Using ArcGIS to Create 'Living Documents' with Archaeological Data:
A Case Study from Svalbard, Norway

The role of GIS in archaeology began more than twenty years ago with pioneering work by Kenneth L. Kvamme (University of Arkansas) who used the technology to develop predictive modeling. Since then, it's uses have continued to expand, increasingly so in recent years. Discussions of the basic structure of GIS have become common place in articles, theses, and books as have high-level considerations about its uses (i.e. viewshed and least-cost pathway analysis). However, the actual ways in which the data is structured remains a less explored topic.

Normally, the way in which data for all GIS projects is restricted by the "point, line, polygon" structure of GIS, meaning that features are stored in files separated by these divisions and what the file is meant to represent (i.e. artifacts, a wall, boundaries, etc.). This result is a jumble of files, called shapefiles in ESRI parlance. These shapefiles, while represented as one file in the GIS itself, are actually made up of between three and nine system files on your computer's hard drive.

Thankfully, ESRI designed and released a new method for structuring GIS data, the geodatabase. Incorporated in versions 8.x of ArcGIS, the geodatabase allows the structuring of data in the GIS in similar ways to that of a traditional text database. This has numerous advantages for the archaeologist. Point, line, and polygon files can now be combined into what are called 'feature datasets' and that behave as one file. Second, the geodatabase allows for the definition of attribute data which creates drop-down menus in the tabular fields stored within the shapefile/feature datasets. Among the other benefits is one very simple, yet amazing, aspect of the geodatabase structure. Instead of storing huge numbers of files on your system's hard drive, an entire field season's worth of archaeological data can be stored in ONE file. The geodatabase is one file, which contains any amount of data you like, all in one easy to access file.

While the above is an admittedly short introduction, I will be expanding this website to document the entire data creation process from field work through completed GIS project in the coming months. I have included two links below about the project thus far. The left-hand link opens a pop-up window that acts as a mini GIS data viewer and showcases some of the interesting features of the GIS (you may need to manually re-size the window a bit). The right-hand link takes you to a flowchart showing the general data processing steps used in the creation of the Svalbard GIS, or see a step-by-step explanation here.

A few compelling comments. The GIS created using data collected during the fieldwork on Svalbard in 2004 originally contained more than 1,000 system files (several dozen shapefiles and numerous background images). Now it consists of one geodatabase and about 12 system files for background images! Also, the structure of the geodatabase behaves just like a regular database, allowing the selection of records based on attributes and the creation of reports from these database queries. Also, features can be selected in a visual environment and reports created! All-in-all, its an amazing tool that is relatively simple to set up and creates an updatable database for future uses.

View my complete MS thesis in PDF format here

Here is a link to a more in-depth article exploring these concepts:
A Sample Geodatabase Structure for Managing Archaeological Data and Resources with ArcGIS. Technical Briefs in Historical Archaeology 2:12-23 (2007). More than 150 archaeological agencies, departments, and individuals have downloaded this article and the accompanying files discussed!

Finally, see the project website for the International 2004 Svalbard Research and Training Program