# Quantitative Reservoir Characterization

## Business context

Introduction: What is interpretation meant to provide

There are several approaches to obtain a model of the subsurface from geophysical data. The overall goal is to get an idea of the subsurface geometries and properties and to visualize them for discussion with other geoscientists and reservoir engineers.

The first approach is to qualitatively interpret seismic data by following contrasts in acoustic impedance (reflections, usually related to rock layers) which will provide a structural view of the subsurface geometry. This has been done successfully for many years and has led to many hydrocarbon discoveries. This should include other geometrical observations such as amplitude anomalies bounded by a closing depth contour or closure against a fault.

The next level is to use quantitative measurements such as
reflection amplitude which is interpreted as being different for
different pore-fluids in the reservoir rock below a seal (for
example shale). This could be done using a rock physics model for
the reservoir and non-reservoir rock. If we had a correct rock
physics model, we could calculate different (geo) physical
quantities for each well-defined rock type. However, only in
exceptional cases is this true and often we work with statistical
relationships between rock properties (V_{p} versus
V_{s}, Porosity versus Acoustic Impedance). These can be
derived from well measurements and we then assume that these
relationships hold when moving away from the well. Although a lot
of progress has been made for clastics we still haven't arrived yet
at a multi-dimensional, multi-parameter model which will give exact
relationships between geophysical parameters and the needed rock
properties (porosity, permeability, saturations, source rock
maturity, etc). Hence, often we still rely on statistical
relationships. For carbonates the rocks physics is even less well
developed and we venture ourselves on more dangerous grounds.

A nice example is the use of elastic impedance and its relationship with AVO. This is even more so for extended elastic impedance which for certain values of its parameter (χ) provides traces that have a surprisingly strong correlation with elastic parameters, such as Lamé's constants.

The final level is to look at different geophysical properties and do a joint interpretation. We hope then that some property such as resistivity is (statistically, but intuitively explainable) related to P-wave velocity or Poisson ratio. Ultimately we need a "physics" explanation for that relationship. So long as we don't have a complete theory to connect the various rock properties we will help ourselves as best as we can with statistical relationships. But note that these always need to be calibrated with well data.

It can be said that the (incorrect) models we use to describe seismic wave propagation or the physics of rocks only need to be appropriate for the purpose at hand and don't need to be more sophisticated than that.

Where does Quantitative Reservoir Characterization fit in?

Quantitative Reservoir Characterization deals with only one
aspect of the total interpretation workflow, which consists
of:

- Seismic acquisition and processing providing the (for the
objectives) most appropriate data.

- Structural interpretation providing the shape of the
reservoir.

- Stratigraphic interpretation dealing with facies changes.

- Quantitative interpretation where we derive the elastic
parameters (V_{p}, V_{s}, ρ) including indications
of anisotropy. From the elastic parameters, we derive the
petro-physical / rock properties which consists of porosity,
permeability, fracture systems (orientation & density),
etc.

- Finally reservoir simulation will be applied for determining the
flow / productivity characteristics, leading to optimum well
placements and productivity.

If possible, we would like to take (seismic) data and invert (seismic to model) them directly to reservoir properties. For these inversions, we need appropriate algorithms for forward (model to seismic) calculations starting with a first estimate of the subsurface. The starting model will iteratively be updated to minimize the difference between the forward calculated synthetic seismic and the observed seismic. Which model to choose is not always obvious, but an effective medium model might be a good starting point.

In the inversion as many sources of information need to be combined: geological models, seismic data, non-seismic data, well data, production data, (you name it). These data should be inverted jointly for an optimum result.

## Who should attend

Geologists, geophysicists, petroleum and reservoir engineers, involved in exploration and development of hydrocarbon fields. That means not only those involved in the production side but also geoscientists designing the acquisition of seismicandnon-seismic data needed for Quantitative Interpretation.

## Course content

Some of the work will be done in teams:

Team work will consist of:

Summary/Presentation of learning points previous day

Formulating 3 multiple-choice quiz questions related to the same
day.

Day1: Monday: Data Preparation, Acquisition, Processing, Resolution (Point Spread Functions)

08:00-09:00 Introduction: Biography, Program, 52 Things, Teams

09:00-10:00 Geophysical methods, Intro Gravity, Intro Magnetics

10:00-11:00 Exercises: Gravity resolution & ambiguity (computer)

11:00-11:15 Refreshments

11:15-11:45 Land & Marine EM

11:45-13:00 Exercise: Scripps EM Modelling (computer)

13:00-14:00 Lunch

14:00-14:30 Videos:

14:30-15:00 Seismic Acquisition

15:00-15:15 Refreshments

15:15-15:45 Seismic Processing

15:45-16:30 Exercises: seismic resolution I & II (computer)

16:30-17:00 Team a: Preparation Summary of day 1

Team b: Formulation of 3 multiple choice questions of day 1

Day2: Tuesday: Interpretation: Structural, Stratigraphic, Quantitative (Gassmann)

08:00-08:45 Team a: Summary of day 1

08:45-09:15 Team b: Quiz questions of day 1

09:15-10:15 Structural Interpretation

10:15-11:00 Stratigraphic Interpretation

11:00-11:15 Refreshments

11:15-12:00 Exercises: Throw, To drill or not (paper)

12:00-13:00 Exercises: DFT, DFT frequencies (computer)

13:00-14:00 Lunch

14:00-14:30 Videos

14:30-14:45 Spectral Decomposition

14:45-15:00 Fractures: Curvature

15:00-15:30 Quantitative Interpretation: Rock Physics

15:30-15:45 Refreshments

15:45-16:15 Quantitative Interpretation: Gassmann fluid replacement

16:15-17:00 Exercise: Gassmann, Dimming (paper)

17:00-17:30 Team b: Preparation Summary of day 2

Team c: Formulation of 3 multiple choice questions of day

Day3: Wednesday: Quantitative Interpretation: Effective Media, Elastic Impedance, Anisotropy

08:00-08:45 Team b: Summary of day 2

08:45-09:15 Team c: Quiz questions of day 2

09:15-10:30 Effective Media, Elastic Impedance

09:30-10:30 Exercises: Effective Media (computer)

10:15-11:00 Inhomogeneity & Anisotropy

11:00-11:15 Refreshments

11:15-12:00 Tuning

12:00-13:00 Appropriate Pre-Processing

Imaging Primaries & Multiples

Imaging Conditions: Correlation, Deconvolution, True-Amplitude

13:00-14:00 Lunch

14:00-14:15 Videos

14:15-14:30 Fine Layering, Fractures, Anisotropy (VTI, HTI, Orthorhombic)

14:30-15:30 Exercise: Tuning Wedge (computer)

15:30-15:45 Refreshments

15:45-16:00 Anisotropic True-Amplitude Imaging Condition

16:00-16:30 Exercises: Anisotropy (paper), Anisotropy (computer)

17:00-17:30 Team c: Preparation Summary of day 3

Team a: Formulation of 3 multiple choice questions of day 3

Day4: Thursday: QI, AVO, AVA

08:00-08:45 Team c: Summary of day 3

08:45-09:15 Team d: Quiz questions of day 3

09:15-09:45 AVO, AVA

09:45-10:15 Rock Templates

10:15-11:00 Exercises: AVO Template (paper)

11:00-11:15 Refreshments

11:15-11:30 Tuning: Simmons & Backus

11:30-12:00 Exercises: Tuning AVO (computer)

12:00-13:00 Inversion

13:00-14:00 Lunch

14:00-15:00 Videos

15:00-15:30 Exercises: AVO (computer)

15:30-15:45 Refreshments

15:45-16:00 Fractures AVAZ

16:00-17:00 Exercises: Fractures (paper)

17:00-17:30 Team x: Preparation Summary of day 4

Team y: Formulation of 3 multiple choice questions of day 4

Day5: Friday: QI Templates, Amplitudes, Bayesian Statistics, VOI

08:00-08:45 Team x: Summary of day 4

08:45-09:15 Team y: Quiz question of day 4

09:15-10:00 VOI: What to spend on a new survey or study?

10:00-11:00 Exercises: VOI (paper), VOI (computer)

11:00-11:15 Refreshments

11:15-11:45 Direct Hydrocarbon Indicators (DHI)

11:45-13:00 Exercises: Amplitude Analysis (paper)

13:00-14:00 Lunch

14:00-15:00 Videos

15:00-15:30 Inversion: Full Waveform Inversion (FWI)

15:30-16:00 Sources of Information: SEG and EAGE (demo)

Finalize course: Any remaining questions/issues

16:00-16:15 Refreshments

15:45-16:00 Course evaluation

## Learning, methods and tools

At the end of the course participants will have a solid foundation in Seismic Quantitative Interpretation. The aim of the course is to provide a solid conceptual understanding without going into mathematical detail. It uses a mixture of lectures, practical exercises and direct (workshop-like) participant involvement, complemented with case histories. Use of laptops for exercises and WIFI internet access in the classroom is mandatory.

The course can be customized to meet specific needs of participants.