skip to content
MODIS Image
Cofrin Center for Biodiversity

•  SPATIAL DATABASES  •

The various branches of geospatial technology have one thing in common. They work with georeferenced data in the form of spatial databases. Spatial and aspatial data are similar in many ways. Multiple elements (records), contain values for a set of attributes that describe the "what is" aspect of each record. However spatial databases also include the "where is" aspect of each record.

There are millions of spatial DBs out there and they didn't just appear out of nowhere. Each one is a technical publication in the same sense as a journal article or reference book and each one has an identifiable author who has chosen a data structure and loaded the data into it.
When we use data from an external source, we need to recognize the properties designed into it by its author. When we create new spatial databases ourselves, we need to make thoughtful design choices based on the properties of the data and who will be using it.

Data Organization
The author selects a data organization model based on what will work best for the type of data being mapped.

Datum and Geographic Coordinate System
The author selects a datum and geographic coordinate system.

Projected Coordinate System
The author may decide that the data is more useful if it's mapped in 2-dimensional space. In that case a projected coordinate system is also selected.

  • Map Projections is a web-based technical manual from the Michigan Department of Natural Resources. The section titled "What is a Map Projection?" is one of the clearest, most concise explanations available. Note their use of the word "curvaceous".
  • Our Projected Coordinate Systems page includes examples and application info

Data File Format
The file format used to store spatial data matters a great deal. The job of a database is to store a large amount of data so that it is both accessible and secure. There is a lot to be said for using a "tried and true" file format but there is also pressure to go with newer data file designs that have more storage capacity and/or run faster and/or are more secure. The database author has to take all this into account when choosing a file format.

Summary Table
The table below summarizes the design choices made in the course of creating spatial databases.

Data Organization Model Database File Format Datum & Coordinate System
Raster
  • Digital Elevation Model (DEM)
  • Satellite Image
  • Digital Orthophoto (DOP)
  • Digital Raster Graphic (DRG)
Traditional
  • ArcInfo Grid
  • Geotiff (.tif)
  • ERDAS Imagine (.img)
  • MrSID (.sid)
Geodatabase
  • Esri raster

Datum - there are hundreds, but only a few are in common use in this part of the world

  • WGS 84 (World Geodetic System of 1984)
  • NAD 27 (North American Datum of 1927)
  • NAD 83 (North American Datum of 1983)
  • NAD 83 HARN aka NAD 83/91 (North American Datum of 1983 with 1991 adjustments)

 

Geographic CS - the geographic coordinate system is defined by the datum it is based on. The same name is used for the datum and the associated geographic coordinate system.

 

Projected CS - a projected system is an optional enhancement added by a DB author to make the product more user friendly. There are thousands of projected systems. Each one is designed for a very specific task on very limited portion of the earth surface so there are only a few that we care about:

  • County systems for counties we are working in.
  • Wisconsin Transverse Mercator for regional-scale mapping in the state
  • Universal Transverse Mercator Zone 16 for regional-scale mapping falling between 84° and 90° west longitude
  • USA Contiguous Albers Equal Area Conic for work spanning the contiguous 48 states.
Vector (point geometry)
  • Point shapefile
  • Point feature class
Traditional
  • ArcView shapefile (.dbf, .prj, .shp, .shx)
  • ArcInfo coverage
Geodatabase
  • Feature class
Vector (line geometry)
  • Line shapefile
  • Line feature class
Vector (polygon geometry)
  • Polygon shapefile
  • Polygon feature class
Other
  • LiDAR point cloud
  • Triangulated Irregular Networks (TIN)
 

Authorship and Metadata
For  his/her mapping work to have lasting value, the author needs to develop and package it in accordance with published standards. The data files should be accompanied by one or more files containing "data about the data" (metadata).