A Review of Geological

Modeling
Short
Short Course
Course503
503
Three-Dimensional
Three-Dimensional
Geologic
GeologicMapping
Mapping for
for
Groundwater
GroundwaterApplications
Applications
October
October27,
27,2007
2007
A. Keith Turner
Colorado School of Mines
&
Carl W. Gable
Los Alamos National Laboratory
Why Is Modeling & Visualization Important?
The World of the Geoscientist Is Multi-dimensional

 Current interpretation methods limit this view

 Digital versions of traditional maps are not sufficient
 Increased efficiency demands computer-based
methods to:
– Integrate and Manage the data
– Interpret geological features
– Visualize attributes spatially and temporally
– Model dynamic Earth processes
 Importance of This Topic …
 3-D subsurface modeling first became feasible in late
1980’s with the introduction of high-performance
Unix-based graphical workstations

 Developing digital representations of the subsurface

does not ensure high-quality and efficiently managed
projects

 Society is increasingly demanding:

 multi-scale, multidisciplinary, integrative projects
 a shift from passive data collection and archiving to
dynamic information management and dissemination
 Over The Past Decade
Enormously more powerful computers and
data storage have vastly reduced costs!
(Source: C-Tech
 Continuing, rapid advances Development Corp.)
Costs MIPS
in computer HARDWARE
and SOFTWARE
technologies
 Modeling & visualization Time
increasingly integrated
 Increasingly realistic
models possible
3-D Model Created/Visualized by PC-based EVS
Problems in Subsurface Investigation
are Unique...
 Subsurface information is often incomplete and
conflicting;
 The subsurface is naturally complex and
heterogeneous;
 Sampling is most often insufficient to resolve all
uncertainties; and
 Scale effects on rock , fluid, and other properties
are usually unknown.
Why We Need Special Modeling and
Visualization Tools and Not Just
CAD
GEO-OBJECT

ENGINEERING OBJECT
 A Typical Modeling Project
Problem
ProblemStatement
Statement
Spatial
Spatial Properties
Properties
Raw
RawData
Data Abstraction
Abstraction Raw
RawData
Data

Geometry
GeometryModel
Model Predictive
PredictiveModel
Model

Create
CreateGeometry
Geometry Define
DefineProperties
Properties

Grid
GridGeneration
Generation Analytical
AnalyticalComputation
Computation

Visualization
Visualization
Interpretation
Interpretation

‘‘Customer’
Customer’Applies
Applies Results
Results
 Geometry (Descriptive) Modeling
Definition: Geometry (Descriptive) Modeling
involves visually describing, through various
means such as computer graphics and modeling:
Spatial
Spatial
Raw
RawData
Data  The geologic framework
Geometry
GeometryModel Distribution and propagation
Model

Create
CreateGeometry
Geometry of attributes

Grid
GridGeneration
Generation

Visualization
Visualization
 3-D Model Involves Two
Stages
Framework Definition Discretisation and
 Borehole and isolated
sample data Property Distribution
 Triangulated surfaces  3-D grids and meshes
 2-D grids and meshes – regular hexahedral
 Iso-volumetric models – octree variable
– from triangulated surfaces – geocellular
– from cross-sections – tetrahedral unstructured
meshes
– from grids and meshes
– parametric (NURBS, etc)
– Boundary Representations
Modeling Often Begins with Borehole Data

(Lynx Geosystems)
Geometry Models can be
Constructed Using Cross-
Sections
 3-D Solid Models can be
developed from Multiple Surfaces
Three
Final Solid
Volume Triangulated
ModelTwo
Between Surfaces
Surfaces

(Lynx Geosystems)
Layered Models may involve many surfaces
Complex channels and “pinch-outs” are difficult to model
 Regional (Volumetric) Subdivision
Feasible for Non-stratigraphic Cases

(C. Gable – LANL)

 Framework Models require Grids or
Meshes to assign Property Distributions

Volumes
Discretised are “empty”
Model
or uniform
Geological Framework Defined First –
then Grid Resolution

(C. Gable – LANL)

 Discretisation may involve
“structured” and “unstructured” meshes
Regular
Cells

Examples

Fault

Unstructured Mesh Quadtree Cells

“ Geocellular” Volumetric Model
Yucca Mountain Represented by a
Tetrahedral Mesh Model

(C. Gable – LANL)

 Accurately Modeling Faults is a
Challenge
 Near-horizontal thrust faults form additional surfaces
Near-horizontal thrust faults form additional surfaces
 Steeply inclined Faults commonly shown as vertical

Edwards Aquifer Model

In Texas – by USGS
 Advanced Fault Modeling
Blocky Faults using
Faults represented Octree model
by unstructured mesh (Carl Gable – LANL)
(Carl Gable – LANL)

3D Unstructured Mesh
of faulted layered sequence
(Carl Gable – LANL)
 Models may be “Nested”
from Regional to Local Scales
Purpose of Modeling is
Prediction…
Prediction has an
extrapolative rather than
interpolative character…
 Involves risk
 Leads to Decision-making
 Predictive Modeling
Examples:
Examples: Predictive Modeling involves prediction
Definition:
 and/or
economic simulations,
simulation
economic of events, dynamic and other
simulations,
 types
reservoir performance,
of processes
reservoir performance, occurring in the
 geological
fluid subsurface:
fluid migration
migration and
and transport
transport models,
models, Properties
Properties
 seismic
seismic velocity
velocity modeling,
modeling, Raw
Raw Data
Data
 Solve equations, or other
 structural
structural restoration
restoration
numerical analyses Predictive
PredictiveModel
Model
 Forward and inverse Define
DefineProperties
Properties
modeling techniques
Analytical
AnalyticalComputation
Computation

Visualization
Visualization
The geologic framework controls
the application modeling
Contamination plume model
shown within geology
Full contamination plume model
The Bottom Line …
Trends in 3-D Modeling
 Continuing, rapid advances in computer technologies
 Increasingly realistic models possible
– “ Coarse-” vs. “Fine-scale” Models
 Cannot easily incorporate geologic knowledge (i.e.
geologic interpretations)
 Difficulties representing uncertainties
 Hampered by imprecise data and inability to
adequately sample the prototype
 The Modeling Challenge
How Can We Do Better?

 Build 3-D models from isolated bore data and

“soft” information
 Connect known conditions between the
boreholes
 Model real-world complexity
 Verify the resulting models
 Represent the uncertainties of the geologic
framework
 Reduce cost and time of model development
An Integrated Data Management
Process
 What Benefits will an Integrative
Approach Provide??

 To the Individual Scientist –

 To the Scientist’s Organization –
 To the “Client” Organizations and Their Staff –
 To Society at Large –
Benefits to the Individual Scientist –

 More time for science, so….

 More interactions with colleagues
 More interesting projects/studies
 More job satisfaction
 More publications
 More rapid promotions
Benefits to the Scientist’s
Organization –
Society
Maximize Organization’s
Value to “Customers”
 Maximize Strategic Value – BETTER!
 Maximize Financial Value – CHEAPER!
 Maximize Operational Effectiveness –
FASTER!
Benefits to the “Client”
Organizations and Their Staff –

Model Consistency

Data versioning, audit trail, documentation

QA compliance

Improved Predictive Capability & Model
Maintenance

Efficiency: streamlining processes, updating
databases

Auto-updating decision support tools
Benefits to Society at Large –

 Relevant science that supports

decision-making
 Clearly communicated results
 Cost reductions
Questions???