GIS Files 2: Geographical data

2.3: Position matters (2)

Methods of referencing data

There are two fundamental methods of referencing spatial data:

The first – known as the geo-centric system – uses a 3-D coordinate system with the centre of the earth acting as the origin of the three axes. This method is universally used in scientific applications, but it is user unfriendly when applied to points on the earth's surface. This is because the axes are made parallel to the spin axes of the earth rather than north (or some other arbitrary direction). The system can be expressed in two ways: either as a 3-D Cartesian coordinate of the form x,y,z; or as longitude (), latitude () and height (H) above a known reference surface – the ellipsoid (see following paragraph: The third dimension: height). It is important to note that the x and y do NOT refer to east and west or north and south.

The second, more common, system is the projection. This takes the 3-D coordinates and expresses them as a plane plus height above it. In other words, it flattens out the curved earth in a small region to a flat surface. Coordinates on the plane tell us where a particular point within the projection is being used and are measured as distances, east and north from the starting point or origin.

To avoid errors, the extent of the projection is usually limited to a small part of the earth's surface. Choice of projection is dependent on the area of the world being considered.

Different projections have different properties. Spatial data users choose the projection that provides the least distortion – in distance, direction, scale, area and so on – within the region being considered.

Bear in mind that there are dozens of different types of projection and each can have hundreds of different definitions depending on where they are used. In Great Britain the chosen projection is known as Ordnance Survey National Grid.

< 2.3: Position matters (1) | 2.3: Position matters (3) >