyze Project Performance Feb 18 '0 4d 0% 104
116 End ProjPro, Project Goal Attained Feb 24 ' 0d 0% 115
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132

12-10 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Project Management

During the Second International Reservoir Characterization Conference held

in Dallas, a work group offered the three following suggestions to increase
cooperation on teams (underscoring added for emphasis):
1) The mission of the team should be based on producing an economically
successful effort. The objectives should be focused and defined in
relation to the realities of dollar/time constraints and personnel allotted to
the effort. Common objectives should be defined. The kinds and amounts
of data that will be needed to arrive at the results should be identified.
The project should be organized into a series of milestones and expected
results. There must be built-in flexibility such that the effort is sensitive
and adaptable to changing economic and management objectives.

2) The team should be located in close proximity (i.e., clustered in a series of

offices that share a common workroom), but should be physically
separated from individual team member's management. The project
leader should be a working member of the team. The team members
should be dedicated to the project for an extended period (6 months to
years). A core group should be maintained at all times for continuity;
however, individuals may be shifted into and out of the team as their
specialty is needed. Allowance for creativity (i.e., brainstorming/innovative
approaches) should be encouraged if this will contribute to project cost
efficiency and technical competitiveness. The team should organize their
own efforts so that there is a timely flow of data to maximize group
efficiency. Team members should be picked such that individuals are
amenable to teach/learn more about each other's discipline and data
format requirements. Frequent, informal meetings of team members
should occur to keep all members informed of any changes in project
emphasis or direction resulting from management decisions. Regular
meetings with management should be scheduled and these should
include most team members when discussing progress or making
recommendations.

3) Database development and management should be computer based and

developed for the long-term. Both historical and currently developed
information should be incorporated into the database. The data format
should be broad in that it incorporates data from all disciplines and at all
scales of reservoir heterogeneity. The format should be flexible enough to
allow a wide variety of end user products that are a compilation of data
from several disciplines/scales. Analog field data for comparison of
simulations to actual successful efforts based on geological depositional
environmental models should be available.

© PetroSkills, 2013. All rights reserved. 12-11

Prospect and Play Assessment
Chapter 12

SETTING RESERVOIR CHARACTERIZATION TEAM GOALS

FACTORS INFLUENCING TEAM FUNCTIONING

A useful scheme for analyzing team functioning is to ask the following questions:

 What is the team trying to do - what are its tasks? (goals)

 Who - which people, resources - are required in order to accomplish those tasks?
(roles)

 How should people work together in performing their roles? (procedures)

 How do the goals, roles, procedures and relationships affect people's work?
(interpersonal)

In determining answers to these questions, it is important to remember that there is no

one best answer or set of answers that applies to all teams. There are, however, some
qualities associated with the factors listed above that represent issues to work on in
order to improve team functioning:

GOAL SETTING
Only after the team has considered its vision and mission statement, identified its
needs, identified available and required resources, should work begin on goal setting.
Any move to set goals without the appropriate background is premature.

CHARACTERISTICS OF GOALS
 Specific
 Measurable
 Achievable
 Realistic
 Timely

Items to consider when setting goals. Each goal should:

 Relate to the statement of purpose.

 Address the needs of the organization and the team.
 Be achievable.
 Be valued, shared and desired by the team and the organization.
 Specific, measurable and observable.
 Include support and field personnel in the goal setting process.

12-12 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Project Management

CRITERIA FOR EFFECTIVE JOB GOALS

The potential of reservoir characterization teams is maximized when the team and each
individual fully understands what is to be accomplished and how the team and
individuals are to be held accountable. The job goals, therefore, should have the
following characteristics:

 Realistic
 Requires optimal effort
 Specific
 Comprehensive
 Understandable

ONE WAY TO SET SUCCESSFUL GOALS

A method of writing goals that has worked successfully in several organizations for
many years is outlined below:

 Management sets overall goals for the organization compatible with its shared
vision and available assets (technical, financial, people)
 The reservoir characterization team writes goals for the project based on the
management guidelines
 The management may outline goals independently of the reservoir
characterization team
 During a working session, the reservoir characterization team presents a plan
that is accepted or modified based on information from the management about
priorities, budgets, resources, etc.
 The reservoir characterization team and management agree to a performance
contract with specific action plans to achieve the goals and periodic review
sessions (including specific dates and times during which the goals will be
modified as needed or reviewed against performance milestones)
 At the end of the period the reservoir characterization team evaluates itself (the
management will also conduct an evaluation, but will not share that evaluation
except where disagreements occur)
 During a forward-looking meeting, the management and reservoir
characterization team will review the project, analyze opportunities and threats,
strengths and weaknesses - this is a good time to reaffirm the shared vision and
to discuss professional growth opportunities for individuals in a constructive,
collaborative manner

This process works equally well with individuals as well as with teams; it focuses the
accountability to perform directly on the individual (or team) and builds an ownership in
performance and output driven work.

© PetroSkills, 2013. All rights reserved. 12-13

Prospect and Play Assessment
Chapter 12

TYPES OF GOALS
There are many types of goals that can be used with teams and individuals:

 Fixed target goals - usually output or production based (example - number of

maps produced, well histories reviewed, etc.)

 Improvement goals - changes in rates (example - reduced lifting cost, quicker

action plan design, etc.)

 Continuing challenge goals - (example - bring field recovery rate to 45%, etc.)

 Moving target goals - something to aim for, but is probably not achieved

 Behavioral goals - personal or team interpersonal skills (example - develop

capability to listen to other viewpoints before forming conclusions, etc.)

ACTION PLANS
In order to attain goals, specific action plans are needed. Action plans should include
timetables, personnel responsible for each task, tasks to be accomplished and the order
in which the tasks are to be completed. These plans will also provide a method for
evaluating progress toward the goals. Project planning software can assist the process
significantly.

When developing and implementing action plans consider:

 How action plans relate to mission, needs, resources and goals

 Specific details of what is to be accomplished
 Assign action plans to individuals
 Coordinate and share timelines for action plans with other related tasks
 Evaluate and revise action plans based on progress

SELLING YOUR PROJECT

 Meet minimum economic parameters
 Clarity of purpose - business rational
 Alignment with organizational goals
 Needs and criteria of your customers
 Support from the rest of the organization
 Multi-disciplinary approach
 State-of-the-art
 Balance detail and enough data
 Well planned presentation
 Clearly articulated recommendations

12-14 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Project Management

COMMUNICATING WITH THE ORGANIZATION

 Most of organization doesn’t understand what you do
 Some “magic” applied to RC
 “Don’t live in real world”

 Share conclusions to allay these fears

 Always plan presentations

 Present findings based on their needs
 Spare the details (unless asked)

 Keep customer in mind

PRESENTING YOUR CASE

 General structure
 What you will say
 The message
 What you told them

 How did you build models?

 Explain what you numbers mean

 Mean reserves
 Range of reserves
 Adequacy
 Minimum economic size

SUMMARY OF CHAPTER 12
Successful projects create value for the organization. Most value is created in the early
stages when projects are envisaged and designed.

Team based planning of project ensures that optimum inter-disciplinary sharing is

realized.

Projects must fulfill the SMART criteria.

Projects are carried out for the benefit of the organization. Therefore, clarity about the
projects carried out must be visible in the organization.

© PetroSkills, 2013. All rights reserved. 12-15

Prospect and Play Assessment
Chapter 12

12-16 © PetroSkills, 2013. All rights reserved.

Chapter 13 P ROBLEMS , P ITFALLS , AND
T HE F UTURE OF R ESERVOIR
C HARACTERIZATION
In geology, the innovative, creative act is to identify the question - the answer will follow
with hard work.
Michael I. Treesh - personal philosophy (2011)

Reservoir characterization is an activity that is rapidly evolving and in constant change.

New advances in technology will continue to modify approaches to understanding
reservoirs and developing producing properties. It will be critical to continue to focus on
the processes and principles that govern oil and gas accumulations - geological,
geophysical and engineering principles.

PROBLEMS, PITFALLS

Some of the major pitfalls and problems that may be encountered are discussed below.

Failure to keep customer in mind

Many otherwise successful RC team forget who their customers are. Effective
teams frame their recommendations and reports in terms useful to the customers of
the process – whether this is the most satisfying way to present the results or not.

Identifying critical controls

It is difficult - sometimes impossible - to identify the controlling parameter(s) in an oil
and gas accumulation. Two apparently identical accumulations may have separate
mechanisms that in one case create an excellent field and in another yield a
mediocre field.

Dependence on technology
Good geologic and engineering reasoning are more important than the dependence
upon the latest technology without a thorough understanding of the technology -
there are no “black box” solutions.

Understanding scales
As indicated above, the reservoir characterization team must be using common
terms with common definitions. Care in understanding the scale represented by a
data set will increase understanding and avoid misunderstandings.

© PetroSkills, 2013. All rights reserved. 13-1

Prospect and Play Assessment
Chapter 13

Fractures
The influence of fractures in reservoirs is just beginning to be understood, but the
knowledge is not widespread or well understood.

Narrowing ranges of expected parameters

Too frequently RC professionals feel pressured to choose a single value rather than
a range of values. The range may represent the level of understanding much better
and may capture subtle variation in the reservoir.

Diagenesis
The influence of diagenesis is profound is many reservoirs, but many professionals
simply map primary depositional environments and then apply interpretations based
on those primary environments.

Dependency of risk
Not understanding the significance of dependency and independence in evaluating
risks can yield significant errors.

Risking versus reality

Risking is a useful tool to help plan the business of an oil and gas company, but
when considering an individual accumulation, there is only one correct answer - no
matter how many scenarios we may run (the true answer may not even be
represented by one of those realizations).

Flow physics
Another area that is yet not adequately understood. Significant potential exists to
improve interpretations with concentration on identifying and understanding flow units.

Inadequate use of analogues

Analogous production can provide many insights into an accumulation, but care
needs to be exercised to apply the analogue appropriately. Use the comparable
field to understand processes and as a source of questions to ask - not as a source
of answers.

Inadequate organizational structure/culture

Organizations that do not empower reservoir characterization teams to complete
their responsibilities without interference - direct or subtle - are losing a great tool to
enhance profitability and efficiency for the organization.

Neglecting priorities
The tendency to solve every problem to the ultimate degree is wasteful of time and
effort. Some activities require a 95% solution others need only an 80% effort;
identifying priorities is a critical project management skill.

13-2 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Problems, Pitfalls, and the Future of Reservoir Characterization

THEMES FOR SUCCESSFUL RESERVOIR CHARACTERIZATION

 Data to information to knowledge

 Identify and understand critical elements
 Multi-disciplinary input to data planning, collection and interpretation
 Data in a usable, accesible form
 Measured versus interpreted data
 Multiple working hypotheses
 Check perceptions against reality
 Integrated planning to achieve goals of RC team
 Understanding economic impact of activity

THE FUTURE OF RESERVOIR CHARACTERIZATION

Reservoir characterization will continue to bean important tool for developing oil and gas
accumulations. The future will probably contain a great number of technical
advancements - particularly in the manipulation, integration and visualization of data -
as well as organizational efficiencies. Advances in our understanding of:

 flow within reservoirs

 affects of fractures (both natural and induced)
 prediction and mapping of diagenesis
 increasing resolution of seismic data
 geostatistical modeling
 attribute extraction from seismic data
 Models that avoid upscaling
 Evaluation of marginal resources (shale gas, tight reservoirs, etc.)
 Integration of data bases and computer platforms

Organizational advances have bees underway for many years; continuation of the trend
to recognize the value, contribution and impact of the working professionals will
enhance the reservoir characterization effort. Outsourcing and the formation strategic
alliances will expand the cross-disciplinary communication demands on working
professionals. A new set of skills will have to be developed at all levels in the
organization. In summary, a combination of procedural, organizational and technical
advances will drive the oil and gas companies of the next century.

Future growth was summarized in the training manual Oil and Gas Teams: How to
make them work (Treesh):

The efficiency of oil and gas organizations is dependent, in part, on the

efficiency of the technology that is applied to each problem, the processes
through which the work is done to solve problems and create wealth and

© PetroSkills, 2013. All rights reserved. 13-3

Prospect and Play Assessment
Chapter 13

the efficiency of the individuals who are performing the critical tasks.
Traditionally there has been a great deal of attention paid to the upgrading
of technical applications in the search for, development of and production
of oil and gas reserves. In addition, in recent years there has been a
dedicated effort in most companies to assure that the most "efficient"
processes are applied to each phase of exploration/production. The area
that has received little attention is that of the application of human
resources; team building through its associated human efficiency
elements is an important method through which to significantly improve
the human element. …concurrent improvement in each of these area can
yield significant magnitudes of improvement in the productivity and
profitability of oil industry companies.

To conclude this treatise on Reservoir Characterisation, the recommended RC

approach is summarized below:

The icons represent the following:

- The jigsaw states that the geological, petrophysical, dynamic and economic model
are one integrated model.

13-4 © PetroSkills, 2013. All rights reserved.

 Prospect and Play Assessment
Problems, Pitfalls, and the Future of Reservoir Characterization

- The triangle signifies that in the business decisions have to be taken under risk
based on evidence produced by activities in the Reservoir Charaterisation (RC)
process.
- The P-model says that in the business a profit needs to be made, that the business
must have a purpose, that in the business three aspects need to be addressed,
namely the technical plan, the people issues and the organization with processes,
and finally that the business has to be sustainable.
- The steps represent the six steps in the RC process. Note that in step three the
agreement of the RC plan is concluded.
- In the RC process, critical elements are studied that support the decisions in an
optimum way as shown by the exclamation mark.
- The solution space is represented by the circle, in which multiple realisations (M1
etc.) are shown. If reality (R) is also within the solution space, then the RC modeling
will be meaningful.
- The inter-connected circles manifest that an integrated effort is required with
scientists and engineers from different disciplines as required for the particular RC
study. Note that not only the traditional subsurface disciplines should be present,
but also people from disciplines like legal, environmental etc. as required by the
study.
- The RC team should work as a super-team as shown by the graph with the different
phases in team formation.
- Last but not least, everything in RC must be SMART: specific, measurable,
achievable and agreed, realistic and time-bound.

In words this RC model is given by:

© PetroSkills, 2013. All rights reserved. 13-5

Prospect and Play Assessment
Chapter 13

SUMMARY OF CHAPTER 13
Reservoir Characterisation is continuously developing in order to meet the challenges of
the modern time and the future. As resources become more difficult to develop and
more marginal, continuous improvement in technology will be required.

The integrated reservoir characterization process is an intelligent process. Technical

professionals who need to carry out this process, need a great deal of flexibility so that
the business, the critical elements and all the uncertainties are always addressed
appropriately. Anything done for this purpose must be SMART. Then, the risks of any
decision will be fully understood and indeed the right decisions will be taken.

13-6 © PetroSkills, 2013. All rights reserved.

Chapter 14 R EFERENCES

GENERAL REFERENCES

Geostatistics in Petroleum Geology, Dubrule, Olivier, 1998,AAPG Continuing Education

Course Note Series #38Oil and Gas Teams: How to Make Them Work, 1995
revised 1999, Treesh, M. I., Training manual published by OGCI.
Integrated Petroleum Reservoir Management: A Team Approach, 1994, Satter, Abdus,
and Thakur, Ganesh, Pennwell Books, 335p.
Reservoir Characterization 1986, edited by Lake, L.W. and Carroll, Jr., H.B., Academic
Press.
Reservoir Characterization II, 1991, edited by Lake, L.W., Carroll, Jr., H.B., and
Wesson, T.C., Academic Press.
Reservoir Characterization III, 1993, edited by Linville, Bill, Pennwell Book
Reservoir quality prediction in sandstones and carbonates, 1997, J.A Kupecz, I. Gluyas,
and S. Bloch, eds., AAPG Memoir 69.
Stochastic Models And Geostatistics: Principles, Methods, And Case Studies, 1994, edited
by Yarus, J.M. and Chambers, R.L, edited by Yarus, J.M. and Chambers, R.L.
The Integration of Geology, Geophysics, Petrophysics and Petroleum Engineering in
Reservoir Description and Management, 1991Proceedings of the First Archie
Conference, Oct 22-25, 1990, published by AAPG.
Well Logs, Cores, And Outcrops, 1990, AAPG Methods in Exploration Series, no. 7, 55p.

REFERENCES CITED

Ainsworth, R. Bruce, Sanlung, Montree and Duivenvoorden, S. Theo C.,, 1999

Correlation Techniques, Perforation Strategies, and Recovery Factors: An
Integrated 3-D Reservoir Modeling Study, Sirikit Field, Thailand, AAPG Bulletin,
V.83, No. 10 (October 1999), P. 1535—1551.
Antonellini Marco, and. Mollema, Pauline N, 2000, Dolomite: Triassic Sella Group,
Northern Italy, AAPG Bulletin, V. 84, No.3 (March 2000), P. 314—344.
Araktingi, U.G., Bashore, W.M., Tran, T.T.B., Hewett, T.A., 1993, Integration of Seismic
and Well Log Data in Reservoir Models, pp. 515-555, in Reservoir Characterization
III, edited by Linville, Bill, Pennwell Books.
Anselmetti, Flavio S. and Gregor P. Eberli, 1999, The Velocity-Deviation Log: A Tool to
Predict Pore Type and Permeability Trends in Carbonate Drill Holes from Sonic
and Porosity or Density Logs, AAPG Bulletin. V 83. No 3 (March 1999). P 450-466
Arestad, J. F. et al., 1995, 3D, 3C Seismic Characterization of Joffre Field, Alberta Canada,
(abs.), Proceedings 1995 Society of Exploration Geophysicists, Annual meeting.

© PetroSkills, 2013. All rights reserved. 14-1

Reservoir Characterization
Chapter 14

Bashore, W.M., Araktingi, U.G., Levy, M., Schweller, W.J., 1994, Importance of a
Geological Framework and Seismic Data Integration for Reservoir Modeling and
Subsequent Fluid-Flow Predictions pp. 159-176, in Stochastic Models And
Geostatistics: Principles, Methods, And Case Studies, edited by Yarus, J.M. and
Chambers, R.L, edited by Yarus, J.M. and Chambers, R.L.
Begg, S.H., A. Kay, E.R. Gustason and P.F. Angert, 1996, Characterization of a
Complex
Fluvial-Deltaic Reservoir for Simulation, SPE Formation Evaluation, Sep. 1996, p. 147-153.
Bracco, Guido L Gartner and Wolfgang Schlager, 1999, Discriminating Between Onlap
and Lithologic Interfingering in Seismic Models of Outcrops, AAPG Bulletin, V 83.
No 6 (June 1999), P. 952-971.
Chambers, R.L., Zinger, M.A., and Kelly, M.C., 1994, Constraining Geostatistical
Reservoir Descriptions with 3-D Seismic Data to Reduce Uncertainty, pp.143-158,
in Stochastic Models And Geostatistics: Principles, Methods, And Case Studies,
edited by Yarus, J.M. and Chambers, R.L.
Cross, T.A., Field-Scale Reservoir Characterization. 1986, Pp. 493-496, in Reservoir
Characterization II, edited by Lake, L.W., Carroll, Jr., H.B., and Wesson, T.C.,
Academic Press.
Dehghani, Kaveh, Paul M. Harris, Kelly Edwards, and William T. Dees, 1999, Modeling
a Vuggy Carbonate Reservoir, McElroy Field, West Texas, AAPG Bulletin, V 83.
No 1 (January 1999). P 19-42
Deutsch, C.V., and Journel, A.G., 1994, Integrating Well Test-Derived Effective
Absolute Permeabilities in Geostatistical Reservoir Modeling, pp. 131-142, in
Stochastic Models And Geostatistics: Principles, Methods, And Case Studies,
edited by Yarus, J.M. and Chambers, R.L.
Dubrule, Olivier, 1998,Geostatistics in Petroleum Geology; AAPG Continuing Education
Course Note Series #38Eschard, R, P. Lemouzy, C. Bacchiana, G. Desaubliaux, J.
Parpant and B. Smart, Combining Sequence Stratigraphy, Geostatistical
Simulations, and Production Data for Modeling a Fluvial Reservoir in the Chaunoy
Field (Triassic, France), AAPG Bulletin. V 82, No 4 (April 1998).
Girgis, John, Sneider, R.M., and Thomas, B., 1995, Multi-disciplinary Teams - What is
the “Right” Structure? Based on Ten Years of MTS’s in Indonesia, pp. 545-564,
Indonesian Petroleum Association, Proceedings.
Grotsch Jurgen and Christophe Mercadier, 1999, Integrated 3-D Reservoir Modeling
Based on 3-D Seismic: The Tertiary Malampaya and Camago Buildups, Offshore
Palawan, Philippines; AAPG Bulletin, V.83, No. 11 (November 1999), P. 1703—
1728.
Haldorsen, H. H. and. Damsleth, E, 1990, Stochastic Modeling, pp. 404—112, Journal
of Petroleum Technology, (April 1990).
Haldorsen, H.H., and Chang, D.M., 1986, Notes on Stochastic Shales; From Outcrop to
Simulation Model, pp. 445-486, in Reservoir Characterization edited by Lake, L.W.
and Carroll, Jr., H.B., Academic Press.
Hardage, Bob A,. Carr, David L Robert, Finley, J., Lancaster, David E., Elphick, Robert
and Ballard, James R, 1995, Secondary Natural Gas Recovery: Targeted
Applications For Infield Reserve Growth In Midcontinent Reservoirs, Boonsville
Field, Fort Worth Basin, Texas; GRI-95/0454.2, VOLUME II: APPENDIX

14-2 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
References

Hardage, B.A., and Fertl, W.H., 1991, Synergism in Formation Evaluation Enhances
Detailed Reservoir Description, pp. 123-129, in The Integration of Geology,
Geophysics, Petrophysics and Petroleum Engineering in Reservoir Description and
Management, Proceedings of the First Archie Conference, Oct 22-25, 1990,
published by AAPG.
Hart, Bruce S, 1999, Geology plays key role in seismic attribute studies, Oil and Gas
Journal, July 12, 1999, P 76-80.
Hart, Bruce S., David M. Sibley, and Peter B. Flemings, 1997, Seismic Stratigraphy,
Facies Architecture, and Reservoir Character of a Pleistocene Shelf-Margin Delta
Complex, Eugene Island Block 330 Field, Offshore Louisiana, AAPG Bulletin, V
81.No 3 (March 1997). P 380-397
Hohn, Michael Ed., Ronald R. McDowell, David L. Matchen, and Ana G. Vargo, 1997,
Heterogeneity of Fluvial-Deltaic Reservoirs in the Appalachian Basin: A Case
Study from a Lower Mississippian Oil Field in Central West Virginia, AAPG Bulletin.
V 81. No. 6 (June 1997). P. 918-936
Holland, D.S., Leedy, J.B., and Lammlein, D.R., 1990, Eugene Island Block 330 Field -
USA, Offshore Louisiana, pp. 103-144, in Structural Traps III, Treatise of
Petroleum Geology, Atlas of Oil and Gas Fields, published by AAPG.
Hoover, Andrew R., Burkhart, Tucker, and Flemings, Peter B.; 1999, Reservoir and
Production Analysis of the K40 Sand, South Timbalier 295, Offshore Louisiana,
with Comparison to Time-Lapse (4-D) Seismic Results; AAPG Bulletin, V. 83, No.
10 (October 1999), P. 1824—1641.
Jones, Alistair, McGill, Colin, and Williams, John, 1993, Renormalization: a Multilevel
Methodology for Upscaling, pp. 823-828, in Reservoir Characterization III, edited
by Linville, Bill, Pennwell Books.
Journel, A.G.,1994, Geostatistics and Reservoir Geology, pp. 19-20, in Stochastic
Models And Geostatistics: Principles, Methods, And Case Studies, edited by
Yarus, J.M. and Chambers, R.L.
Kuhn, Tibor, and Kolosi, Zsolt, 1993, How to Estimate the Reliability of Reserves? A
Quick and Practical Solution, pp. 769-780, in Reservoir Characterization III, edited
by Linville, Bill, Pennwell Books.
Kupfersbergerz, H. and C. V. Deutsch, 1999, Methodology for Integrating Analog
Geologic Data in 3-D Variogram Modeling, AAPG Bulletin. V. 83. No 8 (August
1999) P 1262- 1278
Lasseter, T. J., Waggoner, J. R. and Lake, L. W., 1986, Reservoir Heterogeneities and
Their Influence on Ultimate Recovery, in Reservoir Characterization edited by
Lake, L.W. and Carroll, Jr., H.B., Academic Press.
Lia, Oddvar, Henning Omre, Hakon Tjelmeland, Lars Holden, and There Egeland, 1997,
Uncertainties in Reservoir Production Forecasts, AAPG Bulletin. V 81.No 5 (May
1997). P 775-802
Loren J.D, Kuhla, J.T., Renbarger, K.S., and Sneider, R.M., MacDonald, A. C. and
Aasen, T.O., 1994, A Prototype procedure for Stochastic Modeling of Facies Tract
Distribution in Shoreface Reservoirs, pp.91-108, in Stochastic Models And
Geostatistics: Principles, Methods, And Case Studies, edited by Yarus, J.M. and
Chambers, R.L.

© PetroSkills, 2013. All rights reserved. 14-3

Reservoir Characterization
Chapter 14

Mansoori, John, 1994, A review of Upscaling Procedures: Advantages and

Disadvantages, pp. 65-74, in Stochastic Models And Geostatistics: Principles,
Methods, And Case Studies, edited by Yarus, J.M. and Chambers, R.L.
Martin, Alden J., Stephen T. Solomon, and Dan J. Hartmann, Characterization of
Petrophysical Flow Units in Carbonate Reservoirs, AAPG Bulletin V 81.No 5 (May
1997), P 734-759
Montgomery, Scott L., 1996, Stewart Field, Finney County, Kansas: Seismic Definition
of Thin Channel Reservoirs, AAPG Bulletin. V. 80, No. 12 (December 1996). P.
1833-1844
Mountjoy, E.W., and X.M. Marquez, 1997, Predictin1 reservoir properties in dolomites:
Upper Devonian Leduc buildups, Deep Alberta Basin, in J.A Kupecz, I. Gluyas,
and S. Bloch, eds.. Reservoir quality prediction in sandstones and carbonates
AAPG Memoir 69, p. 267—306.
North Colin P. and Katy S. Taylor, 1996, Ephemeral-Fluvial Deposits: Integrated
Outcrop and Simulation Studies Reveal Complexity, AAPG Bulletin. V 80; No 6
(June 1996). P 811-830
Olea, R., ed., 1991, Geostatistical glossary and multilingual dictionary: Oxford, Oxford
University Press, 177p.
Paulsson, B.N.P., 1991, Cross-Well Seismology - A tool for Reservoir Geophysics, pp.
460-477, in Reservoir Characterization II, edited by Lake, L.W., Carroll, Jr., H.B.,
and Wesson, T.C., Academic Press.
Peijs-van Hilten, Madeleine, Timothy R. Good, and Brian A. Zaitlin, 1998, Heterogeneity
Modeling and Geopseudo Upscaling Applied to Waterflood Performance Prediction
of an Incised Valley Reservoir: Countess YY Pool, Southern Alberta, Canada,
AAPG Bulletin, V 82. No. 12, (December 1998). P. 2220-2245.
Peebler, R.P., 1997, Innovation at the Intersections, Offshore Magazine, v. 57, no. 2,
Feb.1, 1997. Press, 177 p.
Peterson Richard E, Paul T. Branagan, Norman R. Warpinski, Timothy B. Wright, 1996,
Seismic: Imaging Develops New insight into Fracture Growth and Dimensions, gas
Tips, Spring 1996, Gas Research Institute
Prather, B. E, J. R. Booth, G. S. Steffens, and P. A. Craigs, 1998, Classification,
Lithologic Calibration, and Stratigraphic Succession of Seismic Facies of Intraslope
Basins, Deep-Water Gulf of Mexico, AAPG Bulletin. Y. 82. N. 5A (May 1998 Pan
A). P 701-728.
Raza, S. H., 1992, Data Acquisition and Analysis: Foundational to Efficient Reservoir
Management, pp. 466-468, Journal of Petroleum Technology.
Reynolds, A D, 1999, Dimensions of Paralic Sandstone Bodies, AAPG Bulletin. V 83.
No 2 (February 1999). P 211-29
Reynolds, A. D. M. D. Simmons, M. B. J. Bowman, J. Henton,. C. Brayshaw, A. A. Ali-
Zade, I. S. Guliyev, S. F. Suleymanova, E. Z. Ateava, D. N. Mamedova, and R. O.
Koshkarly, 1998, Implications of Outcrop Geology for Reservoirs in the Neogene
Productive Series: Apsheron Peninsula, Azerbaijan, AAPG Bulletin. V 82. No 1
(January 1998). P 2-9
Reynolds, Rodney, 1999, Waterflood in Kansas field should boost recovery by five
million bbl; September 1999, Petroleum Technology Digest, Supplement to World
Oil

14-4 © PetroSkills, 2013. All rights reserved.

 Reservoir Characterization
References

Sailer, Arthur F, Dickson, J.A.D. and Matsuda, Fumiaki, 1999, Evolution and Distribution
of Porosity Associated with Subaerial Exposure in Upper Paleozoic Platform
Limestones, West Texas; AAPG Bulletin, V.83, No. 11 (November 1999), P.
1835—1854
Satter, Abdus, and Thakur, Ganesh, 1994, Integrated Petroleum Reservoir
Management: A Team Approach, Pennwell Books, 335p.
Sneider, R. M., 1996, Multidisciplinary Teams: How and Why They Make Your Money,
AAPG Continuing Education Course Notes #37, 140 p.
Sneider, R. M., and Johnson, H.D.,1996, Development Geology, . Training manual
published by OGCI.
Sprunt, E.S., 1993, Quality Control of Special Core Analysis Measurements, pp. 254-
269, in Reservoir Characterization III, edited by Linville, Bill, Pennwell Books.
Srivastava, R. Mohan, 1994, An Overview of Stochastic Methods for Reservoir
Characterization, pp. 3-16, in Stochastic Models And Geostatistics: Principles,
Methods, And Case Studies, edited by Yarus, J.M. and Chambers, R.L.
Stalkup, F.I., 1986, Permeability Variations Observed at the Faces of Crossbedded
Sandstone Outcrops, pp. 141-180. in Reservoir Characterization edited by Lake,
L.W. and Carroll, Jr., H.B., Academic Press.
Tillman, R.W. and Pitman, E.D., 1991, Reservir Heterogeneity In Valley-Fill Candstone
Reservoirs, Southwest Stockholm Field, Kansas, pp. 51-105, in Reservoir
Characterization III, edited by Linville, Bill, Pennwell Books.
Tinker, C.N., Chambers, L.D., Ritch, H.J., McRae, C.D., and Keen, M.A., 1991, The
Benefits of a Synergistic Approach to Reservoir Characterization and Proration,
Rose City Prairie du Chien Gas Field, Ogemaw County, Michigan, pp.- 360-372, in
The Integration of Geology, Geophysics, Petrophysics and Petroleum Engineering
in Reservoir Description and Management, Proceedings of the First Archie
Conference, Oct 22-25, 1990, published by AAPG.
Tobin, R.C., 1997, Porosity prediction in frontier basins: a systematic approach to
estimating sub- surface reservoir quality from outcrop samples, in JA. Kupecz, J.
Gluyas, and S. Bloch, Reservoir quality prediction in sandstones and carbonates:
AAPG Memoir 69, p. 1-18
Tucker, Karla E., Paul M. Harris, and Richard C. Nolen-Hoeksemn, 1998, Geologic
Investigation of Cross-Web Seismic Response in a Carbonate Reservoir, McElroy
Field, West Texas, AAPG Bulletin, V 82, No 8 (August 1998), P. 1463-1503
Treesh, M. I., 1995 revised 1999, Oil and Gas Teams: How to Make Them Work.
Training manual published by OGCI.
Tyler, K., Henriquez, A., and Svanes, T., 1994, Modeling Heterogeneities in Fluvial
Domains: A Review of the Influence on Production Profiles, pp. 77-90, in
Stochastic Models And Geostatistics: Principles, Methods, And Case Studies,
edited by Yarus, J.M. and Chambers, R.L.
Tyler, K., Svanes, T, and Omdal, S., 1993, Faster History Matching And Uncertainty In
Predicted Production Profiles With Stochastic Modeling, SPE Paper 26420, SPE
Annual Technical Conference and Exhibition, Proceedings, p. 3143.
Tyraskis, P.A., C. Tsingas, B. B. Toelle, T. H. Keith, B.L. Nebrija and G. C. Robinson,
1996, Integrated Reservoir Characterization Process Model: a planning and

© PetroSkills, 2013. All rights reserved. 14-5

Reservoir Characterization
Chapter 14

scheduling tool for E&P organization, First Break, Vol.14, No 9, September

1996/359
Valusek, J.E., 1995, Bridging the PC/UNIX Gap, pp. 14-17, Petro Systems World,
September/October 1995.
van Wagoner, J. C., Mitchum, R.I., Campion, K. M and Rahmanian, V D., 1990,
Siliciclastic Sequence Stratigraphy in Well Logs, Cores, And Outcrops, AAPG
Methods in Exploration Series, no. 7, 55p.
Weber, K.J., 1986, How Heterogeneity Affects Oil Recovery, pp. 487-544, in Reservoir
Characterization edited by Lake, L.W. and Carroll, Jr., H.B., Academic Press.
Weber, K.J. and L.C van Guens, 1990, Framework for Constructing Clastic Reservoir
Simulation Models, SPE, October 1990.
White, D.W. 1986, World Conventional Oil and Gas Resources, from Dahlem
Conference On Resources And World Development, West Berlin.
Wolf, D.J., Withers, K.D., Burnaman, M.D., 1994, Integration of Well and Seismic Data
Using Geostatistics, pp. 177-200, in Stochastic Models And Geostatistics:
Principles, Methods, And Case Studies, edited by Yarus, J.M. and Chambers, R.L.
Worthington, P.F., 1991, Reservoir Characterization at the Mesoscopic Scale, pp. 123-
165, in Reservoir Characterization II, edited by Lake, L.W., Carroll, Jr., H.B., and
Wesson, T.C., Academic Press.
Yilmaz, Oz, 1995, Thinking in Depth, (abs.), Proceedings 1995 Society of Exploration
Geophysicists, Annual meeting.
Yuan, Li-Ping and Rudy Strobl, 1991, Variograms and Reservoir Continuity, pp. 437-
452, in Reservoir Characterization III, edited by Linville, Bill, Pennwell Books.

14-6 © PetroSkills, 2013. All rights reserved.