Methods

The entire spring season was devoted to electrical resistivity survey (Fig. 10) with a Geoscan RM15 machine. Data was processed using Geoplot while image enhancement was done with Surfer software (Fig. 11). The memory in the RM15 permits the survey of up to a maximum of nine 20m by 20m grids per day with a sample density of readings every 0.50m at 0.50m traverse intervals. A combination of inclement weather, areas of difficult terrain and hardware problems reduced the average rate of coverage to slightly more than seven grids per day. A total of 175 grids, totalling 70,000m2 or seven hectares, were surveyed in the course of 24 working days.

Aims

In spring, when the ground is still wet, conditions are ideal for resistivity survey. There were three aims for 2005 (Figs 10 and 11):
1. To link the two main areas of resistivity survey done in previous seasons;
2. To survey one particularly large rectangular compound where the boundaries were visible on balloon photographs as well as on the ground, but where magnetometer survey does not produce clear images of internal structures (Fig. 12, NE corner);
3. To survey the urban block and its environs in the northern sector of the city where test trenches TT15, TT20 and TT21 were dug in 1996 and 1998.

In the descriptions and interpretations that follow, use has been made of all methods of survey carried out at Kerkenes, i.e. balloon photography, resistivity and gradiometer survey, and verification on the ground. Where relevant, results from test excavations in 1996 and 1998 have also been taken into account in this report. Coordinates, where given, always refer to the south-western corner of a grid, structure or feature.

Test excavations confirmed that visible differences in results between both methods of geophysical survey are indications of different buried surfaces, notably burnt clay floors of roofed structures and external stone pavements. Further, comparison of the two sets of data make apparent some of the ways in which burning has had an effect on the magnetism.

Central Area

The Central Area (Fig. 13) has been the main target of the resistivity survey since 2001 when Harry von der Osten obtained images of outstanding quality. During the 2005 survey, the area east of the Büyük Göl, surveyed in 2002, was expanded on either side of the seasonal stream flowing westwards towards the main watercourse and linked to the main survey.

The Urban Fabric

The maps of the Central Area obtained from the resistivity data have revealed the urban fabric of the Iron Age city with remarkable clarity (Figs 14 and 15). The size and shape of structures, even when not visible on the ground, can be determined from the images and digitised for analysis with GIS software. Studies related to building typology, transportation and water management are ongoing at METU and the CAMEL Laboratory in Chicago.

The Large Urban Compound

Advantage was taken of very damp conditions at the start of May to investigate one particularly large urban block on the rising eastern margins of the Central Area, located by its southern corner at E1088, N2015 (Figs 11, 12 and 15). The boundaries could be seen on the surface and it was evident that it was bounded by a broad street on the NE side and, most probably, by streets at either end. At the outset it was not certain if the level walled area adjacent to the NW edge of the block was an extension of the block itself or a street. The main objective was to establish whether or not this block contained a large two-roomed hall. The excellent results (Fig. 15) showed that at Kerkenes considerable potential for resistivity survey exists, even over stony areas, when there is sufficient ground moisture.

The entire block measures about 63m NW-SE by approximately 50m NE-SW, making an area of c. 3,150m2. The extent to which the terrain has been artificially levelled is uncertain. There is no visible entrance into the block on the NW, NE or SE sides. Thus the most probable location for an entrance is towards the NW end of the SW boundary wall where there are two circular piles of stone. These stone features, together with a number of smaller but similar features in the vicinity, postdate the destruction of the Iron Age city. Some of these stone piles would appear to mark tumuli, and in this they resemble the tumuli on the high southern ridge, some of which contain stone cists and almost all of which have been robbed. Other piles of stone seem to be associated with the activities of shepherds. Within the block there is no single building which stands out as having been of some special importance. Rather, the block appears to be subdivided into several subunits which, while discrete, seem to have shared a common entrance into the block.

The North End

There were three objectives to the resistivity survey at the north end (Fig. 16). Of these, the first was to obtain resistivity imagery (Fig. 17 bottom) of the large two-roomed hall and associated structures that had been investigated by excavation in 1996 (trench TT15) and 1998 (trenches TT20 and TT21). This was in order to provide the maximum amount of information concerning subsurface remains with respect to the possibility of conducting further excavation in this area in the future. In addition, it was of particular interest to see if resistivity survey would locate the twin rows of (uncut) granite bases for large wooden columns, the existence of which was demonstrated by excavation but did not show on the gradiometer shade plot (Fig.17 top). Unfortunately, neither this recent survey nor previous geophysical survey of similar large halls at other locations within the city has yielded such evidence and, therefore, the reconstruction of rows of columns has been necessarily conjectural.

A second objective was to see if there were similar large halls in what appeared, on the basis of gradiometer survey and balloon photography, to be urban blocks divided by streets. It will take several additional seasons to cover the entire area of interest since at this welldrained north end the ground dries faster than in lower central parts of the city, thereby restricting the length of the survey season.

The third objective was to examine the shallow grass-covered valley where previous work had suggested the existence of sophisticated water management systems followed by the encroachment of relatively slight structures onto previously wet and marshy areas.

The Hall

The hall and anteroom (Figs 16 and 17) were shown to be 24 metres in width, needing beams or trusses to span 8 metres between two rows of columns. The building was rectangular and the hall, unusually, seems to have been square. These proportions might partly reflect the restrictions imposed by the sloping terrain and outcropping bedrock. It was also revealed that there was a wide central doorway, presumably for wooden double doors, in the wall between the anteroom and the main hall. However, in the front wall of the building no doorway can be discerned on either the magnetic or the resistivity map.

Water Management and Encroachment

The 2005 resistivity survey has provided a little additional clarity while confirming the general picture of features that are very probably to be associated with water management. The new image makes certain that there was not a reservoir directly in front of the block just described, and perhaps provides further evidence in support of the idea that stone channels were used to drain water and direct it into pools and a large reservoir to the WSW. Located at E820, N2440 one pool, if such it is, has parallel sides and rounded ends. Orientated east-west it is 35m long and 8m wide. Verification on the ground confirms that this was not a standing structure. On the other hand, a sinuous feature showed by test excavation to be made of stone, appears to run into it and makes a sharp turn to the SSE at its E end. Features stretching NWSE across the centre of this enclosure are surely buildings of Iron Age date.