Transcript Wk3_lec - Spatial Information Systems (Basis)

Introduction to GIS Modeling
Week 3 — Reclassifying and Overlaying Maps
GEOG 3110 –University of Denver
Presented by
…but we
didn’t finish
last week’s
material!!!
Joseph K. Berry
W. M. Keck Scholar, Department of Geography, University of Denver
Suitability Modeling Logic
Reclassifying Maps (position, initial value, size, shape and contiguity)
Overlaying Maps (point-by-point, region-wide and map-wide)
Display Settings for Grid-based Maps
<Exercise 3 Question #1>
Let’s look at some example displays in the Bighorn.rgs data base…
(Berry)
Reclassify and Overlay Operations
<Exercise 3 Question #2>
<Exercise 3 Question #3>
Covertype
(Input map)
Size
(Command)
SIZE Covertype FOR Covertype_size
Covertype_size
(Output map)
…make sure your map display is “appropriate” –not simply the default
(Berry)
Evaluating Habitat Suitability
<Exercise 3 Question #4>
Assumptions – Hugags like…
gentle slopes,
southerly aspects, and
lower elevations
Generating maps of animal habitat…
(digital slide show Hugag)
Manual Map Overlay
(Binary)
(See Beyond Mapping III online book, “Topic 23” for more information)
Ranking Overlay
(Binary Sum)
Rating Overlay
(Rating Average)
(Berry)
Conveying Suitability Model Logic
gentle slopes
Elevation
Slope
Rows Model Criteria
Columns Analysis
Slope
Reclassify Preference
Bad 1 to 9 Good
Levels
(Times 1)
southerly aspects
Elevation
Aspect
Aspect
Reclassify Preference
Habitat
Rating
(1)
Overlay
Bad 1 to 9 Good
lower elevations
Lines
Bad 1 to 9 Good
Processing Steps
(Commands)
(1)
Elevation
Reclassify Preference
Elevation
Bad 1 to 9 Good
Overlay
Algorithm
Base Maps
Calibrate
Derived Maps
Interpreted
Maps
Fact
Covertype
0= No-go
1 to 9 Good
Weight
Solution
Map
Judgment
Reclassify
Habitat
Rating
Water
Mask
…while Reclassify
and Overlay
commands are not very
exciting, they are some
of the most
frequently used
operations
0= No, 1= Yes
(See Beyond Mapping III online book, “Topic 22” for more information)
Constraint Map
(Berry)
Extending Model Criteria
gentle slopes
Elevation
Slope
Rows Model Criteria
Columns Analysis
Slope
Preference
Bad 1 to 9 Good
Levels
(Times 10)
southerly aspects
Elevation
Aspect
Aspect
Preference
(1)
Bad 1 to 9 Good
Habitat
Rating
Bad 1 to 9 Good
Lines
Processing Steps
(Commands)
lower elevations
Elevation
Preference
Elevation
(1)
Bad 1 to 9 Good
forests
Forests
Forest
Proximity
Forest
Preference
(10)
Bad 1 to 9 Good
water
Water
Water
Proximity
Water
Preference
Bad 1 to 9 Good
(10)
Additional criteria can be
added…
—Hugags would prefer to
be in/near forested areas
—Hugags would prefer to
be near water
—Hugags are 10 times
more concerned with slope,
forest and water criteria
than aspect and elevation
(Berry)
Optional Questions
Using PowerPoint as
3-1)
Flowchart of Binary Habitat Model
3-2)
“Simply” and “Completely” Crosstab Tables
3-3)
Average Suitability Rating for each Covertype parcel
a “graphics package”
Average and Coefficient of Variation maps for the
200 foot contour polygons of Elevation (statistical model)
3-4)
Average Wildfire Fuel Loading index for each
management District (mathematical model)
3-5)
(Berry)
Creating a Flowchart (using PowerPoint)
Enter map title
Enter yet another
map title
2
Slope
Enter map title
1
Enter other map
title
Under the Home tab, 1) use the Rectangle drawing tool to create
and size a box to represent a map in the flowchart and 2) use the
Text box drawing tool to enter the map name in the Rectangle.
Select both and use Format Group to group the two objects.
Copy and Paste the Rectangle to form other maps.
Use the Line drawing tool to connect the boxes. Use
the Text box drawing tool to enter the command,
rotate and place over the Line.
Repeat the process to add additional maps (boxes)
and processing steps (lines) to complete the
flowchart containing the logic of the GIS model.
There are a lot of other things you can do to make the graphic a bit more unique, such as
borders, transparency, shadowing and animation (viewed as a slide show). Also, keep in mind
that Format Align can be used to align multiple graphic objects if things get out of whack.
(Berry)
Compound Graphic (Campground model results)
Campground
Suitability
Use SnagIt to capture one of the graphic elements, such as the S_Pref map then Paste and Size
at the appropriate location on the “canvas” (white background shape). Repeat for all of the
other graphic elements. Use the Text Box tool to embed text as appropriate.
Group the figure elements in logical groupings and then use the Custom Animation tool to
control their sequencing for display if you intend to present as a PowerPoint slide deck.
(Berry)
Big Picture of Map Analysis and Modeling
Nanotechnology
GEOTECHNOLOGY
Biotechnology
Global Positioning System GEOGRAPHIC INFORMATION SYSTEMS Remote Sensing
Mapping and Geo-query
(Vector-based)
SPATIAL ANALYSIS
(Geographic Relationships)
Week 3
ANALYSIS and MODELING
(Grid-based)
GISer’s
Perspectives
Reclassify and Overlay
Distance and Neighbors
Week 4
(Numeric Relationships)
Week 8
Surface Modeling
Spatial Data Mining
Week 5
Week 9
Statistician’s Perspective:
Mathematician’s Perspective:
Basic GridMath & Map Algebra
Advanced GridMath
Map Calculus
Map Geometry
Plane Geometry Connectivity
Solid Geometry Connectivity
Unique Map Analytics
SPATIAL STATISTICS
GIS Modeling
Weeks 6 and 7
Future Directions
Week 10
Math/Stat
Perspectives
(SpatialSTEM)
See Topic 30, Beyond Mapping III
Basic Descriptive Statistics
Basic Classification
Map Comparison
Unique Map Statistics
Surface Modeling
Advanced Classification
Predictive Statistics
(Berry)
Grid-Based Map Analysis
Spatial analysis investigates the “contextual” relationships in mapped data…
 Reclassify— reassigning map values (position; value; size, shape; contiguity)
 Overlay— map overlay (point-by-point; region-wide; map-wide)
 Distance— proximity and connectivity (movement; optimal paths; visibility)
 Neighbors— ”roving windows” (slope/aspect; diversity; anomaly)
Weeks 3,4,5
Spatial statistics
Surface modeling maps the spatial distribution and pattern of point data…
 Map Generalization— characterizes spatial trends (e.g., titled plane)
 Spatial Interpolation— deriving spatial distributions (e.g., IDW, Krig)
 Other— roving window/facets (e.g., density surface; tessellation)
Data mining investigates the “numerical” relationships in mapped data…
 Descriptive— aggregate statistics (e.g., average/stdev, similarity, clustering)
 Predictive— relationships among maps (e.g., regression)
 Prescriptive— appropriate actions (e.g., optimization)
Weeks,8,9
(Berry)
An Analytic Framework for GIS Modeling
…recall Map Analysis organization and evolution discussion
from Week 1 class presentation/reading (GIS Modeling Framework paper)
Reclassify operations involve
reassigning map values to reflect new
information about existing map features
(Berry)
Reclassifying Maps
(Berry)
Reclassifying Maps
<Homework Question #2>
Contiguity
Shape
Initial
Value
Size
Initial
Value
CLUMP -- Assigns new values to contiguous groups of cells within each
map category.
CONFIGURE -- Assigns new values characterizing the shape of the area
associated with each category.
The most frequently used map analysis operation
…and one of the most dangerous!!!
RENUMBER -- Assigns new values to the categories of a map.
SIZE -- Assigns new values according to the size of the area associated
with each map category.
SLICE -- Assigns new values by dividing the range of values on a map into
specified intervals (contouring).
(Berry)
Renumber Operation
Renumber— assigns new values to the categories of a map.
…any real number from -3.4E38 to + 3.4E38 can
be assigned to any existing value on a map
…often integer values are assigned based on user
reasoning (as in this example)
Note: PMAP_NULL is a special value that can
be assigned indicating “no data” and the grid
location will be ignored in processing and display
RENUMBER Elevation
ASSIGNING 1 TO 500 THRU 1800
ASSIGNING 0 TO 1800 THRU 2500
FOR E_Pref
…context Help
provides information
on function and use
of operations
<more info in the MapCalc Manual>
(Berry)
Slice Operation
Slice— assigns new values by dividing the range of values on
a map into specified intervals (“Equal Ranges” contouring)
…Map Range = Max_Value – Min_Value
= 2500 – 500= 2000 feet
…Contour Interval = Map Range / # ranges
= 2000 / 20 = 100 feet
SLICE Elevation into 20 FOR Relief_100ft
…a user can specify the minimum
and maximum values
of the range–
SLICE Elevation INTO 20
FROM 1000 THRU 1200 ZeroFill
FOR Relief_10ft
…Slice <mapName> is often used to
collapse a map with a large set of map values
to just a few intervals for a quick view of the
pattern of the spatial data distribution
(Berry)
Size Operation
Size— assigns new values according to the size of the area
associated with each map category.
…the Size operation assigns the number of cells
comprising each map region (category/value)
…in this instance there are three map regions
(Open Water= 1, Meadow= 2, Forest= 3)– note
that Water occurs at two different places.
SIZE Covertype FOR Coverype_size
…to calculate the actual area
of each map region multiply
the size map times the area
per grid cell– 10,000 m2
or 1 ha in this case
…to calculate the size/area of each occurrence
you must first Clump the map “regions”, then
use the Size command
(Berry)
Clump Operation
Clump— assigns new values to contiguous groups of cells
within each map category.
…a map “category” identifies all locations with
the same characteristic or condition– e.g., Open
Water, Meadow, Forest
…a “clump” is a contiguous group (individual
spatial instance of a map “category”)–e.g., five
cover type clumps with two instances of Open
Water
CLUMP Covertype AT 1 DIAGONALLY
FOR Coverype_clumps
“At” identifies how
far to reach in
defining clumps
“Orthogonally” reaches
horizontally and vertically
only; “Diagonally” includes
off angles
(Berry)
Configure Operation
Configure— assigns new values characterizing the shape
and integrity of the area associated with each map category.
Boundary Configuration
Convexity is the ratio of the Edge
to the Area (Size) and normalized
to that of a circle of the same area
CONFIGURE Covertype Edges
FOR Covertype_edges
Edge cells
Spatial Integrity
Euler = (# Holes) – (1-#Fragments)
(Berry)
Some Reclassifying Things to Keep in Mind
The Covertype map contains Nominal data that is Discretely (Choropleth)
distributed in geographic space. As such, it is best displayed in 2D using
cells (Grid) and with layer mesh on as the stored values do not form
gradients in either numerical or geographic space.
The Size command assigns new values according to the size
of the area (# of cells) associated with each map category. In
this instance the input map is Covertype (Nominal/Discrete
data) and the output map is Covertype_size (Ratio/Discrete
data). The size algorithm “counts” the number of cells for
each map category (stored map value).
Size works on Nominal data (Categorical) but usually is not
appropriate for ratio data as far to many values (decimal
places); results in most of the map being assigned the cell
size value of 1 because elevation values with decimal points
rarely are the same. For example, sizing the Elevation
map…
…identifies that “one cell in size” elevation values occur in
64% of the map area (384/1= 384 times; each value is
unique). “Three cells in size” areas occur in 2.88% of the
analysis window (18/3= 6 times; six sets of the same value).
(Berry)
An Analytic Framework for GIS Modeling
…recall Map Analysis organization and evolution discussion
from Week 1 class presentation/reading (GIS Modeling Framework paper)
Overlay operations involve characterizing
the spatial coincidence of mapped data
(Berry)
Overlaying Maps (conceptual approaches)
(Berry)
Overlaying Maps (accessing the data)
Overlaying maps involves one of four basic techniques for
accessing/organizing
geo-registered data for analysis—
(Tomlin’s organizational framework)
Local
Focal
Zonal
Global
Template
Map
Entire
Area
Map1
Map1
Map1
Map1
Map2
Map2
Map2
Map2
“Spearing”
…collects data on a
cell-by-cell basis
and reports a single
value on a
cell-by-cell basis
“Funneling”
…collects data on a
neighborhood basis
and reports a single
value on a
cell-by-cell basis
“Lacing”
…collects data on a
region-wide basis
and reports
summary on a
region-wide basis
“Plunging”
…collects data on a
map-wide basis and
reports results on a
map-wide or
cell-by-cell basis
(Berry)
Overlaying Maps
<Homework Question #3>
Regionwide
COMPOSITE -- Creates a map summarizing values from one map which
coincide with the categories of another.
Point-byPoint
CALCULATE and COMPUTE -- Creates a map as the mathematical or
Point-byPoint
COVER -- Creates a new map where nonzero values of the top map
Point-byPoint
CROSSTAB -- Generates a spatial coincidence table of two maps.
Point-byPoint
combinations of values on two maps.
statistical function of two or more maps.
…true “map-ematics”
replace the values on the previous (bottom) map, or stack of maps.
INTERSECT -- Creates a map that assigns new values to pair-wise
…Map-wide overlay involves spatial statistics (spatial data mining)
(Berry)
Compute/Calculate Operation
Compute/Calculate— creates a new map as the mathematical
function of two or more maps.
All of the basic mathematical operations on a typical pocket
calculator can be performed on grid maps…
…including Add, Subtract, Multiply, Divide, Exponentiation, Square,
Square Root, Max, Min, And, Or, & Trig functions
…other math operations?
CALCULATE ((Covertype * 10)
+ Water) FOR Covertype_Water
COMPUTE Covertype TIMES 10
Plus Water FOR Coverype_Water
…in this example one map is
multiplied by 10 then added to
another map, thereby creating
a 2-digit code indicating the
first map’s categories as the tens digit followed
by the second map’s categories as the one’s digit
(Berry)
Crosstab Operation
Crosstab— generates a spatial coincidence table of two
maps.
Map1 = Districts
…the maps are compared “cell-by-cell” and
the number of joint occurrences
between the map categories are
summarized in a table
Optional Question 3-2
Map2 = Covertype
CROSSTAB Districts WITH Covertype
Simply TO Newtextfile.txt
In this example there are…
• 58 cells classified as District 1 on the
Districts map
• 82 cells classified as Open Water on the
Covertype map
• 58 cells identified as having the joint
condition of District 1 and Open Water
representing 9.28 percent of the entire map
area.
Note: all of the District 1 cells are in Open Water
(Berry)
Intersect Operation
Intersect— creates a map that assigns new values to pairwise combinations of values on two maps.
Map 1 = Districts
The maps are compared “cell-by-cell” and a user
specified number is assigned to designated
category combinations…
…“zerofill” assigns 0 to all
combinations that are not specified
Map 2 = Covertype
… “oldfill” retains 1st map values
for non-specified combinations
INTERSECT Districts WITH Covertype
ASSIGN 1 to 1, 1 Zerofill FOR Districts1_Cover1
…if “completely” is specified all
combinations are automatically
identified using unique sequential
numbering for map values
Note: Intersect is similar to geo-query
operations in desktop mapping packages by
identifying all locations having specified
category (map-value) combinations
(Berry)
Cover Operation
Cover— creates a new map where the non-zero values of the top map
replace the values on the previous (bottom) map, or stack of maps.
Water
4Water 4
4
0
1Cover 1
3
Covertype
…the maps are compared “cell-by-cell” and the
value in the top-most cell replaces the previous
values unless that value is zero, then the top most
non-zero value in the map stack is retained
(0,3 3)
(4,1 4)
“Zero” is treated as transparent as
maps are staked; non-zero values
treated as opaque
COVER Covertype WITH Water IGNORE 0
FOR Districts1_Cover1
…in the example coincidence
4,1 4 because 4 is non-zero
and replaces what is beneath it
…0,3 3 because zero is ignored and does not
replace the previous value in the map stack
(Berry)
Composite Operation
Composite— creates a map summarizing values from one map that
coincide with the categories (termed regions) identified on another map
Template Map
“Cookie-cutter”
…the regions identified by category values
on one map serve as cookie-cutter shapes
(Template map) for summarizing data
contained on another map (Data map)
Data Map
COMPOSITE Districts WITH Slope Average
IGNORE PMAP_NULL FOR Districts_avgSlope
…data summary procedures include
Average, Standard deviation,
Coefficient of variation, Total,
Maximum, Minimum, Median
Majority, Minority, Diversity,
Deviation and Proportion
(Berry)
Thematic Mapping (Average elevation by district)
Worst
“Thematic Mapping”
(Discrete Spatial Object)
Average Elevation
of Districts
500
1539
(0)
(39)
2176
(9)
653
(29)
1099
(21)
1779
1080
(24)
(9)
Best
…include
Coffvar in
Thematic
Mapping
results
(Berry)
Map Accuracy (Error Propagation– simple overlay)
Topological Overlay
Intersect
Polygons
Spatial Table
Attribute Table
Soil Type
Forest Type
Join
Data
Coincidence Search
100% certain everywhere
in the derived polygon?
Believable?
(digital slide show Honest)
Vector
Raster
Joint Probability — likelihood of
two conditions occurring together…
P(x,y)
P(x)
P(x,y) = P(x) * P(y)
P(y)
Joint
Probability
…evaluated cell-by-cell
(See Beyond Mapping II, “Topic 4, “Toward and Honest GIS”)
(Berry)