Acquiring the appropriate data for a project involves creating a database of layers that encompasses the study area. A database created with ERDAS IMAGINE can consist of:
- continuous layers (satellite imagery, aerial photographs, elevation data, and so forth)
- thematic layers (land use, vegetation, hydrology, soils, slope, and so forth)
- vector layers (streets, utility and communication lines, parcels, and so forth)
- statistics (frequency of an occurrence, population demographics, and so forth)
- attribute data (characteristics of roads, land, imagery, and so forth)
The ERDAS IMAGINE software package employs a hierarchical, object-oriented architecture that utilizes both raster imagery and topological vector data. Raster images are stored in image files, and vector layers are shapefiles or coverages based on the Esri ArcInfo and ArcView data models. Using the seamless integration of these two types of data, you can reap the benefits of both data formats in one system.
Raster data might be more appropriate in the following applications:
- site selection
- natural resource management
- petroleum exploration
- mission planning
- change detection
On the other hand, vector data may be better suited for these applications:
- urban planning
- tax assessment and planning
- traffic engineering
- facilities management
The advantage of an integrated raster and vector system such as ERDAS IMAGINE is that one data structure does not have to be chosen over the other. Both data formats can be used and the functions of both types of systems can be accessed. Depending upon the project, only raster or vector data may be needed, but most applications benefit from using both.
Themes and Layers
A database usually consists of files with data of the same geographical area, with each file containing different types of information. For example, a database for the city recreation department might include files of all the parks in the area. These files might depict park boundaries, county and municipal boundaries, vegetation types, soil types, drainage basins, slope, roads, and so forth. Each of these files contains different information—each is a different theme. The concept of themes has evolved from early GIS systems, in which transparent overlays were created for each theme and combined (overlaid) in different ways to derive new information.
A single theme may require more than a simple raster or vector file to fully describe it. In addition to the image, there may be attribute data that describe the information, a color scheme, or meaningful annotation for the image. The full collection of data that describe a certain theme is called a layer.
Depending upon the goals of a project, it may be helpful to combine several themes into one layer. For example, if you want to propose a new park site, you might create one layer that shows roads, land cover, land ownership, slope, and so forth, and indicate through the use of colors and/or annotation which areas would be best for the new site. This one layer would then include many separate themes. Much of GIS analysis is concerned with combining individual themes into one or more layers that answer the questions driving the analysis. This chapter explores these analysis techniques.