Vector data model

Jump to: navigation, search

The Vector Data Model is a strategy for describing distinct features in a GIS. It is one of the two main types of GIS data models, the other being the raster data model.

The Vector Model

The basic framework of the model is to describe each needed feature by its geometry (location and shape) and attributes (also known as properties or characteristics). Usually, related features (for example, all of the road features in a county) are grouped into a single dataset. Most Vector data formats use standard relational database technology to store the attributes, but need GIS-specific data structures to store the geometry.

The basic structures (geometric primitives) available for describing the represented geometry of a feature depend on the Dimension of the feature (or at least, the dimension chosen to represent it). Traditionally, there are three geometric primitives: points, lines, and polygons.[1]

  • Point (0-dimensional): represented by a single coordinate: p = (x,y). In some vector formats, multipoint features are allowed, in which a single feature consists of multiple coordinates; three-dimensional points (x,y,z) are another common feature.
  • Line (1-dimensional): represented by an ordered list of points (known in this context as |vertices): l = [p1, p2, p3, ..., pn]. Some vector formats are able to store more complex linear structures, such as multilines (single features consisting of multiple line segments) and various forms of parametric curves.
  • Polygon (2-dimensional): only the boundary is represented (the software is expected to accurate interpolate all of the interior points) as a line that closes: P = [p1, p2, p3, ..., pn, p1]. Some vector formats can store structures such as multi-part polygons (e.g., the State of Hawaii).
  • Volume (3-dimensional): there are actually several strategies for representing three-dimensional shapes, but the most common is a polyface, consisting of a set of three-dimensional polygons, that together form a surface (the software is expected to interpolate the interior of the solid).

To be meaningful, the coordinates used in these geometric primitives are usually measured within a standard coordinate system, such as the Geographic coordinate system (GCS) or Universal Transverse Mercator (UTM).

The attributes of the features to be represented are typically structured using the concepts of the relational database, in which a row (or tuple) represents the full description of an individual feature, consisting of columns for each attribute type, and all of the rows for a given topic are grouped together in a table.

Vector Data Formats

The Vector Data Model is only a general strategy for representing objects; there are dozens of physical data structures (file format) that organize vector geometry and attributes in different ways, with unique capabilities. For example, some formats are able to store topology while others cannot; some formats store the geometry and attributes in separate files, while others (typically called spatial databases) store the geometry as one column in the attribute table.

Examples include:

Non-GIS graphical software, including CAD and drawing programs such as Adobe Illustrator, are also based on the vector model.

Advantages of Vector Data

Vector data can can better represent topographic features than the raster data model. Vector data models can represent all types of features with accuracy. Points, lines, and polygons, are accurate when defining the location and size of all topographic features. The vector data model is often the data model of choice for GIS because it can contain information about topology which underlies a large number of GIS operations.[2]

Limitations of Vector Data

Compared to a raster data model, vector requires more time and technology to input the data. Errors are likely to occur when inputting vector data.[3]


  1. Bolstad, Paul (2008). GIS Fundamentals. White Bear Lake, Minnesota: Eider Press. 978-0-9717647-2-9.
  2. Ibid.
  3. Lo, C.P. and Albert K. W. Yeung. Concepts and Techniques of Geographic Information Systems. Prentice Hall. 2002, 84-91.

See Also