Shell Refresher - Petrel 2004

Day 2 – Geology, Model building and

Process Manager
Contents

Introduction 5

Fault Modeling – Slides 9

Fault Modeling – Exercise 18

Define a new model 18
Automatic generation of faults 18
Auto connection of faults 19
Automatic Adjustment of Key Pillars 19

Pillar Gridding – Slides 21

Pillar Gridding – Exercise 33

Before Pillar Gridding 33
Create a new 3D grid 33
Create a boundary based on a surface area 34
Automatic assignment of faults 35
Defining segments using trends 36
Set fault not part of segment boundary 38
Specify number of cells 39
Pillar Gridding 40

Make Horizons – Slides 42

Make Horizons – Exercise 57

Define the 3D grid domain 57
Insert Surfaces into the 3D skeleton grid 57
Conform to existing Horizons 62
‘Horizons Lines’ – Horizon-Fault Intersection Lines 63
Optional: ‘Horizons Lines’ – Use in Simulation Grids 66

Zonation and Layering – Slides 68

Shell Refresher - Day2 Contents • i

 Make Zones – Exercise 74
Create Zones 74
Minimum Cell Thickness 77
Using Intersection planes for QC 78
Optional: Export the intermediate horizons and zones 78

Layering – Exercise 80
Insert Layering 80

Fault Analysis – Slides 83

Fault Analysis – Exercise 92

Fault Transmissibility Multiplier 92
Theory: Transmissibility Multiplier 92
Exercise – Assumption 1: Constant TM 93
Exercise – Assumption 2: Variable TM 94
Constant Kf: 95
Constant Kf, on similar facies contacts (lithology juxtaposition): 96
Estimated Kf, SGR based on N/G (Standard equation): 97

Geometrical Modeling – Slides 101

Geometrical Modeling – Exercise 106

Creating a simple Facies 106
Creating Concession Boundary properties 108
Cells Cut by Surface 108
Connected Volumes based on Fluvial Facies 109
Calculate volume of Connected Volumes in Process Manager 110
Identify connected bodies coinciding with wells - Process Manager 111

Scale up well logs – Slides 113

Scale Up Well Logs – Exercise 117

Scale Up Well Logs 117

Property Modeling – Slides 119

Property Modeling – Exercise 129

Facies Transition Simulation (FTS) 129

ii • Introduction Shell Refresher - Day2

 Property Calculator 132
Exercise – Simple SW Calculation 135

Volume Calculation – Slides 136

Volume Calculation – Exercise 146

Calculate bulk volume above a contact 146
Creating a STOIIP Map 151
Draping the STOIIP map on a depth surface 152
Optional Exercises – CDF 153

Process Manager – Slides 155

Process Manager – Exercise 168

Become familiar with the Process Manager interface 168
Predefined workflows – Repeat all calculations … 169
Optional: Predefined workflows – Make scaled plots … 172
Optional: Operations – Referring to objects using variables 174
Optional - Using Loops to create a nested ‘If statement’ 177
Monte Carlo simulation of contacts 178

Shell Refresher - Day2 Contents • iii

 Introduction

Petrel 2004 Refresher Course

Course Content

Day 1 Day 2 Day 3 Day 4

„ Introduction „ Fault Modeling „ New Seismic Functionality „ Upscaling to SimGrid

„ Data Import „ Synthetics „ LGR’s

„ Pillar Gridding
„ Ant Tracking „ Simulation Setup
„ Open Spirit „ Make Horizons
„ Other seismic features
„ Edit input data „ Zones & Layers „ Initializing
„ Domain Conversion
„ Geometrical Modeling „ Sim Results View
„ Well Correlation

„ Neural Net „ Upscaling Well Logs

„ Well design „ Property Modeling

„ Plotting „ Make Contacts

„ API Plugins „ Volume Calculation

„ Process Manager

1 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Introduction • 5

 Petrel 2004 Refresher Course
Petrel Modules for Shell
PG G PG P G PG
fo
r rP fo
r or fo
r
fo d. ed. df
d d d
ite d. ite
d
ite ite ite it ite ite
lim lim lim lim lim lim lim lim
Un Un Un Un Un Un Un Un

2 © 2004 Schlumberger Information Solutions. All rights reserved.

Some new modules – Short Description:

UTILITIES
Classification and Estimation
Neural network technology for the estimation of well logs, surfaces, seismic volumes and 3D
property models.
Ocean-Petrel API Developers Kit
The Ocean-Petrel API Developers Kit allows users to create and execute plug-in modules
directly in Petrel. The plug-ins are typically algorithmic workflows that operate directly against
data found within the Petrel Data Explorer, with the results producing new Petrel data objects.

6 • Introduction Shell Refresher - Day2

 Petrel 2004 Refresher Course
Petrel Module Dependency Overview Fault Analysis

Utility Systems
Any Core
Well Design

Data Analysis

Domain Conversion Surface Imaging (Satellite Images)

Classification and Estimation

Seismic Sampling
Seismic Interpretation

OpenSpirit Plug-in

Geophysics
Seismic Volume Rendering and Extraction

Multitrace Attributes

Automated Structural Interpretation

Ocean-Petrel API Developers Kit

Geoscience Core

Bricked Seismic Rendering

Data and Resutls Viewer

Well Correlation
Combined Core

Geology
Facies Modeling

Petrophysical Modeling

Advanced Gridding & Upscaling

Engineering
Reservoir
Engineering
Reservoir

FrontSim Locked
Core

History Match Analysis

3 © 2004 Schlumberger Information Solutions. All rights reserved.

Petrel 2004 Refresher Course

Petrel License Dependency Overview
Module Dependency
Geoscience Core
System
Core

Reservoir Engineering Core (NEW)

Data & Results Viewer (NEW) No other modules allowed with this module
Combined Core (NEW) Includes Geoscience AND Reservoir Engineering Core
Seismic Interpretation Geoscience Core
Geophysics

Domain Conversion Geoscience Core

Seismic Sampling Geoscience Core AND Seismic Interpretation
Multitrace Attributes Geoscience Core AND Seismic Interpretation
Automated Structural Interp (NEW) Geoscience Core AND Seismic Interpretation
Seismic Vol. Rend. & Extract Geoscience Core AND Seismic Interpretation
Well Correlation Geoscience Core
Geology

Facies Modeling Geoscience Core

Petrophysical Modeling Geoscience Core
Fault Analysis Geoscience Core OR Reservoir Engineering Core
Engineering
Reservoir

History Match Analysis (NEW) Reservoir Engineering Core

FrontSim Locked Reservoir Engineering Core

Adv Gridding & Upscaling (NEW) Geoscience Core

Data Analysis Geoscience Core OR Reservoir Engineering Core
Surface Imaging Geoscience Core OR Reservoir Engineering Core
Utility

Classification & Estimation (NEW) Geoscience Core OR Reservoir Engineering Core

Well Design Geoscience Core OR Reservoir Engineering Core
Ocean API Developers Kit (NEW)
Bricked Seismic Rendering (NEW)
Party

Geoscience Core AND Seismic Interpretation AND/OR Seismic Vol. Rend. & Extract
3rd

OpenSpirit Plug-in for Petrel Geoscience Core OR Reservoir Engineering Core

4 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Introduction • 7

 Petrel 2004 Refresher Course
Module Notes
‰ One Geoscience Core or one Reservoir Engineering Core or one Combined Core is required
‰ An upgrade to the Combined Core can be made from either Geoscience Core or Reservoir
Engineering Core
‰ The Data and Results Viewer is an independent module and requires no core module
‰ The Petrel Ocean API Developers Kit is an independent module and requires no core module
‰ Classification and Estimation requires Well Correlation to estimate log curves, Facies Modeling to
classify 3D facies, Petrophysical Modeling to estimate 3D petrophysical properties, Seismic
Interpretation to estimate seismic volumes
‰ Synthetic Seismograms requires the Seismic Interpretation module to load seismic data and Well
Correlation to view the synthetic seismograms.
‰ Most clients will require Multitrace Attributes for Automated Structural Interpretation
‰ Advanced Gridding & Upscaling replaces previous module Advanced upscaling and Structured
Simulation Gridding and includes a copy of ECLIPSE FloGrid
‰ Reservoir Engineering Core replaces Simulation and Production Data Analysis and includes a copy
of ECLPSE Office, Schedule and FloViz
‰ Bricked Seismic Rendering is a visualization performance enhancement tool. It will enhance inline,
crossline, time-slices and random lines created in the seismic interpretation module. If a client owns
the Volume Rendering and Extraction module they will also see dramatic volume rendering
improvements

5 © 2004 Schlumberger Information Solutions. All rights reserved.

Petrel 2004 Refresher Course

Available Upgrades for Obsolete Modules

Obsolete Modules - Maintenance Only Upgrades

Advanced Upscaling Upgrade G-Core or RE- Core to Combined Core
Structured Simulation Gridding (SSG) Adv Upscaling to Adv Gridding and Upscaling
Production and Simulation Data Analysis Production and Simulation Data Analysis and G-
Core to Combined Core

6 © 2004 Schlumberger Information Solutions. All rights reserved.

8 • Introduction Shell Refresher - Day2

 Fault Modeling – Slides

Fault Modeling
What’s New?
• Workflow recap
– Fault types and shapes
– Key pillars
– Function bar – Some New Icons/features
• Fault Modeling Technique recap
• Learn How to Edit a Fault Model
– Edit widget – some changes
– Automated trimming and connection of key pillars - Tips
• Extra
– Tie faults to well cuts
– Ant-tracking - NEW

1 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Fault Modeling – Slides • 9

 Fault Modeling
General Workflow – Recap 2
1

1) Define a Fault Model

2) Model the initial set of faults (generate Key Pillars)
3) Edit any problem areas 3
4) Trim/Extend to the top and base of th emodel
5) Connect faults (required for Pillar Gridding)
4
6) Add more faults by repeating steps 2-5
… then do final edits

6 5

2 © 2004 Schlumberger Information Solutions. All rights reserved.

Fault Modeling
Function Bar – Some New changes
Define a new model:
Before starting Fault Modeling, you need to define
and name a model that will be used to store the
fault model and all 3D grids related to that fault 1
model. Then activate the Fault Modeling process.

1. Increase/decrease Shape pt. size

2. Select Horizon nodes, activates related icon tools

New 2004

3 © 2004 Schlumberger Information Solutions. All rights reserved.

10 • Fault Modeling – Slides Shell Refresher - Day2

 Fault Modeling
Editing Key Pillars – Moving the Widget (Tips)
1. Move the widget parallel to orientation of the 3 Free Movement
cylinder (select cylinder which becomes yellow)
Move in Vertical
2. Move widget parallel to orientation of Plane only
plane(select the plate.
Move along Line
3. Restriction based on 3 tools tangent only
4. Press Ctrl - tangent options will change

1 2 4

4 © 2004 Schlumberger Information Solutions. All rights reserved.

Note: The Widget itself does not change direction anymore. It is the planes in which you move
that will change once you use CTRL.

Shell Refresher - Day2 Fault Modeling – Slides • 11

 Fault Modeling Cut/extend pillars against a Surface or
Automated Trimming of Faults (Tips) a given Constant:

1. Under Operations tab, select the Top

and Base limits, given by 2D grids

2. Choose All or selected Pillars, Cut and

Extend to the limits. Press OK.
1

5 © 2004 Schlumberger Information Solutions. All rights reserved.

Automatic Adjustment of Key Pillars

Philosophy:
• Key Pillars should be extended above and below Top and Base Reservoir.
• Key Pillars should be smooth without abrupt changes in height.
Key Pillars should be extended above and below the reservoir, and just above and below. The
reason for this is:
• If the Key Pillars are too short, then they will be extrapolated by Petrel when the horizons are
inserted later on.
• If the Key Pillars are too long, and you are modeling a complex reservoir, then you might
optimize the adjustment of the Key Pillars in areas outside the reservoir, i.e. areas that are of no
interest. Hence, it is better to make the Key Pillars shorter and fit them in the most optimal way
to the areas of interest.
Under Operations tab on the Fault Modeling process, is an option of adjusting all the Key Pillars
by one operation. You need 2D grids that can be used as “cutting surfaces” as input. These
surfaces should contain no faults. The examples on the slide is using a 2D grid which is an
erosional surface and does not contain faults. If the surface contained faults, then you could first
smooth it (from the Operations tab, under Settings for the surface). Two copies of the smoothed
surface could then be shifted so that they where just above and just below the reservoir.

Note: You may have to change the Minimum Pillar height to get a proper adjustment.

12 • Fault Modeling – Slides Shell Refresher - Day2

 Fault Modeling
When to cut the model and when to cut the sticks (Tips) – 1

The geometry of the pillars are

complex.

But just a small part are inside the

area of interest. By cutting the fault
sticks you can use a simplified
geometry OR have more shape
points in the zone of interest.

6 © 2004 Schlumberger Information Solutions. All rights reserved.

Fault Modeling
When to cut the model and when to cut the sticks (Tips) – 2

The geometry of the fault sticks are

simple.

Doesn't matter if you cut the sticks

or the faults, but in many cases it
will be easier to cut the faults.

7 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Fault Modeling – Slides • 13

 Fault Modeling
Automatic Connection of Faults (Tips)
Automatic fault connection:
Normally used with clean orthogonal fault
data. Also used to find problems with
fault intersections. After those have been
located and edited, auto connectino can
be run again.

Extend distance – will search for

pillars within the given distance and
connect them.

Remove distance – will remove

pillars that are less than the given
distance from the connection.

8 © 2004 Schlumberger Information Solutions. All rights reserved.

Automatic Connection of Visible Faults

Petrel has an option for automatic connection of visible faults. It is located under the operations
tap in the Fault Modeling Process (double-click on the Fault Model Process in the Process
Diagram).

But the automatic fault connection only works well with simple fault models and requires that the
user quality checks the result against the input data afterwards.

The best way to use it, when it comes to more complex models, is to use it on the more simple
part of the model, for example the framework faults. Afterwards the user can insert more
complexity in the model manually.

14 • Fault Modeling – Slides Shell Refresher - Day2

 Fault Modeling
QC of Fault Model and Final Edits (Tips)
Quality Check:
After Key Pillars have been adjusted and faults
trimmed and connected, it is important to
display the original input data to verfy that they
match (if you trust the input...)

If they don’t match, do some final edits and

move to the Pillar Gridding process.

Fault sticks – imported into Petrel

and used as input data

Note – All intersecting faults

must be connected (grey
pillar), before doing the Pillar Petrel Fault Model – Key Pillars
Gridding process. generated from input data and edited.

9 © 2004 Schlumberger Information Solutions. All rights reserved.

Confirming Fault Connections – 3D Window

1) Open 3D window
2) Display all fault models
3) Increase vertical exaggeration a little
4) View from above
5) Click on the Orthogonal on/off icon
6) Open the Fault Model Settings dialog:
•Turn off Shape points
•Turn on Pillars and make pipe width 100 – 200
•Turn on Fill between Pillars as white and 50% transparent
7) Gray pillars indicate joined faults

Confirming Fault Connections – 2D Window

Another way to QC Fault connections is to view them in a 2D Window:
1) Open a 2D window, background black
2) In the Models Tab, click on Fault Model
3) Gray pillars indicate joined faults

Shell Refresher - Day2 Fault Modeling – Slides • 15

 Fault Modeling
Extra – Tying Faults to Well Cuts (Tips)

Lock/Unlock to Well Top:

Select one pillar and one well top,
then click the Lock icon.

Note: Remember that the Fault

Model and the Well Tops must be
in the same domain (Time or
Depth).

10 © 2004 Schlumberger Information Solutions. All rights reserved.

Conditioning to Well Cuts

In order to condition to well cuts, you must define a Key Pillar that should be forced to go
through the position of the well cut.
Display the well cut to be conditioned to together with the fault defined by Key Pillars. Select the
Key Pillar and the well cut by clicking on them in the 3D display window. Then press the
Lock/Unlock to Well Top icon. Each Key Pillar can only be conditioned to one well cut.

16 • Fault Modeling – Slides Shell Refresher - Day2

 Fault Modeling
Extra – Ant tracking; Automated Fault Extraction (New)
Interactive Fault Extraction and
Classification tool

11 © 2004 Schlumberger Information Solutions. All rights reserved.

By focusing on structural geology rather than conventional segment picking, Automated

Structural Interpretation reduces conventional interpretation time while increasing your level of
geological detail, structural awareness and reservoir understanding.

The Ant Tracking workflow consists of four independent steps:

1) Enhance the spatial discontinuities in your seismic data using any edge detection algorithm
(i.e. variance, chaos, edge detection) and optionally, pre-condition your seismic data by reducing
noise
2) Generate the Ant Track Cube and extract the fault patches
3) Validate and edit the fault patches
4) Create final fault interpretation model

In the final step (4), the fault patches are used for further seismic interpretation or as input to the
fault modeling directly.

Shell Refresher - Day2 Fault Modeling – Slides • 17

 Fault Modeling – Exercise

Define a new model

Before you start make sure to open the Day2_Shell_Model.pet project from the Projects folder.
Make a Save As in your Student folder.

Exercise Steps

1. Double click on in the Process Diagram (under Structural

Modeling).
2. Name the model “test” and click OK. The model will be placed under the Models tab in the
First Petrel Explorer.

Automatic generation of faults

In this exercise you will learn how to automatically create Key Pillars from sets of fault sticks in a
folder in the Petrel Explorer Input tab.

Exercise Steps
1. Convert one set of fault sticks to Key Pillars:
a. In the Explorer Input tab, find the fault sticks folder called “For Convert to
fault”. Open the folder and right click on one of the fault sticks. Select Convert
to faults) in Fault Model… from the pull-down menu. Set every n’th pillar
again in the pop-up dialog box and click OK.
b. Note the new faults in the 3D window and in the Faults folder for the active
model (in the Models tab).
2. This operation can be performed for all fault sticks in a folder:
a. To avoid double sets of faults, delete the fault created above by selecting the
whole fault in the 3D window, or selecting it (making it bold) in the Models tab
of Petrel Explorer and then pressing delete.
b. Right click on the folder called “For Convert to fault”. Select Convert to faults
in Fault Model… from the appearing pull-down menu.
3. Continue doing the necessary editing of Key Pillars as described above.

18 • Fault Modeling – Exercise Shell Refresher - Day2

 4. Connect faults where necessary.

Auto connection of faults

It is possible to connect the faults automatically in Petrel.

Exercise Steps
1. Display all the faults you have created in a 3D window.

2. Double-Click on the Fault Modeling process in the Process Diagram.

3. Go to the Operations tab.
4. Click on the Auto connect button and use a Extend distance of 250 m. Toggle automatic
fault connection only for the visible faults.

5. Check the connections and do manual editing if necessary. There is one level of Undo (top
of dialog box).

Automatic Adjustment of Key Pillars

It is possible to adjust all the Key Pillars automatically by cutting them or extending them relative
to a surface or a constant value.
If the Key Pillars are trimmed by a surface then this surface should be smooth or non-faulted
surface. In the exercise below, only the top Shape Points will be trimmed. If the base Shape
Points should be trimmed as well, then a copy could be made of Top Etive surface. This copied
surface should be smoothed to remove most of the structure (but keep the general dip) and then
shifted downwards until it is deeper than the entire Top Etive surface.

Exercise Steps
1. Display all the generated Key Pillars together with the Base Cretaceous surface.
2. Double click on the Fault Modeling process (in the Process diagram).
3. Go to the Operations tab from the dialog box that pops up.
4. Select the Base Cretaceous surface in the Petrel Explorer Input tab (click on the name to
make it bold), toggle on the checkbox next to Top limit in the Operations tab under
Cut/Extend pillars and click on the blue arrow. Fill in the other options as the figure shows.

Shell Refresher - Day2 Fault Modeling – Exercise • 19

 5. Click on the Execute button.
6. Observe that all the Key Pillars will be cut by or extended to the Base Cretaceous level,
generating smooth transitions between Key Pillars.

20 • Fault Modeling – Exercise Shell Refresher - Day2

 Pillar Gridding – Slides

Pillar Gridding
What’s New?
• Understand Basic Terminology/Concepts recap
– Boundaries, Segments, Directions and Trends
• Workflow Process recap
– Define Boundaries
– Settings tab
– Set Directions and Trends
– Define Segments
– Automatic Direction Assignment - NEW
• Editing Options
– Edit Fault Model in Pillar Gridding - NEW
• QC of Skeletons
– Intersections and Segments

1 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Pillar Gridding – Slides • 21

 Pillar Gridding
Overview
Create a grid adjusted to the mid-points of the Extrapolate the pillars to the top and base
key pillars. shape points. This will create a 3D mesh,
represented by the top, mid and base
skeleton grids.

Pillars will be created at

every corner of every
grid cell

2 © 2004 Schlumberger Information Solutions. All rights reserved.

Pillar Gridding - Concept

Pillar Gridding is the process of making the ’Skeleton Framework’. The Skeleton is a grid
consisting of a Top, a Mid and a Base skeleton grid, each attached to the Top, the Mid and the
Base points of the Key Pillars. In addition to the three skeleton grids, there are pillars connecting
every corner of every grid cell to their corresponding corners on the adjacent skeleton grid(s).
The figure to the right shows an example of the three skeleton grids and one intersection
through the grid that shows the Pillars connecting grid cells together. As part of the Pillar
Gridding process, faults defined by Key Pillars are converted to Fault Surfaces.

When creating the Pillar Grid, you will work with the Mid Skeleton grid. The Mid Skeleton grid is
the grid attached to the mid-line that connects the Key Pillars, ref. left figure. The purpose is to
create a grid that looks OK at the midpoint level, with respect to grid cell size, orientation and
appearance of the cells. The next step is to extrapolate this Mid Skeleton upward and downward
in order to create the Top and the Base skeletons.

22 • Pillar Gridding – Slides Shell Refresher - Day2

 Pillar Gridding
Overview
Pillars: Lines that define the corners of 3D Pillars
cells. In the Pillar Gridding process they are
still virtual until the process is completed.

Pillars
+
Edges

Pillars
+
Edges
Note: This illustrates the concepts +
of cells, but skeletons do not Base & Top
define physical cells yet =
Cells

3 © 2004 Schlumberger Information Solutions. All rights reserved.

A 3D grid is a 2D grid mesh extended into the third dimension. A 2D-grid mesh is defined by
rows and columns oriented in the X and Y directions (2 dimensions). A 3D-grid mesh is defined
by rows, columns and pillars oriented in the X, Y, and Z directions (3 dimensions). Another way
to look at a 3D grid is that it is a series of 2D grids stacked on top of one another. The lines
connecting corresponding grid nodes from each of the 2D grids are the pillars.

Pillar gridding is the process of defining the 3D mesh. The process starts with a set of rows and
columns evenly spaced at the specified grid increment. At this point the pillars are vertical lines
passing through each row-column intersection. The Key Pillars previously defined guide the
reorientation of these initial pillars, to create the final 3D grid. Through a series of algorithmic
iterations the pillars, rows, and columns are made sub-parallel to the Key Pillars. Output from the
pillar gridding process are the final pillars displayed as a ‘Skeleton’, i.e. 3 skeleton grids,
representing the top, mid and base of the pillars. Skeleton grids are used for QC rather than the
actual pillars, as it is easier to view 3 grids whose nodes define points in space through which
the pillars pass than to look at hundreds of vertical lines (pillars) in 3D space.

At the completion of pillar gridding, the skeletons (actually the pillars they represent) are not
attached to any Z-values, nor do they represent any surface. They are only a set of pillars
defining each grid cell's lateral shape and size in the 3D model.

Shell Refresher - Day2 Pillar Gridding – Slides • 23

 Pillar Gridding
Workflow Recap (Steps)
Pillar Gridding Workflow
2
1. Start Pillar Gridding process 1
2. Select the faults
3. Define the boundary 4
4. Define Grid name and increment
5. Build mid-skeleton grid (click Apply)
3
6. Tune mid-skeleton grid
7. Extend pillars to top and base
skeletons (click OK)

6 5

4 © 2004 Schlumberger Information Solutions. All rights reserved.

Pillar Gridding Note: It is possible to convert a closed polygon

Process – Defining a Boundary (Tips) from the Input tab to a 3D grid boundary

A boundary must be defined around the area

of interest. This is done by using a boundary
polygon or by creating a more complex
boundary involving faults as part of the
boundary together with boundary segments.
Used in combination

Set part of grid

boundary

Create boundary
segment

Create boundary
(closed polygon)

5 © 2004 Schlumberger Information Solutions. All rights reserved.

24 • Pillar Gridding – Slides Shell Refresher - Day2

 Pillar Gridding
Process – Directions and Trends (Recap)

J-direction

I-trend

I-direction

Arbitrary

J-Trend

6 © 2004 Schlumberger Information Solutions. All rights reserved.

Fault Directions and Trends along the I or J direction

The user can apply trends and give directions to faults. The purpose of this is to guide the
gridding process and/or to orient the cells parallel to the faults, especially for simulation grid.
Always start simple, using no trends or directions and see if the pillar gridding process works! If
the fault geometries are complex, you might need to help the gridding process by giving
directions to faults and/or enter trends in the grid. If you need to use directions and trends, start
with a few and add more once you see how the gridding process works!
Fault directions and trends of the same color should be roughly parallel to each other, and the
red trends/directions should be roughly perpendicular to the green trends/directions.
Give a direction to a fault - select the entire fault by clicking on the line between shape points
or select an interval of a fault by first selecting the starting point of the interval and then the
ending point of the interval by holding the Shift key. Click on either the I or J direction icon.
Enter a trend - first select the I or J trend, and then simply digitize the trend where you want to
place it.

Shell Refresher - Day2 Pillar Gridding – Slides • 25

 Pillar Gridding
Automatic Direction Assignment (NEW)

Main Idea - Allows the assignment of trends on faults

based on an angle and the number of segments
between pillars. The angle is the deviation from the
assigned I or J direction.

Best use – With the Vector Field Method for simulation

grids together with Zig-Zag faults (recommended max
15 degrees).

7 © 2004 Schlumberger Information Solutions. All rights reserved.

Reasons for use:

Good default for simulation grids is a max of 15 degrees and at least 3 segments. It is not a
good idea to have a lot of single segment alignments, as it will be very long to grid.

• It is a good idea to use together with the vector field method and zig-zag faults for simulation
gridding.
• Note that sometimes you may create impossible alignments you will be warned about these
when pressing Apply in the Pillar Gridding process. One example is when you have
intersecting faults with the same alignment.

26 • Pillar Gridding – Slides Shell Refresher - Day2

 Pillar Gridding
Process – Automatic Direction Assignment (Steps)
I/J-direction – Automatic assignment is available in the 2B
More tab of Pillar Gridding dialog.

1. Go to Pillar Gridding settings and set

the deviation angle (A) from North-South
(J-directed faults) and East-West (I-
directed faults). Click APPLY (B) < 22.5 deg

1A > 22.5 deg

1B

2. Click the “Automatic Direction

Assignment” (A) icon in the Function Bar
and inspect the results in 2D Window (B).
Click APPLY 2A

8 © 2004 Schlumberger Information Solutions. All rights reserved.

Automatic Direction Assignment for Faults in Pillar Gridding process:

1.Go to Pillar Gridding Settings and set up the deviation angle from North-South for J-directed
faults and East-West for I-directed faults.
2.Click the “Automatic Direction Assignment” icon in the Function Bar and inspect the results in
2D Window.

The result will be Red and Green directions only along segments of the faults that deviate less
than the specified assignment angle from North-South or East-West.

Shell Refresher - Day2 Pillar Gridding – Slides • 27

 Pillar Gridding
The More Tab – Methods (NEW)

Local Iterative Method (old, but improved):

This will assign trends to groups of connected faults
one at a time. It is a good general algorithm.

Vector field Method (new): trends will be assigned

using a vector field over the whole field at once. This
algorithm is smoother in situations where there are
many unconnected faults and trends.It utilizes FloGrid
algorithms for I/J assignment.

Note: Generally the Local Iterative method is

used, but the Vector field method may give
improved results on Simulation grids (e.g.
together with the mentioned Automatic
Assignment method; trial and error is the Key

9 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Diagram for the Pillar Gridding – Settings and More

There are 2 methods for assigning trends to gridlines:
Local Iterative Method: This will assign trends to groups of connected faults one at a time. It is
a good general algorithm.
Vector field Method: trends will be assigned using a vector field over the whole field at once.
This algorithm is smoother in situations where there are many unconnected faults and trends.

28 • Pillar Gridding – Slides Shell Refresher - Day2

 Pillar Gridding
Defining Segments (Tips)
After changes
With Default
Segments are initially the same as fault were applied
settings
compartments, they are defined by closed
faults. The default settings of faults and 5
trends can be changed:

Set part of
segment
boundary
2

Set no fault
3

Set no
boundary
4 1

Refer to Notes for details regarding the numbered trends.

10 © 2004 Schlumberger Information Solutions. All rights reserved.

Defining Segments (Fault compartments) by using Faults (1-3)

By default, all the defined faults will be included as faults in the 3D model, and will define fault
segments if they have been connected to each other. However, the user has the option to
choose whether a fault should be included as a fault or only be used as a trend for the gridding.
The user can also choose whether a fault should be used as a “segment divider” or not.
Defining Segments by using Trends (4-5)
Trends can be used in order to define segments that are not closed by faults or the grid
boundary. By default, a trend will not work as a “segment divider”, and the color will be darker
than if the fault works as a “segment divider”.

Shell Refresher - Day2 Pillar Gridding – Slides • 29

 Pillar Gridding
Edit Fault Model in the Pillar Gridding Process (NEW; Steps)

Sometimes when adding trends it may be

useful to have more key pillars; this can be
done directly in Pillar Gridding, easier than
reverting to Fault Modeling and then back… 3
1
1. Activate Pillar Gridding
process and select two
neighboring pillars (Yellow).

2
2. Click on the Add Pillar
Between icon.

3. See the result in the Fault

model (in 3D window).

11 © 2004 Schlumberger Information Solutions. All rights reserved.

Edit a Fault in Pillar Gridding process:

After the Fault model has been completed in the Fault Modeling process and you want to run the
Pillar Gridding, there may be times when you want to insert an extra pillar to make the
Directions, Trends and geometries look right. Traditionally, this has been done by going back to
the Fault Modeling process and adding pillars there. Now there is an option to add new key
pillars directly in the Pillar Gridding process to avoid the back-step to Fault Modeling.

30 • Pillar Gridding – Slides Shell Refresher - Day2

 Pillar Gridding
QC – Intersections (Recommend; Steps)
1
1. Open a 3D window and display one of the Skeletons in
addition to an I and/or J Intersection.

2. Change the style settings for Intersections; double click

Intersections folder and toggle show Pillars (A). View 3D (B).

2B

2A

3. Highlight one of the intersections and use the Grid

Player to move in the I or J direction to look for poorly
truncated pillars or spikes in the skeleton grid.

12 © 2004 Schlumberger Information Solutions. All rights reserved.

Pillar Gridding
QC – Segments (Recommended; Steps)
1. Change the style settings for the skeleton; double 2B
click on the Skeleton folder and toggle Show Solid:
As Segments (A). View 3D (B)
1B
2. To check segments, use the Segment Filter to
toggle on/off, or make a visual display of segments
as a legend by right-clicking on the Segment filter,
selecting Create/update template (A), and toggling
it on from the Templates tab (B)

1A
2A

13 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Pillar Gridding – Slides • 31

 Pillar Gridding
Horizon-Fault Intersection Lines (NEW; Tips)

Theory: There is an added feature in the next

process steps of Fault Modeling and Make Horizons.
It will generate a relationship between a fault and a
horizon. This will be covered in detail in those
processes, but the initial activation of the method is
in Pillar Gridding.

Practical: Settings ‘Insert Horizon object….’ in the

Settings of the Pillar Gridding dialog (should be
toggled on by default).

14 © 2004 Schlumberger Information Solutions. All rights reserved.

32 • Pillar Gridding – Slides Shell Refresher - Day2

 Pillar Gridding – Exercise

Before Pillar Gridding

Since Pillar Gridding uses the Key Pillars that define the faults and their intersections to guide
the building of the final pillars, the Key Pillars must be properly constructed. Ensure that the fault
modeling process is complete, then Pillar Gridding can be initialized and the skeleton grids
created.

Exercise Steps
1. Continue using the Day2_Shell_Model.pet project.
2. Now open the Gullfaks 2004 FaultMod Model in Models tab and use this from now on. It
contains only a complete fault model with connected faults.
3. Ensure that all faults intersecting are connected properly. Laterally connected faults should
have a shared (grey) Key Pillar.

Create a new 3D grid

Pillar gridding creates the first component (pillars) of a 3D grid. Because of this the process must
either create a new 3D Grid or overwrite an existing one. When updating a model you should
overwrite an existing 3D Grid because the settings will already be set from previous executions
and make the update easier.

Some key settings such as name of the 3D grid and the grid increment are set when initializing
the Pillar Gridding process, although they can be changed at any time.

Exercise Steps
1. Start the process of creating a new 3D Grid. Double click on Pillar Gridding process in the
Process Diagram. A 2D window opens automatically with your fault model displayed. The
line is the projection line between the Key Pillars mid-points you defined in the previous
exercise. The dots are the mid-point on the Key Pillar it self.

Shell Refresher - Day2 Pillar Gridding – Exercise • 33

 2. Enter a name for the 3D grid and specify the I and J increment (100) in the pop-up dialog
window.
3. Move the Pillar Gridding window out of the way but leave it up, as it will be used repeatedly
in the following exercises.

Create a boundary based on a surface area

Exercise Steps
1. Right-click on the Base Cretaceous surface in Input tab. Select ‘Create Boundary’. This
creates a new object called ‘Edge around Base Cretaceous’.

2. Right-click on the new edge and select ‘Convert to Boundary on the active Fault Model’.
3. Go to the Models tab and see that you have a new 3D Grid boundary.
4. Test the boundary: select I-J increment of 100, do not use zig-zag faults and use the ‘Local
Iterative Method’ set in the More tab. Press Apply to run the Pillar Gridding.

Comments
It is possible to move parts and delete parts of the digitized boundary. The Pillar Gridding
process must be the active process and the boundary is displayed in the 2D window.
• Select the Select and Edit/Add points icon. Click on the point to be moved on one of the
digitized lines and moving it to the desired position.
• To create a new point click on the line between two points and move a little.

34 • Pillar Gridding – Exercise Shell Refresher - Day2

 • Points are deleted by selecting the Set select/pick mode icon. Click on the point to be
deleted and press the Delete key. To delete several points hold the Shift key down and select
points with the set select/pick mode. To delete a whole boundary click on the line between the
digitized points, and press delete.
• Faults have to be deleted and edited in the 3D window using the Fault Modeling process. But
trends can be edited and deleted as described above.

It is possible to remove a fault from being part of the boundary.

• Select the Select and Edit/Add points icon. Click on the entire fault by clicking on the line
between the shape points or select part of a fault by clicking on a shape point (start point),
holding the Shift key and pressing another shape point (end point).
• Click on the Set Segment Boundary icon. The selected portion of the fault will turn from
blue to white, indicating that it is no longer part of the boundary but still defines a segment
boundary .

Automatic assignment of faults

If you have tried a few times to assign directions to faults and the result is not too good, you
could always try the Automatic assignment of faults option. By specifying an angle of deviation
from N-S and E-W, Petrel will assign appropriate red (I) and green (J) directions to the faults in
the Pillar gridding process.

Exercise steps
1. In the Pillar gridding process dialog, go to the More tab. At the bottom there is a section
specifying the assignment angle(deviation form I-J). Default is 22.5 degrees.

2. Try the default angle first. Click the Apply button for the More tab (angle settings saved).
Now go to the Function bar and press the Automatic direction assignment icon ; in the
2D window the red and green directions will update accordingly. For the real assignment to
take place, click Apply for the Pillar Gridding process.

Shell Refresher - Day2 Pillar Gridding – Exercise • 35

 3. If you are not happy about the result, try decreasing the angle a bit and test out.

Defining segments using trends

By default all faults in the model are set as segment boundary. If the faults are connected
properly an area limited by faults in all directions defines a segment. Hence, a fault block is
referred to as a segment. The figure below shows a skeleton grid colored with a different color
for each segment. Note how the segments are limited by connected faults (and boundaries) and
have a specified color.

36 • Pillar Gridding – Exercise Shell Refresher - Day2

 This exercise will show how trends can be added to separate portion of a segment from another
(the following optional exercise will cover setting faults as non-segment boundaries). If, for
example, the fluid contact is significantly different in parts of the model, care should be taken to
define the segments. In each segment different input data can be used at a later stage, including
different fluid contacts. After creation, the trend is set to be part of a segment boundary (see
figure below).

Exercise steps
1. In the Pillar Gridding process window, under the Settings tab, choose ‘Create new, named’
and give the grid a different name, for instance “3D grid making segments”.
2. Find a fault that does not terminate against another fault (or boundary). Digitize an
appropriate trend using the New J-trend or the New I-trend icon from a point of

Shell Refresher - Day2 Pillar Gridding – Exercise • 37

 the fault and attach it to the intersecting shape point on the other fault or boundary, like the
figure below. This trend will now work as any other trend and guide the gridding process.

3. To make the trend a “segment divider”, click on the Set Part of Segment Boundary
icon. After this feature is applied the trend will have a brighter color.
4. Click Apply and the OK to generate a new 3D grid with Skeleton grids
5. Display the skeleton grid colored with different colors for the segments. Open settings for
the skeleton folder in the previous 3D grid, and check show solid as segments.

6. Do the same for the “3D grid making segments”. Note the difference in the segments.

Set fault not part of segment boundary

If a fault is set to be a non-segment divider, the selected fault will not separate fault
compartments.

38 • Pillar Gridding – Exercise Shell Refresher - Day2

 Exercise steps
In the 3D window select the fault or part of the fault.
1. Click on the Set No Boundary icon.
2. The selected part of the fault will become grey. If the fault already has a J-direction, it will
show as a solid line, with a dark red color.
3. Do a Save As of your project. You will carry on using it in the next exercise.

Note: You have now completed the Pillar gridding process, and in the Models tab you have the
outline of the new grid, represented by the Key Pillars (from the Fault Modeling process) and the
inserted pillars (from the Pillar Gridding process). Remember however that you DO NOT have a
3G grid yet, as the Skelton is only a visual representation of Pillars. If you double click on the 3D
grid in Models tab and go to the Statistics tab, you will see that the amount of 3D Grid cells and
nodes are still zero.

Specify number of cells

Sometimes too few points are created in a certain part of the model. Often this is in a narrow
space between 2 faults where it is desired to have more cells when interpolating attributes.
Additional cells give the attribute model more interpolation freedom. The same situation can
occur between two closely spaced wells. Adding a trend between the two faults and specifying
the number of cells does this.
Occasionally defining the number of cells in one area will rob cells from another area, creating a
very poor geometric model. Therefore QC the mid skeleton before you creates the top and base
skeleton grids.

Shell Refresher - Day2 Pillar Gridding – Exercise • 39

 Exercise steps
1. In the 2D window, find an area where you have two faults with same direction next to each
other. If you don’t have this configuration on your faults then create it.
2. Define a trend between the two faults, if you already have one, you can use that.
3. Define the number of cells that should exist along one of the defined trends using the tool
. Make it significantly different from those currently existing along the trend.
4. Press Apply and observe how Petrel inserts the specific number of cells in the entire interval
between the directed faults.
5. If the number of cells looks acceptable then leave it, if it looks like a problem then alter the
number and re-Apply.

Pillar Gridding
After the Boundary has been defined and the 2D cell geometry tuned to the point of acceptability
(trends and directions may be applied to help tune the 2D cell geometry), the 3D grid can be
constructed. The result of this construction is the Skeleton, which is a series of pillars, one for
the corner of each cell. Top, middle and base skeleton grids are used to view these pillars easily
in the X-Y dimensions. The integrity of the pillars themselves can be viewed in an I or J
intersection plane.

Exercise steps:
1. Under the Pillar Geometry tab in the Pillar Gridding process window, toggle off ‘Curved’ for
the ‘Non-Faulted Pillars’. This will create a simpler 3D Grid geometry with less chance for
problems.

40 • Pillar Gridding – Exercise Shell Refresher - Day2

 2. Click Apply.
3. When the result is satisfactory, press OK to make the top and base skeleton grid. Say
"Yes" to the window that pops up (asking if you want to build the top and base skeleton
grids).

Shell Refresher - Day2 Pillar Gridding – Exercise • 41

 Make Horizons – Slides

Make Horizons
What’s New?
• Stratigraphic Subdivisions in Petrel
• Make Horizons process - Recap
– Define number of horizons (depending on input data)
– Define horizon type (geological)
• Settings Tab
– Algorithms
– Conform to another horizon – NEW
• Faults Tab
– Fault matrix setup and options – New 2003SE
– Horizon Lines - NEW
• Quality Check Methods
• Extra – Allan Diagram QC
• Extra – Fault Property (advanced QC)

1 © 2004 Schlumberger Information Solutions. All rights reserved.

42 • Make Horizons – Slides Shell Refresher - Day2

 Make Horizons
Process

Add the main horizons that will

create the framework of the
model. Normally seismic
interpretations or gridded
surfaces.

Fault modeling Pillar Gridding Make Horizons Make Zones Layering

2 © 2004 Schlumberger Information Solutions. All rights reserved.

The Stratigraphic subdivision process consist of four steps:

1.Make Horizon process - Insert the input surfaces into the 3D Grid. The inputs can be
surfaces from seismic or well tops, line interpretations from seismic, or any other point or line
data defining the surface.
2.Depth Conversion – If the inserted horizons and faults are built in time, the grid must be
depth converted. This process is skipped if the faults and input horizons are already in depth
3.Make Zones – Additional horizons are inserted into the 3D Grid by stacking isochores up or
down from the previously input horizons.
4.Layering – The final step is to make the fine-scale layering, necessary for property modeling.
These layers define the top and base of the cells of the 3D Grid.

Shell Refresher - Day2 Make Horizons – Slides • 43

 Make Horizons
Horizons Tab (Steps)

1. Append the desired number of items in table OR +

2. Use Multiple Drop; select input data (top of folder)

3. Drop in surfaces/interpretations for Input#1 using blue arrow

3 © 2004 Schlumberger Information Solutions. All rights reserved.

Make Horizons
Stratigraphy and Conformity (New)

Calculation based on Input data only

– assumes a lot of Input data.

Smooth difference between horizons

– assumes limited Input data.

Confirms to another horizon with the

given index (code) number
– assumes no Input data.

4 © 2004 Schlumberger Information Solutions. All rights reserved.

44 • Make Horizons – Slides Shell Refresher - Day2

 Make Horizons
Settings Tab
When running the Make Horizons process, Petrel re-samples the nodes of the input surface and uses
Minimum Curvature as a default to extrapolate horizon nodes that were not found in the re-sampling.

A. Select an algorithm for filling in

gaps (towards faults /undefined
areas); toggle Convergent Gridder
(better for faulted horizons).

B. Collapse zones to zero thickness if A

less than a specified thickness.

5 © 2004 Schlumberger Information Solutions. All rights reserved.

Settings tab:
The overall processes that occur when generating Horizons from input data (e.g., Surfaces) are:
1) Calculate and re-sample Horizon nodes on input.
2) Extrapolate values for horizon nodes that were not found in the previous step by Minimum
Curvature.
3) If Fault extrapolation distance is set in the faults tab (see next slide), or if there is no input
data and “Force horizons to be calculated for all segments” is toggled on, Petrel will use
Convergent Gridder to fill in the gaps of undefined areas adjacent to faults.
4) If Surfaces are in time or there is a T/D relationship with well tops, Well adjustment will be
performed.
5) Finally, Horizons are truncated according to the Horizon Type settings.

Shell Refresher - Day2 Make Horizons – Slides • 45

 Make Horizons
Faults Tab – Displacement 1

There are several settings for how the

displacement fault will act along the fault(s):

1. A fixed Min. and Max. displacement can be

set along a fault; offset below min. and above
max. will be changed, else input data is used
2
2. A constant offset along the fault is given
when Min. = Max.
Note: If off; Petrel will
3. Allow hinge allows the fault to swap between decide which side is
normal and reverse movement along the fault. 3 downthrown based on
Input data only.
4. Smooth; set smoothing by amount of
iterations (recommended 3-10).
Note: If Min./Max.
displacement is set,
5. Smooth; set smoothing by max change per 4
smoothing is
100 horizontal units (m/ft) along a fault. recommended

5
6 © 2004 Schlumberger Information Solutions. All rights reserved.

Make Horizons – Settings

Displacement – This option allows the user to set the maximum and minimum displacement
along the fault. Toggling on “Allow hinge” permits both reverse and normal displacement along
the fault. The last option is to set a smoothing on the displacement.

46 • Make Horizons – Slides Shell Refresher - Day2

 Make Horizons
Faults Tab – Inactive Fault and Growth Fault (New)

Inactive Fault: If a fault or all faults should have Growth Fault: This setting is only used for
zero displacement, there may still be some horizons conforming to another. When this option
artifacts due to extrapolation. is selected the thickness of the zone between the
2 conformable horizons may vary across fault(s).
Hint: If the fault has no activity on a horizon (e.g.
erosional), toggle off active fault to make it inactive Hint: Use this option when you want to preserve
instead. thickness variations due to syn-tectonic
sedimentation or other geological processes.

7 © 2004 Schlumberger Information Solutions. All rights reserved.

Make Horizons
Faults Tab – Distance Setup
Set a distance to the faults, specified in project Units;
Input data closer than the given distance is not used
in the Interpolation Algorithm.

Petrel then back-extrapolates to the fault plane to fill

in the gap caused by removing the data.

Fault distance can be different in front or

behind the fault (YELLOW or BLUE).

Press the icon to view the

Blue/Yellow faults in a 3D window.

8 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Make Horizons – Slides • 47

 Make Horizons
Faults Tab – Result
If distance is set to 0, the horizons may show
an artificial “drag-effect” on the fault plane

By specifying an appropriate fault distance,

the “drag-effect” may be removed

9 © 2004 Schlumberger Information Solutions. All rights reserved.

Faults Tab – Getting the parameters correct

When specifying an appropriate Distance, it is important to avoid both over-simplification and
artificial “drag-effect” detail. If your horizon and/or fault geometry is complicated, it is important
to experiment with various distances to achieve the best, most geologically reasonable settings.

Adjusting Distance and Displacement settings for individual faults

For best results, the Distance and Displacement settings can be specified for each fault:
1. Make sure that the “Use default” option is toggled off.
2. Expand the “Default for each fault” folder in the Make Horizons work window and select the
fault you want to adjust.
3. With that fault selected, adjust the Distance and Displacement settings; click “Apply” to see
the effects of the changes.
*Note that for a complicated model, modifying each fault could add significant time; it should be
used only when you suspect a problem using the general parameters.

48 • Make Horizons – Slides Shell Refresher - Day2

 Horizon Extrapolation
Distance to fault – Theory (Tips)
Horizon from input data
Distance from fault = 10 Distance from fault = 30
60 50 40 30 20 10 10 20 30 40 50 60 50 40 30 20 10 10 20 30 40 50 Horizon extrapolated along
the fault

Note: Better to use good (if available)

seismic Interpretation data directly in
Make Horizons to avoid Grid-drag
effects in the first place!
Fault Fault

Important:
The distance must not exceed the influence
area for the data, e.g. the fault compartment.
Set the sum distance </= fault block width.

10 © 2004 Schlumberger Information Solutions. All rights reserved.

Horizon Extrapolation
Principles

Why a problem?
False drag
Pinch out cells
Solution
Change distance to fault
Fault horizon intersection
line (Recommended)
Before After
Edit 3D grid (Not adjustments adjustments
Recommended)

11 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Make Horizons – Slides • 49

 Horizon Extrapolation
Horizon-Fault Intersection Lines (’Horizon Lines’) – NEW

Horizon lines are extremely useful as they

represent the desired relationship between a
fault and a horizon.

As Horizon lines are stored on the Fault Model

they can be used to generate multiple 3D
grids with consistent structural relationships.

Typically you would use the horizon lines

stored on the Fault model during Scale Up
Zones or to refine the geological grid to
ensure that the structural relationship remains
constant between the fine and coarse grids.

Note: The main Power of ‘Horizon Lines’ is that the intersection lines are physical
and transferable between 3D grids (as compared to normal 3D grid local edits).

12 © 2004 Schlumberger Information Solutions. All rights reserved.

’Horizon Lines’
Background

HINT: Delete unnecessary pillars

to ease editing!
Can be used as input in the
Make Horizon and Scale Up
Zones processes
Created from :
– Re-sampling of fault model
– Input polygons
– Digitizing

13 © 2004 Schlumberger Information Solutions. All rights reserved.

50 • Make Horizons – Slides Shell Refresher - Day2

 ’Horizon Lines’
How To Create and Edit Horizon Lines 1

Re-sampling to Fault Model – 1. Process:

2A
1. Open the settings for the horizon (or the
Horizon folder for all at once) to be edited and
go to Operations tab.
2B
2. Select resample from ‘3D grid to Fault Model’
(A) and press the Resample button (B).

3. The Horizon Lines are stored in a new folder 3

under the active/selected Fault Model. Note
that the colors are given according to the colors
of the original horizons (see next slide for use).

14 © 2004 Schlumberger Information Solutions. All rights reserved.

’Horizon Lines’
How To Create and Edit Horizon Lines
1A
Re-sampling to Fault Model – 2. View:

1. Toggle on the Horizon Lines. Go to

Style settings for the Fault model and 1B
set up to see Horizon nodes and
Pillars(A) and the fault plane (B)

2. To remember which horizon lines

are which, create a Horizon template
by right-clicking on Horizon folder and
selecting ‘Create/Update ‘Horizon’
property template

2
Note: Use the Fault filter to
display faults one-by-one.

15 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Make Horizons – Slides • 51

 ’Horizon Lines’
How To Create and Edit Horizon Lines

Re-sampling to Fault Model – 3. Edit: Add/Move Horizon node at Front side of fault
Add/Move Horizon node at Back side of fault

1. Activate the Fault Modeling process, Edit the Create Horizon Lines from selected polygons
shape points up/down along tangent of Pillars. 2
Smooth selected horizon nodes
Collapse Horizon nodes
2. Use the different tools from the Function bar.
Expand collapsed horizon nodes

16 © 2004 Schlumberger Information Solutions. All rights reserved.

’Horizon Lines’
How To Create and Edit Horizon Lines 1

Re-sampling to 3D Grid – 4. Process:

1. Go to Operations tab in the settings of the

Horizon folder or Horizon. 2

2. Select Resample from Fault Model to 3D Grid

(A). This will now incorporated the edits done in
step 3. Press the Resample button (B).
3
4. Go to the Make Horizon process and
toggle ‘Use fault Lines’. Press Apply.

Note 1: Fault lines are input data and can

be used as only source of input.
4
Note 2: The fault lines can now be used on
a 3D Simulation grids!..or any other grid.

17 © 2004 Schlumberger Information Solutions. All rights reserved.

52 • Make Horizons – Slides Shell Refresher - Day2

 ’Horizon Lines’
Create and Edit Horizon Lines by Polygons (Extra)
1. Input polygon:
Have the Fault Modeling
Process active.
Display the fault model and the
input polygon in a 3D window.
Use the Pick/Select tool and
select the input polygon .

18 © 2004 Schlumberger Information Solutions. All rights reserved.

’Horizon Lines’
Create and Edit Horizon Lines by Polygons (Extra)
2. Digitize polygon:
Click on the “Select horizon nodes”
icon.
Click on the “Create horizon lines from
a selected polygon on the active fault”
icon.
The new horizon line will be added to
the horizon lines folder.

19 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Make Horizons – Slides • 53

 ’Horizon Lines’
Create and Edit Horizon Lines by Polygons (Extra)
3. Digitize horizon lines directly:
Display the fault model in a 3D
window.
Click on the “Select horizon nodes”
icon.
Activate an existing horizon line or
create a new one.
Decide on creating the the horizon line
for the “front side” or the “back side” of
the fault.
Front side
Back side

20 © 2004 Schlumberger Information Solutions. All rights reserved.

’Horizon Lines’
Create and Edit Horizon Lines by Polygons (Extra)

3. Digitize horizon line directly

(cont.):
Click on the “Add/move horizon point
on the active horizon” icon.
Start digitize the horizon line.

Tips: Select Auto-select Horizon nodes to

add nodes between the picked and selected
horizon node.

21 © 2004 Schlumberger Information Solutions. All rights reserved.

54 • Make Horizons – Slides Shell Refresher - Day2

 Make Horizons
Make Horizon Output – QC visually in 3D (Tips)
Once the dialog settings are set, click
Apply and view the results in a 3D
Window

Remember that refining the Distance

and Displacement parameters in the
Fault tab settings is an iterative process
which may need to be done again after
QC in 3D Window

Use available Displays and Filters:

1) Edges – to see Zone division
2) Zone Filter – to strip zone by
zone

18 © 2004 Schlumberger Information Solutions. All rights reserved.

Make Horizons
Make Horizon Output – QC by Isochores (Tips)

When using Growth Faults option in Make

horizons, it may be useful to output Isochores
to see more subtle changes than a 3D view
would give:

1. Go to Settings, Output tab for the Zone in

Zone Filter

2. Press the Make Isochore button – a

thickness map is stored in Input tab; view in a
map/2D window.

19 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Make Horizons – Slides • 55

 Make Horizons
Extra – Allan Diagram (Simple QC)
An Allan Diagram in Petrel can be used for simple fault seal analysis, and to
QC juxtaposition of layers/zones across faults after Fault Modeling and Make
Horizon processes are performed:

1. Go to the Settings for

the Faults folder under the
3D grid and toggle on
Show Horizon Lines in
Lines tab.

2. Go to the Solid tab and

use the As Allan
(separation) diagrams
option.

20 © 2004 Schlumberger Information Solutions. All rights reserved.

1
Make Horizons
Extra – Fault Property (Advanced QC) 2

Fault Properties like a Transmissibility Multiplier or Fault

Displacement property can be used as a visual QC. Normally
early in a Structural modeling process we do not have
properties in 3D yet (like I and J permeabilities) so we
cannot generate the TM, but we can look at displacement:

1. Activate the Fault Analysis process

(Property Modeling).

2. Select the Calculated icon and press ok

to both dialog windows.

3. A Fault Properties folder is stored under

Faults. View the Fault Displacement

21 © 2004 Schlumberger Information Solutions. All rights reserved.

56 • Make Horizons – Slides Shell Refresher - Day2

 Make Horizons – Exercise

In this exercise you will insert horizons into the 3D Grid (Skeleton framework). This is the first
step in the vertical layering workflow.

Define the 3D grid domain

The domain (time or depth) setting guides Petrel in performing certain actions, such as whether
to tie to well tops at this step or in the Depth Conversion step. Therefore, the 3D Grid's domain
needs to be set by you to match the units of the surfaces it will contain. In this exercise the
inputs surfaces are in the time domain.

Exercise Steps
1. Use the same project as you had in the previous chapter (after fault modeling and pillar
gridding).
2. Open the settings window of the 3D grid by doubling clicking on the 3D Grid name.
3. Make sure the 3D grid has a yellow grid-icon (domain = time), if not set the Domain to
Elevation time under the Info tab.

Insert Surfaces into the 3D skeleton grid

Since the Skeleton grids are just a representation of pillars, we will now insert horizons into the
Skelton framework. That will be our first attempt of a coarse grid with physical cells. These input
surfaces or horizon interpretations may be the main first pass interpretations outlining the project
or reservoir area. Once surfaces / Interpretations are inserted, Petrel will output Horizons in the
3D Grid.

Exercise Steps
1. Double-click on the Make Horizon process.

2. Use either the Append item in the table icon or the Set number of items in table
icon located near the top of the dialog to insert one or several new rows into the table.
Create as many rows as you have horizons to build (check the Surfaces (time) folder in the

Shell Refresher - Day2 Make Horizons – Exercise • 57

 Input tab of the Petrel Explorer).

3. Select the data to create the horizons. Insert four horizons as shown in the figure above.
Use the 2D time surfaces (Base Cretaceous, Top Tarbert, Top Ness, and Top Etive),
depress the Multiple drop icon . This allows you to drop a range of data by only
selecting the topmost one in a folder..

4. For each horizon do the following:

Define the horizon's geological character (stratigraphy) - Do this by clicking on the box
under the "Horizon Type" column for a specific horizon. Set Base Cretaceous to be
Erosional and the other surfaces to be Conformable.
5. Select the Settings tab - to define parameters controlling interpolation and extrapolation
parameters. This is also where the influence of points that have been locked during editing
is specified. Refer to the Petrel online manual for details of each parameter. In this exercise
use the default settings.
6. Go to the Faults tab. All the faults that you incorporated in the Skeleton grid when doing the
Pillar Gridding are listed here. Set the fault distance to 100 for all the faults (for further
information see comments below). However it is better to set Base Cretaceous as ‘Not

58 • Make Horizons – Exercise Shell Refresher - Day2

 Active’, so that there will be no artificial throw related to the 100 m extrapolation.

7. To do this: Expand Base Cretaceous in the white dialog window and select all the faults.
Toggle off the Use default option to the left in the dialog. Now toggle off the Active fault
option as well; all the faults for Base Cretaceous will be indicated as not active.

8. In the Models tab you will now find four new horizons in the horizon folder. Display the
generated horizons and QC the fault cuts.
9. Show each zone and related horizons to spot artificial drag effects. Then go back to the
Make Horizon dialog and rerun the process.
10. In the Faults tab, toggle off the Allow Hinge option. With this option on, parts of the faults in
the eastern side (Boundary Fault) of the area will shift from Normal to Reverse. Display only
Top Ness and Top Etive Horizons and Zone 3 in the Zone filter. Re-run the process (also
with and without ‘smooth’ on) and see the result.

Shell Refresher - Day2 Make Horizons – Exercise • 59

 Before: Using Hinge option

After: Not using Hinge option

Stratigraphic Horizon types available in Petrel:

• Erosional (Base Cretaceous) - It can be used as the highest or as an intermediate surface
but not the lowest surface in the table. It will truncate all surfaces below it including the
base surface. This horizon can be thought of as the top of a sequence. It is the one surface
that can be used to terminate the upward extension of faults.

60 • Make Horizons – Exercise Shell Refresher - Day2

 • Conformable (Top Tarbert, Top Ness, Top Etive) - Any horizon in the table can be set to
Conformable. These horizons are truncated by all erosional horizons above them (type =
Erosion or Discont). They are also truncated by all horizons of type Conformable that lie
above them and below the next higher Discontinuity or Erosion horizon.
• Base - If used it must be the lowest horizon in the model. All horizons of type Conformable
above it will lap onto it. All horizons of type Erosion or Discont will truncate it. This horizon
can be thought of as the base of a sequence.
• Discont (Discontinuity) – represents an erosional surface in the middle of the stack of
horizons. It can never be the highest or lowest surface in the table. Horizons above it up to
the next discontinuity or erosional surface will lap onto it. All horizons below it are truncated
by it. These horizons can be thought of as the top or base of a sequence.

“Drag” into the fault plane

Not using all available data to build Key Pillars or sloppy construction of the input data will create
poorly defined faults. A poorly defined fault will often cut into structure data on one side of the
fault, incorrectly placing that data on the other side of the fault. When the data is used to build a
horizon it creates what appears to be structural ‘drag’ into the fault plane.

The Fault Distance can be used to blank input structure data near faults. This will eliminate the
problem although the correct solution is to fix the fault model. Fault Distance is defined in project
units. Petrel sets data in the area, from the fault plane out to a specified distance, to undefined
(deleted). During the gridding process the horizon is in these areas are extrapolated into the fault
based on the trend outside.
If you get a drag towards the fault plane, it will usually help to increase the extrapolation distance
for that fault. Caution should be taken as data along the entire length of the fault is disregarded.

Shell Refresher - Day2 Make Horizons – Exercise • 61

 Conform to existing Horizons
Now lets assume we have a 3D grid in time with only a top erosional horizon and a base faulted
horizon. Then we want to insert new horizons in between but there are no more surfaces to use
as input. We will use well tops as input data, and let Petrel use a new option for letting the new
horizons conform to a pre-existing horizons.

Exercise Steps
1. In models tab, under Gullfaks2004 TestMod, make a copy of the blue Conform to TE (DC)
depth grid. This has already been depth converted from the same time grid. Since we do not
have sufficient checkshots, the conforming process needs to be done in depth.
2. Activate the copy and rename it.
3. Open the Make Horizons process. The input fields are cleared as the original Make
Horizons process was done on the time grid.

4. Toggle off the ‘Calculate’ options for the already made horizons and insert two new horizons
in between. Drop in Top Tarbert and Top Ness well tops as new Input.
5. Now, in the ‘Conform to Another Horizon’ section select Yes. This is a useful option with
limited input data. Make sure Top Ness conforms to the underlying Top Etive (index code 4)
and Top Tarbert conforms to the new created Top Ness (index code 3).

6. Click Apply
7. Visually check the new grid, strip off horizons and use the zone filter. Note that for some
segments, that did not have well data, the segment was not calculated. Check with well
tops.

62 • Make Horizons – Exercise Shell Refresher - Day2

 ‘Horizons Lines’ – Horizon-Fault Intersection Lines
Horizon Lines is a representation of the Fault and Horizon intersections. The lines are physically
stored in Petrel and can be used independent of grids.

Exercise Steps
1. In models tab, under Gullfaks2004 TestMod, activate the yellow Gullfaks (skeleton) 3D
time grid.
2. Go to Operations tab under settings for the Horizons folder (note that you can do this step
for each horizon individually).
3. Resample from the 3D grid to the fault Model (using the attached Fault Model stored in the
same models folder as the 3D grid).
4. Select Match faults ‘By Id number’ (this is recommended; if the name of the fault(s) has
been changed or you are unsure of the history of the project the ID Number should be used
to match the faults. The ID Number is hidden from the user but is used internally in the Fault
Model and the 3D grid).
5. Press the Resample button. The horizon lines are stored under the Fault Model.

Shell Refresher - Day2 Make Horizons – Exercise • 63

 6. Open a 3D Window and view all the faults from the Fault model and the new Horizon lines.
7. Then visualize only Main fault East 1. Go to the Fault Modeling process and press the
‘Select Horizon Node’ icon in the function bar.
8. To visualize better the horizon lines, go to Style settings for the Fault Model. Make the Line
width 3 under ‘Lines between Pillars’. Toggle the Straight Line option and see the
difference. This can be good for subtle edits where the spline function is too curved.

9. To keep track of which horizon line belongs to which horizon, right-click on the Horizons
folder and select ‘Create/Update horizons property template’.
10. Go to templates tab and toggle on the new property template under ‘Discrete property
templates’.

64 • Make Horizons – Exercise Shell Refresher - Day2

 11. Start editing the Horizon line nodes using the widget along pillars. Try Top Ness only.
12. After editing, you want to resample the edited horizon lines back to the 3D grid.
13. First, visualize Top Ness horizon (if you did edits on that horizon), and the fault.
14. Go to Operations tab for the Horizons again. Now, select Resample from fault Model to 3D
Grid. Press the Resample button again.

15. The horizon is updated immediately at the horizon fault intersections. Now re-run the Make
horizons process. Make sure to toggle on ‘Use Fault Lines’.

Shell Refresher - Day2 Make Horizons – Exercise • 65

 16. See the update in the 3D window.
17. Now go to the Settings tab in the Make Horizons process dialog. Toggle on the ‘Lock all
Resampled Horizon Nodes’. The nodes will be locked when re-sampled from the Fault
Model and will not be changed during the Make Horizons process. This means they will
remain unchanged during smoothing, well adjustment and fault displacement constraints.

18. Press Apply and see the changes in the 3D window.

Optional: ‘Horizons Lines’ – Use in Simulation Grids

The main Power of ‘Horizon Lines’ is that the intersection lines are physical and transferable
between 3D grids (as compared to normal 3D grid local edits). Typically you would use the
horizon lines stored on the Fault model during Scale Up Zones or to refine the geological grid to
ensure that the structural relationship remains constant between the fine and coarse grids.

Exercise Steps
1. In models tab, under Gullfaks2004 TestMod, activate the yellow Sim Grid 3D grid. This is
based in the geologiocal grid but has I-J increments in Pillar gridding increased to 200 by
200 and is using zig-zag faults.
2. Go to the Scale up Zones/Select Grid process under Upscaling folder in Process Diagram.
3. The Sim Grid will automatically be selected as the coarse grid. Input the Gullfaks
(skeleton) time grid as the Fine Grid.
4. In the ‘Resample Horizons/Fault Intersections’, select ‘Use faults from the Fine Grid’.
5. Press Apply

66 • Make Horizons – Exercise Shell Refresher - Day2

 Optional QC
1. Go to the Make Horizons process for the Sim Grid and re-run the Make Horizons process.
Make sure the Use fault Lines options in toggled. Press Apply and see the results.
2. Go to settings tab and toggle on ‘Lock All resampled horizon nodes’. Press Apply and see
the difference.

Note: It may not always be a good match, related to the larger grid increment and the fact that
Zig-Zag faults are used on some faults.

Shell Refresher - Day2 Make Horizons – Exercise • 67

 Zonation and Layering – Slides

Zonation and Layering

What’s new?

• The Make Zones and Layering Process workflow - Recap

• Stratigraphic Intervals
• Parameter Settings
– Build zones from and along options
– Minimum cell thickness to “kill” artificial non-zero values - NEW
– Tolerance
• Well Adjustment

1 © 2004 Schlumberger Information Solutions. All rights reserved.

68 • Zonation and Layering – Slides Shell Refresher - Day2

 Make Zones and Layering
Process

Fault modeling Pillar Gridding Make Horizons Make Zones Layering

2 © 2004 Schlumberger Information Solutions. All rights reserved.

Make Zones
Note: Using
Stratigraphic Intervals (Steps) you may specify
directly no.of zones
and Input type.
1. Select the Stratigraphic interval.

2. Append the number of zones, select Isochore as

input type (other types available). 1

3. Press Multiple drop icon and insert top isochore 2 3

from the input data folder.

4. Input well tops if any.

3 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Zonation and Layering – Slides • 69

 Make Zones
Parameter Settings (Tips)

1. Select Build from: 2. Select Build along:

Top Horizon – isochores Vertical Thickness (TVT)

added from top of interval

Base Horizon – isochores

added from base of interval Stratigraphic Thickness (TST)

Both Base and Top Horizon –

only active if a Rest zone is
defined (all zones build towards Along Pillar – use when pillars are
it) close to vertical

3. Select Volume Correction:

Proportional Equal None – all volume error added to last zone

4 © 2004 Schlumberger Information Solutions. All rights reserved.

Make Zones
Minimum Cell Thickness (New)
Sometimes when gridding thin layers a problems Note: The calculations is done
with cells with artificial zero thickness (that in reality on the pillars, not the cell.
have remaining thickness) may occur. To solve this:

Use min. cell thickness

to collapse all thin cells less than
the given length (project units) to
zero thickness

No min.cell thickness

Min.cell thickness
Start from Base

Min.cell thickness
Start from Top

5 © 2004 Schlumberger Information Solutions. All rights reserved.

70 • Zonation and Layering – Slides Shell Refresher - Day2

 Make Zones
Unconformable Handling – (New; Modified 1)
New options for Conformable and 3D Grid Settings dialog
Unconformable handling.

Make Zones process dialog

6 © 2004 Schlumberger Information Solutions. All rights reserved.

Make Zones
Unconformable Handling – (New; Modified 2)
New options for Conformable and
Unconformable handling.

Make Zones process dialog

New algorithm in Make Zones
uses Zone Log to account for
missing tops - geologically
sound Isochores

7 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Zonation and Layering – Slides • 71

 Make Zones 1

Well Adjustment (Steps)

1. Make sure the well tops are included as input in the
Zones settings.

2. Toggle off Well adjustment is according to…, to

2
activate the Well Adjustment tab.

3. Select Across segment, or Inside segment only

(latter typically used for faulted reservoirs). Specify
influence radius.

4. Click Apply, and go to next Stratigraphic interval. 3

8 © 2004 Schlumberger Information Solutions. All rights reserved.

Layering
Settings (Steps)
1. Select the Zone Division (four different types).

2. Specify the number of layers (Proportional), cell thickness

(Follow top/Base) or relative proportions (Fractions).

3. Optional input of a Reference Surface (e.g., Clinoforms) Proportional

Follow Base

Follow Top
1 2
Fractions
3
Follow Base
with Reference

9 © 2004 Schlumberger Information Solutions. All rights reserved.

72 • Zonation and Layering – Slides Shell Refresher - Day2

 Layering
Minimum Cell Thickness (New)
Minimum cell thickness can be set in the Make Zones process,
and in the Layering process!

There is an option to include minimum cell thickness for

proportional and fractional zone division.

No min.cell thickness
Use Min. Cell thickness
to collapse all thin cells less Min.cell thickness
than the given length Start from Base
(project units) to zero
thickness Min.cell thickness
Start from Top

10 © 2004 Schlumberger Information Solutions. All rights reserved.

Layering
Results

Types of zone division:

Follow base

Proportional

Follow top

Fractions

Follow top + using a

depositional surface

Note: It is always a good idea to display

edges in 3D before applying make Zones
and Layering Processes; to see updates.

11 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Zonation and Layering – Slides • 73

 Make Zones – Exercise

The Make Zones process is the next step in defining the vertical resolution of the 3D grid. The
process creates zones between each horizon. Zones can be added to the model by introducing
thickness data in the form of isochores, constant thickness and percentages. Well tops can also
be used to tie top structures to the well picks. This process step may be skipped when no
zonation is given.

Create Zones
This is the process of inserting geological zones in the stratigraphic intervals above, in-between
and below the horizons that were inserted in the Make Horizons process. The zones are typically
created based on isochore grids, constant values or built proportional from existing horizons.
Well tops can be used for well adjustment of the horizons that will be created.

Exercise Steps
1. In the Gullfaks 2004 TestMod, activate the blue depth converted Gullfaks (skeleton)(DC)
grid.
2. Double click on the Make Zones Process step in the Process Diagram. Select the
Stratigraphic interval to be worked first.
3. For the Top Tarbert - Top Ness interval there are three isochors (therefore 3 zones). For the
Top Ness - Top Etive interval there are two isochors (2 zones). There will be no zonation for
the other stratigraphic intervals listed.
4. For each stratigraphic interval:

a. Use the new option Set number of items in table . Create as many
zones as you have isochores to insert. Note that for the Set number of items, a
separate pop-up dialog appears, making it easier for you to specify number of
zones, type of zone etc.

74 • Make Zones – Exercise Shell Refresher - Day2

 b. In the Input tab of Petrel Explorer, select the Isochores from the Isochores
folder and the subfolder that corresponds with the stratigraphic interval you are
working on.
c. Insert the isochores and well tops.
d. The horizon name defaults to "Horizon" unless well tops are used, and then it
will adopt that name. The isochore name defaults to "Zone" unless a 2D
Isochore Grid is selected and then it adopts that name. The names can be
edited at any time.
e. Select Build from base horizon and distribute the volume correction as
Proportional correction among the various sub intervals.
f. Select to build along Vertical Thickness (TVT).

g. Go to the Settings tab and make sure: ‘Well adjustment is according to the
settings in the “Make Horizons” process’ is toggled off. This makes the Well
Adjustment tab active
h. Under the Wells Adjustment tab, set well adjustment Inside segment only
(see figure below).

Shell Refresher - Day2 Make Zones – Exercise • 75

 .
i. Press Apply to generate the intermediate horizons and zone.
5. Repeat the procedure for the next Stratigraphic interval (Top Ness – Top etive). Press
Apply and see the updates in the 3D Wiindow.

Comments
In the Input Type column you have choices for the input data types. The choices include:
• Isochore – Using an isochore to assign the thickness for the zone.
• Percent - a number between 0 and 100 representing the percent of the stratigraphic
interval to assign to this sub-interval. This percent as a ratio is multiplied by the
stratigraphic interval to create an isochore. This isochore is then treated as if it were a 2D
Isochore Grid input to the process.

76 • Make Zones – Exercise Shell Refresher - Day2

 • Constant - a positive number representing the thickness of this zone everywhere. The
constants are in project units
• Conformable - no isochore, percentage, or thickness is available for this sub-interval.
Instead, top picks are provided for the horizon at the top of the sub-interval (if it is at the
bottom of the stratigraphic interval) or the base of the zone (if it is at the top of the
stratigraphic interval). These top locations are then interpolated on the existing horizon
above or below the sub-interval, thickness is calculated between the given pick and the
interpolated pick, and a 2D Isochore Grid created from that thickness. It does not appear
that sub-intervals other than the highest and lowest within the group can be assigned
conformable.
• Rest – This will take what’s left after e.g. assigning a constant zone thickness.

Minimum Cell Thickness

Sometimes when gridding thin layers there may be problems with cells with zero thickness that
have an artificial thickness. These can be removed using a minimum Cell Thickness option, that
looks at the length of each pillar and collapses them within a given window range.

Exercise steps
1. Zoom in on an area that seems to have a problem, normally towards a fault or an erosional
truncation. Use an Intersection plane.
2. In the Settings tab of the Make Zones process dialog, toggle on both the settings under
‘Minimum Cell Thickness’. Use 1 and Auto, then increase to 2 (in project units along pillar).

Shell Refresher - Day2 Make Zones – Exercise • 77

 3. Test some different options

Using Intersection planes for QC

The Intersection Plane is a vital tool for visualization and quality controlling Petrel models.
Playing through the 3D grid using an intersection plane is a very efficient method for quality
checking models.

Exercise Steps
1. Display the base horizon Top
Etive) of the Depth model in a 3D
window.
2. Go to Intersections folder and
activate the I or J Intersection
(click on its name to make it
bold).
3. Check the box to view the
intersection in the 3D window.
4. Click on the Player

5. to move the intersection. Note! The player will only appear if the active intersection is
displayed in the display window.

Comments
The colors of the zones can be changed from the Settings menu for each zone, found under the
Zones filter in the bottom of the 3D Grid (DC).

Optional: Export the intermediate horizons and zones

Horizons and zones made in Petrel can be exported as regular 2D grids in several different
format types. Export of 3D grids may also be performed for different formats. When exporting an
item, Petrel lists the available format types. The format types are also listed in the Getting
Started Manual.

Exercise steps
1. Under the Models tab, click on a horizon name (in the Horizons folder of the depth grid) with
the right mouse button and select Settings.
2. Go to the Operations tab. Define increments and x, y co-ordinates by inserting a surface
(optional).
3. Check Fill in faulted areas.

78 • Make Zones – Exercise Shell Refresher - Day2

 4. Press Make Surface - the 2D surface will be generated and placed in the Input tab window
(Petrel Explorer).
5. Click with the right mouse button on the converted surface and select Export. Define name
and format.
6. To export a zone, double click on a zone under the zone filter in the model and follow the
same procedure as above. The only difference is that you have to go to the Output tab for
Zones settings.

Note: If you want to create surfaces form all Horizons and isochors from all Zones, simply
double-click on the Horizons and Zone Filter folders and press Make surface and Make
Isochore. This will auto generate surfaces of all the horizons and put them in a folder in
the Input tab. The same goes for the Zones.

Shell Refresher - Day2 Make Zones – Exercise • 79

 Layering – Exercise

The final step in building the structural framework is to define the thickness and orientation of the
layers between horizons of the 3D Grid. These layers in conjunction with the pillars define the
cells of the 3D Grid that are assigned attributes during property modeling.
The Layering process will only make a finer resolution of the grid and no input data are used for
this process. The user can define the vertical resolution of the grid by setting the cell thickness,
define a number of cells or use a fraction code. When specifying the cell thickness the zone
division can either follow the base or the top of the zone.
The layers should be defined based on the properties to be modeled. Usually, the layer
thickness should be the thickness of the thinnest facies to be modeled. However, it is important
to keep in mind that the number of cells increases when the layer thickness decreases, so you
should not put in more detail than necessary.

Insert Layering

Exercise steps
1. Make sure that the model that includes the geological zones is active.
2. Double-click on the Layering process.
3. For each zone (identified by the name in the left column of the row), define your layering.
Use a variety of Zone Divisions, as seen in the figure below.
4. In the Common Settings, select Build along Vertical Thickness (TVT).

80 • Layering – Exercise Shell Refresher - Day2

 5. In the Ness-1 zone, select Follow Top as Zone Division. It is optional to input a reference
surface that will further subdivide this zone in addition to the cell thickness layering defined:
Go to the ‘Other data’ folder in Input tab and select DepositionalSurface and drop it in the
Reference Surface field using the blue arrow.
6. Press Apply and check the Statistics settings for the 3D grid. Should be more than 3 million
3D cells. This is due to the DepositionalSurface; remove it and press Apply again. Now you
should be down to around 800.000 cells.
7. Press Apply to see the result in the 3D window. If you are not happy about the way the
layers terminate (truncate/onlap) e.g like below:

8. …then go back to the Layering process dialog and toggle Use minimum cell thickness.
Toggle also Include Proportional/fractions, start from Base.

Shell Refresher - Day2 Layering – Exercise • 81

 9. Some of the strange truncations should be eliminated, since there is a minimum thickness
of the layers. Also converging effect will be avoided in proportional or fractional layers as
they can start from Base or Top as well.

Comments
Available zone division types in Petrel:
• Proportional - Define the number of layers to be created. The same number of layers will
be inserted in the entire interval, independent of thickness variations. However, if a horizon
is being truncated, than the layers will be truncated as well.
• Follow Top - the layers are the same and defined by a constant thickness. Based on this
thickness, the layers are built downwards and on-lapped when they meet the horizon
below. A depositional surface (Ref.Surface) can be used. In this case the layers are built
from this surface downwards instead of from the top surface. The depositional surface must
exist in the Petrel Explorer Input tab and it can be entered into the table by selecting it and
clicking on the blue arrow.
• Follow Base - just as for the Build from top option, except the layers are built from the
base and are truncated by the horizon above.
• Fractions - the number of layers are specified as well as the relative thickness between
them. Example: If you want to insert 4 layers into a zone and the second layer should be
twice as thick as the other three layers, then write: 1,2,1,1.

82 • Layering – Exercise Shell Refresher - Day2

 Fault Analysis – Slides

Fault Analysis
What’s New?

• Fault analysis – New 2003 SE

• Transmissibility Multiplier
– Fault seal analysis background
• Shale Gauge Ratio (SGR)
– Inherent shale content of the fault
• Grid Permeability
– I and J direction
• Fault Displacement and thickness
– Property relationship

© 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Fault Analysis – Slides • 83

 Fault Analysis
New in Petrel 2003SE
Transmissibility Multiplier
Depending on the geologic history of the fault, the TM can be assumed to be a
constant for a fault or all faults in a field (TM=1 is open, TM=0 is closed). The fault
can also be partially sealing (0<TM<1) or variably sealing (calculated).

Properties are calculated for each unique pair of touching cells from opposite sides
of the fault.
TM is input to a Simulator (e.g. ECLIPSE) or used simply as a visual assessment of
the sealing potential of faults.

© 2004 Schlumberger Information Solutions. All rights reserved.

Fault Analysis in Petrel allows the user to generate fault transmissibility multipliers, either directly
or by modelling fault properties, providing grid permeabilities and calculating the multiplier.
These are then used as input to the simulation or simply as a visual assessment of the sealing
potential of faults.

84 • Fault Analysis – Slides Shell Refresher - Day2

 Fault Analysis
Workflow
The overall aim would be to get a transmissibility multiplier (TM) for faults which could be
exported to a Simulator. However, some other fault properties like Fault displacement and Shale
Gouge Ratio could be looked at independently in a geological perspective.

Process steps: Activate the Fault Analysis process, set the settings/calculations and output a fault
property folder. Display the different properties. Export TM if calculated and desired

© 2004 Schlumberger Information Solutions. All rights reserved.

Create Transmissibility Multiplier property:

Constant for entire fault
Based on geological criteria/3D grid properties
Output is Transmissibility Multiplier property:
Review in 3D window
Add to Simulation Case

Shell Refresher - Day2 Fault Analysis – Slides • 85

 1
Fault Analysis
Displacement – 3D Structural Grid QC (Tips) 2

Fault Properties like a Transmissibility Multiplier or Fault

Displacement property can be used as a visual QC. Normally
early in a Structural modeling process we do not have
properties in 3D yet (like I and J permeabilities) so we
cannot generate the TM, but we can look at displacement:

1. Activate the Fault Analysis process

(Property Modeling).

2. Select the Calculated icon and press ok

to both dialog windows.

3. A Fault Properties folder is stored under

Faults. View the Fault Displacement.

© 2004 Schlumberger Information Solutions. All rights reserved.

Fault Analysis
Constant TM (Steps)
TM can be set constant or it can be a variable along a fault:

For Constant TM; simply type in:

1. Select Constant and type in a number.

2. To use different constants, simply select the

fault , tick off ‘use default’ and type in a different
1
constant..

3. Press Apply; the output is stored in the Fault

Properties folder under Faults in the 3D grid.
2

© 2004 Schlumberger Information Solutions. All rights reserved.

86 • Fault Analysis – Slides Shell Refresher - Day2

 Fault Analysis
Variable TM (Terminology)
TM can be set constant or it can be a variable along a fault:

For variable TM one has to calculate it in Petrel:

Permeability: the permeability of the fault contents.

Shale: the shale ratio (SGR) of the material in the fault. Only required if the standard equation
is used for fault permeability.
Thickness: the thickness of the fault body/zone.
Grid Permeability: the permeability of the 3D grid. Used for the calculation of the
transmissibility multiplier.

© 2004 Schlumberger Information Solutions. All rights reserved.

The way Petrel will calculate the TM is: First it will calculate the transmissibility between centers
of the two cells on each side of the fault (not looking at the fault itself), then it will calculate the
same transmissibility again taking into account the permeability (Permeability tab) and thickness
(Thickness tab) assigned to the fault cell face (two sides). The second transmissibility is divided
by the first and gives the TM which is ready for input to the simulator.

Shell Refresher - Day2 Fault Analysis – Slides • 87

 Fault Analysis
Variable TM (Steps)

1. Select Calculated.

2. Hit the calculate icon to launch the dialog

2
box.
1
3. Append a new equation (if several) and
set the settings correctly in the tabs,
depending on available input and desired
output. Press Apply to store the equation.
3

© 2004 Schlumberger Information Solutions. All rights reserved.

Fault Analysis
Fault Permeability (Permeability tab) – Lithology Dependent

Method 1 – Lithology Proportional

Define permeability factors for each lithology of the

selected facies model; values on either side of the
fault is calculated using a displacement-weighted
average (arithmetic/harmonic).

© 2004 Schlumberger Information Solutions. All rights reserved.

88 • Fault Analysis – Slides Shell Refresher - Day2

 Fault Analysis
Fault Permeability (Permeability tab) – Lithology Dependent

Method 2 – Lithology Juxtaposition

Define permeability factors for each lithology by

entering the permeability values for each
combination of adjacent lithology types.

© 2004 Schlumberger Information Solutions. All rights reserved.

Fault Analysis
Fault Permeability (Permeability tab) – Equation

Method 3 – Equation

Manzocchi’s equation:
Log Kf = (-A1*Shale)–(A2*log (Xf)*(1-Shale)expA3)

Kf – fault permeability
A1 – Empirical constant based on shale smear effect.
Shale – Shale Gauge Ratio (SGR) taken from Shale tab,
based on N/G or Vshale.
A2 – Empirical constant based on displacement effects.
A3 – Empirical constant based on A1 and 2.
Xf – Displacement along the fault.

Note: Using the Equation

requires input from the Shale
tab and the Thickness tab

© 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Fault Analysis – Slides • 89

 Fault Analysis
Shale Gouge Ratio (Shale tab)

Option 1 – Grid Property

SGR based on existing grid properties like Vshale or
Net-to-Gross.

Option 2 – Lithology dependant

Define SGR content for each lithology of a chosen
3D facies model.

Common rule for both options:

All cells the fault cell has been displaced by will be
accounted for, and their influence weighted to
amount of displacement. Fault Perm

Note: To be able to use the

Shale tab, the Standard
equation must be selected in
the Permeability tab. Shale Gouge Ratio

© 2004 Schlumberger Information Solutions. All rights reserved.

The SGR is a fraction related to shale content of a fault. When the SGR is 0.15 – 0.2 (or 15 to
20%) the fault is considered to have a complete seal (Watts 1987). Note that the Vshale or NG
property must be distributed in 3D; it will not work having just the upscaled properties so make
sure to select the correct model. The same goes for the Facies model when choosing Lithology
dependant.

90 • Fault Analysis – Slides Shell Refresher - Day2

 Fault Analysis
Fault Displacement and Fault Zone Thickness (Thickness tab)

Constant
Set a constant displacement independent of
the throw of the fault in project units (m/ft).

Proportional to throw
Thickness will be multiplied by the given
multiplier; gives a linear (or log) relationship
between the throw/displacement along fault
and the fault zone thickness.

Fault Perm
Log thickness

Throw / Fault Displacement

Shale Gouge Ratio
Linear Thickness

© 2004 Schlumberger Information Solutions. All rights reserved.

Fault Analysis
TM from I-J Permeability (Grid Permeability tab)

Grid Permeability - TM
The permeability on each side of the fault is
compared to the Fault Permeability (Kf)
calculated in the Permeability tab. Together
they will compute the Transmissibility
Multiplier

Note
If you are happy with only the intermediate
steps of Displacement or SGR, you can tick
off the option to create the TM.

VS.
Throw / Fault Displacement
=
Shale Gouge Ratio

© 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Fault Analysis – Slides • 91

 Fault Analysis – Exercise

Fault Transmissibility Multiplier

Hydraulic properties of faults are incorporated in reservoir flow models using Transmissibility
Multipliers (TM). Fault Analysis in Petrel allows the user to generate fault transmissibility
multipliers, either directly or by modelling fault properties, providing grid permeabilities and
calculating the multiplier. These are then used as input to a Simulator, or simply as a visual
assessment of the sealing potential of faults. The idea of this exercise is merely to introduce you
to the idea of how to use constant and variable TM’s and the different options in Petrel.
Remember that we will only look at methods of determining TM based on static geo-data.

Theory: Transmissibility Multiplier

Faults are known to be transmissibility barriers in reservoirs, affecting the fluid communication
across fault zones. The effect on fluid flow is modeled using a TM. The TM is not the same as
transmissibility; as an example lets assume the TM is 0.2 – this means the fault has reduced the
effective communication across a fault zone to 20% of expected value if the fault was assumed
simple with no thickness or internal fault properties. Normally the horizontal permeability is
reduced across a fault while the vertical connectivity increase. In a simulation with the TM, the
multiplier will act only on flow perpendicular to the fault.

The TM is a function of permeabilities of grid blocks and is also related to fault zone permeability
and thickness. This multiplier has until recent been difficult to model and it has several factors of
uncertainty (short range correlation of fault rock permeability and thickness, phyllosilicate
content of fault rock and its relation to permeability and Shale-Gauge-Ratio (SGR), diagenetic
effects and e.g. error in fault displacements). The sealing potential of faults is very important for
production, and knowledge about fault properties helps reduce the risk of producing faulted
reservoirs. A static seal is known to have a pressure drop over geologic time, while a dynamic
seal spans the life of a producing field. The dynamic sealing capacity was by Yielding (1999)
related to fault gouge permeability combined with capillary entry pressure. Therefore there will
be a dynamic pressure drop across a fault when production on one side exceeds the other side.
Therefore it is also important to have pressure measurements during a later simulation. Most
Simulators (like ECLIPSE) look at the TM combined with a defined threshold pressure. The TM
is applied by the simulator to the standard transmissibility that it calculates.

92 • Fault Analysis – Exercise Shell Refresher - Day2

 Depending on the geologic history of the fault, the TM can be assumed to be a constant for a
fault or all faults in a field (TM=1 is open, TM=0 is closed). The fault can also be partially sealing
(0<TM<1) or variably sealing. In the latter case the TM must be calculated. We will have a brief
look at some of these assumptions and see the result on one fault that we know. Remember
that TM also can be larger than 1 as it will then multiply to increase the transmissibility across
the fault.

Note: It is generally better to work on one fault at a time (faster), or a subset of faults (make
fault folders) that have similar properties (normally boundary faults with only grid blocks on one
side will not have a TM). Properties are calculated for each unique pair of touching cells from
opposite sides of the fault. The two sides of a fault face cell may have experienced different
histories; Petrel will therefore calculate all properties separately for the two sides, until the TM is
established – TM will have a single value independent of fault side.

Exercise – Assumption 1: Constant TM

1. Make sure to activate the grid called Fault Analysis model and open the Fault Analysis
process.
2. Toggle on “Create new property” and set the default Transmissibility Multiplier to Constant =
1. Press Apply.
3. Then highlight a fault and toggle OFF “use default”. Set this fault to sealing by giving it the
constant value 0 (See Figure 1).
4. View the results in the 3D window: Toggle on all faults in the Faults folder.
5. A new Fault properties folder is added under faults. Expand it and toggle on the
Transmissibility Multiplier (TM).
6. Go to the Color tab under settings for the TM and get min and max.
7. Use the Automatic legend from the Tool bar to see the colors update (max should be 1 and
min should be 0).

Figure 1: Settings for constant fault transmissibility in Fault Analysis.

Shell Refresher - Day2 Fault Analysis – Exercise • 93

 8. Export the fault data to simulation. Since this is a simple constant global TM, only the Grid
section will be exported to the simulator (ECLIPSE), containing fault grid info, and
keywords MULTFLT (global TM) and THPRESFT (threshold pressure).
9. Right-click on the faults folder and select Export. Use the format ECLIPSE Transmissibility
Multiplier Data (ASCII), and select only the fault you chose to model. Only one TM is
available to export.
Note: The exported fault info can consist of two sections, Grid and Edit. Edit is only required for
calculated transmissibility multipliers.

We will now look at how to calculate TM, assuming it varies across the fault.

Exercise – Assumption 2: Variable TM

There are different ways of looking at variable TM’s; We have four main inputs to the calculation
of TM, but we will only vary the fault permeability (Kf) while keeping the other settings the same:
Permeability: the permeability of the fault contents.
Shale: the shale ratio (SGR) of the material in the fault. Only required if the standard equation is
used for fault permeability.
Thickness: the thickness of the fault body/zone.
Grid Permeability: the permeability of the 3D grid. Used for the calculation of the
transmissibility multiplier.

The way Petrel will calculate the TM is: First it will calculate the transmissibility between centers
of the two cells on each side of the fault (not looking at the fault itself), then it will calculate the
same transmissibility again taking into account the permeability (Permeability tab) and thickness
(Thickness tab) assigned to the fault cell face (two sides). The second transmissibility is divided
by the first and gives the TM which is ready for input to the simulator.

We will try to model Fault Permeability (Kf) in three ways:

- Using Kf as a constant,

94 • Fault Analysis – Exercise Shell Refresher - Day2

 - using a juxtaposition approach between facies
- using a SGR approach based on a Standard Equation (Manzocchi) to estimate Kf.

Constant Kf:
1) Activate the Fault Analysis model and go to Fault Analysis process under Property
Modeling.
2) Open the Faults folder and display only Main Fault West 3.
3) Highlight Main Fault West 3 under fault name in the Fault Analysis dialog and toggle off
“Use Default”.

4) Choose “Calculated” Transmissibility Multiplier. Press the calculation icon and a new
dialog window pops up.

5) In the Permeability tab (Fault permeability - Kf), select Constant from the drop down menu
and type in 300 (mD).

Shell Refresher - Day2 Fault Analysis – Exercise • 95

 6) Shale tab is inactive (only used for SGR with standard equation option) so go to Thickness
tab. Select “Proportional to throw” (fault thickness linearly related to fault displacement),
with default multiplier 0.01 and no log relationship.
7) In the Grid Permeability tab, select the Permeability for the 3D grid Perm model1 (U). Click
OK.

Note: The grid permeability has to be populated in the 3D grid - if you select only a log upscaled
permeability the TM will not show. The best is to use tensor upscaled permeabilities (with
direction) for I and J direction, otherwise use the same permeability property. It is important that
the selected grid permeability is the same that is imported to the simulator. If you do not have a
permeability yet, toggle on “Do not calculate Transmissibility Multipliers”.

8) Once back in the first dialog window, click apply and display the Main Fault West 3 in a 3D
window.
9) Expand the Fault Properties folder under Faults folder. Toggle on the Fault Permeability
(Kf),; this should be constant 300. Then toggle on TM, refresh the colors and turn on the
Auto legend to see the color scale. You will notice that the TM has values close to 1.

10) Now go again to the Permeability tab in the dialog and type 2 instead of 300. Use the same
Equation 1 and click Ok and Apply. This will update the Kf to constant 2 mD. View the TM
again in 3D (refresh colors) and some values are now closer to 0.

Note: The observations above concludes that by reducing the Kf to a very low value (2 mD) the
TM was reduced, implying that the fault has reduced the effective communication compared to
the expected permeabilities looking only at the grid permeability.

Constant Kf, on similar facies contacts (lithology juxtaposition):

The Kf will vary along the fault, but will be constant on defined facies contacts that face
eachother.
1) Display the Main Fault West 2 in the 3D window. Go to the Fault Analysis dialog and select
“update existing property”.
2) Select Main Fault West 2, toggle off “Use default” and go to the Calculation dialog.
3) Append a new Equation and rename it in the Run name window like shown below (rename
also equation 1).

96 • Fault Analysis – Exercise Shell Refresher - Day2

 4) Select Lithology – juxtaposition from the drop down menu for Kf in the Permeability tab.
Select Facies model1 (U) as Lithology property. Input the data as shown in the table
above for the different facies contacts (assuming Units in mD).
5) Set the same settings as previous exercise in the Thickness and Grid Permeability tabs.
Click OK.
6) Make sure to that the correct equation is selected for Main Fault West 2. Click APPLY.
7) Display the fault in 3D window with the TM fault property. Refresh the colors and then
display both faults to compare.
Note: Main Fault West 3 has uniform high values (around TM=1) while Main Fault West 2 has a
color scale ranging from 0 to 1.

Estimated Kf, SGR based on N/G (Standard equation):

1) Display Main fault NS 1, go to Fault Analysis dialog and highlight it. Take off “Use Default”
and go to Calculation.

Shell Refresher - Day2 Fault Analysis – Exercise • 97

 2) Select Standard Equation as method for estimating Kf and use the default parameters set
for Manzocchi’s equation.

Log Kf = (-A1*Shale)–(A2*log (Xf)*(1-Shale)expA3)

Kf – fault permeability
A1 – Empirical constant based on shale smear effect.
Shale – Shale Gauge Ratio (SGR) taken from Shale tab, based on N/G or Vshale.
A2 – Empirical constant based on displacement effects.
A3 – Empirical constant based on A1 and 2.
Xf – Displacement along the fault.

Note: If permeability (Kf) is only dependant on displacement, set A1 and A3 to zero.

3) In the Shale tab choose “From grid property” and select NetGross model (U).

Note: the shale ratio is based on the N/G property in the 3D grid. This accounts for cells which
have moved past the fault face cell during faulting. Properties are displacement weighted, i.e. a
value is assigned to each facies type and the value in the fault face cell is an average of the
values of all the cells that the fault face has moved past. This method describes the smearing of
properties along the fault cells that a shale has moved past. It assumes that the fault movement
is parallel to the fault pillars.

4) Let Thickness and Grid Permeability tabs remain the same as previously set.

98 • Fault Analysis – Exercise Shell Refresher - Day2

 5) Click OK and make sure to select the correct function for TM before clicking APPLY in the
next window.
6) Display the Main Fault NS1 in the 3D window and toggle on the new property Shale gouge
ratio (SGR) found under the Fault Properties folder.

Shell Refresher - Day2 Fault Analysis – Exercise • 99

 Note: The SGR method use Manzocchi’s equation in the Standard Equation method for
estimating Kf. Like the other methods, we assumed a linear relation between fault zone
thickness and displacement (look at those properties as well and compare). In this case Kf is a
function of lithology displaced past a certain point along the fault, this leads to the TM
decreasing with both increasing fault displacement and more shale being involved in the fault
zone.

The SGR is a fraction related to shale content of a fault. When the SGR is 0.15 – 0.2 (or 15 to
20%) the fault is considered to have a complete seal (Watts 1987).

It is worth mentioning that Manzocchi et al 1999 stated that the influence of fault displacement
on fault zone permeability (Kf) was not good at low shale content – therefore the equation may
not give reliable estimates of permeability as SGR goes to zero.

Hint:
A) When fault displacement is small, both using the Lithology juxtaposition method or Standard
equation method will give TM close to 1 (fault zone is not very thick). However the lithology
juxtaposition method has a larger effect on transmissibility across the fault since there is
assumed no shale on the large permeability contacts.
B) When fault displacement is large, both methods reduce the TM.

100 • Fault Analysis – Exercise Shell Refresher - Day2

 Geometrical Modeling – Slides

Geometrical Modeling
What’s New?

• Geometrical Modeling Methods - Overview

• Use Different Property Filter Options - Recap
• Example Method
– Simple Facies assignment – NEW
– Connected Volumes – Moved
– Cells cut by Surface
– Zone and Segment models

1 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Geometrical Modeling – Slides • 101

 Geometrical Modeling
Property Modeling Overview

Property modeling involves a number of processes:

• Upscaling (blocking) of well logs
• Geometrical Modeling
• Facies Modeling
• Petrophysical modeling
• Data Analysis

General tools for working with properties :

• Property Calculator
• Property Filter
• Property Player

2 © 2004 Schlumberger Information Solutions. All rights reserved.

Geometrical Modeling
Methods
After a 3D grid has been made, one can assign values to
each cell. One cell can have ONE value only!:
Geometrical modeling can be done when no input data is
available. Values are assigned based on the grid geometry.
Note that if logs are available, you can upscale the well logs and
do Facies/Petrophysical modeling directly.

Examples of use:
• Create Bulk Volume to check for
negative cells
• Create Cell angles to inspect
non-orthogonal cells
• Resample seismic into the 3D
grid cells
• Create Above Contact property
to use in SW-calculation

3 © 2004 Schlumberger Information Solutions. All rights reserved.

102 • Geometrical Modeling – Slides Shell Refresher - Day2

 Geometrical Modeling
1A
Property Filter Options
There are three types of property filters:

1) The I-J-K filter (A), filters in I/ J direction and K

direction. With this filter there is also a Property
Player (B). It is useful for QC of grid.
1B 1A

2) The Index filter skips a user defined amount of

cells in I-J-K direction. It is useful for viewing a
fence diagram.

4 © 2004 Schlumberger Information Solutions. All rights reserved.

Geometrical Modeling
Value Filter and Zone/Segment Filter
3) Value Filter - Inspect a bulk volume
property. Good for finding negative grid
cells:

4) Zone Filter and Segment Filter. These

can be used during property modeling and
also after modeling to filter values in
Histograms etc.

5 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Geometrical Modeling – Slides • 103

 Geometrical Modeling
Simple Facies Assignment - New
Assign Facies value between two surfaces:

1. Go to Geometrical modeling and select

Assign Facies between surfaces. 1

2. Input two surfaces, one as top and one as 2

base.If you work with a depth converted grid the
surfaces should be depth converted.

3
3. Set Value to assign between surfaces (here 1;
according to facies code of template) and
remaining cells can be undefined or get another
facies code (here 0).

6 © 2004 Schlumberger Information Solutions. All rights reserved.

Geometrical Modeling
Connected Bodies – New place
Connect volumes of a discrete property:

1. Go to Geometrical modeling and select

Connected volumes.

2. Select e.g. a Fluvial facies model where a 1

channel object is created.
2
3. Select the facies code 3

7 © 2004 Schlumberger Information Solutions. All rights reserved.

104 • Geometrical Modeling – Slides Shell Refresher - Day2

 Geometrical Modeling
Cells Cut By Surface - Tips
Assign a value to all cells that are cut by a surface:

1. Go to Geometrical modeling and select Cells

cut by Surface.
1
2. Select a General Discrete template. 2
3
3. Insert a surface and assign a value. Choose to
keep the remaining values undefined or assign
also a value to them.

8 © 2004 Schlumberger Information Solutions. All rights reserved.

Geometrical Modeling
Zone and Segment Index vs. Constants in Segments and Zones
Zone and Segment index: Constants or Surface in Segments and Zones:

Useful when creating a property for use in Useful for a first approach to assign property
property calculator, or when used as values (continuous) in the zones and/or segments.
Region Index Property in Define Simulation
case (Eclipse).

9 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Geometrical Modeling – Slides • 105

 Geometrical Modeling – Exercise

Geometrical properties are models created by using pre-defined system variables such as Cell
Height, Bulk Volume, Depth and Above Contact.
Each cell will get a numerical value corresponding to the selected system variable. These
properties can be important in volume calculations and in mathematical operations between
petrophysical properties (e.g. for Sw transforms).
The Geometrical Property Modeling process allows the users to create models of geometric
properties in addition to simple modeling operations.

Creating a simple Facies

1. Activate the Final (DC) 3D grid.
2. Since we have a depth converted grid, we want to use some of those Horizons, but they
have to be converted to Surfaces first. Double-click on Top Tarbert and Top Ness horizons
in the Horizon folder of the 3D grid. Go to Operations tab and make sure to toggle Fill in
faulted areas (default), and press Make surface.
3. The two horizons are now converted to surfaces and stored in the Input tab.
4. In the Geometrical Modeling process dialog, select Create new property.
5. Select Assign facies between surfaces as Method. Drop the two new surfaces into Top
and Base surface in the dialog window.
6. Now select Fluvial Facies as Property template.
7. Drop in the Simple Facies polygon located in the “Geometrical Modeling input” folder in the
Input tab.

106 • Geometrical Modeling – Exercise Shell Refresher - Day2

 8. Select which facies code to use; Since we selected Fluvial facies as template we wish to
use only two facies types: We want to assign Fluvial channel inside the polygon and
Background floodplain outside.

9. It is a good idea to open the Fluvial facies template from the Templates tab to check that the
codes are correct.

10. The result would look like below, with only two facies distributed in one zone. Filter away
Zone 1.

Shell Refresher - Day2 Geometrical Modeling – Exercise • 107

 Creating Concession Boundary properties
1. Activate the Final (DC) 3D grid.
2. Select again Assign facies between surfaces method.
3. Input ‘Base Cretaceous’ as Top Surface and ‘Top Etive’ as Base Surface and input
‘Concession Boundary’ as polygon (all stored in Geometrical Modeling folder).
4. Assign a value and let everything outside be undefined.

Cells Cut by Surface

1. Activate the Final (DC) 3D grid.
2. Select Cells Cut by Surface as method.
3. Use a General Discrete property and insert the Top Tarbert surface from the’ Geometrical
Modeling input’ folder.
4. Assign value 1 and let the other cells be Undefined.
5. Press Apply. Compare the new property with the surface used.

108 • Geometrical Modeling – Exercise Shell Refresher - Day2

 Connected Volumes based on Fluvial Facies
1. Activate the Final (DC) 3D grid.
2. Select Connected Volumes as method.
3. Use the Fluvial Facies property already created. It has channel bodies in Zone Tarbert 1.
Model the Channel facies only.
4. Go to Segment filter and toggle on only Segments 1,2 and 3. In the process dialog, toggle
on the ‘Use filter’ option.
5. Press Apply and see the result.

Note: If you already have an existing property, overwrite this one, use the filter option and select
‘Unchanged’ as value to assign and ‘Undefined’ as value to other cells.

Shell Refresher - Day2 Geometrical Modeling – Exercise • 109

 Calculate volume of Connected Volumes in Process Manager
1. Activate the Final (DC) 3D grid.
2. Use the Connected Volumes created and calculate the volume of the ten first (largest)
body volumes of the channels.
3. Insert a new Workflow and set it up like below.

4. The workflow will generate a new property (define Bulk Volume as template for ‘Boddyvol’)
based on the Calculator statement. It is stored in the Properties folder. The second
calculator statement will then sum the volume of each body and output the results in a log
window.
Note: The workflow is already made and stored under Workflow tab as ‘Body Volume’. This can
be run if there is no time to create the workflow.

110 • Geometrical Modeling – Exercise Shell Refresher - Day2

 Identify Connected Bodies coinciding with Wells - Process Manager
The Workflow will identify connected bodies where they coincide with wells.
If a body is split by a fault, this method takes the body as still connected.

Note: If there is too little time to generate th eworkflow, open the already made workflow called
Connectivity test. Run this instead.

1. Activate the Final (DC) 3D grid.

2. Insert a new Workflow. Use the 3D grid Final (DC) and make sure a geometrical modeling
step is done so a connected volume property called ‘Body’ is used. This body I salready
generated and should contain about 200 connected volumes of Fluvial facies code 2
(Channels).
3. Three bodies needs to be generated:
i. OneBody - isolates a body and determine if the cells coincide with upscaled cells (if so
then cells=1)
ii. Upscale - contains value = 1 for upscaled cells and U for all other cells.
iii. ConnectedBodies - where resulting bodies connected to wells are connected.
4. Set the properties undefined first, then determine how many bodies are present so as to use
them in a loop.

5. Set up a loop. Note that the loop starts at 0 = and goes trough to maximum body number.
Select the body with current loop number and copy to a new property (ie. isolate body).
Determine all cells where the single body coincides with an upscaled cell and write cells to a
new property.

6. Sum number of cells the single body coincides with and upscaled cell (if sum=0 then not
intersected by a well).

Shell Refresher - Day2 Geometrical Modeling – Exercise • 111

 7. If body is intersected by well, then add body to ConnectedBodies property.
ConnectedBodies property contains results with all bodies connected to wells. Visualize the
final ConnectedBodies property.

The final workflow should look something like this:

112 • Geometrical Modeling – Exercise Shell Refresher - Day2

 Scale up well logs – Slides

Upscaling of Well Logs

What’s New?
• Understand and Use Discrete Logs
– Additional Averaging methods
• Understand and Use Continuous logs
– Additional Averaging methods
• QC of Upscaled Logs
– Well section window button – New 2003SE
– Making synthetic logs of properties

1 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Scale up well logs – Slides • 113

 Upscaling of Well Logs
Discrete Logs – Averaging

Discrete logs like for example facies

have only integer values (0,1,2 etc.).

1. The Averaging method is normally

Most of (will use the value which is most
represented in the cell).

2. In some situations it may be necessary

to use the weighting option. Toggle the
Use weighting and fill in Weighted tab.

2 © 2004 Schlumberger Information Solutions. All rights reserved.

Upscaling of Well Logs

Continuous Logs – Averaging

Continuous logs like for example porosity have

real number values (decimals).

1. Select the Average method. It is

normally Arithmetic for porosity (averages
all the values in the cell equally).

2) It is usually necessary to bias to facies.

Use the biasing option, to bias against a
discrete upscaled log (see next slide).

3) You can specify that there

should be a Min. number of
points passing through a
cell for it to be populated
with a value.

3 © 2004 Schlumberger Information Solutions. All rights reserved.

114 • Scale up well logs – Slides Shell Refresher - Day2

 Upscaling of Well Logs
Averaging Methods for Continuous Logs
Arithmetic mean - Typically used for properties such as porosity, saturation and
net/gross because these are additive variables.

Harmonic mean - Gives the effective vertical permeability if the reservoir is layered
with constant permeability in each layer. It works well with log normal distributions. It is
used for permeability because it is sensitive to lower values.

Geometric mean - Normally a good estimate for permeability if it has no spatial

correlation and is log normally distributed. It is sensitive to lower values.
Generally:
RMS (Root Mean Squared) - Will provide a strong bias towards high values.
RMS > Arithmetic
Minimum/ Maximum - Samples the min/max value of the well log for the cell.
Arithmetic > Geometric
Mid Point Pick – Will pick the log value where the well is halfway through the
cell. Essentially a random choice and is therefore more likely to give a property with the Geometric > Harmonic
same distribution of values as the original well log data.

Random Pick – Picks a log point at random from anywhere within the cell. This random option avoids
the smoothing tendency of other methods and is therefore more likely to give a property with the same New 2004
distribution of values as the original well log data.

4 © 2004 Schlumberger Information Solutions. All rights reserved.

Upscaling of Well Logs

QC of Upscaled Logs – Button (2003SE)

Once upscaling is done it is a good habit to QC

the upscaled cells with the original log:

Press the Show results in Well Section or set up

a Well Section display on your own (discussed
below).

If setting up a well Section; toggle raw logs from

Global well logs and upscaled properties from
Properties folder.

View the logs and Upscaled cells of interest; check

for inconsistencies. Review Upscaling if necessary.

5 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Scale up well logs – Slides • 115

 Upscaling of Well Logs
QC of Upscaled Logs – Synthetic Logs (Tips)
Create Synthetic logs; these will resemble the
properties (Upscaled logs) but are stored as
physical logs under Global Well Logs

1. Go to the Make logs tab of settings for Wells.

2. Select the property to make log from.

3. Press Make logs button; the log is stored in

the Global Well Logs folder and can be toggled
on from there.

6 © 2004 Schlumberger Information Solutions. All rights reserved.

Synthetic Logs
These logs can be created from any property in the Properties folder of a 3D Grid.
It is most useful as a visual QC of continuous data.
In the Well Section window it may be faster toggling on the properties directly form the
Properties folder.
However, if you need to display the synthetic logs in a 3D Window along a Well path, you need
to create synthetic logs first. Then toggle them on from the Global Well Logs folder.

116 • Scale up well logs – Slides Shell Refresher - Day2

 Scale Up Well Logs – Exercise

In order to distribute property values for each cell within the geometry of the 3D grid, scaling up
of well logs or well top attributes is required. The scale up of well logs process performs an
averaging of original log values within each cell cut by the well path.

Scale Up Well Logs

There are no major changes in the Scale up well logs process. A couple of new averaging
algorithms are added.

Exercise Steps
1. Activate the 3DGrid Final (DC). In the Petrel process window open the Scale Up Well Logs
and use the ‘Create new Property’ option.
2. Select the wells to be included in the process (use all wells for this exercise), and select
input from: Well Logs.
3. Select the Fluvial facies log to be upscaled from the drop-down menu.
4. Toggle Create new property again and select Porosity as the log to be used.
5. Press Apply to create the upscaled property. Now Press the Show result in Well Section
button, which turned active after you generated the Porosity model.

Shell Refresher - Day2 Scale Up Well Logs – Exercise • 117

 6. A Well Section window will open and automatically display the original log and the upscaled
log side-by-side. This is a great way to QC the upscaling process parameters and settings,
and it is easy to do changes and re-create the upscaled logs. Toggle on Well Tops and
use the Create/edit curve fill tool form the function bar to fill the original porosity log.

118 • Scale Up Well Logs – Exercise Shell Refresher - Day2

 Property Modeling – Slides

Property Modeling
Edit of Input Seismic Well Fault Pillar Depth Conversion
Correlation Modeling Gridding
Edit Input Data
Intro to Petrel
Import data

Facies Petrophysical Plotting Volume Well Design

Modeling Modeling Calculation

Process Manager

Shell Refresher - Day2 Property Modeling – Slides • 119

 Property Modeling
What’s New?
• Interface Unchanged
• General - Use variables for many parameter settings - NEW
- Variable Orientation etc for Channel facies
- Variable for channel fraction
- > used with Process Manager
• Facies Modeling
- Neural Nets - NEW
- Transition Boundary editing in Facies Transition Modeling - NEW
• Petrophysical Modeling
- Neural Nets - NEW

2 © 2004 Schlumberger Information Solutions. All rights reserved.

Facies Modeling
Common Settings - Review

Two Modeling Settings buttons are

available:
1. Common 1 2
2. Zone Settings

Define Common settings for all zones: 3

3. Use filter
4
4. Ensure that all cells get a value
5. Number of realizations
5

3 © 2004 Schlumberger Information Solutions. All rights reserved.

120 • Property Modeling – Slides Shell Refresher - Day2

 Facies Modeling
Zone Settings - Review
Define Zone settings for ALL
zones:
1. Leave the Zones button off. 1
2. Press the Leave Zone.
2
Unchanged button. You will now
use one algorithm for All Zones.

Define Zone settings for EACH 5

zones:
3. Press the Zones button. 3
4. Select the zone of interest from 6
the drop-down menu. 4
5
5. Press the “padlock” icon to
deselect Leave Zone Unchanged
to activate the zone settings.
6. Select a method from the drop-
down menu Method for
zone/facies.

4 © 2004 Schlumberger Information Solutions. All rights reserved.

Facies Modeling
Object Modeling – Fluvial Channels

1. Property and zone selection 1A

A. Make sure to pick the correct property; must be
upscaled i.e. have (U) as suffix.
B. Select Object Modeling as method for one zone.
1B
2. Facies body:
A. Select the Fluvial channels icon to insert a channel
2A
body.
B. Select facies properties to match Channel and Levee.

2B
3. Fraction:
• Use fraction of Channels and Levees from upscaled 3
cells.
• …or type in a fraction.

5 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Property Modeling – Slides • 121

 Facies Modeling
Object Modeling – Fluvial Channels

1. Layout:
• Specify Orientation, Amplitude and Wavelength.

Note: Drift will Apply randomness to each parameter.

2. Channel:
• Specify the width and thickness of the channel
• Thickness can be in distance units or as a fraction of
the width.

3. Levee:
• Levees are “gull wing” shaped deposits on the side
of the channel.
• Specify width and thickness (smaller than channel).

6 © 2004 Schlumberger Information Solutions. All rights reserved.

Facies Modeling
Fluvial Channels – Variables in Process Manager (New)

Run Uncertainty on Channel

direction in Process Manager
Variables must be defined $d in PM;
then run several realizations

7 © 2004 Schlumberger Information Solutions. All rights reserved.

122 • Property Modeling – Slides Shell Refresher - Day2

 Facies Modeling
Object Modeling – General Objects

Geometry:
• Select the Body Shape from drop down menu.
• Set the Orientation, Width and Thickness

Rules:
• Specify whether the facies
will replace other facies or
not.

Probability:
• Vertical (function curve)
• Areal (probability map)

8 © 2004 Schlumberger Information Solutions. All rights reserved.

Facies Modeling
SIS – Global Facies Distribution Control
From Data Analysis:
• Attribute Probability curves or
• Vertical Proportion curves

From Upscaled Logs:

Based initially on upscaled facies

… Or choose between
distribution based on:
• Probability cube
• Probability surface
• Vertical probability function

9 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Property Modeling – Slides • 123

 Facies Transition Simulation
Truncated Gaussian Simulation Theory

0.4
0.6
0.81.0
1.21.4
+
1.6 1.8 2.0 2.2 2.4 2.6

Deterministic trend based on well Residual field based on variogram

data and interpretation (to keep and well data (for interfingering
the facies ordering and effect)
boundaries)

The final model by re-truncating the combined field

10 © 2004 Schlumberger Information Solutions. All rights reserved.

The deterministic trend is based on the point or line source location, orientation, and the position
and angle of the facies boundaries. The facies boundaries can also be edited.

The residual field is computed between the deterministic trend and the upscaled cell values. For
example: if the deterministic trend value is equal to upscaled facies value then the residual value
is equal to 0.0.

If the upscaled cell value is different to the trend value then there is some residual value to
distribute into the cell between well locations.

Residual values are distributed based on interfingering parameters (I.e., a variogram and a
variance value that describes the correlation of residual values). The variance is used when
calculating the residual values in the upscaled wells. The square root of the variance is used as
the standard deviation in the Gaussian algorithm.

124 • Property Modeling – Slides Shell Refresher - Day2

 Facies Transition Simulation
Aggradation Angle

Aggradation angle: angle

between facies boundary and
grid layer

11 © 2004 Schlumberger Information Solutions. All rights reserved.

Facies Transition Simulation

Edit Transition Lines (New)

Edit lines: Normally used

together with interfingering
(variance)

12 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Property Modeling – Slides • 125

 Facies Transition Simulation
Example of Complex Property – Including Channel Bodies (Tips)

13 © 2004 Schlumberger Information Solutions. All rights reserved.

Facies Transition Simulation

Complex Property – Result

Make sure to use the

initial FTS model as a
‘Background property’

14 © 2004 Schlumberger Information Solutions. All rights reserved.

126 • Property Modeling – Slides Shell Refresher - Day2

 Facies Modeling
Interactive Modeling - “Paint” Facies (Tips)
Brush type
Irreversible process:
• Use Simbox view.
• Will overwrite all other facies, including
upscaled cell values. No undo!

Editedfacies
Initial faciesmodel
model
Facies type

Radius

Height

Profile

15 © 2004 Schlumberger Information Solutions. All rights reserved.

Petrophysical Modeling
Zone Settings
1. Define Zone settings for each zone: 2. Select proper modeling technique:
A. Press the Zones button. A. Press the “padlock” button to open the
B. Select zone of interest from the drop- settings for the selected zone
down menu. B. Select a method from the drop-down
menu for the zone

1A

1B
2B
2A

16 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Property Modeling – Slides • 127

 Petrophysical Modeling
Sequential Gaussian Simulation (SGS) – Set Up

1. Property and zone selection 1A

A. Make sure to pick the correct upscaled
property ([U] as suffix).
B. Select SGS as method for the zone. 1B

2. Variogram tab: 2B
A. Specify Range, Nugget and Type
B. …or get a variogram from Data Analysis

3. Distribution tab:
• Select Standard and From Upscaled logs
2A

Used if upscaled logs

Used if no/few upscaled logs

From distribution function 3

17 © 2004 Schlumberger Information Solutions. All rights reserved.

Petrophysical Modeling
Property Calculator – SW Property Example
SW is a function of height above fluid contact,
permeability and porosity. Two main options:

1) Type in a simple expression. This requires:

•
•
Above Contact property from Geometrical modeling.
Function (imported or made in Petrel (SW vs. height)).
=

+ New 2004

2) Load a more advanced expression (.mac/.txt file):

• Toggle From file, browse and Run.

18 © 2004 Schlumberger Information Solutions. All rights reserved.

128 • Property Modeling – Slides Shell Refresher - Day2

 Property Modeling – Exercise

Scaled up facies well logs can be used as input data. Trend data in the form of maps (2D Grids),
3D properties and probability functions can be used to influence and control the distribution of
the different facies. A pair of flow lines can be given in order to direct objects in a specific
direction. A number of source points can be given in order to specify where the channels start.

Both stochastic and deterministic methods are available for modeling the distribution of
petrophysical properties in a reservoir model. Scaled up well logs and/or trend data is used as
input and various user settings are available

Facies Transition Simulation (FTS)

Facies transition simulation is designed to be used in systems where there is a natural transition
through a sequence of facies. Typical examples include carbonate environments and
progradational fluvial sequences.The method involves first choosing which facies codes are to
be included in the sequence and in what order, then defining a trend along which the facies
codes are expected to change. Residuals between the trend and the well logs are then
distributed using the sequential gaussian algorithm and the defined variogram, the trend added
and finally the values converted back to the original facies codes.

Exercise steps
1. Activate the Final (DC) 3D grid.
2. Upscale the Facies log.
3. Under Templates tab, go to the ‘Discrete property templates’ folder and copy the Facies
template. Rename it to ‘Shorline’ and make ‘Backshore Lagoon’ (code 0), ‘Upper Shoreface’
(code 1), ‘Lower Shoreface’ (code 2), ‘Channel’ (code 3) and ‘Levee’ (code 4).

Shell Refresher - Day2 Property Modeling – Exercise • 129

 4. Go to the upscaled Faceis log and rename it to ‘Shoreline facies’, and select the new
‘Shoreline’ template.
5. Go to Facies Modeling, select ‘Shoreline Facies’ as property to model and use Facies
Transition Simulation as method. Remember to put the facies types in the correct order as
they have relative relationship to each other. Use only code 0-2.

6. Go to the Geometry tab and set up an azimuth according to the figure. For simplicity we
keep the rest of the settings default (prograding shoreface at 45 degrees). Do not spend
time matching wells as this is a ‘dummy’ model based on imagined data.

130 • Property Modeling – Exercise Shell Refresher - Day2

 7. In Interfingering tab, set the variance default (0.01). More variance, approaching 1, will give
a lot of interfingering, while a low variance will make the transition boundaries look like the
trend lines set up in the Geometry tab.

8. Press apply and see the result in a 3D window.

9. Go back to the Geometry tab, toggle on ‘Edit Lines’ option. Go to the transition lines and
edit so that they become curved. Rerun the model.

10. Make a copy of the ‘Shoreline Facies’ property and rename it to ‘Complex Shoreline Facies’.
11. Open Facies Modeling and use Object Modeling as method. Use the new property and
insert a Channel object. Make sure to match the Channel and Levee and make the fraction
around 30%.
12. Make sure the orientation is matching that of the Shorline (about 130 degrees).
13. In the Background tab, use ‘Other Property: Shoreline Facies’. Run the Model.

Shell Refresher - Day2 Property Modeling – Exercise • 131

 14. Look at the results in the different zones using zone filter.

Property Calculator
As for the well logs there is a similar calculator associated with the properties. This calculator
can be used to create new 3D properties and to do operations between properties. There is a

132 • Property Modeling – Exercise Shell Refresher - Day2

 whole range of predefined functions associated with the calculator, such as logical and
mathematical functions. It is possible to write nested statements and there is no limit for the
length and the complexity of such statements. If a calculation is getting complex then it is
possible (and useful) to create macros containing the calculation. The macro can be read from
file to do several operations in one go (saved with .mac extension).

Exercise Steps
1. Creating a new property model:
a. Right-click on the Properties folder in the 3D Grid Final (DC) and select
Calculator from the pull down menu.
b. Change the Properties template to Net/Gross and type into the white
formula field:
NG=0.8 (creates a constant Net to Gross property). Press ENTER.

c. Type in NetVol=Bulk_Volume*NG. Select the Net volume template.

Press ENTER. (If you do not have a Bulk Volume property, generate one
from Geometrical Modeling).

Shell Refresher - Day2 Property Modeling – Exercise • 133

 d. Display the new NetVol property in a 3D Window.
2. Calculating values:
a. The Calculator may be used as a normal calculator or for returns of single
values using properties and/or logs.
b. First create a new Pore Volume property, using the Bulk Volume multiplied by
the Porosity property.

c. The new Porevolume property is stored in the Properties folder, and in the
“Select property variable” field in the Calculator. Now do a simple operation
using the sum of all data (stored under Functions in the Calculator) for the
Porevolume property.

d. By pressing ENTER Petrel will output the pore volume in the Calculator field:

134 • Property Modeling – Exercise Shell Refresher - Day2

 Remember that it is given in the project units set for the Petrel project. If you
are uncertain; go to Project Settings in the Menu bar - Unit tab.

Exercise – Simple SW Calculation

The water saturation in the hydrocarbon zone is a function of height above the fluid contact, the
permeability and the porosity. The calculator can be used to make a water saturation property
that takes these parameters into account.
In this case we have water saturation in the oil zone that
can be expressed by the formulae to the right:
Where:
h = height above the OWC
K = permeability
Φ = porosity
a = 0.5, b = 0.2

To make it simple we can make a simple SW property based on a height function and an already
made ‘Above_contact’ property.

Exercise Steps
1. Activate the Final (DC) grid.
2. Right-click the Properties folder and select Calculator.
3. Use the function stored at the bottom of the Petrel Explorer Input tab. Examine it first.
4. Type in the name of the new Water Saturation property and also the name of the function
(as Petrel can only list the properties in the Calculator field).

5. Turn off the filter option .

6. Select SW as template and press the Enter. The new property is stored in Models tab.
Inspect it.

Shell Refresher - Day2 Property Modeling – Exercise • 135

 Volume Calculation – Slides

Make Contacts and Volumetrics

What’s New?

• Define Fluid Contacts remain unchanged

• New Principles of Volume Calculation
– Case tab and Results tab
• Use the Process Dialog To Create a New Case
– Using contacts
– Oil and Gas properties
– Filters and well boundaries
• Create and Use Outputs
– Report
– Distribution functions (CDF)
– Volume maps

1 © 2004 Schlumberger Information Solutions. All rights reserved.

136 • Volume Calculation – Slides Shell Refresher - Day2

 Make Contacts
Process – Define Fluid Contacts
1
Make Contacts is the last process under
Structural Modeling. How to create them:
2

1. Append the number of contacts.

3a

2. Define the contact type and name it.

3. Define the contact level

3b
a) Fixed for all zones and segments
b) Different for zones and/or segments

2 © 2004 Schlumberger Information Solutions. All rights reserved.

Make Contacts
Contact Settings – Visualization

The new contact set is stored in a Fluid

Contacts folder in the 3D Grid.

Go to Style tab under Settings for the Fluid

Contacts folder to control visualization
settings.

1 2

3 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Volume Calculation – Slides • 137

 Volume Calculation
Process
Cases tab
Running a volume calculation will create a Case, which will
have a folder in the Cases tab and a volume calculation sub
folder. This will act as a filter for viewing results..

Results tab
Acts as a filter for outputs of the Volume Calculation. The
standard parameters will be calculated for all volume
calculations as long as the appropriate input is supplied.

Note: The Cases and results tabs are also where Simulation
cases and results will be stored.

4 © 2004 Schlumberger Information Solutions. All rights reserved.

The Volume Calculation process step is located in the bottom of the Process Diagram. Double
click on the icon to start the process. Running a volume calculation will create a Case, which will
have an icon in the case tab and a volume calculation sub icon. This will then become the filter
for showing the results of the run in the plot window.
Once all of the appropriate inputs have been entered the user should press Run to do the
volume calculation. Pressing OK or apply will simply save the case to the cases tab. From
there it may be run at a later date via the right click menu, or via the volume calculation process.

138 • Volume Calculation – Slides Shell Refresher - Day2

 Volume Calculation
Process Dialog – Create a New Case (Properties Tab)

1) Define name for the new case or use an

existing Case name. 1

2) Select 3D Grid from the drop-down menu 2

to run the Volume Calculation on

3
3) Define the Hydrocarbon intervals to use.
You may use the Contacts created earlier.

4) Select the contacts from Fluid Contacts

folder in 3D grid and drop them into the 4
Contacts tab windows using blue arrow

5 © 2004 Schlumberger Information Solutions. All rights reserved.

Contacts – enter the contacts to be used. The contacts must be created in the Make Contacts
process.

Shell Refresher - Day2 Volume Calculation – Slides • 139

 Volume Calculation
Process dialog – Create New Case (Properties Tab)

5) Define Oil and Gas properties. Use either

an already made property (pull down menu)
from the grid, or toggle the Constant property
box and type in a value.
5
6) Under General Properties tab, activate N/G
and Porosity property or use constants.

6 © 2004 Schlumberger Information Solutions. All rights reserved.

Gen.Props – General properties are properties such as N/G, Porosity, Recoverable oil and
Recoverable gas. The properties must exist under the Properties folder as 3D properties. If they
do not exist, then constant values must be given.

Oil Zone Props – Properties specific for the oil zone, i.e. Water saturation or oil saturation for
the oil zone, Bo, Rs.

Gas Zone Props – Properties specific for the gas zone, i.e. Water saturation or gas saturation
for the gas zone, Bg, Rv.

140 • Volume Calculation – Slides Shell Refresher - Day2

 Volume Calculation
Process dialog – Create New Case (Results Tab)

7) In the Results tab; select which properties

and volume height maps to generate if needed.

8) When Volume Calculation is run the

properties are stored in the Properties folder of
the selected grid (useful to create subfolders if
there are many runs).
7
9) Maps are stored 8
in a Volume Maps
folder in the Input
tab.

7 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Volume Calculation – Slides • 141

 Volume Calculation
Process dialog – Create New Case (Boundaries Tab)
10) Click Apply to save settings and Run to 10
run the Volume Calculation. View output

11) Use filters (A) or Well Boundaries (B) if

desired. Click Run again

11A

11B

8 © 2004 Schlumberger Information Solutions. All rights reserved.

Boundary – This is where to specify boundaries and filters. Boundaries could be polygons that
describe a license boundary. Boundaries can also be used around some/all of the wells, but then
a radius is used. If you want to use a filter (Property Filter, Zone filter or Segment filter) you must
toggle on the ’Use Filter’ option under this tab. Note that all active filters will be used if this option
is toggled on.

142 • Volume Calculation – Slides Shell Refresher - Day2

 Volume Calculation 1
Output – Report
1) In the Results tab the filter settings are
stored. Right-click on Volumetrics folder and
2 3
select Settings.

2) Toggle on the Properties to display in the

report.

3) Specify the Detail level (see capture). 6B

5
4) Make sure the correct Volume Calculation is 6A
toggled in the Cases tab.
4
5) Press the Make Report button.

6) View the Report (A) and e.g. copy it to a

text-editor (can also be written directly to file
using Output to file (B).

9 © 2004 Schlumberger Information Solutions. All rights reserved.

To create a report can be done in 3 ways:

1) Make report while running the Volumetrics (Make Report button).
2) Leave it for later, run first, then right-click on the Volumetrics run in the Cases tab.
3) Right-click on Volumetrics in Results tab, go to Reports tab.

Shell Refresher - Day2 Volume Calculation – Slides • 143

 Volume Calculation
1
Output – Report Details
1) The output volumes have set formats
defined under each of the parameters
Template Settings.

2) Open the Templates tab. Expand the

Volume templates folder.

3) Open Settings for the parameters of choice.

3
4) Set number of decimal places. 4

5) Complete the 5
remaining
parameters.

2
10 © 2004 Schlumberger Information Solutions. All rights reserved.

Volume Calculation
Output – Distribution Functions
1
3
1) Open a Histogram Window from
Windows menu. 5
6
2) In the Cases tab toggle the Case or
Cases of interest.
4
3) In the Results tab filter out the
Property to view.

4) You can also use other grid filters

(segments and zones).
2

5) Press the Show Histogram icon.

6) Press the Show CDF curve icon

(Cumulative Distribution Function).

10 © 2004 Schlumberger Information Solutions. All rights reserved.

144 • Volume Calculation – Slides Shell Refresher - Day2

 Volume Calculation
Output – Volume Maps 2
1
1) Volume Maps are stored in Input tab.

2) Maps can be displayed in a Map window

(e.g. STOOIP).

3) The Volume maps can also be shown in 3D

– draped over a topographic surface: 3a
a) Output a depth Surface from Horizon
b) Go to Style settings of new surface and
choose Textured. 3b
c) Drop in the Volume map (blue arrow).
3c
4
4) Click Apply and view the
surface in 3D window.

11 © 2004 Schlumberger Information Solutions. All rights reserved.

Volume Calculation
Output – Volume Maps 2
1
Another new way to output the map is using the
Operations settings for the generated STOIIP
Property.
2
1) Output the depth structure grid

2) Use the depth structure grid as Surface

template in the Operations tab > Settings for
the generated STOIIP property in Models tab

12 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Volume Calculation – Slides • 145

 Volume Calculation – Exercise

Volumes are most commonly calculated in the Volume Calculation Process step. Volumes can
be calculated exactly within zones, segments and user defined boundaries (e.g. License
boundaries). The Contacts defined in the previous process (Make Contacts) are used as input to
the Volume Calculation process.

Volume calculations can be performed using several hypotheses in one operation. Each
hypothesis is called a Run. Running a volume calculation will create a Case, which will have an
icon in the case tab and a volume calculation sub icon. This will then become the filter for
showing the results of the run in the plot window. The Results tab controls created outputs and
allows the user to create extra filters for separating out results.
The design of Cases and Results tab is to allow the same setup for running ECLIPSE and
Streamline simulation runs, as for running simple Volumetrics

All settings related to the Volume Calculation process are found under the Volume Calculation
process window. The user will specify the type of output to generate (3D properties, report,
distribution functions) and which input to be used.
If you want to see the effect of different property models, then you should define several Volume
Runs, each using different inputs (such as property, contacts or boundary).
A report defined by the user will be created after the process has been run, listing all the
volumes per zone, per segment and/or per facies.
The user also has the option of creating a volume height maps (typically a HCPV or STOIIP
map).

Calculate bulk volume above a contact

Exercise Steps
1. Use your existing project or open Day2_Shell_Vols_WFlows.pet.
2. Open the Volume Calculation process, located in the Process Diagram under Utilities.
3. Toggle ‘Create new case’ and call it Case_1. Select the 3D depth converted grid Gullfaks
[DC].
4. Specify Hydrocarbon interval (Contacts tab):

146 • Volume Calculation – Exercise Shell Refresher - Day2

 a. Toggle on the both Oil and Gas. This will activate the Contact options in
Contacts tab. Highlight both contacts already made in Gullfaks [DC] (Models
tab) and drop them in using the Blue buttons.

b. In the General Properties tab, specify 0.8 for Net/Gross and toggle off constant
property and use Porosity [1]. Note that all the properties that can be selected,
should not only be upscaled but also populated in 3D using Petrophysical
methods previously mentioned in the Petrophysical modelling chapter.

c. In Oil properties tab keep the ‘Constant property’ option toggled on and use a
constant (0.3). Use a constant Sg of 0.1 and Formation Volume Factor of Oil
(Bo) of 1.21.

Shell Refresher - Day2 Volume Calculation – Exercise • 147

 d. Go to Gas properties tab and use the ‘Constant property’ also here property
also here (Sw of 0.5). Type in a constant Formation Volume Factor of Gas (Bg)
of 0.0009. Oil Saturation (free) should be 0.4.

e. Now go to Results tab (at the same level as Properties tab). Turn on the
properties as shown in the figure below. All the properties you toggle here will
be created and stored under the Properties folder of your 3D grid in Models
tab.

f. Go to Facies tab and select the Fluvial facies object property already
created. This option makes it possible to locate in which facies type the
properties you create are located (makes sure to output in the Report).

148 • Volume Calculation – Exercise Shell Refresher - Day2

 5. If you want to output a report at the same time as you run the case, then press the Report
Settings… in the Properties tab. This will open a new dialog window for setting up a report.
Set it up as below, then click OK (not Make Report).

Note: Select the same properties as you did for the Make Property option in Output tab. The
format to the right gives you the option of setting up the detail level. A small active ‘Example’
capture below the settings will show you how it will be output.

6. Make sure the Make spreadsheet report is toggled in the Output tab, then click Apply on
the main dialog. Apply will only save the settings and not run the Volumetrics. See both the
Results tab in the First Petrel Explorer and the Cases tab in the Second Explorer:

Shell Refresher - Day2 Volume Calculation – Exercise • 149

 7. Now press the Run button . A report will be created and properties will be stored
in the Models tab. Inspect both. After you have finished, exit the report window. Note that
this text report is not stored anywhere so you either have to save it as a text file or open it
at a later stage. How to do this:

a. Save as text file: Click the save icon and save it as .txt document. You may also
highlight certain columns/rows and use Ctrl C+V to copy and paste them directly in an
opened Excel document.

b. Save directly to file: Instead of visualizing a report in Petrel as you do a Run, you can
toggle the Output to file option in the Report settings… dialog:

c. Open at a later stage: This option is good to have if you run several cases with different
settings. This will create several cases in the Case tab. Now go to the Results tab and
double-click on the ‘Volumetrics’ folder. This will open the Report Settings dialog. Specify

150 • Volume Calculation – Exercise Shell Refresher - Day2

 what you want in the report, then go to Cases tab and toggle on the case of interest. Now
Petrel knows which case to output. Then press Apply and then the Make Report button.

Comments
• If negative volumes are detected a warning will be given. Negative volumes will occur if the
model contains crossing pillars. These are typically associated with vertically truncated
faults or other faults. The Bulk Volume property can be used to filter on negative volumes
to see where they occur. Do this from Properties > Settings > Filters > Values Filter and
filter on the Bulk Volume to only see negative values.
• Cell volumes are calculated exactly across segments, boundary polygons and contacts.
This is possible due to the triangulation of the cells.

Creating a STOIIP Map

A hydrocarbon column height map is the sum of all the values in the same X,Y position. For
instance, a STOIIP map will show the sum of STOIIP for every X,Y position in the entire grid. It
will therefore show you where to expect the highest concentration of oil.

Exercise Steps
1. Close the report window.
2. Open the Volume Calculation process again. Use the same settings as define previously
but toggle on STOIIP in the ‘Make volume height map’ section. If you don’t want to
regenerate the Bulk volume, you should toggle this option off.
3. You can toggle on the option ‘Overwrite existing properties (in the lower left part of the
window) to not create all the same properties over again.
4. Leave the map post-processing section default, then Apply and Run.

Shell Refresher - Day2 Volume Calculation – Exercise • 151

 5. After running the process, the map will be placed in a folder in the bottom of the Petrel
Explorer Input tab.

6. Display the map in a 3D window. You will probably have to click on the View All icon
and the View from Above icon to be able to see it. Then click on the map name in
Input tab and refresh the color scale using the Adjust Color Table on Selected icon.

Draping the STOIIP map on a depth surface

Exercise Steps
1. Open the settings for the Top Tarbert horizon, found under the Horizons folder under 3D
Grid (DC) in the Petrel Explorer Models tab. Under the Operations tab, press the Make
Surface button. The generated surface will be stored under the Input tab in the Petrel
Explorer.

152 • Volume Calculation – Exercise Shell Refresher - Day2

 2. Open the Settings for the generated Top Tarbert depth surface. Under Style tab >Solid >
Color, choose ‘Textured’. Select the STOIIP map by clicking on it in the Petrel Explorer
Input tab (make it bold) and enter it into the empty field by clicking on the blue arrow .

3. Press OK.
4. Display the Top Tarbert depth surface to see the STOIIP map draped over it. Remember
that you are now draping the STOIIP map for all zones on to the Top Tarbert surface.

Optional Exercises – CDF

How to display a Distribution Function

For example we may want to have a look at how much the STOIIP varies if the fluid contact
varies, or simply compare different case outputs.

Exercise Steps
1. Create a new case (Case_2). Just change the N/G under General Properties tab to 0.7, the
Sw under Oil Properties tab to 0.4 and Sw under Gas Properties tab to 0.6. Press Apply
and Run

Shell Refresher - Day2 Volume Calculation – Exercise • 153

 2. Open a New Histogram Window from Window tab.
3. Go to Cases tab and select which cases to compare (Case_1 and Case_2).
4. Go to Results tab and filter which properties to compare. Set up Results tab as in figure
below and press the Show CDF curve icon in Function bar to show the Cumulative
Distribution Curve in the Histogram:

Comments
• If you want to make a distribution function for all segments but only for some of the zones,
use the Zone Filter before running the Volume Calculation. To use the zone filter, toggle it
on under the Boundary tab under the Selected Run tab in the Volume Calculation process.

154 • Volume Calculation – Exercise Shell Refresher - Day2

 Process Manager – Slides

Process Manager
What’s New?

• Slight Interface Make-over

– Interface tools (Utility, Operations, Processes)
• Added options for Operations
• More Processes incorporated
• Variables in all processes
• Variables in calculators
• Output operations to variable (vol below surface/area etc.)
• Write to OUTPUT Sheet
• Create Folders / Move to Folders
• Call a workflow from within another (nested workflows)
• Runs Volumetrics and Simulation
• Uncertainty and Risk

1 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Process Manager – Slides • 155

 Process Manager
Overview

Application
• Provides a record of how a model was
created, extracting the details of which
input data and settings were used.
• Automatically update the model after the
input data has been changed.
• Create a number of different versions of
the model by varying process settings.
• Create different property realizations.
• Automate repetitive mapping\editing
tasks.

3 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Manager
Interface
Top Section - Workflow Description, , Name of workflow,
author name and date.

Left Section – Right Section –

Lists available Contains the
utilities inserted or auto-
(Conditions, selected steps of
expressions, loops the current
etc.), Operations workflow. The
and Processes. grey tips section
above gives info
about each
selected function.

Bottom Section – Contains predefined workflows and

tools for testing and saving workflows.

5 © 2004 Schlumberger Information Solutions. All rights reserved.

156 • Process Manager – Slides Shell Refresher - Day2

 Process Manager
General Tools
Warning
Instructs the Process Manager what to do
if errors occur during execution.

Run
Test will test the syntax of the workflow. If
failed logic – a Failed warning occur, if not
a green Test Ok will appear in the Status
field. Once tested, press Run and
Running will be the status until workflow is
completed with a green OK. The workflow
Visualization is now done, but NOT saved.
The clock writes the elapsed
time of the execution
Save
Sets full screen mode while Apply will save the settings of the
running the workflow workflow, but NOT run it. Ok will save and
Close. Remember to save before hitting
Hides the dialog when running Close.
the workflow

7 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Manager
Left Section Tools – Utility

Utility
(used to be Utility and Logic)

Contains simple actions such as Set active and

Display. Set Reference allows the use of a
variable to define an object.
Also contains logic or typical programming
functions such as loops and if statements,
definition of variables and feedback.

5 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Process Manager – Slides • 157

 Process Manager
Left Section Tools – Operations and Processes

Operations
Lists all the operations available from the
Operations tab of Settings of objects in the
Input tab.

Processes
List all available processes in Petrel that can
be used in the Process manager.

9 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Manager
A Simple Start – Predefined Workflows

Auto generate workflow…

Gives 5 options of predefined workflows already
programmed in Petrel

We will look at 2 of them:

1) Repeat all calculations on the active 3D grid
Creates a workflow which rebuilds the active
Petrel Grid and properties from the input data
using the settings last used for each process.
2) Make a scaled plot of each horizon in the
active 3D grid
The workflow automatically sets the header,
legend, scale bar, info box, etc. visible and ready
for output to the printer.The header and all
information will update with each new horizon that
is displayed.

10 © 2004 Schlumberger Information Solutions. All rights reserved.

158 • Process Manager – Slides Shell Refresher - Day2

 Process Manager
Predefined Workflows – Repeat All Calculations…
Update both Time and Depth grids
including Depth Conversion:

1) Copy the Time grid and highlight 1

the copied grid; select Repeat all
calculations on the active 3D grid. 2

2) Insert the Make Velocity Model 4B

process and the Depth Convert 3D 3
grid process below the Time Grid
process.
4A
3) Make sure to select overwrite:
“Depth grid”.

4) Highlight Depth grid, select Repeat

all… (A) and tick With Copy (B).

13 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Manager
Operations – Referring To Objects (Example)
1. Set Reference; var.a
points to the topmost object in
Example: Cut imported Fault Sticks against a surface. the folder first, then picks the
next for each loop.

2. Make copy so as not to edit

the original fault sticks.

3. Set the objects visible in the

active window.

4. Eliminate fault sticks above

specified surface.

5. An “End Loop” command is

necessary after using the “for
all icons in” command to
terminate the loop.

9 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Process Manager – Slides • 159

 Process Manager
Operations – Defining Variables (Example)
1. Define Variable.
Example: Assign a variable and use it in a redefinition
2. Make a copy of surface,
of itself
represented by (var.a) so as
not to alter the original.

3. Assign variable equal to

variable $elevation.

4. Set selected data objects

visible in active window.

5. Redefine the variable

elevation using itself.

6. Make another copy of

variable surface repr.by

7. Assign the new variable

elevation to the surface.

10 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Manager
Logical Functions and Tests (Example)

Example: Use of nested IF statements to meet several 1. Define Variable; picks a random
conditions. number between 1 and 30.

2. If the number is higher than 10

then it will output the two
messages below.

3. If the above condition is not met

it will look again to see if the next
condition is met; if yes it will output
the messages below, if not it will
look further down.

4. Finally if none of the above

conditions are met, then it will
output the final condition
statement.

11 © 2004 Schlumberger Information Solutions. All rights reserved.

160 • Process Manager – Slides Shell Refresher - Day2

 Process Manager
Looping (With Example)

The command will add a loop that will run a 1. Define a variable to assign an
specified number of times. A variable will be used to count elevation to a surface.
through the loops and the user is asked for two integers to
tell Petrel where to begin and where to stop counting. 2. Insert a ‘Loop’ function and
specify number of times to run
through the loop
An end loop command must be entered to
specify where the loop ends. 3. Assign the the variable
elevation to the surface.

4. Set the surface and intersecting

surface visible

5. Create a contour line where the

two surfaces intersect

6. Adjust the intersection by 50 m

increments.

12 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Manager
Suggestions

The Process Manager is indisputably the most

powerful tool in Petrel, particularly when it comes to
updating existing large models.

The Living Model concept uses Process Manager as

the basis for updating a model with new data
continuously (e.g. while drilling/logging wells).

The Process Manager is the Petrel “Uncertainty

Engine” with the ability to generate loops and scenarios
of key modeling process steps in Petrel.

Note: In the Process Manager saved workflows can be

used as Corporate company templates as they can
be dragged and dropped into any project from a
“Template Project” using the Open secondary project
option. The setup will be there, but the data will be
removed (red and black X).

13 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Process Manager – Slides • 161

 Process Manager
The Living Model Cocept
High uncertainty
Coarse model
Refining the model Economics

Petrel
Exploration / Smart Prospect

The Living Model Solution

Generation Solution

Production

Delineating the
Reservoir
Petrel
Drilling

Smart Prospect
Reducing
Generation Engineering
uncertainty
Lead illumination and
Qualification

Exploration Appraisal Development Production

OpenSpiritTM

GeoFrameTM or OpenWorksTM project environment

14 © 2004 Schlumberger Information Solutions. All rights reserved.

Smart Prospect Generation is made up of two main activities:

The first one is illuminating all possible leads. For this, you need Formula 1 speed in visualizing
massive volumes of seismic data. This also allows us to leverage the dramatically improved
seismic data coming from new acquisition technologies like Q-Seismic.
However, a lead is not a prospect. For a lead to become a prospect, you need to qualify it, and
you need to do it fast. Integrated and specific interpretation qualification functionality is required.

The Living Model

Provides the complete Asset lifecycle support, driving the Model to actual and actual to model
workflows, giving you the ability to replay a complete study in seconds using every new bit of
information and hence drives better business decisions.
The power of Petrel is using the seismic-to –simulation capabilities together with the Process
Manager for automatic updates of existing models. In this way, a model and uncertainties related
to all aspects of it can be established and hence evaluated in “no time”.

162 • Process Manager – Slides Shell Refresher - Day2

 Process Manager
Uncertainty and Risk

Process Manager
Automates creation of stochastic models. All
processes available. Seismic to simulation.
Variables in Processes
All standard numerical input can be replaced by
variables and controlled in the workflow.
Variables for Workflow Control
Extract information during the workflow and steer the
processes.

Reserves
Mean = 10.9 m3 106
Azimuth

Distribution

Structure

Contacts

Width

Faults

-1.5 -1.0 -0.5 0.0 0.5 1.0 1.5 2.0

del ST OIP (M m3)

15 © 2004 Schlumberger Information Solutions. All rights reserved.

Uncertainty analysis
Initial focus on static uncertainties (structure, contact, geology, petrophysical properties)
Volumetric uncertainty tool to collect scenarios, number of realisations, run volumetrics and
create cases.
Graphics to show sensitivities and report results
Reduce uncertainty using geological screening – FRONTSIM in Petrel.
Link to economics
The Process Manager is the engine !

Shell Refresher - Day2 Process Manager – Slides • 163

 Process Manager
Uncertainty – Establishing a Base Case Scenario
Workflow: BuildModel Description: Builds Model
1) A Base Case scenario (a Petrel model)
must exist before starting the uncertainty
evaluation.
2) Capture the Base Case model in a Base
Case Workflow by using the Auto
generate workflow > Repeat all
calculations on the active 3D grid
3) The Base Case Workflow can at the end
be used to rebuild the model from the
original data (after having run multiple
realizations or ‘experiments’). The Base
Case variables must be saved to
accomplish this (next slide)

16 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Manager
Uncertainty – Base Case Variables
Workflow: BaseVari Description: Sets Base Case Variables

$SEED: Property Seed

$SSEED:Structural Seed
$C1: Contact 1
1) Define a variable for each of the $C2: Contact 2
$NG: Net-to-Gross
previously defined Key $AZI: Azimuth (Channel direction;
Uncertainty Parameters Petrophysical variogram azimuth)
2) Create a sub-workflow (BaseVari) $WID: Channel width

which contains a listing of all the $V1: 1 std error on surface 1

variables and their Base Case $V2: 1 std error on surface 2
$ISO: 1 std error on isochore
values
3) In the BuildModel workflow, open
the corresponding processes and
replace the Base Case number by
the variable name (see next slide)

17 © 2004 Schlumberger Information Solutions. All rights reserved.

164 • Process Manager – Slides Shell Refresher - Day2

 Process Manager
Uncertainty – updating all processes affected
Facies Modeling
Update all processes in the BuildModel workflow to
Channel/Levee
use the variable instead of a constant value (Double-
fraction
click each process to access the process diagram)

Contacts

Channel width

Channel Azimuth

Variogram
Azimuth
Facies Modeling Seed

18 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Manager
Nested workflows
Workflows can be called from within
another workflow, i.e. Nested workflows

Example: From within one workflow, the

workflows called Headers, BaseVari and
BuildModel will be run

Nested variable: When this is toggled on, then the variables

defined in one workflow, whether it is within another workflow or
not, will be used by subsequent workflows.

19 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Process Manager – Slides • 165

 Process Manager
Recording the results
Running uncertainties and creating
several realizations/scenarios is only half
the problem. It is also important to
capture the values of the variables that
Example of an Output Sheet consisting of
were used in each case to be able to
Headers + Reported variables
reproduce a specific case.

The Process Manager allows values to be

written to a spreadsheet called an Output
Sheet.

The user controls how to design the

architecture of the output sheet: The
header columns, and which parameters
to be reported

20 © 2004 Schlumberger Information Solutions. All rights reserved.

Process Manager
Recording the results: Writing Headers to Output Sheet

Workflow: Headers Description: Define Headers

1) Define how the Output Sheet
shall look like in terms of number
of columns and which parameter
to be held in each column (The
parameters to be reported are
typically the variables defined in
the BaseVari workflow)

2) Create the Headers by using the 3) Enter the Row and

Output Sheet command Column number and the
Name of the variable to be
reported

21 © 2004 Schlumberger Information Solutions. All rights reserved.

166 • Process Manager – Slides Shell Refresher - Day2

 Process Manager
Recording the results: Writing Variables to Output Sheet
Workflow: Output Description: Writes output
1) After having written the Header
columns, then fill the subsequent 2 1
rows with data from running a
looped workflow.

2) Define the row number by a

variable to not overwrite when
running several loops
3A
3)
A -Example workflow that for
each loop runs the Output
workflow
B - writing all the variables to
the Output Sheet.

3B
22 © 2004 Schlumberger Information Solutions. All rights reserved.

Shell Refresher - Day2 Process Manager – Slides • 167

 Process Manager – Exercise

The Process Manager can provide a record of how a model was created, extracting the details
of which input data and settings were used. The Process Manager can also be used to
automatically update the model after the input data has been changed, create a number of
different versions of the model using different settings, and automatically repeat mapping tasks.
It is also possible to combine different functions, e.g. Operations- and Utility functions in the
Process Manager.

Become familiar with the Process Manager interface

There are for main sections used in Process manager. The top section defines the workflow
name and purpose, the left section stores all utilities and processes used. In the right section the
‘code’ is stored and manipulated while the bottom section lefts you save, test and run the
workflows at hand.

1. Continue to use your project or open the Day2_Shell_Vols_WFlows.pet project under the
Projects directory.
2. Go to the Insert option in Menu bar and select New Workflow. A new Workflow 1 is
stored at the bottom of the Workflows tab, and a workflow window is activated.

3. In the top section, write Test as the name and a short description that you are practising on
using the interface.

168 • Process Manager – Slides Shell Refresher - Day2

 4. Press the Utility button under ‘Available functions:’ Drop in Loop, End Loop and Run by
highlighting each of them, then use the blue/green arrow to drop them into the empty

right-hand window. Now use the arrow to move the Run command up to the middle
of the statement. Drop in some other commands after you have searched up and down and
use the button to test the statement.
5. Probably the test will fail, then remove all the last commands you entered by highlighting
them and then use the delete option. Now test again and the status should be set to a
green OK.
6. Highlight and drop in another workflow called ‘Nested if statements’ in the empty field after
the Run command.
7. Click Apply and then Run. The workflow name will be updated and the workflow will be run
with an output log window (don’t worry about the output since this is just to test the
workflow set-up and options).

Predefined workflows – Repeat all calculations …

Repeat all calculations on the active 3D grid – will create a workflow which rebuilds the current
Petrel Grid and properties from the input data using the settings last used for each object. A list
of icons will appear, each one representing a Petrel process which has been performed during
the building of the model. Double clicking on an icon will display the dialog box for that process
with the settings that were used. These settings may be changed before running the workflow. It
is not mandatory to re-run all of the processes which were performed on the grid.

A Simple start – Using a predefined workflow on the active 3D grid

When repeating all the calculations on the active 3D grid it will only run processes related to that
grid. If you have a depth converted grid (DC) it will not look at any processes before the depth

Shell Refresher - Day2 Process Manager – Slides • 169

 conversion process, i.e. you cannot do any structural changes to the 3D grid (pillar gridding), or
any changes related to the Make Horizons process.

Exercise Steps
1. First insert a New Workflow from the Insert option in Menu bar. Name it Repeat 1.
2. Highlight the 3D depth converted grid (Gullfaks [DC]) so it turns active. Then go to the Auto
generate workflow button and select ‘Repeat all calculations on the active grid’. A list of
most of the processes that were done to the Depth converted grid will appear in the
rightmost window.
3. Toggle ‘With Copy’ and ‘Copy Properties’ after the With 3D Grid command; this will
generate a copy of the entire 3D grid, so that you do not overwrite the old grid (good for
comparison or user mistakes). Also the properties that are not listed will be copied directly
over from the original grid.

4. Press Apply, Test and Run. A new copy of the grid is stored in the Models tab. Inspect it
and see what was re-created and not.
Note: The ‘Copy Properties’ option is good when you for example do not do any changes to the
Zones and Layering but only to the properties. It means that you can deactivate or delete the
properties that should not be re-run from the list – saving you some time. On the other hand if
you do some zone and layering changes you should not use this option. Then insert the
remaining processes you need into the list and toggle off ‘Copy Properties’. Note that if you do
not use the ‘Copy Properties’ option the remaining properties will get a black X in front.

170 • Process Manager – Slides Shell Refresher - Day2

 Update model – Edit the input data to the Autogenerated workflow.
As you may have noticed in the previous exercise not all the processes were captured. We
made copies of the remaining processes that were not captured by the Autogenerate function.
Now we will alter the Zonation and layering.

Exercise Steps
1. First insert a New Workflow from the Insert option in Menu bar. Name it Repeat 2.
2. Highlight the 3D depth converted grid (Gullfaks [DC]) so it turns active. Then go to the Auto
generate workflow button and select ‘Repeat all calculations on the active grid’ again.
3. Toggle ‘With Copy’ after the With 3D Grid command.
4. Double-click on the Make Zones process for Zone 2. The saved set-up dialog for the
process appears.
5. Go to Input tab in Petrel Explorer and expand the folder called “Process Manager input”.
Highlight the Edited Isochore object and drop it into the Tarbert-3 field of the Make Zones
process dialog window.

Shell Refresher - Day2 Process Manager – Slides • 171

 6. Click OK to the process dialog. The new changes are stored but not updated as that
requires you to do a Run.
Note: There are some dependencies here:
i. The Geometrical Modeling steps are stored before the Facies and Petrophyscal
Modeling steps as they are listed sequence under the Process Diagram. However, the
Connected Volume in Geometrical modeling is dependant on an already made facies,
therefore if it is listed above it will not be re-created. Use the arrows and move it down
beneath the Facies Modeling steps.
ii. Volume Calculation is not listed, since Petrel requires knowledge of which case to run.
Simply go to Volume Calculation listed under Processes button, drop it in at the bottom of
the list. Then double-click on it and select the correct Case. Click OK.

Optional: Predefined workflows – Make scaled plots …

This exercise will generate a nice map of all the horizons created in an active 3D model. All of
the settings such as title, scale bar, color scale, axis, etc. are already programmed to show a
meaningful and visually stimulating plot. The maps generated can be automatically printed off,
provided the printer settings have been defined prior to executing the workflow, and they use
Windows drivers.

Exercise Steps
1. Go to Window in Menu bar and open a New Map Window.
2. Go to Insert in Menu bar and insert a New Workflow. Name it “Maps.”

172 • Process Manager – Slides Shell Refresher - Day2

 3. Make sure the 3D grid is active and go to the Process Manager window and hit the
Autogenerate workflow button. Select Make a scaled plot of each horizon in the active
3D grid.

4. A list of all the settings for making plots of all horizons in the active grid is show. Since you
do not want to print in real-time, deactivate the ‘Print’ command using the Disable or

enable selection icon. The Print command turns grey (inactive).

5. Insert a Pause command of 1 second beneath the ‘Set horizon and fault visible’ command.
‘Pause’ is found in the ‘Messages’ folder under the Utility button options.

6. Press the Set Full Screen , and Hide Dialog while running the workflow
options.

Shell Refresher - Day2 Process Manager – Slides • 173

 Optional: Operations – Referring to objects using variables

Use Set Reference to set a ‘child reference’ to represent an object. The child reference must be
selected from the variable list in the Workflow tab.
You can also define variables. Variables are used in a workflow to have a simple word or string
represent a function or command.

Eliminate Fault Sticks Above the Top Surface of a Model

This exercise will create a workflow which performs the same operation on all object located in a
folder. Specifically, this workflow will eliminate the portion of all fault sticks that extend above a
surface, using the ‘refer to objects’ operation. Many new visual options, which can be used for
demonstrations and presentations, are included in this exercise.

Optional Exercise steps

1. Insert a new workflow (Insert > New Workflow).
2. Under the ‘Statements’ folder of the Utility button, insert a command
in line 1 of the workflow.
3. Insert the folder called ‘Fault Sticks’ from the input tab into the empty field and assign a
child reference A from the ‘Variables’ folder of the Workflow tab.

4. Under the Utility button, insert a command in line 2. (This is done so as not to
edit the original fault sticks)
5. Assign to equal in the copy command.
6.
7. Insert two commands from the Utility button (to visualize the execution of the
workflow).
8. Insert and (from the ‘Process Manager input’ folder in Input tab)
into the ‘Set visible’ fields respectively.

9. Insert a command found in the Utility button.

10. Insert a command from the Utility button and type in duration of half (0.5) a
second.
11. Insert a command, found in the Operations button > in a
folder called ‘Eliminate Where’.
12. Insert the reference and the surface respectively into the empty
fields and toggle on the ‘Eliminate outside’ so the command looks as follows:

174 • Process Manager – Slides Shell Refresher - Day2

 (This will eliminate the portion of each fault stick that is located above the elevation of the Top
Surface).
13. Now create a folder under the ‘Process Manager input’ folder in Input tab; right-click on it
and select Insert Folder. Rename the new folder to ‘Edited fault sticks’.
14. In ‘Folder Administration’ folder under Utility button, select the Move command and insert it
below the ‘Eliminate Where’ Operation. Since refers to the edited faults sticks, drop
this in the first empty field, then drop in the new folder you created in the second empty field
using the Blue buttons.

15. Insert a command (found under Utility button).

16. Press the Set Full Screen , and Hide Dialogue while running the workflow
options.
17. Test and Run.
18. Name the workflow and press Apply to save.

Note: The edited fault sticks are created and stored as new objects at the bottom of the Input
tab.

Shell Refresher - Day2 Process Manager – Slides • 175

 Define a Variable and Assign it to a Surface
The purpose of this exercise is for you to become familiar with the concept of defining variables
and assigning them to objects. First we define a variable to equal a certain elevation. Then we
assign that elevation to a surface, and that surface is created and stored in the input tab.

Next, we redefine the variable to equal its current value minus, an additional value. It will then
assign this new elevation to another surface. This surface is also created as an object and
stored in the input tab. As the execution of the workflow runs it will visualize the newly created
surfaces in the 3D window.

Exercise steps
1. Insert a new workflow (Insert > New Workflow).
2. Under the Utility button, insert a command in line 1 of the workflow.
3. Name the new expression ‘$elevation’ and assign it a value of -250.

4. Under the Utility button, insert a command (Folder Administration folder) in line 2
of the workflow.
5. Insert the surface (from Input tab) and have it = child reference
(This is done so as not to edit the original surface)

6. Insert from the Operations button > ‘Arithmetic Operations’ folder.

7. Assign the child reference to equal the constant value represented by the variable
$elevation.
8. Insert a ‘Set visible’ command from the Utility button (Visibility and Plotting folder); make
the new surface represented by visible.

This first portion of the workflow is used to assign a surface represented by child reference
a constant value of –250m. We will now redefine the variable ‘elevation’ to create
an additional surface assigned a constant value –550m.

9. Insert another into the workflow.

10. Name it $elevation again, but this time define it to equal the old $elevation, minus an
additional 300m for a total of –550m.

11. Under the Utility button, insert another command.

12. In the Copy command line have (the new surface) = (the original surface)

176 • Process Manager – Slides Shell Refresher - Day2

 13. Insert ‘Z = Constant ’ from the Operations button > ‘Arithmetic Operations’ folder.
14. Assign the child reference to equal the constant value represented by the new
variable $elevation.

15. Insert a from the Utility button.

16. Test and Run the workflow, name it and press Apply to save.

Optional - Using Loops to create a nested ‘If statement’

In this exercise you will get a feel for using logical functions and tests. These types of commands
can be used in any process where a variable or other data must be distributed according to a
calculator or scientific function. The ‘If Statements’ are used when the value of a variable is
tested, and used in the workflow only when certain conditions are met. In the exercise below,
Petrel will assign the variable to be a random number between 1 and 30. A message box will
appear with a different message depending on the number assigned

Exercise steps
1. Insert a new workflow (Insert > New Workflow).
2. Insert a command from the Utility button (Statements folder) into line 1 or the
workflow, and define it run through from 1 to 6.
3.
4. Under the Utility button, insert a command in line 1 of the workflow.
Name the new
variable $bob and define it to = a random number between 1 and 30.
5. Under the Utility button, insert an ‘if statement’ in line 3 of the workflow. Define the
condition of the ‘if statement’ to be: .

Now we must define the functions to occur in the workflow if the above statement is true.

6. Insert two commands also from the Utility button (Messages folder).
7. In the first one type the name of the variable $Bob.
8. In the second one type a message to be displayed if the value of $Bob is less than 10. For
example: “The number is too low!”

9.
10. Under the Utility button, insert an ‘Else If’ statement in line 6 of the workflow.

Shell Refresher - Day2 Process Manager – Slides • 177

 11. Define the conditions for the ‘Else If’ statement to be if $Bob is > 20.

12. Once again we must define the functions to occur in the workflow if the above statement
also is true.
13. Insert two commands also from the Utility button.
14. In the first one type the name of the variable $Bob.
15. In the second one type a message to be displayed if the value of $Bob is greater than 20.
For example: “The number is too high!”

16. Under the Utility button, insert an statement in line 9 of the workflow. This
command dictates that if the random number assigned to the variable $Bob is neither less
than 10 nor greater than 20, some other operation will proceed in the workflow. This other
operation is defined immediately following the ‘Else’ command.
17. Insert two commands. In the first one type the name of the variable $Bob, in
the second one type a message to be displayed if the value of $Bob something other than
less than 10 or greater than 20. For example: “The number is perfect!”
18. Insert an ‘End If’ command. This command is ALWAYS necessary when using If
Statements.
19. Insert an ‘End Loop’ command to end the loop.
20. Now you can Test and Run the workflow. Each time you run the workflow a message box
will appear with a number and message, depending on the value of the number.
21. Don’t forget to name the workflow and press Apply to save!

Monte Carlo simulation of contacts

In this exercise we will set up a workflow used for inspecting uncertainties on the contact level.
Only the contact levels will be varied, all other parameters in the Base Case model (previous
chapter) will be kept constant.

Exercise Steps
1. Use the same project as in the Volume Calculation chapter Day2_Shell_Vols_WFlows.pet.
The case defined to calculate bulk volume above a contact, will be used as Base Case in this
workflow.
2. Insert a new Workflow.
3. Prepare the output sheet where all results will be stored. Insert ‘Clear output sheet’ and
‘Output sheet’ from the Utility tools. Type in CONTACTS in Row 1 and Column 1 for the
‘Output sheet’ Toggle on Bold.

178 • Process Manager – Slides Shell Refresher - Day2

 4. Input Run with ‘Headers’ and ‘BaseVariables’ nested Workflows (already made in the
startup project) with Nested variables toggled on.

5. Insert a Loop statement with variable $Loop going from 1 to 5.

6. From the Processes tools toggle box, select and insert With 3D Grid. Input Gullfaks [DC].
7. Insert two Numeric Expressions to define the variable contacts ($C1 = RAN(-1870,-1890)
and $C2 = $C1-NORMAL(130,2)) .

8. Insert a Make Contacts process into the workflow. Open it and insert $C1 and $C2 as Gas
Oil Contact and Oil Water Contact, respectively.

9. Insert a Volume Calculation process. Open the Volume calculation process dialog by
double-clicking on it, and make sure Case_1 is used (edit existing case).

Shell Refresher - Day2 Process Manager – Slides • 179

 10. Input Move, make sure the Contacts folder under the Cases tab is created. Insert the
Output variable and Contacts folder into the expression.

11. Input Get name and drop in the Output variable.

12. Input Run with ‘Output Workflow’ nested Workflow (already made in the startup project) with
Nested variables toggled on.
13. Finalize the workflow with an End Loop command.

14. In the Workflow Editor window, depress the hide dialog window button so you can see
the output sheet.
15. Test and Run the workflow and observe the data appearing in the output sheet as the
workflow loops through the volume calculations with different contact levels.

16. Save your changes.

17. Observe the volumetric result found in the Contacts folder under the Cases tab.
18. List the result by opening Settings for Volumetrics in the Results tab. Select what properties
to include and toggle ‘Single line format’.

180 • Process Manager – Slides Shell Refresher - Day2

 19. Select all Cases from the Cases tab and press Make Report.
20. The final workflow should look like the following:

Comments
This exercise exemplifies how to vary the contact levels and the effect on volume calculations.
Any parameter going into the volumetrics or the modeling of the input Base Case model can be
defined by variables.

Shell Refresher - Day2 Process Manager – Slides • 181