RASTER DATA

ANALYSIS
Geographic Information
Systems
GEOG 4047
Raster and vector comparison
Vector Data:
● Represents spatial information
Raster Data:
● Uses a grid of regularly spaced cells using points, lines, and polygons.
● Each feature in the data (point,
or pixels to represent spatial
line, polygon) has attributes
information.
● Each cell in the grid contains a associated with it.
● Well-suited for representing
single value or attribute.
● Well-suited for continuous data such discrete features such as roads,
rivers, and administrative
as elevation, temperature, and
boundaries.
satellite imagery.
● More flexible in terms of
● Can efficiently represent large,
representing complex shapes and
continuous areas.
● Simple and efficient for operations topologies.
● Typically more compact than
like distance and proximity analysis.
● Can be more memory-intensive raster data for storing complex
geometries.
than vector data for large areas or
● Suitable for operations that involve
high resolutions.
● Common formats include overlay, buffering, and network
analysis.
Geodatabase raster, GeoTIFF,
● Common formats include
JPEG, and ERDAS Imagine.
shapefile,Geodatabase feature
class, GeoJSON, KML.
Raster and vector comparison
Raster Vector

Precision Dependent on raster cell Dependent on sampling

size device, e.g. GPS, source
map

Structure Simple Complex

Data type Continuous/discrete Discrete data, TIN for

rasters continue data

Geometric elements Cells or grid matrix Point, lines, polygons

Data sources Temperature, air pressure, Administrative data, roads,

elevation, satellite images GPS, river, incidents, point
of interest

Spatial analysis Surface functions, raster Overlay, spatial join, query

calculator, distance, map , buffer, dissolve
algebra, focal functions
 Geographic properties
• The four geographic properties of raster datasets are
the following:
• A coordinate system
• A reference coordinate or x,y location (typically the upper
left or lower left corner of the image)
• A pixel size
• The count of rows and columns
• The order of rows is opposite to the Y coordinate space
• The geotransform coefficient (gt) for the Y coordinate is
usually negative:
• y_geo = line * gt + y0

https://pro.arcgis.com/en/pro-app/latest/
help/data/imagery/introduction-to-raster-
data.htm
Raster data management tools
• Convert to vector
• Convert from vector
• Copy raster
• Projection
• Resampling
• Reclassification
Converting features to raster

Choose one attribute

Text attribute
Grid value
carried over
Converting raster to features
To complex to convert

OK to convert

Convert only discrete data

without too many polygons
Copy raster
- The tool creates a copy of
the raster dataset
- You can use this tool to
download images from
online image services
- Change pixel type if needed
- Save the data in a different
format is needed
- Change the environment
settings including output
coordinate system, extent,
cell size, snap raster, etc.
Clip raster
• This tools cuts out a portion
of a raster dataset, mosaic
dataset, or image service
layer.
• An existing raster or vector
layer can be used as the
clip extent
• If a vector data is used, the
raster is clipped by the
minimum bounding box or
by the polygon geometry of
the features
• You can interactively draw a
polygon on the raster in the
map to be used as the
clipping extent. Click the
Create new features in the
current map to use as input
Projecting raster data
Similar to projecting vector data
Need to specify the resampling method
Resampling
If input grids do not match, then one
must be resampled to match the other.
Resampling can degrade the accuracy of
a raster even if the difference in cell size
and location is small.
The new cell grid is determined, and the
old cell values must be fit into the new
structure somehow.
Y2 Several methods are used for
resampling.
Resampling methods
Nearest neighbor resampling grabs the value from the old
cell that falls at the center of the new cell. It preserves the
original value and should always be used with categorical
data, or when the original data values need to be preserved.
It is the fastest method.

Bilinear resampling calculates a new value from the four

cells that fall closest to the center of the new cell. It uses a
distance-weighted algorithm based on the old cell centers. It
is best used with continuous data such as elevation.

Cubic convolution resampling calculates a new value from the

sixteen cells that fall closest to the center of the new cell. It
uses a distance-weighted algorithm based on the old cell
centers. It is best used with continuous data such as
elevation. It is the most time-consuming method.
Reclassify
Slope

Boolean: Slope < 10

Spatial Analyst functions
Local Focal functions Zonal functions
functions
LOCAL FUNCTIONS
Local analysis

• Raster analysis uses

cell-by-cell functions on
one or more input grids.
• Cells must be the same
size and line up
spatially.
• Older software required
the user to ensure that
all input rasters had
exactly the same size,
shape, and aligned cell
sizes.
Map Algebra – Local functions
• Map algebra is also called map calculator, or local functions
• The calculation is performed by applying some formulas to the values of
the grid
• For example: C = A + B
• Put every cell in layer C as the sum of the values of the corresponding cell in A
and B
Raster Calculator
• 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
Raster calculator functions listed in the
tools
• Math
• Abs – get the absolute value
• Exp – Exponential
• Float – convert to floating point number
• Int – Convert to integer value
• Logical
• IsNull – Tests for NoData
• A DIFF B - if A and B are different, return A, if not, returns zero
• Conditional
• Con(c, A, B) – if c is true, then return A; otherwise returns B
• SetNull(A) – set the value of A to NoData
Using the condition function
Using the raster calculator for
overlay analysis
• Boolean overlay: use 1 for true, 0 for false.
• The results can be a multiplication of the input layers:
• If all input raster layers have "1" at a cell, the cell will
have the value of 1. Otherwise, the cell will have the
value of 0
• Snail habitat = Elevation * Densetrees * Limestone
FOCAL FUNCTIONS
(NEIGHBORHOOD
OPERATION)
Focal (Neighborhood) Operations
• Also known as moving
window, filter, kernel,
convolution, and focal
functions
• Windows with chosen shape,
size, and values
• The shape could be square,
rectangle, circle, donut, or
wedge, but in most cases, the
neighborhood is a square with
an odd number of cells on each
side to ensure the symmetry
about the central cell
 Focal statistics
• The values of the window will
convolute with the cells that fall into
the window and the new value will
replace the old value of the central cell
• sum, average, standard deviation,
maximum, minimum, median of the
neighborhood
• The center of the window will move to
the next cell after each operation
• In ArcGIS Pro, it is called Focal
Statistics
 Moving (sliding) Window

Point Window Moving

being surrounded window
measured point
Neighborhood functions
Statistics: Mean,
majority, variety,
and so on
Window
Target cell
3 1 4 4 1 2.0 2.5 3.4 2.8 3.6
3 1 4 4 1 3.8 3.2 2.9 3.0 2.3
6 2 1 1 2 3.1 3.3 2.5
5 4 3 3 4
3 1 4 4 1

Input grid Output grid

Majority filter
• Take the majority value
from the moving
window
• Applicable to a
discrete raster
• Majority: 3 out of 4 or
5 out of 8 neighbors
must have the same
value
• Half: 2 out of 4 or 4 out
of 8 neighbors must
have the same value
• The output raster will
be stabilized (will no
longer change) after a
few runs of Majority
ZONAL FUNCTIONS
Zonal functions
• Operations defined by zones
• Two input data:
• Zone layer defines the boundaries of zone
• Value layer defines the value of the phenomena
What is a zone?
• A zone is the area(s) of a raster or feature dataset that
share the same integer value.

Zones need not be contiguous!

Zonal statistics
Watersheds and slope
• Zones defined by the
zone layer (watersheds)
• Generates statistics for
each zone from the
value grid (slope)
• Output is either a raster,
or a table
Average slope in watershed
Zonal statistics of lines Mean slope of
streams

Use Feature-ID for zone

designation so that each
stream is a unique zone.
Zonal statistics of points
Determine elevation of summits

• Use name or FID for each

point so each is unique
zone.
• Calculate stats for each
point.
• All statistics are the same—
the elevation of the point.
Density functions
Density functions
• 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
Occurrences may be features or weighted using an
attributes of features (number of
cities versus city population).
attribute, if desired
• The map shows a
population density
raster resulting from
counting cities within a
500 mile circle, with
 Point density
50km radius

100km radius

A larger radius gives smoother results.

The radius is given in map units.
Units: people/sq km
 Line density

50 km radius

Density of rivers m/sq km

11-41
100 km
Heatmap
• Heat map symbology draws
point features as a
dynamic, representative
surface of relative density.
• Use heat map symbology
when you have many points
that are close together and
cannot be easily
distinguished.
• The density is calculated
using the kernel density
method, the same
algorithm used by the
Kernel Density
geoprocessing tool.
Kernel density
• The Kernel Density tool
calculates the density of
features in a neighborhood
around those features.
• The population field can be
used to weight some
features more heavily than
others or allow one point to
represent several
observations.
• Kernels are defined based
on distance to the points
Global functions
Global functions
• The output value at each cell location is potentially a
function of all the cells combined from the various input
raster datasets.
• Examples of global functions:
• Distance functions
• Least cost path
• Flow accumulation
Distance functions
• Euclidean distance
• Assign to each cell in the output raster dataset its
distance from the closest source cell.
• Raster buffer zone can be created in this way
• Weighted distance
• Cost of moving from a destination cell to the nearest
source cell over a surface that is weighted by some cost in
addition to Euclidean distance
• Given a cost layer, e.g. slope, dense vegetation, swamp,
etc.
• Calculate the shortest path and a cost accumulation
raster
Use "Euclidean distance" function to
create buffer zones

Roaddist < 500

0
1

1. Use distance function to

calculate distances to
features
2. Use reclassification to
delineate the buffer zones
Cost distance
Create a cost surface
• A cost raster identifies the
cost of traveling through
each cell.
• Although the cost raster is
a single dataset, it is often
used to represent several
criteria.
• For example, land use and
slope influence the
construction costs.
• Multiple datasets must be
reclassified to a common
scale to make them
comparable.
• Weighted combination:
• 0.66 * slopeclass + 0.34 *
 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
Terrain (Surface)
Analysis
Digital Elevation Model (DEM)
• Used to refer specifically to a
regular grid of spot heights.
• Simplest and most common
form of digital representation
of topography.
• A DEM is a raster
representation of a continuous
surface, usually referencing the
surface of the earth.
• The elevation numbers are
created by referencing to a
defined virtual surface ( Mean
Sea Level, NAVD88, etc.)
• The unit might be meter or foot
• Vertical accuracies of DEM vary
from tens of meters to
 ➢The simplest way is to use a 3x3
Slope window centered on the point. The
maximum slope on the basis of a
comparison of a central target cell with
its neighbors.
 Aspect
■ The deepest downslope
direction
■ Vegetation, crops, fruits
between slopes facing north
and south
■ Wind generators: face the
prevailing winds rather than be
sheltered from them.
■ The aspect at each location
determines the direction of
water flow over the terrain
surface.
Hill shaded Maps from DEM
• Hill shading, relief shading,
plastic shading, shaded
relief.
• Each pixel's illumination
computed from its slope
relative to a hypothetical
“sun”.
• Assume that light source
infinitely far away from
surface and the light is
coming from a constant
direction and elevation
angle.
• Azimuth: the direction of
the light
• Altitude: the height of the
light
Shaded Relief Map
• The continuous tonal/color
variations give an impression
of shadow produced by the
interaction of the sun with
topographic surface, thus
making the map looks more
like a photograph.
• Fusion of colors with the
hillshade map:
• R, G, B, Hillshade
Viewshed analysis
• Areas visible from a set of
observation points
• A viewshed is the region that
is visible from a given
vantage point in the terrain.
Viewshed Analysis
Cut and Fill (Volumetric) Calculation
• The volume of material related
to cutting and filling.
• Subtract the upper (top) surface
height values from the lower
surface height, a variable
thickness (depth) value can be
obtained. Then, integrate
(accumulate) the thickness
values over the entire area to
obtain the volume.
• Engineers need to establish road
or railroad routes and gradients
that minimize the movement of
earth. It is generally most
economical to balance the
amount of material removed
from the high areas (cut) with
the amount of material required
to fill low areas (fill).
• Reservoir capacity estimation,
ice volume, etc.
Cut and fill on a site

[Initial surface] – [final surface]

Cut
Fill

11-60
Cut fill
Hydrologic functions

Derive streams, watersheds, and other hydrologic features based

on analysis of a DEM.
Flow direction
• Flow direction can be
determined by comparing the
elevation data of a cell to its
neighbors
• Typically, the flow direction is
the aspect (deepest
downslope direction)
• Coding system used in ArcGIS
is 2^n (n = 0 … 8), no flow is
recorded as 0
Flow accumulation
• The Flow Accumulation tool
calculates accumulated
flow as all cells flowing into
each downslope cell in the
output raster
• Cells with a high flow
accumulation are areas of
concentrated flow and may
be used to identify stream
channels.
• Cells with a flow Figure from ArcGIS online
accumulation of 0 are local documentation
topographic highs and may
be used to identify ridges.
Watershed
• A watershed is the upslope
area that contributes to a
common outlet as
concentrated drainage.
• The boundaries between
watersheds are termed
drainage divides.
• The outlet, or pour point, is ArcGIS: how Watershed works
the point on the surface at
which water flows out of an
area.
• An outlet point is usually
defined where a gauge
station is installed