Raster Data Model

• Rasters composed of a matrix of pixels (cells)
• Each pixel contains a value representing the conditions for the area covered by that pixel

Advantages of Rasters

• Representing continuous spatial variation
• More efficient than small polygons (in the vector model)

Georeferencing Rasters

• Defines the correspondence between each pixel and location on the earth’s surface
• Coordinates of top left pixel in the images
• Size of each pixel in X and Y direction
• Rotation angle

Raster in QGIS

Raster in QGIS

Sources of raster data

Remote Sensing Satellites

Remote Sensors

Planet

Earth Engine Timelapse

‘url’

Remote Sensing Imagery Resolution

• Spatial Resolution
• Spectral Resolution
• Temporal Resultion

Spatial Resolution

How much area is represented in each pixel

• MODIS 500m x 500m
• SPOT5 satellites: 10m x 10m
• Areal Photography, drones: 50cm x 50cm

Spatial Resolution and Scale

Spectral Resolution

• The number of spectral bands in which a sensor can collect reflected radiance
• Choice depends upon application of use
• Different types of ground targets have different spectral signatures

Temporal Resolution

Revisit Periods for Satellites

• SPOT5 - 3 days
• Modis - daily
• Airborne - as needed

Conversions

Which Model?

Vector to raster conversion

Raster to vector conversion

Rasters in GIS for Public Policy

• Measures of environmental variables
• Risk surfaces
• Dasymetric mapping and areal interpolation

Environmental variables

Freeman et al. (2017) “Willingness to pay for clean air in China”

FEMA: Risk Surface

Dasymetric Mapping

NSF-Neighborhoods

Interpolation and Harmonization

Tobler

Tobler: Voting Precincts

Tobler: Dasymetric

National Land Cover Database

Next Up

• Coordinate Reference Systems