Home Geospatial Applications Archaeology Ancyra project: The survey of the Augustus’ temple in Ankara

Ancyra project: The survey of the Augustus’ temple in Ankara

Gabriele Fangi, Gianluca Gagliardini
Ancona University, Italy
E-mail: mailto:%[email protected]

Paula Botteri, Piero Piva, Giuliano Rossi
Trieste University, Italy
E-mail: mailto:%[email protected]

Beatrice Pinna Caboni
Sovrintendenza archelogica, Roma, Italy

Following Octavian Augustus’ conquest of Galatia, in central Anatolia, in about 25 BC, a marble temple was built in Ankara. This monument to Roman hegemony celebrated the glory of Augustus, the first roman Emperor; after his death the temple was adorned with coloured engraved letters to recount his res gestae, the most famous epigraph of the roman world. All that remains today of the original building (known as Monumentum Ancyranum) is the pronaos and the lateral walls of the cella, one of which is badly holed. Inside the pronaos, on the intact lateral wall of the cella, there is the Augustan text in Latin, while a Greek translation may be read outside. Alongside the still impressive remains of the temple, almost 12 metres high and 32.5 metres long, there is a 15th-century mosque with minaret. The engraved marble with its remarkable political heritage, testimony to an epoch-making change from one regime to another and from one millennium to another, is now rapidly deteriorating because of the severe atmospheric conditions and heavy pollution. The last known measurements are in Der Tempel in Ankara by M. Schede and D. Krencker, published in Berlin in 1936. In September 1997 the photogrammetric survey of all the engraved walls was carried out, an area of approximately 46 square metres in total. In year 2000 a complete topographic and photogrammetric survey of the whole complex was carried out in view of the restoration project. The used surveying technique took advantage of the new reflectorless theodolites; the benefits are in terms of required equipment and time. Here the results of the survey are shown. The geometric reconstruction of the monument is only a part of a more complex and interdisciplinary study, going from town planning, history, epigraphy, petrography, statics, and architecture. On the basis of the survey, some restoration projects have to be prepared. The virtual reality techniques enable the restoration projects to be observed and monitored before their realisation.

The importance of the Augustus temple in Ankara derives from the epigraph engraved of the wall, the so-called Res Restae, the most important and famous of the roman world. The University of Trieste – Italy- conceived the so-called Ankyra Project, for the safeguard of the epigraph and the temple. The Ancyra Project mostly concentrates on the Monumentum Ancyranum. This is the definition commonly given by scholars of the ancient world to indicate the bilingual text engraved on the walls of the famous temple of Caesar Octavian Augustus, built in Ancyra (today’s Ankara). This political celebration of Augustus’ achievements was copied from the original one, no more existing, engraved on two bronze pillars at the entrance to his mausoleum, in Rome. The temple in the Medial Age was transformed in a byzantine church, and afterwards it was used as a dwelling house. The last known measurements are in Der Tempel in Ankara by M. Schede and D. Krencker, published in Berlin in 1936. The general conditions of the remains of the temple are indeed very poor, from the static and general point of view, because of the very hard environmental state, produced by the severe atmospheric conditions and mainly the terrible air pollution. In September 1997 a scientific mission carried out the photogrammetric survey of all the engraved walls, an area of approximately 46 square metres in total (Fangi et al. Olinda, 1999). The photogrammetric survey provided a virtual cast of the complex, obtained by non-aggressive and non-intrusive techniques. Photogrammetry proved again to be a very efficient technique, enabling a virtual cast to be obtained in very short time compared to six months period needed to the ancient archaeologists to get at real cast. In August 2000 a complete topographic and photogrammetric survey of the whole complex was carried out in view of the restoration project following the steps:

  1. Preserve the inscription.
  2. Conduct a detailed petrographic survey to establish the criteria needed to put an end to the process of dilapidation and decay of the epigraphic surface.
  3. Study the possibility to restore the epigraphs.
  4. Prepare a proposal for intervention to safeguard the precarious condition of the monument, which supports the epigraphs.

In this paper the first results of the project are shown and discussed.

Fig. 1 – The entrance of the temple in an ancient print

Fig. 2 – Aerial view of the remains of the Augustus temple seen from the minaret of the Haci Bayram Camii

Description of The Temple
The plan of the temple is a rectangular shape, the axis directed north-west, the main entrance facing the large square in front of the Haci Bayram mosque. Alongside the still impressive remains of the temple, almost 12 metres high and 32.5 metres long, there is a 15th-century mosque with minaret, the Haci Bayram Camii, perhaps the Turkish capital’s most important Islamic centre. Only the walls of the cella, the inner part of the temple, are still standing; the roof is still covering the temple, thus worsening the conditions of the walls. Large scaffolding surrounds the wall. A colonnade originally bordered the temple. No one column is still there. As concerns the original peristyle, what remains is located below the present road level. In the fig.2 the temple as seen form the minaret of the near mosque. Little can be perceved as modern urban development has drastically altered the original setting: a wide pavement surrounds the mosque and parts of the temple whilst a road open to traffic runs parallel to the lateral side of Roman monument. Only the two longitudinal walls together, with the entrance door, can be seen. In the rear part, the remains of the abse, build when the temple was transformed in a byzantine church. The transversal wall was then destroyed, producing weakening of the whole structure. Below the abse there is a crypt.

The Geodetic Survey
A local geodetic network was established, composed by 11 theodolite points. Almost all the photographic control points have been signalised, taking advantage of the existing scaffoldings. The reached accuracy is rather high, less than one centimetre in the three co-ordinates. The used instrument was a Topcon GTP1002 total station. The computation and the adjustment of the co-ordinates was performed with RETE a least squares computed program (Fangi, 8, 1995), with least constraints. The final local co-ordinate system has been taken parallel to the eastern wall, the most complete and in better condition. 175 points, including 11 theodolite stations compose the network. In planimetry there are 496 observations, 259 angular equations, 243 measured distances, for a redundancy equal to 138 (relative redundancy 1.4). In altimetry the observations are 271, the unknowns 176 for a redundancy of 97, (relative redundancy 1.5). The obtained average for standard deviation both in plan and high are 0.5 cm, mainly due to the signalization of the control points. There is one argument to be remarked about the distances measured by the reflectorless EDM-theodolite: 18 over 245, i.e. 7% of the total, have been rejected because of incorrectness or disagreement with the other measurements. The cause for that is when the reflecting spot is not on a rather large and smooth surface, the obtained measure is not corrected. The operator must avoid to take distance form sharp points such as corners that can be observed only by angular readings. Apart this feature, the new reflectorless instruments have great advantage in terms of time, accuracy, but first of all, in terms of amount of necessary equipment, whose weight be greatly reduced, say more than 50%. In fact the connection among the points was not realised by the traditional traversing procedure, that is to connect the stations points by angular and range measurements. All the control points were observed, taking possibly the distances also. In this manner the station points can be easily and precisely established in the network; we could avoid to connect the stations by measurements of angles and distances. So the amount of equipment was greatly abridged compared with the traditional one. In fact three tripods, one total station, two tribranches and two prisms normally compose the traditional equipment. Only one tripod and one total station form now the required equipment with reflectorless instrument, say less than 50% of the first one, in terms of weight and number of pieces. The range of the distance-meter without prism is limited to hundred meters, more or less, depending very much on the reflecting surface. The range is then enough for the normal task of architectural survey. The internal recording of the measurements further speeds up the work. The efficiency of the survey is enhanced also by the freedom in choosing the station points, liberated from the constraint to be visible to each other. Finally, the possibility to get an uncorrected distance must lead to a particular large value of the redundancy. The type of instrument is then particularly suitable for archaeological missions where the weight and the volume of the equipment are a very limiting factor. In fig. 3 the lay-out of the control network. Finally, in order to be able to pass from one local system to another, we took care to determine also some control points of the previous photogrammetric survey, the epigraphs (Botteri,et al, 2, 1999).

Fig. 3 – The plan of the control network with the error ellipses  

The Photogrammetric Survey
Almost all the photogrammetric control points were signalised by simple paper targets stuck on the scaffolding. The survey is then realised with horizontal stereo strips, taken from the scaffolding for the interior walls and from the pavement for the eastern wall. We used a semi-metric Rollei 6×6 camera and a non-metric camera Fuji Gsw69 large format camera.

Fig.4- Lay-out of the photogrammetric taking

Fig. 5 – The orthophotomosaic of the external Western Wall

Tab.1- Main features of the photogrammetric photographs

Wall n.strips n.photo
x strip
Total Mean
1 interior Est 2 6 12 1/200
2 interior West 2 6 12 1/200
3 interior south 1 2 2 1/150
4 exterior south 1 6 6 1/250
5 exterior south (facade) 2 2 4 1/250
6 abse -interior wall 1 8 8 1/150
7 abse exterior wall 1 2 2 1/250
Total     46  

The film was colour reversal Velvia Fuji 50ASA. The mean photo scale is about 1/200, allowing final plotting at scale 1/50. The existing scaffoldings have been useful for taking the photographs and for targeting, but on the other end hampered the total view al the wall. In fig. 4 the plan of the photogrammetric takings.

The Photogrammetric Plotting
At this moment the photogrammetric plotting is still going on. We use a Galileo-Siscam Stereometric-Pro digital station. All the photographs have been scanned with a calibrated Epson perfection 1200 Photo scanner, with a resolution of 1200 dpi, which seems to be enough good for our purposes. In fig. 5 an orthophomosaic of the western exterior wall is shown.

For the time being only some remarks about the surveying technique can be expressed. The new reflectorless theodolites enhance the efficiency of the survey, allowing the operator to get rid of the constraints of the visibility of any other measuring station. Finally, the possibility to get an uncorrected distance for certain type of points must lead to a particular large value for the redundancy and suggest to avoid determinations without redundancy. The type of instrument is then particularly suitable for archaeological missions where the weight and the volume of the equipment are a very limiting factor.


  • Botteri P. (1997) – Progetto Ankyra. Il tempio di Augusto in Ankara, Turchia. Relazione interna. Servizio di ateneo Gestioni Immagini e Cartografia (Sagic), Università di Trieste
  • Botteri P., Fangi G, Nardinocchi C. (1999)- The photogrammetric survey of the Res Gestae in Ankara XVII Cipa Symposium Recife-Olinda
  • Akurgal (1993) – Salviamo la memoria di Augusto ad Ankara, Archeo n. 1- 1993
  • Canali L. (1993) – Res Gestae divi Augusti – Editori Riuniti
  • Momshen T.H. (1865) – Res Gestae divi Augusti – Critical edition Res Gestae, Berlin, 1865
  • Schede M., Krenken D. (1936) – Der Temple in Ankara, Edition Res Gestae, Berlin Leiptig
  • Web Site:
  • Fangi G. (1995) – Note di Fotogrammetria, Clua, Ancona
  • Botteri P., Res Gestae Divi Augusti, Dictionnaire de l’Antiquité classique, Paris 2000
  • Botteri P. – Fangi G., Ankara. The Augustus’ Temple. Preliminary and Photogrammetrical Report in 1997 (The 20th International Symposium of Excavations, Survey and Archeometry – May 25-29. 1998 – Tarsus, Ankara)
  • Botteri P., Progetto Ancyra: la ricerca e il metodo, Metodi e Ricerche, 1999, p. 195-8
  • Botteri P., Project Ancyra: the Temple of Augustus and Rome in Ankara, Culture in Sustainable Development. An Italian Strategy, Ministero degli Affari Esteri. Roma, 1999, p.136-138