|
|
CARTOGRAPHIC MODELING Data Structures There are different ways to organize the data inside an information system. During your initial GIS set-up it is necessary to decide which type of spatial data structure to use. It is important to understand the different types of data structures since more than one are often in use at the same time. Data storage volume and processing efficiency are affected by the type of data structure being used. A choice can be made depending on how you want to store your data. The three basic models of the world that ArcInfo uses are vector, raster, and TIN. Vector modelsA vector model represents the world in terms of points, lines and areas and is a more precise representation of specific locations. These models are particularly useful for representing discrete features such as buildings, roads, contour lines or geographic boundary lines. Points, lines and areas are stored as coordinates or sets of connected coordinates. The data is stored as features in separate feature classes. Feature data allows you to remove the attribute information from either the polygon, arc or node and place it directly with the feature. Features describe the various components; complex features are features composed of other features. You can select a vector model to represent your data in coverage, shapefile or geodatabase format. See the Vectorizing in ArcInfo page for more information about vector data specifically using ArcInfo software. Raster modelsIn a raster model, the world is represented as a surface divided into a grid of cells. Each cell is attributed a value, which can represent membership in a class, a measurement or an interpreted value. Raster data is a larger data format, but is limited because you can only specify which cell you are in; there is no label for the area in between. The smaller the cell size, the higher the resolution and thus more detailed the map and larger the volume of data to be stored. A satellite photograph is a popular example of information suited to a raster model. TIN modelsIn a triangulated irregular network (TIN) model, the world is represented as a network of linked triangles drawn between irregularly spaced points with x, y, and z values. This model is efficient for heterogeneous surfaces that have a lot of variation in specific geographic areas and little in others. These surfaces are stored as TIN datasets. Specifically for GIS, TIN allows you to efficiently calculate slope and aspect data for displays such as contour or shaded relief maps. |