Looking beneath the surface: Geophysical surveys at Göbekli Tepe

Since recently there has been renewed interest in the results of geophysical survey undertaken at Göbekli Tepe in the years 2003, 2006, 2007, and 2012 we put together this short overview on these works and their results – which helped to understand the extension of the Neolithic site and its monuments even in those parts of the tell not yet excavated.

Without a doubt, the most widely known features of the Göbekli Tepe archaeological site are the monumental buildings, which, due to their ‘outstanding universal value’, were recently inscribed on the UNESCO World Heritage List. Notably, since the very early years of excavations, one of the most pressing questions has been whether these structures, with their characteristic T-pillars, were restricted to certain parts of the mound (where revealed through excavation and suggesting a unique agglomeration of this particular building type) or whether they existed all over the tell.

Archaeological survey methods have changed significantly over the last years. One innovation which has dramatically changed the way field archaeologists work are ground-based physical sensing techniques (for a short introduction into this technology and its application see, e.g. here [external link]). This technology provides us with images of possible archaeological features beneath the surface without even taking a shovel to hand. In 2003, a geophysical survey was undertaken at Göbekli Tepe with the help of GGH – Solutions in Geoscience GmbH. In a first step, large parts of the tell were subjected to extensive magnetic prospection, and later selected areas were studied using georadar and geoelectric tomography.

As already noted by Klaus Schmidt in his 2003 field report which was published the same year (Schmidt 2003, 5), first results already provided a better understanding of the site and served to confirm earlier observations:

“More than ten large enclosures could be located in the geomagnetic map, and some more can be expected. As four enclosures are under excavation (Anlage A-D), in total a minimum of 20 enclosures seem to exist inside the mound of Göbekli Tepe. At every enclosure a number of 12 megalithic pillars can be expected. So, in total more than 200 pillars can be calculated.”

Subsequent surveys which were undertaken in the years 2006, 2007, and 2012 also confirmed the earlier predictions based on archaeological surface investigations, i.e. that the monumental circular enclosures were not restricted to a specific part of the mound but existed all over the tell (cf. Dietrich et al. 2012, 675).

Göbekli Tepe 2014_mit Geophysik

Göbekli Tepe excavations and surveys by ground-penetrating radar (Plan: Th. Götzelt, DAI).

Survey-work also provided a useful tool in the planning of field research strategies, with operations focused in areas of particular interest as indicated by survey results. From 2007, excavations were also conducted in other parts of the site where more monumental structures were suspected, e.g. in the Northwest-Hollow. Here, georadar results showed a large, cloverleaf-shaped accumulation comprising of what appeared to be several circular structures. It is in this part of the site that excavations led to the discovery of Enclosure HAlthough fieldwork is still not completed in this part of the site, the current state of excavation already confirms the geophysical-geoelectric results (Dietrich et al. 2016, 56).

Enclosure H_geo

Geomagnetic survey results in the NW depression, excavation areas superimposed. (Geomagnetics: GGH- Solutions in Geoscience, Plan: J. Notroff, DAI)

Enclosure H

Aerial of Enclosure H at current state of excavations, including pillar numbers. (Photos: N. Becker, compilation: J. Notroff, DAI)


References and further reading:

O. Dietrich, M. Heun, J. Notroff, K. Schmidt, M. Zarnkow, The Role of Cult and feasting in the Emergence of Neolithic Communities. New Evidence from Göbekli Tepe, South-eastern Turkey, Antiquity 86, 2012, 674-695. [external link]

O. Dietrich, J. Notroff, L. Clare, Chr. Hübner, Ç. Köksal-Schmidt, K. Schmidt, Göbekli Tepe, Anlage H. Ein Vorbericht beim Ausgrabungsstand von 2014, in: Ü. Yalcin (ed.) Anatolian Metal VII – Anatolien und seine Nachbarn vor 10.000 Jahren / Anatolia and Neighbours 10.000 years ago. Der Anschnitt Beiheft 31 (Bochum 2016), 53-69.

K. Schmidt, The 2003 Campaign at Göbekli Tepe (Southeastern Turkey), Neo-Lithics 2/2003, 3-8. [external link]

15 thoughts on “Looking beneath the surface: Geophysical surveys at Göbekli Tepe

  1. Pingback: Looking beneath the surface: Geophysical surveys at Göbekli TepeBeware Oh wanderer the road travels too.Don Michael Hudson

  2. it was my understanding a couple of years ago that you did not believe the geophysical results really showed as many as 20 unexcavated enclosures. Pls clarify.



    • Jens is *citing* a 2003 report by Klaus Schmidt saying that more than 10 enclosures are visible in the geophysical map done in this year. In addition to 4 excavated buildings, Klaus was estimating a *total* of 20 enclosures in the whole tell at that moment. There is no reference to 20 unexcavated enclosures in the text.
      That said, with the experience from excavation in the northwestern drepression, it seems that only very large structures show clearly in the survey maps. There are 10+ buildings clearly visible in the map, but more could be hidden in the tell.


  3. ok, tks. they don’t seem very clear in the geophysical map to me but presumably advanced ground-penetration imagery is continuously improving. When will you publish new pics of the entire site (especially the parts you might never excavate)?


    • The images are pretty good actually, it´s more that site structure may limit results. We have published the general overview map, which includes most of the site, excluding some areas, where a survey was not possible due to technical issues, no sediment cover or obstacles on the surface. There is nothing more to publish at the moment.


  4. “At every enclosure a number of 12 megalithic pillars can be expected. So, in total more than 200 pillars can be calculated.”
    Isn’t that 13 pillars actually, 2 central and 11 in the enclosure around them ? That gives a total of 260. There are 266 pillars in Karahan, if I recall correctly, which is peculiar, and there were 8 sites with T-pillars known when I last checked.
    1. How many sites with T-pillars are known today ?

    2. How many T-pillars per site are expected / known ?

    3. Are there any excavations planned for Karahan or ongoing there ?

    4. In an interview (Hancock, Magicians of the Gods, chapter 1), Klaus Schmidt said that the Enclosure D was built at 9,600 BC, at the end of Younger Dryas, which ended at 9,700-9,640 BC, but I also read some other source somewhere on this site, I don’t recall where, whereby C-14 analysis claims 9,700 BC for the age. What is the current best guesstimate ?


    • Please note that the first quote by K. Schmidt comes from an older paper in 2003; excavation and research have made some progress over the last 15 years. The assumption of 12 pillars per enclosure was based on the distance and distribution of known pillars back then (at still incomplete state of excavation).

      1. See for instance here: The current distribution of sites with T-shaped pillars.

      2. Since no significant excavation work was conducted at any of the other sites yet and since we’re working at that site, we can only speak for Göbekli Tepe where at current state of excavations about 126 pillars are excavated (comprising both the older monumental enclosures as well as smaller pillars from the younger layer; not counting miniature pillars among small finds though).

      3. Not by us. That’s a decision completely up to local antiquity authorities.

      4. Regarding 14C-dates see: How old is it? Dating Göbekli Tepe. But keep in mind that these dates are coming with a certain probability range and do not at all depict a fixed moment in time, but some point in the construction and use-lives of these buildings, considering a possible ‘old wood’ effect etc. pp. – that’s why we’re actually not ‘guesstimating’, but follow multiple approaches to arrive at a certain plausibility regarding chronology.


  5. Pls can you tell me how many enclosures you can can see that are similar to or larger than enclosure D?

    It’s my understanding that you might take 150 years to complete the excavations based on the view that future generations will have better technology. Is that true – and, if it is, what makes you confident the the site won’t be destroyed by the unpredictability of war in the next few decades?


    • At least seven, some of these apparently multi-layered – so the estimation given by Klaus Schmidt seems about right.

      It’s impossible to guess how many time future excavation work will take since, naturally, one can’t predict the exact circumstances of things which haven’t taken place yet. I’m afraid I would made a rather bad fortune teller, so I sincerely beg your understanding that I’d prefer to not comment on the future.


  6. If I may, I’d like to suggest something of a small project to you.
    1. Considering that feastivites on Göbekli Tepe were held about at this particular time of the year, 2nd half of August, perhaps this is the best time for photographing the pillars in their most natural illumination conditions, on all sides, from all angles, but particularly at dawn, noon, and dusk, when shadows and colors make special effects, and coming from angles that they were coming when the pillars were young and in use. How about that ?

    How to make a good resolution photo: 2 person required, one holds a single hair before the pillar, and the other who is a photographer tries to detect it in the objective. Once it does, he holds still while the assistant goes away from the cadre. Then the picture is taken. Repeat on all sides, empty or not. Then, leave to future researchers a high resolution documentation if you wish, or study that by yourself. Cheers.

    2. Regarding the radiocarbon data, yes, that was the page that I was referring to, but this is from 2 years ago. Perhaps you are not aware of it, but in the meanwhile there were some finds (I read that then only briefly) that claim that the conversion rates from C-14 to BP dates have to be adjusted by a couple of decades due to some previously unconsidered effect, something exotic, like non-linear accumulation of C-14 molecules by species, by age of organisms and by local weather phenomena. In total I think they suggested an additional 20-50 years of uncertainty come from these particular effects that they were studying. Sorry for no links.

    Realistically, do you think that it would have been possible to build the site during the Younger Dryas, or not ?


    • 1. ‘Natural illumination’ really is anything but clear in this case; considering that the enclosures likely were subterranean and probably covered by roofs, lighting in there could’ve been rather dim. That’s why we’re following strict documentation rules for the photographs taken. But thanks for your suggestions!

      2. And that is the reason archaeology usually is not only relying on one dating method, but using different approaches mutually correcting each other.

      3. All dates so far point post the Younger Dryas.


      • ‘Natural illumination’ really is anything but clear in this case; considering that the enclosures likely were subterranean and probably covered by roofs, lighting in there could’ve been rather dim.

        ‘likely’…’probably’ – what is the exact likelihood, and how did you evaluated it ? That, considering that modern telescopes are subterranean too, for protection from elements, whereby the roof is removed for the sake of making observations.
        Also, any sane artist would carve a pillar out in the open sunlight (of August), and only once the work is completed (under natural illumination that occurs during late August) the pillar would have been used as a ‘brick in the wall’, namely stationed in position, and even if it was first integrated into the wall, can you really prove that they were so dumb to first build the roof, and than in darkness to carve the images that required high skill and intelligence to be carved in the first place ?

        Further, I do not know what your policy and procedures are for making photos, but I must say that out of those 126 pillars not a single one has been exposed to public from all sides by their photos, and regarding quality photos, you made practically less than one per year of excavation thus far. Those few good ones that do exist and are publicly available were made by Schmidt, Wagner, Becker, and Collins, mostly of pillars from the Enclosures D and H, and of smaller artifacts. For that, I give them credit.

        I find it regrettable that there isn’t a single good picture of that sheep with a net of snakes from the Enclosure A, nor a good picture of the belt of the Pillar 31. For instance, I asked you once whether there are ‘H’ or ‘I’ symbols there too, but as usual you avoided to give a clear answer on such a simple question. That is very frustrating.

        Considering that there are ~260 pillars, and that each has 6 visible sides, and the buried bottom, at a rate of one picture per year, the exploration of the site would take not 150 years, but practically the same amount of time as the construction did, almost 2 millenia. There is some epic poetry in that, I admit. If that is your goal, to recreate the feeling of centuries of time, you are progressing quite well.

        That said, I wonder if you ever checked the underside of the top (dash) part of ‘T’ on pillars for images. Not a single photo of any quality exists of those areas…


      • We kindly have to ask your understanding that our documentation standards follow the rules generally applied in our field, but published data are the result of our research questions. However, a detailed catalog of all the monumental Layer III pillars is currently in preparation and will be published in due time, including detailed photos and diagrams of all pillars and relevant contexts. Of course we will announce this here too.


  7. Just to clarify, you can make out 7 enclosures with a larger diameter than Enclosure D?

    It wasn’t the future I was asking you to comment on but whether you are factoring in future technological sophistication into your present excavation plans 🙂

    I am reading your excellent and illuminating paper ‘Establishing Identities in the Proto-Neolithic; “History Making”at Göbekli Tepe from the Late Tenth Millennium’ on Google Books. Why is it not on here?



    • It’s impossible to evaluate the exact size and extent of structures yet not excavated. The surveying method only allows to a localisation of structures to a certain degree. Not more. So, anything beyond that would be out of question until further research.

      Technological progress will have an influence on future excavations, but of course we can only plan with what’s known as of yet. Besides, complete excavation is not an objective in archaeology – we aim to uncover as much as necessary to understand a site and answer certain questions, but at the same time preserve as much as possible of a site and its original context.

      Regarding the latest paper: Thanks a lot for your kind words. Yet, one step after the other: We spend our work days with research to produce such articles; this blog here is private fun in the afternoon. We’re catching up … one step after the other.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s