Introduction to Aerial Archaeology - Aerial Archeology Research Group

A Short Introduction to Aerial Archaeology

More archaeological features have been found worldwide through aerial photography than by any other means of survey. The method also provides means of examining context and larger areas of land than through the traditional site-based focus. Aerial photographs of archaeological features in many parts of the world show there to have been occupation of much of the land from Neolithic times onwards. In some places this is represented by apparently isolated sites, in others there are networks of tracks, fields and settlements that show there to have been a higher local population than at present. The text below first outlines a brief history of archaeological uses and taking of aerial photographs, followed by descriptions of types of aerial images, mapping methods and uses of aerial sources. It ends with references to publications indicated.

A brief history

Archaeologists were working with existing aerial photographs during and immediately after the First World War, studying sites and landscapes in Macedonia, Romania (Roman limes), Mesopotamia and deserts of the Near East. For example, in 1919, in a paper titled ‘Air photography in archaeology’, Lieutenant-Colonel G.A. Beazeley¹ published his discovery of an ancient city on the Tigris as a result of aerial survey for mapping. In a different landscape, with less imposing remains, O.G.S. Crawford had been working with photographs taken by the Royal Air Force to map areas of upstanding archaeological landscapes in southern England. His results were published in an Ordnance Survey Professional Paper (1924)² that included analytical comments about aerial photographs and maps of extensive areas. Also in that book was a summary of Crawford’s work on part of the Stonehenge Avenue that was plough-levelled but had been visible as differential crop growth in RAF photographs taken in 1921. Other photographs taken at around the same time showed indications of buried archaeological features through their effect on cereal crops and Crawford was able to identify banks and ditches through this proxy. He named them ‘crop sites’ or ‘crop marks’ and these phenomena led to the huge success of aerial observation for archaeology in temperate lands in the succeeding 100 years.

Crawford himself undertook an airborne project with Alexander Keiller in 1924, when they set out to photograph many upstanding sites in central-southern England (Wessex). A selection of their aerial photographs illustrated their book (1928)³ that contained detailed analytical field surveys of those sites and was a first display of the ability of specifically-targeted aerial images to illustrate past features. At a similar date, Antoine Poidebard (1934) ⁴La trace de Rome dans le Désert de Syrie. Le limes de Trajan à la conquète arabe. Recherches aériennes, 1925–1932. Paris: Paul Geuthner. was flying and photographing to seek evidence of Roman activity in Syria and was paralleling Crawford’s results in a very different environment. Further aerial photography in the Near East was undertaken by Erich Schmidt (1940)⁵Flights over ancient cities of Iran. Oriental Institute: Chicago. and sponsored by the Oriental Institute of the University of Chicago on whose web site Schmidt’s photographs can be seen.

Between the two World Wars, the journal Antiquity, founded by Crawford in 1927, promoted uses of aerial photographs in a world-wide scope in almost every issue. Through his contacts in Britain, Crawford was able to ensure that many of the sites photographed by the RAF were examined on the ground, often through excavation. This particular gain in knowledge on British sites is one key that may have been responsible for the fact that aerial evidence is accepted as valid by itself there but held in some suspicion in other countries. By the late 1920s, Crawford had an arrangement with the RAF that pilots would target archaeological sites during navigation exercises and that the photographs would be passed to him as Archaeology Officer for the Ordnance Survey. This enabled him to build up what is now known as the Crawford Collection, held now in British national collections.

Still in Britain, in 1933, Major Allen, inspired by one of Crawford’s books, made his own cameras and began flying and photographing sites in the Thames Valley (and later other places). His aim, he wrote, was to discover sites – and this same aim has been one of the driving forces behind many later aerial photographers. Allen used some of his discoveries to create two maps of Thames valley locations which were published in 1938⁶ and 1940⁷. His photographs captured the rural landscape when it looked ‘soft’, before the advent of industrialised farming and they should be valued for this as well as for their archaeological content. Copies of his photographs can be seen on Ashmolean Museum’s website.

The advent of World War Two halted further exploration and development in aerial work for almost everybody, but with two exceptions. In England, Derrick Riley, who was able to use some of his RAF test flights to observe, record and sometimes photographs sites close to his airfield bases. Riley published five papers during or just after the war, one of which laid down the basic rules and definitions about how sites were visible from above⁸. He then faded from the aerial scene until the 1970s. Immediately after the end of the war, John Bradford and Peter Williams-Hunt, both archaeologists and serving photographic intelligence officers in Apulia, Italy, undertook their own aerial survey (using RAF aircraft and equipment) of a considerable area of the Tavoliere which was dense in sites from Neolithic to Roman times. Bradford was responsible for most of the publication as Williams-Hunt had moved to SE Asia where he retained (and published) his aerial interests. Bradford’s work culminated in a book (1957)⁹ that included theory, method and case studies relevant to using aerial photographs to examine ancient landscapes.

In 1948, Cambridge University appointed J.K.S. St Joseph as Curator of Aerial Photography. He had begun his aerial photographic work a few years earlier using RAF aircraft. His department remained active in aerial photography past his retirement, when he was succeeded by David Wilson, and by then had become CUCAP – Cambridge University Committee for (or Collection of) Aerial Photography – with its own aircraft and pilot. For almost twenty years, St Joseph was the only active aerial photographer in Britain with flights extending across Britain, Ireland and later to Northern France, Denmark and the Netherlands. St Joseph and Wilson published extensively in journals and through CUCAP’s own series of books, with Wilson producing a photo reading guide in 1982¹⁰.

From 1960 there was an increase in the number of archaeologists taking aerial photographs in parts of Europe among whom was Irwin Scollar (Rhine Valley in Germany, 1965)¹¹, Roger Agache (Picardy in France, 1975)¹² and several others in France. The journal Antiquity, by then edited by Glyn Daniel, continued to publish notes on ‘aerial reconnaissance’ and included articles by Agache and, notably, a series of ‘recent results’ by St Joseph. The majority of published work from this period was of photographs and accompanying descriptive text, sometimes with a sketch to clarify the photographic information. Mapping, to combine information from several photographs to show extents of past landuse, was rare but the accumulation of photographs in parts of Britain enabled sketch-mapped studies to be made of small areas (eg RCHME 1960¹³, Webster and Hobley 1965¹⁴). More extensive sketch mapping from aerial photographs, combined with field walking, derived from a study of the Roman fenland in England (Phillips 1970)¹⁵. That project began before World War Two and was continued afterwards, making use of a 1:10,000 scale post-war aerial survey and targeted photographs taken by St Joseph. Mapping from existing photographs was never as popular as taking pictures but was revived by John Hampton, who, in 1965, had established the Air Photographs Unit within the English Royal Commission (Hampton 1989)¹⁶. For Hampton, aerial photographs were sources of information that needed to be interpreted and accurately mapped to maximise our understanding of their archaeological content. However, until the advent of computer transformation (Palmer 1977)¹⁷, accurate mapping by hand was a slow and tedious process.

The 1970s saw a continued increase in numbers of active aerial photographers, including the beginning of work in Warsaw Pact countries that had their particular problems of secrecy and censorship. An example is the work in Dobrogea, Romania by Alexandru Simion Ştefan that was discussed in many publications and summarised in 1983¹⁸. This lone work contrasts with the activity in Britain and France from the mid-1970s and through the 1980s when more people were active in the air than at any time before or since. Dry summers encouraged some of this activity, funding was organised, aviation laws were less restrictive than in later years and archaeologists liked to discover new sites. Fortunately, at least in Britain, most of the resulting photographs were copied to national archives and became available for archaeological projects. One of the people active above Britain was Derrick Riley, last heard of just after World War Two, who returned to archaeology and the aerial view to fill his retirement years. Riley had identified a research area in the English midlands which had previously been a blank on the map but within which he identified, photographed and mapped systems of ‘brickwork fields’ and their accompanying farmsteads (1980)¹⁹. Riley extended his aerial activity to Jordan (with David Kennedy, 1990)²⁰ and Israel, edited and published (in 1984)²¹ a manuscript written by Major Allen in the 1930s, wrote his own Air Photography and Archaeology (1987)²² and encouraged the next generation through his teaching at Sheffield University. When mapping the brickwork fields, Riley had made some use of computer mapping, but Rog Palmer’s publication of the 400 sq km of the Danebury environs in 1984²³ was almost entirely based on computer-transformed interpretation. A later publication by Cathy Stoertz (1997)²⁴ achieved the pinnacle of mapping to depict and area of some 1350 sq km and analyse its archaeological content.

The 1980s saw new beginnings in aerial work in Europe that was led and encouraged by Otto Braasch from Germany who had retired from the Luftwaffe and started his second career as an aerial photographer in 1980. Braasch’s early work covered those German states who paid for his survey flights but, with the end of the Cold War, his interests extended to those previously closed countries where aerial work had been forbidden or very difficult. There, he saw his mission to encourage archaeologists to appreciate the value of the aerial view and to use aerial photographs as part of their research projects. That we now have people active in many European countries is largely due to Braasch’s encouragement and to training schools and workshops that were initially begun by him after the first international symposium in Kleinmachnow, Germany, in 1994 (Kunow 1995)²⁵.

AARG also played an active part in teaching in many of the European workshops. Its first meeting had been in 1983 (Stoertz 2013)²⁶ and ten years later its membership included a mixture of aerial photographers, photo interpreters and interested collaborators. The first European workshop was held in Kiliti, Hungary, in 1996 with Braasch organising the airborne aspects and AARG members teaching what became known as ‘ground school’. In terms of continuation of interest, that workshop was the most successful of any that were run in the next 20 years as most of its ‘students’ already held teaching posts or other official archaeology positions in their respective countries. It could be said that the networks generated by Braasch, AARG and the Kiliti workshop helped germinate aerial work in Hungary, Poland, Czechia, Slovenia, Estonia, and Lithuania. In Czechia, in particular, the work of Martin Gojda and his students continues to develop and extend uses of aerial images, combining their information with fieldwork and documentary evidence to produce archaeological narratives (eg. 2010²⁷; 2011²⁸; 2013²⁹).

By 2000 two things had changed – sources of high-resolution images now included those from some satellites and airborne laser scanning (ALS or lidar, which recorded topography at a level and precision previously undreamed of), and use of geographical information systems (GIS) were becoming commonplace as means of stacking and manipulating data, archaeological and otherwise. Satellite sources included the recently-released Cold War photographs taken by the Americans and Russians and much use was made of the easily-available US Corona material, especially to aid Near Eastern studies (eg Ur 2003³⁰; 2013³¹; Hanson and Oltean 2013³²). Initial use of ALS tended to be at the ‘wow’ level, but within ten years publications were emerging from all parts of the world that included it as a primary source of evidence (Opitz and Cowley 2013)³³. Google Earth was initially released in 2001 and, along with Microsoft’s Bing, now provides a first view for anyone examining an area from above. Many countries include aerial and ALS layers in their national geoportal sites and these provide free access to a wealth of images that can help archaeological work. It must be noted, however, that there are many aerial images that are not on the internet and that proper study of any area may benefit from consulting these.

From 2010, use of drones or unmanned aerial vehicles (UAV) was becoming more common in archaeological work. Their applications including recording excavations (in cases replacing the job that kites and balloons had performed) and specific targeted fields or monuments. Close-range photogrammetric software enabled the creation of 3D models and orthophotos from the original drone images and these can provide precision data about a site. Dry summer conditions over much of Europe in 2018 led to the discovery and recording of several new archaeological sites, among which the then-called ‘dronehenge’ in Ireland’s Boyne Valley was the most spectacular (AARGnews 57).

Aerial photography and its uses are not static, and AARG has encouraged several talks and discussions about method and theory which have resulted in a few publications. Among these, work by Wlodek Rączkowski (eg 1999)³⁴ and Martin Gojda (eg 2004)³⁵ have been stimulating and two edited books have questioned why we do what we do (Brophy and Cowley 2005³⁶; Mills and Palmer 2007³⁷).

Types of aerial images and mapping methods

Aerial images are usually classified as either vertical – taken with a camera pointing directly at the ground – or oblique – which records the ground at any angle other than vertical – but for archaeological images, these basic classifications can be expanded.

Firstly, we should introduce the concept of stereo pairs which are overlapping images that enable a viewer to see a scene in 3D. There is, however, more value to use of stereoscopy than just the 3D view, as, for example, examination of a pair of prints will help an interpreter eliminate ‘sites’ that may be caused by processing marks or dust on sensors. The ability to see in stereo is recognized as an essential qualification for professional photo interpreters, therefore it shuold be taken as evidence of good airborne practice that we are provided with such images routinely and that photo interpretation is carried out routinely through examination in three dimensions (Palmer 2013)³⁸.

Vertical photographs are usually taken using a camera that is fixed to an aircraft or satellite that flies a series of straight and parallel courses. Exposures can be set to provide images that overlap by 60% so that stereoscopic viewing can examine these and perceive height. This helps and adds confidence to photo interpretation. Modern images, taken specifically for use as a single layer in GIS are often taken a much smaller overlap and are of lesser value to the photo interpreter. Corona satellite photographs were taken to provide stereoscopic views which, because of image distortion caused by the panoramic cameras, may be better viewed after they have been georectified. Many of the current high-resolution satellites are able to provide stereoscopic images although the cost of these may be prohibitive for many archaeological projects. Typically, vertical images are taken to record an area of land and will show any and all archaeological features that were visible below the camera at that time. Thus, they record the ground in an unbiassed way.

Images in web sources such as Google Earth are a mixture of satellite images and aerial photographs. The accuracy of their position may vary and ought not to be assumed to be correct, especially in more rural areas, and height distortion is readily apparent when building facades can be seen. Most countries also have photographs of an earlier date and these can be a valuable source of information in parts of the world where development of mineral extraction has destroyed archaeological features. Examples include the survey of Britain that was flown by the RAF in the immediate post-war years and land photographed by the USAF in Spain, Italy, Brazil, Jordan, Japan and many other countries (Pérez et al 2014)³⁹.

Oblique photographs are usually taken using a hand-held camera from an aircraft or by tilting an orbiting satellite to take off-nadir scenes. Airborne archaeologists favour this way of recording archaeological sites as the best viewpoint can be chosen to show detail or context. However, to photograph a site it has first to be noticed by an aerial observer and analysis has shown that aerial photographers are not as efficient as they think. Despite this, use of a single-engine aircraft and a hand-held camera remains the way that archaeologists record the ground, and in part this is necessary because of the short notice needed to fly at times when crops are responding to buried features and the weather is good. Oblique photographs are also more easily read by others and have formed the basis of many books. They can also be taken to be viewed as stereoscopic pairs to help their interpretation. Most archaeological oblique photographs are retained as paper prints, slides, or digital images in either national collections (in the UK) or in the museums or university departments of the people who have taken them.

ALS data for many parts of the world can be downloaded from geoportals or national agencies. Software has been written to enhance its uses for archaeological projects (Kokalj and Hesse 2017)⁴⁰ and use of a GIS is essential for more than a static hill-shaded view.

Mapping from aerial images is the step in which those features that have been interpreted are located on to a base map (or orthophoto) to show their location. Mapping can be done from any type of image and can also be used to merge information from photographs taken on different dates. Sketch mapping is used to locate by eye those features and, with practise, can be sufficiently accurate when working at scales of about 1:10,000. Increased accuracy can be gained by generating a network of working lines between the same points on a map and photograph. These create an uneven graph-paper-like mesh which gives greater precision to mapping when transferring information from an aerial photograph. Irwin Scollar illustrated ways of doing this in a paper that also described his early computer method of transferring point information from a photograph to a map (1975)⁴¹. Networks continue to be occasionally useful even in the computer age.

In the 1970s use of optical instruments to transfer information from photograph to map was tried by RCHME (Hampton 1989, 23)⁴² but this was not easy with oblique photographs that provided the majority of archaeological information. Optical transfer was superseded by computer programs that initially were used to transcribe information graphically from interpretations of photographs (Haigh 1991)⁴³ and those in turn were replaced by methods that transformed digital images to match a map or other ground information (Haigh 1996⁴⁴; Scollar 1998⁴⁵). Since those early specialist programs, both written to deal particularly with the problems of oblique image transformation, modern GIS include packages for georectifcation of images that help build a layer of photographic data over which interpretations can be made.

There are other more-sophisticated programs for mapping, some of which include algorithms for creating orthophotos and 3D models from a high-density cover of images such as may be made from drones.

Mapping archaeological information is done for a purpose and this should decide the level of accuracy required and thus the choice of method to be used. Projects requiring accurate depiction of archaeological traces are likely to require line and polygon information that show ditches, banks, walls and possibly also select non-archaeological information. Maps of this kind form an ideal base for research into settlement patterns of an area and can also be sufficiently accurate to match with field-gathered information and to locate question-orientated excavation trenches. Less visually-informative mapping may be suitable for heritage management whose needs may be met in a GIS by enclosing archaeological features in a polygon and creating a written record in its attached metadata.

References

Agache, R. and Bréart, B., 1975. Atlas d’archéologie aérienne de Picardie. La Somme Protohistorique et Romaine. Société des antiquaires de Picardie, Amiens.

Allen, G.W.G., 1938. Marks seen from the air in crops near Dorchester, Oxoniensia 3, 169-171.

Allen, G.W.G., 1940. Cropmarks seen from the air, Northfield Farm, Long Wittenham, Berks, Oxoniensia 5, 164-165.

Allen, G.W.G. (ed D.N. Riley) 1984. Discovery from the Air. Aerial Archaeology 10.

Beazeley, G.A., 1919. Air Photography in Archaeology. Geographical Journal 53, 331-335.

Bradford, J.S.P., 1957. Ancient Landscapes: studies in field archaeology. Bell: London.

Brophy, K. and Cowley, D. (ed), 2005. From the Air: Understanding Aerial Archaeology. Tempus, Stroud.

Crawford, O.G.S., 1924. Air Survey and Archaeology. Ordnance Survey Professional Papers 7. London.

Crawford, O G S and Keiller, A, 1928. Wessex from the Air, Oxford.

Gojda, M., (ed) 2004. Ancient Landscape, Settlement Dynamics and Non-Destructive Archaeology: Czech research project 1997-2002. Academia: Prague.

Gojda, M. a kolektiv, 2010. Studie k Dálkovému Průzkumu Archeologii (Studies in Remote Sensing for Archaeology). Pilzen.

Gojda, M. and Trefny, M. (ed). 2011. Archeologie Krajiny pod Řípem (Archaeology in the Landscape around the Hill of Říp). Opomíjená Archeologie (Neglected Archaeology) 2. Dept of Archaeology: Pilsen.

Gojda, M., John, J., a kolektiv, 2013. Archeologie a letecké laserové skenování krajiny (Archaeology and airborne laser scanning of the landscape). Pilzen.

Haigh, J.G.B., 1991. The AERIAL program, version 4.1, AARGnews 3, 31-33.

Haigh, J.G.B., 1996. Another member of the AERIAL software family. AARGnews 12, 26-33.

Hampton, J.N., 1989. The Air Photography Unit of the Royal Commission on the Historical Monuments of England 1965-1985. In D. Kennedy (ed) Into the Sun: essays in air photography in honour of Derrick Riley, 13-28. Sheffield: J R Collis.

Hanson, W.S. and Oltean, I.A. (eds), 2013. Archaeology from Historical aerial and satellite archives. New York: Springer

Kennedy, D and Riley D, 1990. Rome’s Desert Frontier from the Air. Batsford, London

Kokalj, Ž. and Hesse, R., 2017. Airborne laser scanning raster data visualization: A Guide to Good Practice. Založba ZRC: Ljubljana.
https://iaps.zrc-sazu.si/en/publikacije/airborne-laser-scanning-raster-data-visualization-1#v

Kunow, J., (ed), 1995. Luftbildarchäologie in Ost- und Mitteleuropa (Aerial Archaeology in Eastern and Central Europe). Forschungen zur Archäologie im Land Brandenburg 3.

Mills, J. and Palmer, R. (ed), 2007. Populating Clay Landscapes. Tempus: Stroud.

Opitz, R.S, and Cowley, D. C. (eds), 2013. Interpreting Archaeological Topography: 3D data, visualisation and observation. Oxbow, Oxford.

Palmer, R., 1977. A computer method for transcribing information graphically from oblique aerial photographs to maps. J Archaeological Science 4, 283-290.

Palmer, R., 1984. Danebury: an aerial photographic interpretation of its environs. RCHM Sup Series 6.

Palmer, R., 2013. Stereo photography for airborne observers. AARGnews 47, 39-44.

Pérez, J.A., Bascon, F.M. and Charro, M.C., 2014. Photogrammetric usage of 1956–57 USAF aerial photography of Spain. The Photogrammetric Record, 29/145, 108-124.

Phillips, C.W. (ed), 1970. The Fenland in Roman Times. Roy Geog Soc Res Ser 5.

Poidebard, A., 1934. La trace de Rome dans le Désert de Syrie. Le limes de Trajan à la conquète arabe. Recherches aériennes, 1925–1932. Paris: Paul Geuthner.

Rączkowski, W., 1999. Power of image: some ideas on post-processual aerial archaeology. AARGnews 19, 10-14.

RCHME, 1960. A Matter of Time: an archaeological survey of the river gravels of England. London.

Riley, D.N. 1946. The technique of air-archaeology. Archaeol J 101, 1-16.

Riley, D.N., 1980. Early Landscape from the Air. Collis: Sheffield.

Riley, D.N., 1987. Air Photography and Archaeology. Duckworth: London.

Schmidt, E.F., 1940. Flights over ancient cities of Iran. Oriental Institute: Chicago.

Scollar, I. 1965. Archäologie aue der Luft. Arbitsergebnisse der Flugjahre 1960 un 1961 im Rheinland. Rheinland Verlag: Köln.

Scollar, I., 1975. Transformation of extreme oblique aerial photographs to maps or plans by conventional means or by computer. In D.R. Wilson (ed) Aerial Reconnaissance for Archaeology. Counc Brit Archaeol Res Rep 12, 52-59.

Scollar, I., 1998. AirPhoto – a WinNT/Win95 program for geometric processing of archaeological air photos. AARGnews 16, 37-38.

Ştefan, A-S., 1983. Cercetări aerofotografice în anni 1978-1980. Materiale şi Cercetări Arheologice – a XV-a sesiune anuală de rapoarte, Muzeul Judeţean Braşov – 1981, 15, 178-192.

Stoertz, C., 1997. Ancient landscapes of the Yorkshire Wolds, RCHME, Swindon.

Stoertz, C., 2013. Anniversary Reflections from a Founder Member. AARGnews 47, 8-12.

Ur, J. A. 2003. CORONA Satellite Photography and Ancient Road Networks: A Northern Mesopotamian Case Study. Antiquity 77, 102-115.

Ur, J. A. 2013. Spying on the Past: Declassified Intelligence Satellite Photographs and Near Eastern Landscapes. Near Eastern Archaeology 76, 28-36.

Webster, G. and Hobley, B., 1965. Aerial reconnaissance over the Warwickshire Avon. Archaeol J 121, 1-22

Wilson, D.R., 1982. Air Photo Interpretation for Archaeologists. Batsford, London.