Almost without exception, all GIS users apply integrated raster layers as a fundamental data source in their GIS. In many cases, users only think of the raster data as a static "snapshot" of information that builds compressed static catalogs of their imagery. However, users are beginning to conceive of raster data as a transactional layer undergoing continuous update. This is increasingly relevant as new data sources are readily available to be plugged into their GIS. This is due in large part to vast new sensors coming on line as well as the emergence of strong raster support in ArcGIS and the geodatabase.

Raster data sources are integral information layers in all GIS applications including GIS use in local government and hydrology.

Key data management benefits are provided by the combined use of GIS and the database management system (DBMS)—not only for storing and managing vector data sets but also for managing massive raster databases. New sensors are increasing the resolution, frequency, and flow of raster information into GIS systems. Integrated raster data management is useful for access by many users, managing and maintaining raster updates, and controlling information flow and access. In addition, the number of individual and small GIS users continues to expand along with their growing need to ingest and manage large raster data collections.

Raster data sources are integral information layers in all GIS applications, including GIS use in local government and hydrology.

The geodatabase and ArcGIS 9 provide comprehensive support for managing and using raster data, including

• Large continuous raster mosaics
• Collections of multiple images in raster catalogs
• Catalogs that reference image files on disk
• Pictures and attributes in relational tables
• Support for approximately 50 formats, including direct use and conversion
• On-the-fly raster projections
• Raster pyramids and partial updates
• Raster and image processing tools

These four sample data models [PDF-8.36 MB] illustrate some commonly used raster data representations and how GIS users manage raster data sets as integral GIS data layers. The case study examples include

• A raster catalog containing orthoimagery covering the state of Texas at one-meter resolution
• Use of a raster catalog with a complete mosaic of the scanned United States Geological Survey (USGS) map sheets (one image in the catalog for each state)
• The nationwide elevation database for the United States as a single continuous raster data set developed by USGS and known as the National Elevation Dataset
• A time series catalog of Geostationary Satellite Server (GOES) imagery of Hurricane Mitch from 1998

Esri has tested these and other global databases for high performance and usability with excellent results—subsecond performance, access by hundreds of simultaneous users without loss in performance, ability to manage global data collections, and other critical benefits. High performance and scalability have been demonstrated across all DBMSs—Oracle, as well as SQL Server, Informix, and DB2 databases.

For more information on these raster and other ArcGIS data models, see the new Esri Press book Designing Geodatabases by David Arctur and Michael Zeiler. You can also visit the companion Web site for the book to download and use sample databases and data models at support.esri.com/datamodels.