Mastering_ArcGIS_Pro_CH11_ADA_AU_Accessible.pptx

Because learning changes everything.®
Chapter 11
Raster Analysis
Mastering ArcGIS Pro
Second Edition
Maribeth H. Price
© 2023 McGraw Hill, LLC. All rights reserved. Authorized only for instructor use in the classroom.
No reproduction or further distribution permitted without the prior written consent of McGraw Hill, LLC.

© McGraw Hill, LLC 2
Raster analysis
Source: USGS
• Raster data have a unique storage model
base on numeric arrays georeferenced to
a location on earth
• Each cell has a size (resolution) and
stores only numeric values
• This format permits a wide variety of
analysis techniques
• Raster analysis is often quicker than
vector analysis
• Because raster analysis tools often rely on
distances or areas, rasters should be
stored in a projected coordinate system
that preserves these properties
Access the text alternative for slide images.

Raster types
Source: USGS
Rasters come in two forms
• Discrete rasters that store objects
or categorical data, like these land
use polygons.
• Continuous rasters that store a
numeric measurement that occurs
everywhere, like this elevation
raster.
Because raster analysis tools often
use distances or areas, rasters should
be stored in a projected coordinate
system that preserves area and or
distance

Map algebra
precip_in * 2.54
[ ]
• As arrays of numbers, rasters with
the same extent and resolution can
be analyze using cell-by-cell
methods called map algebra
• Two rasters can be added by
summing the values of the
corresponding cells
• Or a raster may be multiplied by a
constant, for example to convert
precipitation values in inches to
values in cm
• A new raster is created to store the
map algebra result

Raster queries
Source: USGS
• Rasters can also be queried
using logical expressions
• The purple area in this DEM
shows where the elevation is
grater than 1400 meters
• The output raster has a 1
(True) where the condition is
met and a 0 (False) where
the condition is not met
• This type of raster is called a
Boolean raster

Boolean operators1
• Boolean operators can be used in
map algebra
• This diagram shows how each
operator (AND, OR, XOR, and
NOT) combines conditions A and
B, where yellow indicates true and
gray indicates false
• The results are shown as a Venn
diagram and a four-cell raster
example
• Each output cell receives a true or
false value based on the input
conditions and the Boolean
operator

Boolean operators2
• The AND operator is equivalent to the
vector intersect function
• The OR operator is equivalent to the
vector union function
• The NOT operator is equivalent to the
vector erase function
• XOR has no simple vector equivalent
(although it can be achieved using
intersect followed by erase)
• Notice that the AND operator yields
the same result as multiplying the
rasters; the two can be used
interchangeably

Raster overlay
Source: USGS (a), South Dakota Geological Survey (b), Black Hills National Forest (c)
• Overlay can be
performed using vector
or raster data
• To do a raster version of
the snail habitat analysis,
Boolean rasters are
created for each
condition
• Then the AND operator
is applied to find where
all conditions occur

Raster Calculator
Source: Esri
• Map Algebra and
Boolean overlay are
performed using the
Raster Calculator tool
• Here, the three snail
habitat condition rasters,
ElevRange,
LimestoneBoo, and
DenseTrees, are being
multiplied to produce the
snail habitat raster

Boolean overlay
To determine suitable areas for lodgepole pine, two Boolean
rasters are generated
a) areas where precipitation is greater than 60 cm
b) areas where elevation is above 1500m
The AND operator is applied and results in 1 (true) where
both conditions hold, defining the lodgepole habitat (c)

Additive Boolean overlay
• In Boolean overlay, the inputs
can be added to provide a
ranking of the habitat instead of
a stark yes or no
• For lodgepole pine, 2 indicates
that two conditions are present,
and 1 indicates that one
condition is present
• In the snail habitat example, a
total of three different conditions
may be present

Boolean overlay drawbacks
One drawback of Boolean overlay is that
conditions may be more gradational than
warrants a true/false treatment
In the lodgepole example, an elevation of
1500 meters is considered suitable, but
1501 meters is considered unsuitable
Weighted overlay ranks conditions for a
more realistic approach
0 to 1000 m is unsuitable
1000 to 1500 m is somewhat suitable
>1500m is ideal

Weighted overlay
• In weighted overlay, K rasters represent K conditions
• Each condition raster is ranked on a scale of 1 to N from
worst to best
• Each raster is also assigned a weight W to indicate its
relative importance. The K weights must sum to 1
1 2 k
W + W +…+ W = 1
• The output is then calculated as
1 1 2 2 k k
W R + W R +…+ W R
• to yield an output raster with values from 1 to N

Siting a landfill
Weighted overlay can be illustrated by the problem of siting a
landfill based on these conditions
• Lower slopes are better, to minimize site bulldozing.
• Low soil infiltration is important, to prevent leakage of
contaminants into groundwater.
• Closer is existing roads is better, to minimize the cost of
building an access road.
• Further from streams is better, to minimize contamination
of streams by runoff from the landfill.
We have K=4 conditions and will rank them on a scale of 1 to
3 (N=3)

Landfill suitability ranks and weights
The four conditions (slope, infiltration,
distance to roads, and distance to
streams) are reclassified into three
categories where 3 is best and 1 is worst
Each condition is also assigned a weight
based on importance
• Infiltration is considered most
important to prevent leaking of waste
into groundwater.
• Slope was considered least important.
• The weights must sum to 1.

Landfill suitability calculation
The ranked rasters are added,
with the weights, to produce a
Landfill Suitability Index (LSI)
LSI = 0.1*slope+ 0.4*infil+ 0.2*roaddist + 0.3*streamdist
The resulting raster shows the
suitability for a landfill site on a
scale of 1 to 3

Distance functions
Source: USGS, Black Hills National Forest
Distance functions are used
to
a) calculate distances from
features
b) determine direction of
travel to the closest
feature
c) create buffers
d) determine least cost
paths

Euclidean distance
Source: Black Hills National Forest
Source: Esri
• This tool is used to calculate
distances from a set of vector
features like streams
• It creates a raster in which
each cell represents the
distance to the closest stream
• The distance units match the
units of the raster coordinate
system (usually meters or
feet)
• A Euclidean direction raster
may be created if desired

Euclidean direction
Source: Esri
• The Euclidean distance tool
can also calculate a
direction raster if one is
requested
• It shows the direction one
would travel to get to the
nearest feature (stream) in
degrees from 0 to 360
• In this map, red indicates
north, yellow east, cyan
south, and blue west

Buffers
• A Euclidean distance raster (a)
can be reclassified to create raster
buffers
• A logical expression is applied,
such as [streamdistance]<500
• The closer cells that meet the
criteria are assigned a value of 1
and the further cells are assigned
a 0, creating a Boolean raster
representing the buffers (c)
• The raster buffers can then be
used for Boolean overlay or other
analysis

Least cost path
• A straight line is the shortest
distance between two points, but it
is not always the easiest
• A hiker would likely prefer the
longer path that climbs gradually
rather than going up and down
several ridges
• The least cost path function uses
a cost raster (such as slope) to
quantify the cost of travel and
combines it with a distance raster
to determine the easiest route
between the points

Density functions
Source: Esri
• Density functions count the
number of features within a
specified circle and divide by
the circle area to yield a density
value
• Each feature can be weighted
using an attribute, if desired
• The map shows a population
density raster resulting from
counting cities within a 500 mile
circle, with each city weighted
by population

Types of density functions
Source: Esri
A simple density function assumes
that the weight value occurs exactly
at the feature location
• Counting impact craters on a
planet’s surface.
A kernel density function spreads
the value over an area using a
Gaussian distribution before counting
• This method is better for
population, which is not
concentrated at the single point
representing the city.

Interpolation
Source: Black Hills National Forest, National Climatic Data
Center
• Interpolation estimates values in
between point measurements
• This raster contains annual
precipitation values estimated
between the climate stations
• The inverse distance weighted (IDW)
method is the most common form of
interpolation, where nearby stations
are given more weight than those
further away
• Interpolation is a complex art, and
those who use it should learn more
about it

Surface analysis
Source: USGS
A large suite of functions are
available for analyzing
surfaces
The most common types are
applied to elevation surfaces
like this DEM (a) and
include
b) calculating slope
c) determining aspect
d) creating hillshade and
viewshed rasters

Slope
Source: USGS
Source: Esri
The Slope tool calculates the
steepness of a surface such as a
DEM (a)
The output raster (b) may be
expressed in degrees or percent
Calculated in degrees θ using
tanθ = rise/run
Calculated in percent using
% slope = rise/run*100
In (b), the darker brown areas
have higher slope

Aspect
Source: USGS
• Aspect is the direction of
steepest slope
• Aspect is expressed in degrees
from 0 to 360, where North = 0
• Flat areas are assigned a value
of −1
• In this map, north is red,
northeast is orange, east is
green, southeast is chartreuse,
south is cyan, southwest is blue,
west is purple, and northwest is
magenta

Hillshade
Source: USGS
• A hillshade raster (gray) mimics
the appears of a uniform
topographic surface under an
illumination source
• The source is placed at a
designated azimuth (direction)
and zenith angle
• Hillshade maps excel at
displaying fine topographic
surface details and are good
base maps

Viewshed
Source: USGS
• A viewshed shows the areas that
are visible from a designated
point such as this trail lookout
• The height of the observer and
heights of obstructions such as
trees can be factored in if
desired
• The orange shades in the map
show the areas that are visible
from this lookout point

Neighborhood statistics
Source: Esri
Neighborhood statistics functions operate on a neighborhood, or
window, around a target cell
• The neighborhood is usually square, but other shapes can be used
as well, such as circles or wedges.
Statistics for the neighborhood, such as the mean value, are calculated
and placed in the target cell
The neighborhood then moves to the next target cell for evaluation, until
an entire new raster is produced

Types of neighborhood functions
A block neighborhood function
evaluates one set of cells and then
moves to an entirely new set
• Cells are never in more than one
neighborhood block.
A focal neighborhood function
moves one cell at a time to create
overlapping neighborhoods
• Each cell is evaluated in multiple
neighborhoods.

Neighborhood function example
Source: USGS
This land cover map (a) has been subjected to a majority
neighborhood statistics function with a 4×4 cell window
• The cover type that occurs most frequently in the neighborhood
is assigned to the target cell.
• If no cover type has the majority, a white cell (No Data) results.
Map (b) shows the output of a focal majority function
Map (c) shows the output of a block majority function

Simplifying a raster
Source: USGS
• Rasters can be converted to features, but a complex
raster may produce many tiny polygons
• This land cover raster (a) has been subjected to two
passes of the focal majority tool with a 4×4 window
• The output is simpler and more suitable for conversion to
polygon features

Zonal functions
Source: South Dakota Geological Survey
• Zonal functions are also
neighborhood statistics
functions, but the neighborhood
is determined by features in a
data set
• A zone is a set of cells or
features with a unique value
• Note that zones need not be
contiguous regions
• There are six zones (geologic
units) in this map

Using zonal statistics
Source: USGS; Black Hills National Forest
• In this map (a), the watershed polygons define the zones
• The Feature ID field was used to ensure that each
watershed had a unique value and would be a different
zone
• The zones were applied to a slope raster to calculate the
average slope for each watershed (b)

Reclassify function
Source: Esri
• A reclassify function
changes the values in a
raster to a different set of
values
• The elevation ranges in this
raster are being reclassified
as suitable snail habitat (new
value of 1) or unsuitable
habitat (new value of 0)
• The output can then be used
in a Boolean overlay to find
snail habitat

Spatial Analyst
Source: Esri
• Spatial Analyst is an
extension to ArcGIS that
performs raster analysis
• It extends the standard tools
available in the
Geoprocessing pane
• Its many functions are
organized into toolboxes
• An additional license is
required to run these tools

Environment settings
Source: Esri
In raster analysis, it is helpful to
set several environment
settings to ensure correct and
consistent processing
• workspace.
• processing extent.
• cell size.
• mask.

Workspace settings
Source: Esri
• The Workspace settings control where output is placed
(defaulting to the project geodatabase)
• Raster analysis produces many rasters along the way, some of
which are temporary and exist in the scratch workspace only
while the tool is running
• Sometimes it is wise to direct the output and temporary rasters
to a “junk” database created to hold these results, to make
cleanup of unwanted files easier

Output Coordinates settings
Source: Esri
• The Output Coordinates settings control the output coordinate
system of the processing results
• Because projecting rasters involves resampling, it is NOT a
good idea to change coordinate systems during processing
• Doing so may degrade the quality and accuracy of the results
• This setting should be blank for raster analysis

Processing Extent settings
Source: Esri
• Rasters are always rectangles
• If the extents of two input rasters
don’t match, this setting
determines the output extent
• The default is the intersection of
the inputs (yellow area) because
these are the cells that would have
meaningful values
• To avoid resampling, the best
practice is to choose one of your
rasters or layers (Roads in this
example) and force all outputs to
match it

Cell size setting
Source: Esri
• If the cell sizes of raster inputs do not match, they will be
resampled to match each other
• This setting controls what cell size will be used
• It defaults to the maximum cell size of the inputs (since the
largest resolution controls the accuracy of the output)
• To avoid automatic resampling, which degrades the data sets,
the best practice is to set a consistent cell size that matches
your inputs

Mask setting
Source: Esri
Source: USGS
• A mask can be used to clip
all output rasters to the area
of interest, such as this
watershed
• Either a raster or polygon
feature class can be used as
a mask (a raster must have
NoData values outside the
desired region)
• The output rasters will
contain NoData values for
cells outside the mask

Resampling
• Raster processing requires cell sizes to match; if they
don’t it will resample the input rasters
• This process is invisible to the user, who is often not
aware that the data are being degraded, and the analysis
result may not be as accurate as it should be
• Avoid automatic resampling by being careful to use
consistent extents and cell sizes throughout an analysis

Because learning changes everything.®
www.mheducation.com
End of Main Content
© 2023 McGraw Hill, LLC. All rights reserved. Authorized only for instructor use in the classroom.
No reproduction or further distribution permitted without the prior written consent of McGraw Hill, LLC.

Mastering_ArcGIS_Pro_CH11_ADA_AU_Accessible.pptx

More Related Content

Similar to Mastering_ArcGIS_Pro_CH11_ADA_AU_Accessible.pptx

Recently uploaded

Mastering_ArcGIS_Pro_CH11_ADA_AU_Accessible.pptx