Read Edit History

Processing Survey Data

Processing Survey Data


Full text

Survey methodology is not exclusively confined to fieldwork techniques and strategies. The processing of raw survey data, as well as the analytical and interpretive steps between fieldwork and final publication, are all part of the methodological discourse of the discipline. As this section cannot treat all issues in full detail, it will concentrate on providing an overview of the most important steps and possibilities in the processing and analysis of survey data. The wiki pages on the background of Geographical Information Systems (GIS) explains the analytical potential of GIS for survey archaeology in greater detail.

One of the first, and perhaps most elementary aspects of processing the survey data is the pottery analysis . Most survey/collection strategies produce an enormous amount of finds. All materials have to be counted, weighed and classified according to a framework of local/regional pottery classes. A selection of diagnostic finds (i.e. rims, bases, handles, etc.) can be drawn, described and compared with published pottery sequences from nearby stratigraphical excavations. This provides the survey material with relative dates, and can offer crucial evidence for the (functional) interpretation of sites.

A great risk, however, lies in the problem of periodization. Surface artefact scatters can rarely be dated more precisely than to within broad periods or even centuries. This does not necessarily mean that a site has been frequented during this entire period of time. Survey maps tend to be plotted per, again relatively broad, period, producing an overview of sites which might have been used contemporaneously, but might also have been subsequent to each other.

Another important step in the processing of survey data is the digitization of the recorded field data. At present most survey projects use PDAs with mobile GIS-based applications to record data and spatial references digitally directly in the field. Therefore, it is possible to start the processing and analysis of this data immediately. The backgrounds and functionality of GIS are explained in greater detail in the technical wiki. To illustrate the analytical potential of GIS within survey archaeology, three of its frequently used applications are described in this section.

To begin with, some of the statistical methods of quantification used to analyse survey data will be briefly explained. The most important preparations for this process, namely counting, weighing and classifying the archaeological materials, are executed during the pottery analysis. This data can, however, be treated in a number of ways, none of them necessarily better or worse, depending heavily on the research design of a project (i.e. its objectives, research questions, field-walking techniques, collection strategies, etc.). Almost invariably, the first analytical step is the calculation of sherd density figures so as to be able to plot the quantities of archaeological material on a map. Often, based on the classification of pottery wares, these quantities are divided into assemblages dating to different chronological periods, creating maps of the density of archaeological materials for each period. Hence, essentially quantitative maps, based on a qualitative pre-selection.

Some survey projects use the procedure of visibility correction to weigh sherd density figures based on the recorded visibility conditions. In other words, where survey results might have been obscured by bad visibility conditions, a relatively smaller amount of encountered finds is potentially biased and this may be corrected by being multiplied, using a predetermined formula. Critics of this correction method have stated that various revisit experiments have proved that the relationship between visibility conditions and sherd density is not fixed (i.e. quantifiable), and that visibility is only one of a wide array of possible factors which could bias survey results.

The second example is viewshed analysis (sometimes referred to as line-of-sight analysis), a GIS-based method to examine the visibility of different points (e.g. archaeological sites) in the survey area, which might have been a crucial factor for the lay-out and experience of the archaeological landscape in antiquity. The digitized survey data is plotted on a three-dimensional map of the survey area in a GIS-environment, using a digital elevation model (DEM). Subsequently, the software can compute all parts of the landscape which are visible from any given point in the landscape.

Another, more elaborate GIS-based application is cost surface analysis. This method can calculate the effort (‘friction’) it takes to move throughout a landscape, again using a three-dimensional digital elevation model (DEM), to create models of least-cost pathways and to analyse connectivity between different points (e.g. archaeological sites) in the survey area.

Bibliography and further reading

Carver, M., 1990, ‘Digging for data: archaeological approaches to data definition, acquisition and analysis’, in: Francovich, R. & D. Manacorda (eds.), Lo scavo archeologico: dalla diagnosi all’edizione, Firenze, pp. 45-120.

Conolly, J. & M. Lake, 2006, Geographical information systems in archaeology (Cambridge manuals in archaeology0, Cambridge.

Fletcher M. & G. Lock, 2005, Digging Numbers: Elementary statistics for archaeologists, Oxford.

Gillings, M., Mattingly, D & J. Van Dalen, 1999, The Archaeology of Mediterranean Landscapes 3: Geographical Information Systems and Landscape Archaeology, Oxford.

Millet, M., 2000, ‘Dating, quantifying and utilizing pottery assemblages from surface survey, in: Francovich, R., Patterson, H. & G. Barker (eds.), The Archaeology of Mediterranean Landscapes 5: Extracting Meaning from Ploughsoil Assemblages, Oxford, pp. 53-59.

Shennan, S., 1997, Quantifying archaeology, Edinburgh.

Research topics: Survey Methodology