Exploring ArcGIS Analysis methodology and GeoProcessing tools
Exploring ArcGIS Analysis methodology and GeoProcessing tools
Matthew Baker ESRI Educational Services Redlands, CA
UC 2006 Tech Session
1
What is GIS analysis? • Analysis is one of six GIS functions • Process for highlighting patterns and relationships in geographic data • Most people are still only using GIS to make maps • GIS can do much more - Analysis
Capture Store Query Analyze
Customers Buildings Streets
Di l Display Output p
Reality UC 2006 Tech Session
2
Early spatial analysis • John Snow, Snow 1854 • Cholera via polluted water, not air • “John Snow’s pump” Learn More at www.jsi.com http://www.ph.ucla.edu/epi/snow.html
Broad Street water pump UC 2006 Tech Session
3
Why do analysis? • Gives insight into places of interest • Helps identify trends and patterns to focus actions better
Schools
Determine the relationship between crime locations and schools
0.1 mile buffer
Crimes
UC 2006 Tech Session
4
What questions can analysis answer
X,Y
Where things are located
what is contained within
How dense things are clustered
How things are changing
Analysis What the differences
what is in
between features are
the proximity of something
UC 2006 Tech Session
5
What are the 5 steps of the analytical process
2
1 Formulate your question
5
Collect, explore and pre-process data
Interpret results
gy Refine methodology
3 Choose analysis tools and methods
4
Create Models
y Implement analysis
UC 2006 Tech Session
6
What are the 3 basic analysis operations
Analysis Operations Attribute queries
Spatial queries
Geoprocessing (Generation of new datasets based on data attribute information and/or spatial relationships)
Start simple evolve to more complex UC 2006 Tech Session
7
What are the most common analysis tasks
Single-layer operations
Point pattern analysis Surface analysis
Multiple-layer operations
Network analysis
Raster analysis
UC 2006 Tech Session
8
What is a model? • A representation of reality
Onions
Celery
Carrots
Chop
Whole Chicken
Boil
Chicken Broth
Garlic
Wild Rice
Chopped Vegetables
Chicken Meat
B il Boil
Chicken Soup
Salt and Pepper
Boiled Chicken Separate
Chicken Bones UC 2006 Tech Session
9
What are spatial models • Models that deal with geographic features
Ag globe is a model of the Earth
An aerial p photograph g p is a model of surface features
A wildfire map p is a model of wildfire locations
UC 2006 Tech Session
10
Modeling spatial problems • Models help us understand and solve complex problems –Simplify reality –Combine geographic layers to answer questions • e.g., “Where should you build your next store?” Reality
GIS layers
Suitability for store
Model criteria: - Zoned commercial - Near N target population l i - Away from competition UC 2006 Tech Session
11
Why use models? • • • •
Share process with others Framework for understanding realreal-world processes Document work Easy to modify – Rerun – Explore scenarios
Land cover
Select
Forest Areas
View
Select
Good View Intersect
Lakes
Buffer
Distance to Lakes
Trails
Buffer
Distance to Trail
Natural Setting
UC 2006 Tech Session
12
What are the different types of models Representation models
Suitability models GIS data layers
Best store location
Process models
Automated workflows
Filling a reservoir UC 2006 Tech Session
13
Suitability models - Binary • Use for simple problems – Like a query
Snow
• Classify layers into good (1) and bad (0) – Combine with AND, addition, or multiplication: Ski = S S Snow o And d S Slope ope And d Su Sun
• Advantages: – Easy
1
0
0
And
Slope 0
0 1
And
Sun 0 1
0
• Disadvantages: – No ‘next‘next-best’ sites – All layers have same importance – All good values have same importance
Ski 0 1
UC 2006 Tech Session
14
Suitability models - Weighted • Use for complex problems • Classify layers into suitability 1– 1–9 (9 = best)
Snow
– Weight and add together: Ski = (Snow * 0.5) + (Slope * 0.3) 0 3) + (Sun * 0.2)
• Disadvantages:
5 Slope 5
1 9
Sun
• Advantages: – All values have relative importance – All layers have relative importance – Returns suitability on a scale 1 1– –9
9
1
5
1.8
1
9
6.6
Ski 4.2
5.0 9 7.0 70
– Preference assessment is harder
UC 2006 Tech Session
15
Weight and combine the layers
Snow
• For each submodel
9
1
5
– Multiply suitability layers by weights
Slope
• Weights must add up to one
5
1 9
– Add the weighted layers together
Sun
• Repeat to combine submodels
5
1.8
1
9
6.6
Ski 4.2
5.0 9 7.0
Ski = (Snow * 0.5) + (Slope * 0.3) + (Sun * 0.2)
UC 2006 Tech Session
16
Modeling & Analysis tools in ArcGIS • Mostly GeoProcessing tools • Used within ArcGIS GeoProcessing framework – Toolbox Toolbox:: Container for tools and toolsets – Toolset Toolset:: Logical container of tools and other toolsets – Tool Tool:: Single GeoProcessing operation
UC 2006 Tech Session
17
Geoprocessing framework • Multiple environments
Command Line
Models
Tool dialog
Scripts
UC 2006 Tech Session
18
Creating a model 1.
Add ArcToolbox dockable window
3. Right Right-click click new toolbox; click New > Model
2.
Create a new user toolbox
4.
ModelBuilder window opens
UC 2006 Tech Session
19
Constructing a model • Add data d t and d tools t l
Add Data
•Drag D and d drop d from f ArcCatalog A C t l •Use Add Data button in ModelBuilder •Add empty variables in ModelBuilder
Add Tool
•Drag and drop •Use Add Data button
UC 2006 Tech Session
20
Helpful Literature • • • •
The Th ESRI Guide G id to t GIS Analysis, A l i Volume V l 2 The ESRI Guide to GIS Analysis, Volume 1 GIS, Spatial Analysis, and Modeling Advanced Spatial Analysis
UC 2006 Tech Session
21
Questions ?
UC 2006 Tech Session
22
Matthew Baker ESRI Educational Services Redlands, CA
UC 2006 Tech Session
1
What is GIS analysis? • Analysis is one of six GIS functions • Process for highlighting patterns and relationships in geographic data • Most people are still only using GIS to make maps • GIS can do much more - Analysis
Capture Store Query Analyze
Customers Buildings Streets
Di l Display Output p
Reality UC 2006 Tech Session
2
Early spatial analysis • John Snow, Snow 1854 • Cholera via polluted water, not air • “John Snow’s pump” Learn More at www.jsi.com http://www.ph.ucla.edu/epi/snow.html
Broad Street water pump UC 2006 Tech Session
3
Why do analysis? • Gives insight into places of interest • Helps identify trends and patterns to focus actions better
Schools
Determine the relationship between crime locations and schools
0.1 mile buffer
Crimes
UC 2006 Tech Session
4
What questions can analysis answer
X,Y
Where things are located
what is contained within
How dense things are clustered
How things are changing
Analysis What the differences
what is in
between features are
the proximity of something
UC 2006 Tech Session
5
What are the 5 steps of the analytical process
2
1 Formulate your question
5
Collect, explore and pre-process data
Interpret results
gy Refine methodology
3 Choose analysis tools and methods
4
Create Models
y Implement analysis
UC 2006 Tech Session
6
What are the 3 basic analysis operations
Analysis Operations Attribute queries
Spatial queries
Geoprocessing (Generation of new datasets based on data attribute information and/or spatial relationships)
Start simple evolve to more complex UC 2006 Tech Session
7
What are the most common analysis tasks
Single-layer operations
Point pattern analysis Surface analysis
Multiple-layer operations
Network analysis
Raster analysis
UC 2006 Tech Session
8
What is a model? • A representation of reality
Onions
Celery
Carrots
Chop
Whole Chicken
Boil
Chicken Broth
Garlic
Wild Rice
Chopped Vegetables
Chicken Meat
B il Boil
Chicken Soup
Salt and Pepper
Boiled Chicken Separate
Chicken Bones UC 2006 Tech Session
9
What are spatial models • Models that deal with geographic features
Ag globe is a model of the Earth
An aerial p photograph g p is a model of surface features
A wildfire map p is a model of wildfire locations
UC 2006 Tech Session
10
Modeling spatial problems • Models help us understand and solve complex problems –Simplify reality –Combine geographic layers to answer questions • e.g., “Where should you build your next store?” Reality
GIS layers
Suitability for store
Model criteria: - Zoned commercial - Near N target population l i - Away from competition UC 2006 Tech Session
11
Why use models? • • • •
Share process with others Framework for understanding realreal-world processes Document work Easy to modify – Rerun – Explore scenarios
Land cover
Select
Forest Areas
View
Select
Good View Intersect
Lakes
Buffer
Distance to Lakes
Trails
Buffer
Distance to Trail
Natural Setting
UC 2006 Tech Session
12
What are the different types of models Representation models
Suitability models GIS data layers
Best store location
Process models
Automated workflows
Filling a reservoir UC 2006 Tech Session
13
Suitability models - Binary • Use for simple problems – Like a query
Snow
• Classify layers into good (1) and bad (0) – Combine with AND, addition, or multiplication: Ski = S S Snow o And d S Slope ope And d Su Sun
• Advantages: – Easy
1
0
0
And
Slope 0
0 1
And
Sun 0 1
0
• Disadvantages: – No ‘next‘next-best’ sites – All layers have same importance – All good values have same importance
Ski 0 1
UC 2006 Tech Session
14
Suitability models - Weighted • Use for complex problems • Classify layers into suitability 1– 1–9 (9 = best)
Snow
– Weight and add together: Ski = (Snow * 0.5) + (Slope * 0.3) 0 3) + (Sun * 0.2)
• Disadvantages:
5 Slope 5
1 9
Sun
• Advantages: – All values have relative importance – All layers have relative importance – Returns suitability on a scale 1 1– –9
9
1
5
1.8
1
9
6.6
Ski 4.2
5.0 9 7.0 70
– Preference assessment is harder
UC 2006 Tech Session
15
Weight and combine the layers
Snow
• For each submodel
9
1
5
– Multiply suitability layers by weights
Slope
• Weights must add up to one
5
1 9
– Add the weighted layers together
Sun
• Repeat to combine submodels
5
1.8
1
9
6.6
Ski 4.2
5.0 9 7.0
Ski = (Snow * 0.5) + (Slope * 0.3) + (Sun * 0.2)
UC 2006 Tech Session
16
Modeling & Analysis tools in ArcGIS • Mostly GeoProcessing tools • Used within ArcGIS GeoProcessing framework – Toolbox Toolbox:: Container for tools and toolsets – Toolset Toolset:: Logical container of tools and other toolsets – Tool Tool:: Single GeoProcessing operation
UC 2006 Tech Session
17
Geoprocessing framework • Multiple environments
Command Line
Models
Tool dialog
Scripts
UC 2006 Tech Session
18
Creating a model 1.
Add ArcToolbox dockable window
3. Right Right-click click new toolbox; click New > Model
2.
Create a new user toolbox
4.
ModelBuilder window opens
UC 2006 Tech Session
19
Constructing a model • Add data d t and d tools t l
Add Data
•Drag D and d drop d from f ArcCatalog A C t l •Use Add Data button in ModelBuilder •Add empty variables in ModelBuilder
Add Tool
•Drag and drop •Use Add Data button
UC 2006 Tech Session
20
Helpful Literature • • • •
The Th ESRI Guide G id to t GIS Analysis, A l i Volume V l 2 The ESRI Guide to GIS Analysis, Volume 1 GIS, Spatial Analysis, and Modeling Advanced Spatial Analysis
UC 2006 Tech Session
21
Questions ?
UC 2006 Tech Session
22
