Spatial Data

There are two types of spatial (or GIS) data:

1. Vector Data
2. Raster Data

Vector Data Model

• Type of data represented: DISCRETE data
  • Data consisting of clearly defined/delineated entities
  • Examples: road networks, houses, population counts, etc
• Basic data types:
  • Points - identify locations (e.g., hospitals, retail stores, etc)
  • Lines (or Arcs) - connect points (e.g., roads)
  • Polygons (or Areas) - consist of connect line segments
• Data Storage:
  • Topological data and attribute data are stored in separate files
  • Each vector file contains only one basic data type and usually a single theme (e.g., roads, rivers, hospitals, schools)
    • Note: the theme of a vector file depends on who is defining it and for what purpose; roads and bus lines may be considered different themes for one application, but may be grouped together in a more general theme of transportation for another application
    • The key is that each vector file can represent only "line" spatial features, "point" spatial features, or "area" spatial features, but no combination of these!
• File types:
  • Shapefiles
  • Coverages
• Issues:
  • Vector data MUST have a projection
    • A projection is a mathematical model that translates the locations of features on the Earth's surface (3-dimensional) to locations on a 2-dimensional surface; in essence, a projection converts the location of a spatial entity from spherical coordinates (i.e., latitude, longitude) to Cartesian coordinates.
    • It is necessary for doing any sort of spatial analysis
    • Converting between two different projections is possible; it just requires knowledge of the formula used to translate from spherical coordinates (latitude, longitude) to each projection, then an inverse tranformation can be done to convert one of the projections back into spherical coordinates (and then into the other projection)

Raster Data Model

• Type of data represented: CONTINUOUS data
  • Data that exists everywhere in space and varies continuously
  • Examples: satellite images, aerial photographs, digital elevation models (DEMs), etc
• Basic data types: Cell
  • Note: lines and areas in the raster data model are represented by a collection of cells
• Data Storage:
  • Each cell contains two pieces of information:
    1. Location (co-ordinate)
    2. Attribute value
       • i.e., raster data stores all of its attribute data implicitly in the raster file unlike vector data, in which attribute data is stored separately
    • Note: Unlike the vector data model, topology is stored implicitly in a raster file
    • The spatial location of each cell is embedded in the ordering of cells in the matrix
• Issues:
  • The cell is the MINIMUM MAPPABLE UNIT in raster data; therefore, when combining multiple raster files, the overall scale will be the largest one

-- AprilWebster - 05 Mar 2007