Raster Data

why use raster’s

some data is only available as a raster

what is an example of data only available as a raster

imagery

what type of data is typically stored in raster format

continuous data

what are advantages of rasters 

1. a simple data structure 

2. ability to represent continuous surfaces

3. ability to store points, lines, polygons and surfaces

what does it mean rasters are simple data structure

they are a matrix of cells with values representing a coordinate and value

examples of continuous surfaces

elevation

rainfall

temperature

vector data tends to be ______ in a raster 

coarser 

what happens when you try to make vector data finer

it takes more memory to store and do

what are some disadvantages of raster data

1. spatial inaccuracies due to limits imposed by pixel size

2. can be very large datasets

3. loss of precision accompanying restructuring of data into raster

what is each cell known as in raster data 

known as a pixel that has a value 

what do cell values represent

represents the phenomenon portrayed by the raster dataset (categorical, magnitude, height or spectral values)

what are four requirements for rasters

1. coordinate system

   1. reference coordinate or XY location (ex. 0,0)

2. pixel size

3. count rows or columns

how are coordinates calculated

counting number of rows and columns from corner with coordinates and multiplying by cell size

what does the area represented by the cell have

the same width and height

how does cell size vary 

by product (some are small and others large)

what do satellite images or digital images HAVE? what do they NOT have?

have - raster format

lack - internal information beyond cell value

how many values per cel

ONE

what type of value is a raster cell

brightness value

how are the brightness value recorded

as a Digital Number (DN)

what makes it easier to see but NOT changes the cell value

contrast stretching

what value represents dark brightness? light brightness?

dark = closer to 0

light = closer to 255 (upper limit)

what do satellites measure

how bright the reflectance of a target is

what is reflectance recorded in 

as a DN

what does the range of the DN values depend on

bit depth of the sensor

what is the range of DN values for 8 bit data

2^8 = 256

range is 0 to 255

what defines the radiometric resolution of the sensor

number of DN values

what is radiometric resolution 

measure of a sensor’s ability to detect and differentiate small variations in energy intensity (brightness) from a target 

are all raster’s satellite images

NO - vectors can be raster

what can cell values be

positive or negative, integer, floating point

what are integer values used to represent 

categorical (discrete) data

what are floating point values used to represent 

continuous surfaces

what kinds of data have the value at the center of the cell? the whole area of the cell?

center = DEM

whole area = soil PH, rainfall, land classification

what is count

the number of cells equal to value

what number must values be under for a table to be created

under 500

types of data 

1. nominal 

2. ordinal 

3. continuous 

nominal vs ordinal vs continuous 

nominal 

• categorical or qualitative

• have names or codes

• NO quantitative value

• ex. land use or soils

ordinal

• categorized classes

• categorical or ordered

• has rank

• ex. risk, land suitability

continuous 

• spatially continuous phenomenon

• each pixel holds a measurement with any value in a range

• ex. elevation, temp, rainfall

what are rasters well suited for representing

data that changes continuously across a landscape

how do rasters provide a method of storing continuity of a surface

using regularly spaced grid cells to represent the surface values

_____ values are a common application of surface maps

elevation

can rasters have math preformed on them

YES

what should be done BEFORE doing raster math

1. cell sizes are the SAME

2. matrix size is the SAME (columns and rows)

3. rasters align/snap to each other

what is the extent 

geographic bounding area within which spatial analysis can occur 

what is the extent bounding area defined as

the maximum x and y coordinates of opposite corners

what is a mask

identifies areas to be included in analysis either as a raster or feature layer

mask vs extent

extent - a rectangular bounding box for the entire map or the analysis,

mask - a more specific area within the extent that limits where analysis occurs

rasterization vs vectorization 

rasterization: vector → raster

vectorization: same raster → vector

what can cause rasterization to be spatially inaccurate

if the rid cell size is TOO large

(points too large, polygon edges too coarse, lines too wide)

what defines how coarse or fine the feature will appear in a raster (rasterization)

the cell size

smaller cell size in rasterization means? larger cell size?

smaller = the smoother or more detailed the raster will be BUT needs more processing power

larger = coarser or less detailed the raster BUT needs less processing power

contrast the FOUR types of GIS analysis functions 

local - preformed on SINGLE cell at a time where neighboring cells DO NOT influence the result

(ex. trig functions, exponential functions)

focal - preformed on single cell AND its neighboring cells

(ex. sd, mean, sum or values)

zonal - calculation on a zone/set of cells with common value

(ex. area, perimeter)

global - computation on the raster as a WHOLE

(ex. watershed delineation)

slope identifies

the maximum rate of change in value from each cell to its neighbours

what can the output slope grid be calculated as

percent slope

what does the slope use when calculating values 

revolving 3×3 window 

what are some stats calculated by the 3×3 window

1. mean

2. max

3. range

4. slope

5. aspect 

6. terrain roughness 

percent of slope formula

(rise/run) * 100

degree of slope

tran-1(rise/run)

what happens to the slope as grid cell size increases

the mean slope decreases and the surface smooths out

aspect identifies

the downhill direction of a slope at a specific location

what can aspect be thought of

the slope direction

what are aspect values

compass direction

what does a zero slope mean and what is the aspect

flat slope, aspect = -1

what makes up zonal statistics 

zone layer - defines the zones 

value raster - the input values used in calculating the output for each zone

output table - can be joined to the zone layer to display stats per zone 

what is a distance raster

raster file where each cell’s value is its distance from the specific feature (point, line, polygon)

what is Euclidean distance

the "straight-line" distance between two points calculated using the Pythagorean theorem

what is reclassify

assign new values to old cells which put wide range of values into meaningful classes