GIS - Geographic Information System (Archaeology)

It has been proposed that rather than being classified as a ‘tool’, GIS might more accurately be labelled a sub-discipline in its own right, complete with its own competing methodologies and camps endorsing one approach over another in relation to the vast amount of functionality that GIS encompasses. Conolly and Lake (Conolly,J., Lake, M.(2006), Geographical Information Systems in Archaeology, Cambridge University Press) list five categories of activity that GIS systems can address and it is apparent that the aggregation of all of these activities represents a significant proportion all of the ICT related activities that archaeology is likely to provoke.

  • Data acquisition
  • Spatial data management
  • Database management
  • Spatial data visualization
  • Spatial data analysis

Bearing in mind the very broad functionality that GIS brings to archaeological practice, it may be useful to concentrate on the signature function that demarcates these systems from, for example, CAD and visualization packages. GIS is designed to allows users to describe some form of entity (situated in a landscape) in terms of its geospatial coordinates and to then make connections between that very explicit instance of data and any other information that might be pertinent to the description or analysis of that entity. This related information can take the form of text, images, statistics, graphs, multi-media; in fact anything that can be stored or referenced by a database. This enables GIS to act as both a visualization tool (displaying the database information spatially) and an analysis tool (displaying the spatial information quantitatively). Whilst it could be argued that some CAD packages can also be used to anchor displayed objects to real-world geospatial co-ordinates, their principle function is to enable users to produce very sophisticated graphical representations and the analysis of geospatial data is beyond the remit and functionality of such systems.

In the past, GIS systems came in two mutually exclusive ‘flavours’ and the different types, raster and vector-based systems, meant that users had to determine which approach more closely suited their requirements. Current GIS systems can deal with both types of data but even so, it is still of practical use to know when the use of raster images (also known as bitmaps) might be more appropriate than their ‘resolution independent’ vector counterparts, and vice-versa. (See the TASI website for a detailed explanation of various properties of the two formats http://www.tasi.ac.uk/advice/creating/fformat.html).

The tables in fig.2 represent some sample database information that might relate to the areas shown on the images. The coordinates would reference either a national grid system (e.g. the British National Grid ), the Universal Transverse Mercator, or more usually, a latitude and longitude reference. Once this link is established, information can be analysed in layers featuring other representations of the same spatial areas, linking to alternative associated data. At that point, more complex queries are possible such as, ‘are features on one layer near to features on another?’ Information about the location of specific items could, for instance, be analysed in terms of: the altitude of the terrain; their distance from surface water; their proximity to evidence of buildings, and so forth.

Fig. 2 Vector and raster representationsFig. 2 Vector and raster representations

Whilst commercial GIS packages such as ArcGIS are universally popular and have very advanced functionality, there are also free and open source options available. GRASS (Geographic Analysis Support System) GIS is a very popular open source package and is widely used by academic and commercial organisations. It is a fully functional package featuring tools for geospatial data management and analysis, image processing, graphics/maps production, spatial modelling and visualization and is supported by a community of users who contribute information to the GRASS wiki.

Syndicate content