Skip to main content


The nature of spatial data relationships are important to understand within the context of GIS. In particular, the relationship between geographic features is a complex problem in which we are far from understanding in its entirety. This is of concern since the primary role of GIS is the manipulation and analysis of large quantities of spatial data. To date, the accepted theoretical solution is to topologically structure spatial data.

It is believed that a topologic data model best reflects the geography of the real world and provides an effective mathematical foundation for encoding spatial relationships, providing a data model for manipulating and analyzing vector based data.

Most GIS software segregate spatial and attribute data into separate data management systems. Most frequently, the topological or raster structure is used to store the spatial data, while the relational database structure is used to store the attribute data. Data from both structures are linked together for use through unique identification numbers, e.g. feature labels and DBMS primary keys. This coupling of spatial features with an attribute record is usually maintained by an internal number assigned by the GIS software. A label is required so the user can load the appropriate attribute record for a given geographic feature. Most often a single attribute record is automatically created by the GIS software once a clean topological structure is properly generated. This attribute record normally contains the internal number for the feature, the user's label identifier, the area of the feature, and the perimeter of the feature. Linear features have the length of the feature defined instead of the area.