Assessing the benefit of a program of geophysical surveys in the Black Sea area

Akkerman Fortress, Bilhorod-Dnistrovsky, Ukraine

A geophysical pilot study was undertaken, during August 2007, of the Akkerman fortress, in the Ukraine, by Alex Turner and Richard Haddlesey, from the University of Winchester (UK). The primary aim of this initial survey was to ascertain the effectiveness of a geophysical survey, to aid and advance the understanding of the Ottoman phase of the site.

As the geography of the area was unknown, the geophysical instrument chosen for this initial survey was the resistivity meter, as the nature of this instrument offers the highest probability of results under these conditions. A vital requirement of such a survey is a small degree of ground moisture, as this provides a path for the electrical current to flow, from one probe on the meter, to the other. Although the aridity of the southern coast of the Ukraine would normally have an adverse effect upon the instrument readings, a thunderstorm during the initial stages provided the required ground conditions. Toward the end of the survey, this had virtually dried out, making viable results more difficult to achieve.

Due to the dry terrain and compressed earth, from many years of tourist access, it was decided to carry out a brief survey, employing 1m spacing over a 20m grid. This was purely to determine whether it was possible to collect any meaningful data (fig. 1 illustrates the result). This particular grid was chosen after walking the site, as it seemed to offer the best possibility of achieving good results. This was due to a visible inserted roof line and remains of a wall; the only contemporary structural evidence not yet under archaeological investigation. The resulting grid data was downloaded onto a laptop and processed in Geoplot 3. A jpeg image was then exported for geo-referencing in the Geographical Information System (GIS) program, ESRI ArcView 9.2 (fig 2). The image was also downloaded onto a Nokia N80 mobile phone to facilitate the comparison of data analysis, with on-site evidence, in real time. The results were very encouraging and a further ten grids were surveyed at a higher resolution of 0.5m intervals in 20m grids, rendering higher quality results.

Official Akkerman Fortress Project Home Page

Following the Akkerman survey, an expedition to three other fortresses in the Southern Ukraine was undertaken. No geophysical surveys were carried out during the expedition, but their plausibility was assessed by Richard Haddlesey, one of the geophysics team. The following is a report on this expedition.

Kerch Fortress, Crimea, Ukraine (20/08/07)
There is good potential at this site, for a geophysical survey to shed light on the interior of the fortress. I would suggest that both resistivity and magnetometery and, possibly, ground penetrating radar (GPR), would yield good results, though the jagged undulations of the site might make a GPR survey difficult. As with Akkerman, the ground is very dry during the summer months and I would suggest that an earlier Easter date would provide better environmental conditions for the survey. The site is situated on top of an exposed hill, covered in rough scrub approximately 0.6m high. This conceals numerous holes and jagged stones, potentially creating a hazardous working environment. It would be safer if the scrub could be cleared, prior to any survey. Alternatively, I suggest that protective clothing should be provided, such as safety boots and protective gaiters and teams should operate in groups of two or three.
Very little of this fortress survives beyond the main external wall and, therefore, a geophysical survey would provide a valuable insight as to what remains of the interior. Large areas of this site could be surveyed (c 100mx100m), potentially yielding positive results, providing that the ground has adequate moisture. Because of the rough terrain, I would estimate an average work rate of four to five 20m grids, in a straight line, per day; less, if the grids need to be set out along differing alignments.

Arabat, on the Crimean Spit, Southern Ukraine (21/08/07)
A straightforward survey could be problematic at this site. As with the above fortress, only the outer walls remain. However, the interior has experienced recent burning and, in some areas, is covered in deep deposits of sand and small shells. The resulting lack of vegetation has revealed more of the site and, as only the top layer was burnt, it will not adversely affect the instrument readings. What would hinder a survey is the site aridity and depth of the sand, though remains of some buildings can be traced in the topography and structural remains are visible in some areas. The survey methods I would recommend are, as before, resistivity, magnetometery and GPR but, in conjunction with GPS. Due to the terrain, large grid areas may prove difficult and, therefore, I would suggest that only three to four 20m grids would be possible per day. Again, the ground is very dry and I would advise that the survey be carried out in wetter periods. From an archaeological perspective, the exterior of the site could also yield some very interesting data, based on the adjacent topography and previous experience of the author.

Dyke Bastion, Crimea, Southern Ukraine (21/08/07)
This site appears to be the remains of a single defensive bastion, built at the end of a long dyke, over-looking the sea. To the south of the bastion is a possible entrance, cut into the dyke to provide access. The topography and visible earthworks suggest that some form of building once stood there and would prove an interesting area to conduct a resistivity survey, conditions permitting. As there is little archaeological knowledge of this area, some small test pits would assist in further interpretation.

Dyke Fortress, Crimea, Southern Ukraine (21/08/07)
This fortress is earlier in date than the others visited and shows little evidence of stone-built fortifications. There appears to be an earthen ditch and bank construction, with possible mud and earth, or wooden structures, in the interior. For this reason I would recommend the use of resistivity and GPR to locate any other ditches and possible stake, or postholes. The lack of stone-built structures and, some evidence of archaeological stratigraphy, requires a limited programme of excavation, in order to verify any anomalies suggested by the survey data.
The site was very dry and the scrub had been burnt, providing an ideal surface to survey, in sufficiently damp conditions. The majority of the expansive interior was flat and I estimate that four to five 20m grids could be undertaken per day