Geographic Information Systems and Regional Survey at the American School

Jack Davis and I will present an introduction to Regional Survey and GIS this morning at the American School.  I will talk about using GIS in the context of survey.  My seminar will be pretty basic and focus mainly on the conceptual framework behind such software as ESRI ArcGIS.  My feeling is that grasping the principles of most GIS programs (as well as the basic ideas behind such applications as databases) and understanding potential of GIS application is more valuable for the average archaeologist than developing a technical mastery of the applications.  In the last decade the number of GIS labs has proliferated on US university campuses (UND has two!), many of which are familiar in general with archaeological uses of GIS.  Consequently people with technical expertise are available to develop GIS platforms to a project's exact specifications.  The issue then is knowing exactly what GIS can do and communicating the needs of the project effectively to the technical experts (a kind of modern-day, technological, "ritual experts"). 

To this end, I am going to start with what most of us know -- relational databases -- and then describe GIS software as basically a spatial database.  After a short introduction to what a database is (the tendency among a certain kind of Classical Archaeologist is to conflate databases and spreadsheets), I'll talk about how GIS creates and queries spatial relationships between data sets.  I hope by introducing some of the basic technical vocabulary of GIS (shape file, vector/raster, DEM/DTM, georeferencing, orthorectifying) and, indeed, cartography (geodetic markers, projections, UTM/Coordinate systems) that I can demystify the process of producing GIS data for a site or a region and reading reports that rely heavily on GIS data . 

To this end, I've given them a suggested reading list with some case studies.  The seminar is optional, so I couldn't be too rigorous (but these are all good studies that use GIS data in a productive way). 

Four Case Studies

M. Given, H. Corley and L. Sollars, “Joining the Dots: Continuous Survey, Routine Practice and the Interpretation of a Cypriot Landscape,” Internet Archaeology 20; URL: http://intarch.ac.uk/journal/issue20/4/1.htm

A. Bevan and J. Conolly, ‘GIS, Archaeological Survey and Landscape Archaeology on the Island of Kythera, Greece’, JFA 29 (2004), 123-138; Stable URL: http://links.jstor.org/sici?sici=0093-4690%28200221%2F200422%2929%3A1%2F2%3C123%3AGASALA%3E2.0.CO%3B2-D

A. Bevan. The Rural Landscape of Neopalatial Kythera: a GIS perspective, JMA 15 (2002), 217-256.

Y. Lolos, B. Gourly, and D.R. Steward, "The Sikyon Survey Project: A Blueprint for Urban Survey," JMA 20 (2007), 267-296

I'll supplement the these case studies with one from the Eastern Korinthia Archaeological Survey and one from the Pyla-Koutsopetria Archaeological Project which have the added advantage of having light-duty, java-based, GIS interfaces available on line (EKAS GIS, PKAP GIS).

One thing that struck me as I was going back through my mass of GIS data from EKAS, PKAP, and various other projects is how easy it is to forget the processes that produced the data.  Fortunately, I've managed to keep decent metadata over the years and I was able to reconstruct the processes that created my current data sets.  In the past, keeping good metadata has helped me (and my collaborators) to "fix glitches (always with a smile) " and, perhaps more importantly from a conceptual standpoint, to preserve a record of the intermediate analyses that created out seemingly "stable" data sets.  Given et al. (see above) talk about "verbing" the landscape; it is vital for the archaeologists to see their work of creating the data sets as part of this verbing process.  Data doesn't simply exist, it's constantly being created by the process of archaeological analysis itself whether it be in the field or late at night with a laptop. Metadata is the narrative of data creation in our increasingly digital age and has all the fragility and ephemeral qualities of an ancient text.

This sets up the conclusion of my presentation which will deal with the issue of data preservation, archiving, and presentation.  Regional survey projects increasingly rely on GIS to produce both stable images of the landscape and interactive interfaces for analysis.  It is crucial, then, that we as archaeologists communicate our assumptions in creating these images effectively even if it remains explaining such murky concepts as "corrected for visibility" or explaining how the values are assigned to the color increments that show artifact densities (many projects use Jenks/K-Means to show densities, but this, like any form of analysis , creates breaks in the artifactual landscape that are based on statistical formulae).  Finally, the issue of creating stable archives for GIS data -- both for the "final results" of analysis (shape files, raster images, GPS point data), but also the intermediate steps that preserve a paper trail and form the evidence for the interpretative narratives embedded in the digital and archaeological metadata. The potential loss of repositories like the Arts and Humanities Data service in the UK is a reminder of how far we still have to go in creating the kind of long term storage facilities for our digital data as our libraries and archives afford paper notebooks.