522 
Aydin, Toprak, Baturayoglu 
By combining these three maps, a result map of cumulative scores was produced in TNTMips. In order to produce this zonation map, 
a relative score system was developed. Those areas where the elevation-value is greater than 1290m and less than 1320m, the slope- 
value is greater than 4 degrees and less than 10 degrees and the aspect-value is greater than 120 and less than 160 or greater than 40 
and less than 70 were all counted as scoring 1. In the resulting map, all the areas that have a score of 3 are the most preferred areas in 
relation to elevation, slope and aspect; these are shown in black. The moderately preferred areas, which have a score of 2, are shown 
in white. The least preferred areas, which have a score of 1, are shown in gray. (Fig. 12) 
Figure 12: The preference zonation of the study area with respect to the cumulative scores of elevation, slope and aspect. 
4. CONCLUSIONS AND FUTURE RESEARCH DESIGN 
Results of interpretation and evaluation of this data can yield the following conclusions: Length, elevation, direction and slope values 
of sub-surface features are determined from digitised maps using GPS data. Comparison of the histograms of these parameters with 
the landscape has shown that the features are preferably located at the following intervals: at elevations of 1290-1320m, at slopes of 
4- 10 degrees, and at NW-SE and NE-SW directions. The lengths of the features range from l-73m with a maximum concentration of 
5- 6m. Density map of the walls is prepared using the lengths of the features, maximum density being 800m/2500msq in the northern 
central part of the study area. 
Elevation, slope and aspect histograms of both sub-surface features and the landscape are used to define the threshold values for the 
“undesired” parts of the area. Analyses suggest that the central part of the area is preferred because of the elevation parameter and the 
eastern part of the area is preferred due to the aspect parameter. Slope, on the other hand, does not indicate a certain preferred area 
but shows continuous variation within the study area. 
With this type of methodological approach, large-scaled sites can be surveyed much more effectively in terms of both generating a 
site plan and revealing the relationship between the occupation and the landscape. Since the last decade, GIS is being used in 
archaeology on a regional level, for instance for detecting settlement distribution or as a spatial database for cultural heritage 
management, but there has been no previous applications in this scale at a single site. On the other hand, presently, GIS has found a 
wide application especially in city planning and hence, it may be used at ancient cities to reveal the urban dynamics such as water 
resource management and urban zoning. 
The next step should be to produce a complete geomagnetic map of the whole site, which is estimated to be finished in the 2003 field 
season except for the Kale. After that point the same types of analyses can be applied to the whole site and the results will be 
sufficient. Moreover, other databases can be integrated into this study, such as balloon photographs and the ground-truthing database, 
so that the sub-surface features that were digitised from the geomagnetic map can be correlated with both of them.