The Visibility Issue

Surface visibility is one of the main methodological challenges in this region. This is due to the mountainous character of the Cide and Şenpazar landscapes, dense vegetation cover, and relative scarcity of cultivated fields. In many parts of our survey region, from forests to grasslands, surface visibility is almost nil.

Initially we had envisaged an intensive field walking strategy modelled on successful Mediterranean surveys such as the Boeotia Survey Project, the Kythera Island Project and Antikythera Survey Project. The main methodology of these projects is intensive tract walking, followed by even more intensive collection of materials at potential sites.

We tested the first stage of this approach – 5 walkers spaced 20 metres apart, walking 100-metre tracts – in selected, mostly low gradient, areas in the first days of the 2009 season. The first difficulty we encountered was the identification of suitable tract walking areas, due to the scarcity of large field systems. More importantly, visibility in those areas that we were able to cover was very poor due to vegetation cover (lush green grass), and overall recorded sherd densities were extremely low, ie. the sherd count was zero in eight out of ten tracts.

Photograph of extreme low visibility of ground surface typical of many area

Extreme low visibility of ground surface typical of many area

That these low pottery counts were related to visibility rather than the absence of archaeological remains became clear when we focused on areas with high ground visibility, such as small ploughed fields and eroded patches. We therefore concluded that standard tract-walking was a methodology not suited for most landscapes within our survey area.

Intensive Unit-Walking

In 2009 we began to develop an intensive methodology better suited to local topographic conditions and ground visibility, which we call unit-walking. We are still in the process of experimenting with aspects of this method such as the ideal number of walkers in each unit and its suitability for different types of terrain. The amount of ground covered and the range and density of sites discovered using this method, however, provide a good indication of its effectiveness (see Results).
Roaming field walking

Roaming field walking

Map of unit grids

Unit grids on a map

The first step in this methodology is the selection of a suitably large area with higher than average ground visibility – usually an area affected by erosion. Onto this area a grid of 50 x 50m squares, or ‘field units’ is superimposed on our GIS map. The corner points of units are fed into GPS receivers and located and marked in the field, as the team is walking. Alternatively, in some cases, existing field boundaries are used for defining units, especially where field are separated by substantial fences or brushes or where landuse varies considerably for adjacent fields.

Information recorded for each unit includes data on general topography, slope, vegetation, and type and degree of erosion. Ground visibility is recorded for each walker individually. Units are investigated ideally by five walkers, who are free to roam within parallel bands, inspecting the entire strip but focusing on areas with good visibility. Walker numbers for each unit, however may differ depending on factors such as accessibility of fields and gardens. Walkers use counters to record sherd densities and collect diagnostic or unusual ceramics, lithics and small finds. While allowing greater flexibility for dealing with low visibility environments, this methodology retains the possibility of analysing walker-specific data.


During the 2010 season we used gridding as a follow-up methodology in order to investigate areas of special interest such as the fields below Okçular Kale more intensively. We adapted this method from the Antikythera Survey Project. Choosing a base point, which is recorded using a GPS receiver, a grid consisting of 5 x 5m or 10 x 10m squares is flagged out with tape measures. In each standard grid, one walker spends a fixed period of time (eg. 5 minutes) counting all pottery and collecting diagnostic ceramics, lithics and any other finds. This is followed by an additional fixed time ‘vacuum’: in a 1m radius all sherds and other finds are collected as a representative sample of each grid-square.

Tapping into Local Knowledge (extensive survey)

A complementary method for improving our understanding of Cide’s archaeological record is tapping into local knowledge. Certain local residents have a wealth of knowledge about their environment and have taken us on extended hikes into the mountains, guiding us to hidden and difficult to access caves. Other sites, although close to villages or roads, are hidden by shrub and forests, and we would not have been able to locate them without their help.

Recai Yilmaz shows us one of many caves

Recai Yilmaz shows us one of many caves

Generally, we are usually shown two types of sites by local residents: caves and castles. Many of these have proven to be of archaeological importance and had not been recorded previously by archaeologists. However, it is also clear that this methodology is biased towards ‘spectacular’ and special purpose sites. Artefact scatters are less likely to be found in this way.

As might be expected, our best results come from areas in which we are able to place the caves and castles shown to us by local residents into a wider context of observed patterns of landscape use and artefact scatters recorded by intensive survey.

Read the current project Results.

Comments are closed.